[urbit] Whitepaper + video drafts

I think I love you.

First thing that jump at me is how terrible of a term "pserver" is. Is
there really nothing better?
Maybe introduce the semantic origins of Ur-bit and refer to it as "an
urbit" like we do currently?

Although you quickly explain what's wrong with AWS, and how private servers
are eating ur dataz, that's only the "OS" part of the "OS and Internet"
rewrite. The introduction doesn't touch on what's wrong with the web until
later, except for "it integrates better".

There's a typo at "inuitiion" under "A simple typed functional language",
along with "but but nonmemorable bases remain a serious handicap." under
PKI/Identity.

Might want to explain explicitly, under "Definition", that the frozen spec
(Nock) is merely used as a bootstrapping core for Urbit and is not the
entire ecosystem...it took me quite a while to understand that the first
time I encountered Urbit. It's also very scary that it jumps directly into
Nock formulas for booting the kernel. All of the "see lower for
implementation details" might be better as direct links, also, for the
impatient or curious. No promises they will understand everything, but jets
are pretty easy to understand without having to wade through %vanes first.

I'd put an explanation of the groupings of runes before "Syntax geometry".
It's quite important, and can explain the insane-sounding idea of runes.

The overview of how Urbit boots mentions events and the %noun tag for them,
but it doesn't actually introduce events until "Input and output". Not that
big of a deal, but it's quite a gap. Maybe a quick sentence about what you
mean by an event, until the real intro?

"Every Urbit event log starts by setting its own unique and permanent base."
This isn't actually correct, right? Promoting a sub adds on another entry
in your will and resets your base. Will this be ripped out before the
relaunch?

The overview of vanes mentions how they take the current time to cheat
referential transparency. It kinda makes it sound like it's just a sly con,
though, and that the current time is always passed in. Possibly mention how
that allows replaying of the event log to be consistant? Or at least make
it sound less dirty :p

"Certificate transfer and key exchange are aggressively piggybacked, so the
first packet to a stranger is always signed but not encrypted." Thanks for
mentioning this! When I was reading over the crypto code a couple weeks
back I tripped over this and thought it was a flaw. Glad that it's cleared
up immediately, along with the downside.

"%clay is a sort of simplified, reactive, typed git." Ah. Maybe we should
steer away from the word "reactive", no? ;D
I-I only just got the %ford -> builder joke. Man, now I feel dumb.

"Marks are like MIME types, except with executable specifications, which
define how to validate, convert, diff, and patch content objects, they
would be marks." I don't think this sentence makes any sense? Cut off the
"they would be marks". /tip also isn't currently a thing, right?

"Urbit is a personal server; it can handle an HN avalanche on your blog"
[citation needed] ;D

%jael sounds quite unsafe. Do we actually currently use it for anything?
Arbitrary apps being able to open each other's secrets doesn't sound like
something you want. In fact, some mechanism that has apps behind a %vane
whitelist would be great - I don't care if +facebook can read my %jael if
it can't touch anything else, although that's pretty hard to define when
%clay and %gall all hit the network. The topic of :dojo also touches on
this with its generators, but it doesn't seem that isolated - %clay is
exposed by .^, for example, and so all of the classes can request .^(%cx
/~evil-attacker/msg) and communicate over the network to ~evil-attacker,
even if they aren't of the web-scraper type.

In %dill, "Every system needs its"...? Sentence cut-off?

Reading the Jets section when suddenly: "Another approach would be a safe
systems language, such as Rust."
Ouch, that hurt. It's a bit hard to write a jet in Rust when >50% of u3 are
C macros. I kinda gave up, unfortunately.\

Might want to point out a bit cleaner about replacing galaxy keys and
forking the network. It's a bit hard to read right now, and how we promise
that Urbit isn't designed to be decrypted by dictators, regardless of what
websites say. You can ignore all ships (wait, what's the generic name for
urbits now? It's not ships, but all the computer classes aren't one
specific thing. Planetary body is too verbose...) signed by a galaxy and
accept ones signed by your new hard-coded public key, there's just no
promise the rest of the network will agree with you.

"The abstraction does not leak" is my new favorite thing. Sorry for the
rambling.

On Thursday, September 17, 2015 at 4:28:57 PM UTC-4, Galen Wolfe-Pauly

Post by Galen Wolfe-Pauly
Psst.
http://happut-fopnys.urbit.org/home/tree/pub/release-0
Weâd love to hear feedback from the list before posting anywhere else.

Curtis Yarvin

2015-09-18 01:41:33 UTC

I think I love you! This is awesome feedback, haptem-fopnys... (You are haptem, right? Nicks coming soon to :talk.)

The "pserver" (the p is silent, as in psychology or Psmith) is the generic product category, like tissue vs Kleenex. Substitute doxology welcome!

Sent from my iPhone

Post by c***@gmail.com
I think I love you.
First thing that jump at me is how terrible of a term "pserver" is. Is there really nothing better?
Maybe introduce the semantic origins of Ur-bit and refer to it as "an urbit" like we do currently?
Although you quickly explain what's wrong with AWS, and how private servers are eating ur dataz, that's only the "OS" part of the "OS and Internet" rewrite. The introduction doesn't touch on what's wrong with the web until later, except for "it integrates better".
There's a typo at "inuitiion" under "A simple typed functional language", along with "but but nonmemorable bases remain a serious handicap." under PKI/Identity.
Might want to explain explicitly, under "Definition", that the frozen spec (Nock) is merely used as a bootstrapping core for Urbit and is not the entire ecosystem...it took me quite a while to understand that the first time I encountered Urbit. It's also very scary that it jumps directly into Nock formulas for booting the kernel. All of the "see lower for implementation details" might be better as direct links, also, for the impatient or curious. No promises they will understand everything, but jets are pretty easy to understand without having to wade through %vanes first.
I'd put an explanation of the groupings of runes before "Syntax geometry". It's quite important, and can explain the insane-sounding idea of runes.
The overview of how Urbit boots mentions events and the %noun tag for them, but it doesn't actually introduce events until "Input and output". Not that big of a deal, but it's quite a gap. Maybe a quick sentence about what you mean by an event, until the real intro?
"Every Urbit event log starts by setting its own unique and permanent base." This isn't actually correct, right? Promoting a sub adds on another entry in your will and resets your base. Will this be ripped out before the relaunch?
The overview of vanes mentions how they take the current time to cheat referential transparency. It kinda makes it sound like it's just a sly con, though, and that the current time is always passed in. Possibly mention how that allows replaying of the event log to be consistant? Or at least make it sound less dirty :p
"Certificate transfer and key exchange are aggressively piggybacked, so the first packet to a stranger is always signed but not encrypted." Thanks for mentioning this! When I was reading over the crypto code a couple weeks back I tripped over this and thought it was a flaw. Glad that it's cleared up immediately, along with the downside.
"%clay is a sort of simplified, reactive, typed git." Ah. Maybe we should steer away from the word "reactive", no? ;D
I-I only just got the %ford -> builder joke. Man, now I feel dumb.
"Marks are like MIME types, except with executable specifications, which define how to validate, convert, diff, and patch content objects, they would be marks." I don't think this sentence makes any sense? Cut off the "they would be marks". /tip also isn't currently a thing, right?
"Urbit is a personal server; it can handle an HN avalanche on your blog" [citation needed] ;D
%jael sounds quite unsafe. Do we actually currently use it for anything? Arbitrary apps being able to open each other's secrets doesn't sound like something you want. In fact, some mechanism that has apps behind a %vane whitelist would be great - I don't care if +facebook can read my %jael if it can't touch anything else, although that's pretty hard to define when %clay and %gall all hit the network. The topic of :dojo also touches on this with its generators, but it doesn't seem that isolated - %clay is exposed by .^, for example, and so all of the classes can request .^(%cx /~evil-attacker/msg) and communicate over the network to ~evil-attacker, even if they aren't of the web-scraper type.
In %dill, "Every system needs its"...? Sentence cut-off?
Reading the Jets section when suddenly: "Another approach would be a safe systems language, such as Rust."
Ouch, that hurt. It's a bit hard to write a jet in Rust when >50% of u3 are C macros. I kinda gave up, unfortunately.\
Might want to point out a bit cleaner about replacing galaxy keys and forking the network. It's a bit hard to read right now, and how we promise that Urbit isn't designed to be decrypted by dictators, regardless of what websites say. You can ignore all ships (wait, what's the generic name for urbits now? It's not ships, but all the computer classes aren't one specific thing. Planetary body is too verbose...) signed by a galaxy and accept ones signed by your new hard-coded public key, there's just no promise the rest of the network will agree with you.
"The abstraction does not leak" is my new favorite thing. Sorry for the rambling.

Post by Galen Wolfe-Pauly
Psst.
http://happut-fopnys.urbit.org/home/tree/pub/release-0
Weâd love to hear feedback from the list before posting anywhere else.

--
You received this message because you are subscribed to the Google Groups "urbit" group.
For more options, visit https://groups.google.com/d/optout.

Andrew Harkness

2015-09-18 02:57:53 UTC

I noticed another typo in addition to "intuitiion" - there's an "iff", in
"This statement succeeds iff it accurately communicates..." under
Requirements.

Also, another vote for "pserver" being an unwieldy sort of word - I kept
wanting to read it as "pee-server", despite efforts otherwise.

Otherwise, quite an interesting read :)

(Please excuse any failure of mailing list etiquette - I'm new to this..)

Post by c***@gmail.com
I think I love you.
First thing that jump at me is how terrible of a term "pserver" is. Is
there really nothing better?
Maybe introduce the semantic origins of Ur-bit and refer to it as "an
urbit" like we do currently?
Although you quickly explain what's wrong with AWS, and how private
servers are eating ur dataz, that's only the "OS" part of the "OS and
Internet" rewrite. The introduction doesn't touch on what's wrong with the
web until later, except for "it integrates better".
There's a typo at "inuitiion" under "A simple typed functional language",
along with "but but nonmemorable bases remain a serious handicap." under
PKI/Identity.
Might want to explain explicitly, under "Definition", that the frozen spec
(Nock) is merely used as a bootstrapping core for Urbit and is not the
entire ecosystem...it took me quite a while to understand that the first
time I encountered Urbit. It's also very scary that it jumps directly into
Nock formulas for booting the kernel. All of the "see lower for
implementation details" might be better as direct links, also, for the
impatient or curious. No promises they will understand everything, but jets
are pretty easy to understand without having to wade through %vanes first.
I'd put an explanation of the groupings of runes before "Syntax
geometry". It's quite important, and can explain the insane-sounding idea
of runes.
The overview of how Urbit boots mentions events and the %noun tag for
them, but it doesn't actually introduce events until "Input and output".
Not that big of a deal, but it's quite a gap. Maybe a quick sentence about
what you mean by an event, until the real intro?
"Every Urbit event log starts by setting its own unique and permanent
base." This isn't actually correct, right? Promoting a sub adds on
another entry in your will and resets your base. Will this be ripped out
before the relaunch?
The overview of vanes mentions how they take the current time to cheat
referential transparency. It kinda makes it sound like it's just a sly con,
though, and that the current time is always passed in. Possibly mention how
that allows replaying of the event log to be consistant? Or at least make
it sound less dirty :p
"Certificate transfer and key exchange are aggressively piggybacked, so
the first packet to a stranger is always signed but not encrypted."
Thanks for mentioning this! When I was reading over the crypto code a
couple weeks back I tripped over this and thought it was a flaw. Glad that
it's cleared up immediately, along with the downside.
"%clay is a sort of simplified, reactive, typed git." Ah. Maybe we should
steer away from the word "reactive", no? ;D
I-I only just got the %ford -> builder joke. Man, now I feel dumb.
"Marks are like MIME types, except with executable specifications, which
define how to validate, convert, diff, and patch content objects, they
would be marks." I don't think this sentence makes any sense? Cut off the
"they would be marks". /tip also isn't currently a thing, right?
"Urbit is a personal server; it can handle an HN avalanche on your blog"
[citation needed] ;D
%jael sounds quite unsafe. Do we actually currently use it for anything?
Arbitrary apps being able to open each other's secrets doesn't sound like
something you want. In fact, some mechanism that has apps behind a %vane
whitelist would be great - I don't care if +facebook can read my %jael if
it can't touch anything else, although that's pretty hard to define when
%clay and %gall all hit the network. The topic of :dojo also touches on
this with its generators, but it doesn't seem that isolated - %clay is
exposed by .^, for example, and so all of the classes can request .^(%cx
/~evil-attacker/msg) and communicate over the network to ~evil-attacker,
even if they aren't of the web-scraper type.
In %dill, "Every system needs its"...? Sentence cut-off?
Reading the Jets section when suddenly: "Another approach would be a safe
systems language, such as Rust."
Ouch, that hurt. It's a bit hard to write a jet in Rust when >50% of u3
are C macros. I kinda gave up, unfortunately.\
Might want to point out a bit cleaner about replacing galaxy keys and
forking the network. It's a bit hard to read right now, and how we promise
that Urbit isn't designed to be decrypted by dictators, regardless of what
websites say. You can ignore all ships (wait, what's the generic name for
urbits now? It's not ships, but all the computer classes aren't one
specific thing. Planetary body is too verbose...) signed by a galaxy and
accept ones signed by your new hard-coded public key, there's just no
promise the rest of the network will agree with you.
"The abstraction does not leak" is my new favorite thing. Sorry for the
rambling.
On Thursday, September 17, 2015 at 4:28:57 PM UTC-4, Galen Wolfe-Pauly

Post by Galen Wolfe-Pauly
Psst.
http://happut-fopnys.urbit.org/home/tree/pub/release-0
Weâd love to hear feedback from the list before posting anywhere else.

Jeremy Wall

2015-09-18 13:35:31 UTC

that "iff" might be intentional. It's often used in mathematical papers to
mean "if and only if".

On Thu, Sep 17, 2015 at 9:57 PM, Andrew Harkness <

Post by Andrew Harkness
I noticed another typo in addition to "intuitiion" - there's an "iff", in
"This statement succeeds iff it accurately communicates..." under
Requirements.
Also, another vote for "pserver" being an unwieldy sort of word - I kept
wanting to read it as "pee-server", despite efforts otherwise.
Otherwise, quite an interesting read :)
(Please excuse any failure of mailing list etiquette - I'm new to this..)

Post by c***@gmail.com
I think I love you.
First thing that jump at me is how terrible of a term "pserver" is. Is
there really nothing better?
Maybe introduce the semantic origins of Ur-bit and refer to it as "an
urbit" like we do currently?
Although you quickly explain what's wrong with AWS, and how private
servers are eating ur dataz, that's only the "OS" part of the "OS and
Internet" rewrite. The introduction doesn't touch on what's wrong with the
web until later, except for "it integrates better".
There's a typo at "inuitiion" under "A simple typed functional language",
along with "but but nonmemorable bases remain a serious handicap." under
PKI/Identity.
Might want to explain explicitly, under "Definition", that the frozen
spec (Nock) is merely used as a bootstrapping core for Urbit and is not the
entire ecosystem...it took me quite a while to understand that the first
time I encountered Urbit. It's also very scary that it jumps directly into
Nock formulas for booting the kernel. All of the "see lower for
implementation details" might be better as direct links, also, for the
impatient or curious. No promises they will understand everything, but jets
are pretty easy to understand without having to wade through %vanes first.
I'd put an explanation of the groupings of runes before "Syntax
geometry". It's quite important, and can explain the insane-sounding idea
of runes.
The overview of how Urbit boots mentions events and the %noun tag for
them, but it doesn't actually introduce events until "Input and output".
Not that big of a deal, but it's quite a gap. Maybe a quick sentence about
what you mean by an event, until the real intro?
"Every Urbit event log starts by setting its own unique and permanent
base." This isn't actually correct, right? Promoting a sub adds on
another entry in your will and resets your base. Will this be ripped out
before the relaunch?
The overview of vanes mentions how they take the current time to cheat
referential transparency. It kinda makes it sound like it's just a sly con,
though, and that the current time is always passed in. Possibly mention how
that allows replaying of the event log to be consistant? Or at least make
it sound less dirty :p
"Certificate transfer and key exchange are aggressively piggybacked, so
the first packet to a stranger is always signed but not encrypted."
Thanks for mentioning this! When I was reading over the crypto code a
couple weeks back I tripped over this and thought it was a flaw. Glad that
it's cleared up immediately, along with the downside.
"%clay is a sort of simplified, reactive, typed git." Ah. Maybe we
should steer away from the word "reactive", no? ;D
I-I only just got the %ford -> builder joke. Man, now I feel dumb.
"Marks are like MIME types, except with executable specifications, which
define how to validate, convert, diff, and patch content objects, they
would be marks." I don't think this sentence makes any sense? Cut off the
"they would be marks". /tip also isn't currently a thing, right?
"Urbit is a personal server; it can handle an HN avalanche on your blog"
[citation needed] ;D
%jael sounds quite unsafe. Do we actually currently use it for anything?
Arbitrary apps being able to open each other's secrets doesn't sound like
something you want. In fact, some mechanism that has apps behind a %vane
whitelist would be great - I don't care if +facebook can read my %jael if
it can't touch anything else, although that's pretty hard to define when
%clay and %gall all hit the network. The topic of :dojo also touches on
this with its generators, but it doesn't seem that isolated - %clay is
exposed by .^, for example, and so all of the classes can request .^(%cx
/~evil-attacker/msg) and communicate over the network to ~evil-attacker,
even if they aren't of the web-scraper type.
In %dill, "Every system needs its"...? Sentence cut-off?
Reading the Jets section when suddenly: "Another approach would be a
safe systems language, such as Rust."
Ouch, that hurt. It's a bit hard to write a jet in Rust when >50% of u3
are C macros. I kinda gave up, unfortunately.\
Might want to point out a bit cleaner about replacing galaxy keys and
forking the network. It's a bit hard to read right now, and how we promise
that Urbit isn't designed to be decrypted by dictators, regardless of what
websites say. You can ignore all ships (wait, what's the generic name for
urbits now? It's not ships, but all the computer classes aren't one
specific thing. Planetary body is too verbose...) signed by a galaxy and
accept ones signed by your new hard-coded public key, there's just no
promise the rest of the network will agree with you.
"The abstraction does not leak" is my new favorite thing. Sorry for the
rambling.
On Thursday, September 17, 2015 at 4:28:57 PM UTC-4, Galen Wolfe-Pauly

Post by Galen Wolfe-Pauly
Psst.
http://happut-fopnys.urbit.org/home/tree/pub/release-0
Weâd love to hear feedback from the list before posting anywhere else.

You received this message because you are subscribed to the Google Groups "urbit" group.
To unsubscribe from this group and stop receiving emails from it, send an
For more options, visit https://groups.google.com/d/optout.

--
Jeremy Wall
http://jeremy.marzhillstudios.com
***@marzhillstudios.com

Curtis Yarvin

2015-09-18 18:08:17 UTC

A bunch of replies / questions:

(-1) Does anyone *not mind* "pserver"? Anyone? Bueller? Bueller? Or is
it just lame?
(0) The word "base" is becoming "plot." As in, "plot of real estate," but
also "plot a course," ie, a route. Also sounds a little like "planetary
bodies" :-)
(1) Yes, I am the sort of geek that uses "iff" without thinking about it :-)
(2) There are a bunch of things in the white paper that are...
aspirational. This includes the nifty little boot sequence, and
"single-homing" (you can't change your base).
(3) Major props for noticing the crypto first-packet tradeoff yourself
before it was documented!
(4) I'm not quite sure what you mean about "bootstrapping core" versus
"entire ecosystem"? You just mean that no one actually writes code in Nock?
(5) Yes, %jael only works when you have the correct approach to security
across the system. I know this approach, but it's not in the paper at
present... :-(

Post by c***@gmail.com
I think I love you.
First thing that jump at me is how terrible of a term "pserver" is. Is
there really nothing better?
Maybe introduce the semantic origins of Ur-bit and refer to it as "an
urbit" like we do currently?
Although you quickly explain what's wrong with AWS, and how private
servers are eating ur dataz, that's only the "OS" part of the "OS and
Internet" rewrite. The introduction doesn't touch on what's wrong with the
web until later, except for "it integrates better".
There's a typo at "inuitiion" under "A simple typed functional language",
along with "but but nonmemorable bases remain a serious handicap." under
PKI/Identity.
Might want to explain explicitly, under "Definition", that the frozen spec
(Nock) is merely used as a bootstrapping core for Urbit and is not the
entire ecosystem...it took me quite a while to understand that the first
time I encountered Urbit. It's also very scary that it jumps directly into
Nock formulas for booting the kernel. All of the "see lower for
implementation details" might be better as direct links, also, for the
impatient or curious. No promises they will understand everything, but jets
are pretty easy to understand without having to wade through %vanes first.
I'd put an explanation of the groupings of runes before "Syntax
geometry". It's quite important, and can explain the insane-sounding idea
of runes.
The overview of how Urbit boots mentions events and the %noun tag for
them, but it doesn't actually introduce events until "Input and output".
Not that big of a deal, but it's quite a gap. Maybe a quick sentence about
what you mean by an event, until the real intro?
"Every Urbit event log starts by setting its own unique and permanent
base." This isn't actually correct, right? Promoting a sub adds on
another entry in your will and resets your base. Will this be ripped out
before the relaunch?
The overview of vanes mentions how they take the current time to cheat
referential transparency. It kinda makes it sound like it's just a sly con,
though, and that the current time is always passed in. Possibly mention how
that allows replaying of the event log to be consistant? Or at least make
it sound less dirty :p
"Certificate transfer and key exchange are aggressively piggybacked, so
the first packet to a stranger is always signed but not encrypted."
Thanks for mentioning this! When I was reading over the crypto code a
couple weeks back I tripped over this and thought it was a flaw. Glad that
it's cleared up immediately, along with the downside.
"%clay is a sort of simplified, reactive, typed git." Ah. Maybe we should
steer away from the word "reactive", no? ;D
I-I only just got the %ford -> builder joke. Man, now I feel dumb.
"Marks are like MIME types, except with executable specifications, which
define how to validate, convert, diff, and patch content objects, they
would be marks." I don't think this sentence makes any sense? Cut off the
"they would be marks". /tip also isn't currently a thing, right?
"Urbit is a personal server; it can handle an HN avalanche on your blog"
[citation needed] ;D
%jael sounds quite unsafe. Do we actually currently use it for anything?
Arbitrary apps being able to open each other's secrets doesn't sound like
something you want. In fact, some mechanism that has apps behind a %vane
whitelist would be great - I don't care if +facebook can read my %jael if
it can't touch anything else, although that's pretty hard to define when
%clay and %gall all hit the network. The topic of :dojo also touches on
this with its generators, but it doesn't seem that isolated - %clay is
exposed by .^, for example, and so all of the classes can request .^(%cx
/~evil-attacker/msg) and communicate over the network to ~evil-attacker,
even if they aren't of the web-scraper type.
In %dill, "Every system needs its"...? Sentence cut-off?
Reading the Jets section when suddenly: "Another approach would be a safe
systems language, such as Rust."
Ouch, that hurt. It's a bit hard to write a jet in Rust when >50% of u3
are C macros. I kinda gave up, unfortunately.\
Might want to point out a bit cleaner about replacing galaxy keys and
forking the network. It's a bit hard to read right now, and how we promise
that Urbit isn't designed to be decrypted by dictators, regardless of what
websites say. You can ignore all ships (wait, what's the generic name for
urbits now? It's not ships, but all the computer classes aren't one
specific thing. Planetary body is too verbose...) signed by a galaxy and
accept ones signed by your new hard-coded public key, there's just no
promise the rest of the network will agree with you.
"The abstraction does not leak" is my new favorite thing. Sorry for the
rambling.
On Thursday, September 17, 2015 at 4:28:57 PM UTC-4, Galen Wolfe-Pauly

Post by Galen Wolfe-Pauly
Psst.
http://happut-fopnys.urbit.org/home/tree/pub/release-0
Weâd love to hear feedback from the list before posting anywhere else.

Raymond Pasco

2015-09-18 19:44:38 UTC

What *is* the correct approach to security?

(pserver totally sucks, especially if you can't pronounce the 'p'...)

Yours,
r

Philip Monk

2015-09-18 19:49:02 UTC

I recommend everyone who dislikes "pserver" to come up with an
alternative. Two hard things, and whatnot.

Post by Raymond Pasco
What *is* the correct approach to security?
(pserver totally sucks, especially if you can't pronounce the 'p'...)
Yours,
r
--
You received this message because you are subscribed to the Google Groups "urbit" group.
To unsubscribe from this group and stop receiving emails from it, send an
For more options, visit https://groups.google.com/d/optout.

Raymond Pasco

2015-09-18 19:57:21 UTC

The neologism-free versions like "personal cloud computer" or even "personal server" work for me. If you have to go neologism, it's better to be entirely straight-faced and call it a "federated Yarvin node" or something.

Yours,
r

Henry Ault

2015-09-18 20:06:45 UTC

i like just calling it an "ur-bit", but I think personal cloud computer
could work just fine too.

Post by Raymond Pasco
The neologism-free versions like "personal cloud computer" or even
"personal server" work for me. If you have to go neologism, it's better to
be entirely straight-faced and call it a "federated Yarvin node" or
something.
Yours,
r
--
You received this message because you are subscribed to the Google Groups "urbit" group.
To unsubscribe from this group and stop receiving emails from it, send an
For more options, visit https://groups.google.com/d/optout.

Galen Wolfe-Pauly

2015-09-18 20:17:49 UTC

Yeah it's tough to see how the category can exist separately from urbit since we claim to be defining it at the same time. I do think there is a category here though.

Personal server seems fine to me too. Less things to remember.

i like just calling it an "ur-bit", but I think personal cloud computer could work just fine too.

Post by Raymond Pasco
The neologism-free versions like "personal cloud computer" or even "personal server" work for me. If you have to go neologism, it's better to be entirely straight-faced and call it a "federated Yarvin node" or something.
Yours,
r
--
You received this message because you are subscribed to the Google Groups "urbit" group.
For more options, visit https://groups.google.com/d/optout.

--
You received this message because you are subscribed to the Google Groups "urbit" group.
For more options, visit https://groups.google.com/d/optout.

Anton Dyudin

2015-09-18 20:18:54 UTC

Tool-assisted speedruns on Federated Yarvin Nodes: party 5

Post by Henry Ault
i like just calling it an "ur-bit", but I think personal cloud computer
could work just fine too.

Post by Raymond Pasco
The neologism-free versions like "personal cloud computer" or even
"personal server" work for me. If you have to go neologism, it's better to
be entirely straight-faced and call it a "federated Yarvin node" or
something.
Yours,
r
--
You received this message because you are subscribed to the Google Groups "urbit" group.
To unsubscribe from this group and stop receiving emails from it, send an
.
For more options, visit https://groups.google.com/d/optout.

--
You received this message because you are subscribed to the Google Groups "urbit" group.
To unsubscribe from this group and stop receiving emails from it, send an
.
For more options, visit https://groups.google.com/d/optout.

Curtis Yarvin

2015-09-18 20:52:11 UTC

Before "pserver" we were appropriating the old Vannevar Bush word, "memex"...

Sent from my iPhone

I recommend everyone who dislikes "pserver" to come up with an alternative. Two hard things, and whatnot.

Post by Raymond Pasco
What *is* the correct approach to security?
(pserver totally sucks, especially if you can't pronounce the 'p'...)
Yours,
r
--
You received this message because you are subscribed to the Google Groups "urbit" group.
For more options, visit https://groups.google.com/d/optout.

Philip Monk

2015-09-18 20:53:31 UTC

I'm pretty sure the only reason I'm not utterly opposed to pserver is
because at least it's better than memex.

Post by Curtis Yarvin
Before "pserver" we were appropriating the old Vannevar Bush word, "memex"...
Sent from my iPhone
I recommend everyone who dislikes "pserver" to come up with an
alternative. Two hard things, and whatnot.

Galen Wolfe-Pauly

2015-09-18 20:59:44 UTC

I think it's just unlikely that we can prefigure the name of the product category. I sincerely hope that it does get named out of necessity, but we may not be the ones to do it.

I'm pretty sure the only reason I'm not utterly opposed to pserver is because at least it's better than memex.

Post by Curtis Yarvin
Before "pserver" we were appropriating the old Vannevar Bush word, "memex"...
Sent from my iPhone

I recommend everyone who dislikes "pserver" to come up with an alternative. Two hard things, and whatnot.

Post by Raymond Pasco
What *is* the correct approach to security?
(pserver totally sucks, especially if you can't pronounce the 'p'...)
Yours,
r
--
You received this message because you are subscribed to the Google Groups "urbit" group.
For more options, visit https://groups.google.com/d/optout.

--
You received this message because you are subscribed to the Google Groups "urbit" group.
For more options, visit https://groups.google.com/d/optout.

Nik Black

2015-09-18 21:07:51 UTC

While you can't say for certain what the category will be called, you
definitely have some influence since you're defining it right now.
"Personal Cloud Computer (PCC)" Seems like it would work well, it's
similar to PC but it isn't quite. (And it's not pserver :P).

Post by Galen Wolfe-Pauly
I think it's just unlikely that we can prefigure the name of the product
category. I sincerely hope that it does get named out of necessity, but we
may not be the ones to do it.
I'm pretty sure the only reason I'm not utterly opposed to pserver is
because at least it's better than memex.

Post by Raymond Pasco
What *is* the correct approach to security?
(pserver totally sucks, especially if you can't pronounce the 'p'...)
Yours,
r
--
You received this message because you are subscribed to the Google Groups
"urbit" group.
To unsubscribe from this group and stop receiving emails from it, send an
For more options, visit https://groups.google.com/d/optout.

--
You received this message because you are subscribed to the Google Groups "urbit" group.
To unsubscribe from this group and stop receiving emails from it, send an
For more options, visit https://groups.google.com/d/optout.
--
You received this message because you are subscribed to the Google Groups "urbit" group.
To unsubscribe from this group and stop receiving emails from it, send an
For more options, visit https://groups.google.com/d/optout.

Anton Dyudin

2015-09-18 21:10:43 UTC

And when in a decade that inevitably gets corrupted, PC3 (Personal
Closed-Cloud Computer)

Post by Nik Black
While you can't say for certain what the category will be called, you
definitely have some influence since you're defining it right now.
"Personal Cloud Computer (PCC)" Seems like it would work well, it's
similar to PC but it isn't quite. (And it's not pserver :P).

Post by Galen Wolfe-Pauly
I think it's just unlikely that we can prefigure the name of the product
category. I sincerely hope that it does get named out of necessity, but

Post by Galen Wolfe-Pauly
may not be the ones to do it.
I'm pretty sure the only reason I'm not utterly opposed to pserver is
because at least it's better than memex.

Groups

Post by Raymond Pasco
"urbit" group.
To unsubscribe from this group and stop receiving emails from it, send

an
<javascript:;>.

Post by Raymond Pasco
For more options, visit https://groups.google.com/d/optout.

--
You received this message because you are subscribed to the Google Groups
"urbit" group.
To unsubscribe from this group and stop receiving emails from it, send an

<javascript:;>.

Post by Galen Wolfe-Pauly
For more options, visit https://groups.google.com/d/optout.
--
You received this message because you are subscribed to the Google Groups
"urbit" group.
To unsubscribe from this group and stop receiving emails from it, send an

<javascript:;>.

Post by Galen Wolfe-Pauly
For more options, visit https://groups.google.com/d/optout.

--
You received this message because you are subscribed to the Google Groups "urbit" group.
To unsubscribe from this group and stop receiving emails from it, send an
<javascript:;>.
For more options, visit https://groups.google.com/d/optout.

Steve Dee

2015-09-18 23:06:49 UTC

Fwiw I don't mind pserver.

Post by Anton Dyudin
And when in a decade that inevitably gets corrupted, PC3 (Personal
Closed-Cloud Computer)

Post by Galen Wolfe-Pauly
I think it's just unlikely that we can prefigure the name of the product
category. I sincerely hope that it does get named out of necessity, but

Post by Galen Wolfe-Pauly
may not be the ones to do it.
I'm pretty sure the only reason I'm not utterly opposed to pserver is
because at least it's better than memex.

Groups

Post by Raymond Pasco
"urbit" group.
To unsubscribe from this group and stop receiving emails from it,

send an

Post by Raymond Pasco
For more options, visit https://groups.google.com/d/optout.

--
You received this message because you are subscribed to the Google

Groups

Post by Galen Wolfe-Pauly
"urbit" group.
To unsubscribe from this group and stop receiving emails from it, send

Post by Galen Wolfe-Pauly
For more options, visit https://groups.google.com/d/optout.
--
You received this message because you are subscribed to the Google

Groups

Post by Galen Wolfe-Pauly
"urbit" group.
To unsubscribe from this group and stop receiving emails from it, send

Post by Galen Wolfe-Pauly
For more options, visit https://groups.google.com/d/optout.

Cameron Higgins

2015-09-18 23:33:32 UTC

is PCC out? That's my favorite so far.

Post by Steve Dee
Fwiw I don't mind pserver.

Post by Anton Dyudin
And when in a decade that inevitably gets corrupted, PC3 (Personal
Closed-Cloud Computer)

Post by Raymond Pasco
What *is* the correct approach to security?
(pserver totally sucks, especially if you can't pronounce the 'p'...)
Yours,
r
--
You received this message because you are subscribed to the Google Groups
"urbit" group.
To unsubscribe from this group and stop receiving emails from it, send an
For more options, visit https://groups.google.com/d/optout.

--
You received this message because you are subscribed to the Google Groups
"urbit" group.
To unsubscribe from this group and stop receiving emails from it, send an
For more options, visit https://groups.google.com/d/optout.
--
You received this message because you are subscribed to the Google Groups
"urbit" group.
To unsubscribe from this group and stop receiving emails from it, send an
For more options, visit https://groups.google.com/d/optout.

Anton Dyudin

2015-09-18 23:35:03 UTC

*went to a PCC for a semester once*

Post by Cameron Higgins
is PCC out? That's my favorite so far.

Post by Steve Dee
Fwiw I don't mind pserver.

Post by Anton Dyudin
And when in a decade that inevitably gets corrupted, PC3 (Personal
Closed-Cloud Computer)

Post by Galen Wolfe-Pauly
I think it's just unlikely that we can prefigure the name of the product
category. I sincerely hope that it does get named out of necessity,

but

Post by Galen Wolfe-Pauly
we
may not be the ones to do it.
I'm pretty sure the only reason I'm not utterly opposed to pserver is
because at least it's better than memex.

Post by Raymond Pasco
What *is* the correct approach to security?
(pserver totally sucks, especially if you can't pronounce the

'p'...)

Post by Raymond Pasco
Yours,
r
--
You received this message because you are subscribed to the Google Groups
"urbit" group.
To unsubscribe from this group and stop receiving emails from it, send an

<javascript:;>.

Post by Raymond Pasco
For more options, visit https://groups.google.com/d/optout.

--
You received this message because you are subscribed to the Google Groups
"urbit" group.
To unsubscribe from this group and stop receiving emails from it,

send

Post by Galen Wolfe-Pauly
an

<javascript:;>.

send

Post by Galen Wolfe-Pauly
an

<javascript:;>.

Post by Galen Wolfe-Pauly
For more options, visit https://groups.google.com/d/optout.

--
You received this message because you are subscribed to the Google

Groups

Post by Nik Black
"urbit" group.
To unsubscribe from this group and stop receiving emails from it, send

an
<javascript:;>.

Post by Nik Black
For more options, visit https://groups.google.com/d/optout.

--
You received this message because you are subscribed to the Google

Groups

Post by Anton Dyudin
"urbit" group.
To unsubscribe from this group and stop receiving emails from it, send

an
<javascript:;>.

Post by Anton Dyudin
For more options, visit https://groups.google.com/d/optout.

--
You received this message because you are subscribed to the Google Groups
"urbit" group.
To unsubscribe from this group and stop receiving emails from it, send an

<javascript:;>.

Post by Steve Dee
For more options, visit https://groups.google.com/d/optout.

Philip Monk

2015-09-18 23:36:46 UTC

As long as we don't call it PC^2, because OS/2 was traumatic enough.

Post by Anton Dyudin
*went to a PCC for a semester once*

Post by Cameron Higgins
is PCC out? That's my favorite so far.

Post by Steve Dee
Fwiw I don't mind pserver.

Post by Anton Dyudin
And when in a decade that inevitably gets corrupted, PC3 (Personal
Closed-Cloud Computer)

Post by Galen Wolfe-Pauly
I think it's just unlikely that we can prefigure the name of the product
category. I sincerely hope that it does get named out of necessity,

but

Post by Galen Wolfe-Pauly
we
may not be the ones to do it.
I'm pretty sure the only reason I'm not utterly opposed to pserver

Post by Galen Wolfe-Pauly
because at least it's better than memex.

Post by Curtis Yarvin
Before "pserver" we were appropriating the old Vannevar Bush word,
"memex"...
Sent from my iPhone
I recommend everyone who dislikes "pserver" to come up with an
alternative. Two hard things, and whatnot.

Post by Raymond Pasco
What *is* the correct approach to security?
(pserver totally sucks, especially if you can't pronounce the

'p'...)

Post by Raymond Pasco
Yours,
r
--
You received this message because you are subscribed to the Google Groups
"urbit" group.
To unsubscribe from this group and stop receiving emails from it, send an
For more options, visit https://groups.google.com/d/optout.

--
You received this message because you are subscribed to the Google Groups
"urbit" group.
To unsubscribe from this group and stop receiving emails from it,

send

Post by Galen Wolfe-Pauly
an
For more options, visit https://groups.google.com/d/optout.
--
You received this message because you are subscribed to the Google Groups
"urbit" group.
To unsubscribe from this group and stop receiving emails from it,

send

Post by Galen Wolfe-Pauly
an
For more options, visit https://groups.google.com/d/optout.

--
You received this message because you are subscribed to the Google

Groups

Post by Nik Black
"urbit" group.
To unsubscribe from this group and stop receiving emails from it,

send an

Post by Nik Black
For more options, visit https://groups.google.com/d/optout.

--
You received this message because you are subscribed to the Google

Groups

Post by Anton Dyudin
"urbit" group.
To unsubscribe from this group and stop receiving emails from it, send

Post by Anton Dyudin
For more options, visit https://groups.google.com/d/optout.

--
You received this message because you are subscribed to the Google

Groups

Post by Steve Dee
"urbit" group.
To unsubscribe from this group and stop receiving emails from it, send

Post by Steve Dee
For more options, visit https://groups.google.com/d/optout.

Curtis Yarvin

2015-09-19 00:43:38 UTC

dilnem-nodfun suggested "stead," which I thought was interesting. Any opinions in the peanut gallery?

Base is definitely "plot," it's final.

Sent from my iPhone

Post by Philip Monk
As long as we don't call it PC^2, because OS/2 was traumatic enough.

Post by Anton Dyudin
*went to a PCC for a semester once*

Post by Cameron Higgins
is PCC out? That's my favorite so far.

Post by Steve Dee
Fwiw I don't mind pserver.

Post by Anton Dyudin
And when in a decade that inevitably gets corrupted, PC3 (Personal
Closed-Cloud Computer)

Post by Curtis Yarvin
Before "pserver" we were appropriating the old Vannevar Bush word,
"memex"...
Sent from my iPhone
I recommend everyone who dislikes "pserver" to come up with an
alternative. Two hard things, and whatnot.

Post by Raymond Pasco
What *is* the correct approach to security?
(pserver totally sucks, especially if you can't pronounce the 'p'...)
Yours,
r
--
You received this message because you are subscribed to the Google
Groups
"urbit" group.
To unsubscribe from this group and stop receiving emails from it,
send an
For more options, visit https://groups.google.com/d/optout.

--
You received this message because you are subscribed to the Google Groups
"urbit" group.
To unsubscribe from this group and stop receiving emails from it, send an
For more options, visit https://groups.google.com/d/optout.
--
You received this message because you are subscribed to the Google Groups
"urbit" group.
To unsubscribe from this group and stop receiving emails from it, send an
For more options, visit https://groups.google.com/d/optout.

Philip Monk

2015-09-19 00:51:20 UTC

@thinks planets should be homesteads and moons should be steeds.

Post by Curtis Yarvin
dilnem-nodfun suggested "stead," which I thought was interesting. Any
opinions in the peanut gallery?
Base is definitely "plot," it's final.
Sent from my iPhone
As long as we don't call it PC^2, because OS/2 was traumatic enough.

Post by Anton Dyudin
*went to a PCC for a semester once*

Post by Cameron Higgins
is PCC out? That's my favorite so far.

Post by Steve Dee
Fwiw I don't mind pserver.

Post by Anton Dyudin
And when in a decade that inevitably gets corrupted, PC3 (Personal
Closed-Cloud Computer)

Post by Galen Wolfe-Pauly
I think it's just unlikely that we can prefigure the name of the product
category. I sincerely hope that it does get named out of

necessity, but

Post by Galen Wolfe-Pauly
we
may not be the ones to do it.
I'm pretty sure the only reason I'm not utterly opposed to pserver

Post by Galen Wolfe-Pauly
because at least it's better than memex.

Post by Curtis Yarvin
Before "pserver" we were appropriating the old Vannevar Bush word,
"memex"...
Sent from my iPhone
I recommend everyone who dislikes "pserver" to come up with an
alternative. Two hard things, and whatnot.

Post by Raymond Pasco
What *is* the correct approach to security?
(pserver totally sucks, especially if you can't pronounce the

'p'...)

Post by Raymond Pasco
Yours,
r
--
You received this message because you are subscribed to the

Google

Post by Raymond Pasco
Groups
"urbit" group.
To unsubscribe from this group and stop receiving emails from it,
send an
For more options, visit https://groups.google.com/d/optout.

--
You received this message because you are subscribed to the Google Groups
"urbit" group.
To unsubscribe from this group and stop receiving emails from it,

send

send

Post by Galen Wolfe-Pauly
an
For more options, visit https://groups.google.com/d/optout.

--
You received this message because you are subscribed to the Google

Groups

Post by Nik Black
"urbit" group.
To unsubscribe from this group and stop receiving emails from it,

send an

Post by Nik Black
For more options, visit https://groups.google.com/d/optout.

--
You received this message because you are subscribed to the Google

Groups

Post by Anton Dyudin
"urbit" group.
To unsubscribe from this group and stop receiving emails from it,

send an

Post by Anton Dyudin
For more options, visit https://groups.google.com/d/optout.

--
You received this message because you are subscribed to the Google

Groups

Post by Steve Dee
"urbit" group.
To unsubscribe from this group and stop receiving emails from it, send

Post by Steve Dee
For more options, visit https://groups.google.com/d/optout.

Curtis Yarvin

2015-09-19 00:55:08 UTC

@cautions Comrade Monk that this is a serious matter! A revolution is not a garden-party, comrade...

Sent from my iPhone

Post by Philip Monk
@thinks planets should be homesteads and moons should be steeds.

Post by Curtis Yarvin
dilnem-nodfun suggested "stead," which I thought was interesting. Any opinions in the peanut gallery?
Base is definitely "plot," it's final.
Sent from my iPhone

Post by Philip Monk
As long as we don't call it PC^2, because OS/2 was traumatic enough.

Post by Anton Dyudin
*went to a PCC for a semester once*

Post by Cameron Higgins
is PCC out? That's my favorite so far.

Post by Steve Dee
Fwiw I don't mind pserver.

Post by Anton Dyudin
And when in a decade that inevitably gets corrupted, PC3 (Personal
Closed-Cloud Computer)

Post by Galen Wolfe-Pauly
I think it's just unlikely that we can prefigure the name of the
product
category. I sincerely hope that it does get named out of necessity, but
we
may not be the ones to do it.
I'm pretty sure the only reason I'm not utterly opposed to pserver is
because at least it's better than memex.

Post by Curtis Yarvin
Before "pserver" we were appropriating the old Vannevar Bush word,
"memex"...
Sent from my iPhone
I recommend everyone who dislikes "pserver" to come up with an
alternative. Two hard things, and whatnot.

Post by Raymond Pasco
What *is* the correct approach to security?
(pserver totally sucks, especially if you can't pronounce the 'p'...)
Yours,
r
--
You received this message because you are subscribed to the Google
Groups
"urbit" group.
To unsubscribe from this group and stop receiving emails from it,
send an
For more options, visit https://groups.google.com/d/optout.

--
You received this message because you are subscribed to the Google
Groups
"urbit" group.
To unsubscribe from this group and stop receiving emails from it, send
an
For more options, visit https://groups.google.com/d/optout.
--
You received this message because you are subscribed to the Google
Groups
"urbit" group.
To unsubscribe from this group and stop receiving emails from it, send
an
For more options, visit https://groups.google.com/d/optout.

jpt4

2015-09-19 01:00:03 UTC

I am glad my soul is still susceptible to stirring, and that of such Urbit remains capable.

Post by Curtis Yarvin
@cautions Comrade Monk that this is a serious matter! A revolution is
not a garden-party, comrade...
Sent from my iPhone

Post by Philip Monk
@thinks planets should be homesteads and moons should be steeds.

Post by Curtis Yarvin
dilnem-nodfun suggested "stead," which I thought was interesting.

Any opinions in the peanut gallery?

Post by Curtis Yarvin
Base is definitely "plot," it's final.
Sent from my iPhone

Post by Philip Monk
As long as we don't call it PC^2, because OS/2 was traumatic

enough.

Post by Anton Dyudin
*went to a PCC for a semester once*

On Friday, September 18, 2015, Cameron Higgins
is PCC out? That's my favorite so far.

Post by Steve Dee
Fwiw I don't mind pserver.

Post by Anton Dyudin
And when in a decade that inevitably gets corrupted, PC3

(Personal

Post by Anton Dyudin
Closed-Cloud Computer)

Post by Nik Black
While you can't say for certain what the category will be

called, you

Post by Nik Black
definitely have some influence since you're defining it right

now.

Post by Nik Black
"Personal Cloud Computer (PCC)" Seems like it would work

well, it's

Post by Nik Black
similar to PC but it isn't quite. (And it's not pserver :P).
On Fri, Sep 18, 2015 at 4:59 PM, Galen Wolfe-Pauly

Post by Galen Wolfe-Pauly
I think it's just unlikely that we can prefigure the name

of the

Post by Galen Wolfe-Pauly
product
category. I sincerely hope that it does get named out of

necessity, but

Post by Galen Wolfe-Pauly
we
may not be the ones to do it.
On Sep 18, 2015, at 1:53 PM, Philip Monk
I'm pretty sure the only reason I'm not utterly opposed to

pserver is

Post by Galen Wolfe-Pauly
because at least it's better than memex.
On Fri, Sep 18, 2015 at 1:52 PM, Curtis Yarvin

Post by Curtis Yarvin
Before "pserver" we were appropriating the old Vannevar

Bush word,

Post by Curtis Yarvin
"memex"...
Sent from my iPhone
On Sep 18, 2015, at 12:49 PM, Philip Monk
I recommend everyone who dislikes "pserver" to come up

with an

Post by Curtis Yarvin
alternative. Two hard things, and whatnot.
On Fri, Sep 18, 2015 at 12:44 PM, Raymond Pasco

Post by Raymond Pasco
What *is* the correct approach to security?
(pserver totally sucks, especially if you can't pronounce

the 'p'...)

Post by Raymond Pasco
Yours,
r
--
You received this message because you are subscribed to

the Google

Post by Raymond Pasco
Groups
"urbit" group.
To unsubscribe from this group and stop receiving emails

from it,

Post by Raymond Pasco
send an
To post to this group, send email to
For more options, visit

https://groups.google.com/d/optout.

Post by Galen Wolfe-Pauly
--
You received this message because you are subscribed to the

Google

Post by Galen Wolfe-Pauly
Groups
"urbit" group.
To unsubscribe from this group and stop receiving emails

from it, send

Post by Galen Wolfe-Pauly
an
To post to this group, send email to
For more options, visit https://groups.google.com/d/optout.
--
You received this message because you are subscribed to the

Google

Post by Galen Wolfe-Pauly
Groups
"urbit" group.
To unsubscribe from this group and stop receiving emails

from it, send

Post by Galen Wolfe-Pauly
an
To post to this group, send email to
For more options, visit https://groups.google.com/d/optout.

--
You received this message because you are subscribed to the

Google Groups

Post by Nik Black
"urbit" group.
To unsubscribe from this group and stop receiving emails from

it, send an

Post by Nik Black
To post to this group, send email to
For more options, visit https://groups.google.com/d/optout.

--
You received this message because you are subscribed to the

Google Groups

Post by Anton Dyudin
"urbit" group.
To unsubscribe from this group and stop receiving emails from

it, send an

Post by Anton Dyudin
To post to this group, send email to
For more options, visit https://groups.google.com/d/optout.

--
You received this message because you are subscribed to the

Google Groups

Post by Steve Dee
"urbit" group.
To unsubscribe from this group and stop receiving emails from

it, send an

Post by Steve Dee
To post to this group, send email to
For more options, visit https://groups.google.com/d/optout.

--
You received this message because you are subscribed to the Google Groups "urbit" group.
To unsubscribe from this group and stop receiving emails from it, send
For more options, visit https://groups.google.com/d/optout.

Curtis Yarvin

2015-09-19 01:03:34 UTC

Moons would be steads, of course. All plots above 2^32 would just be *pronounced* "steed..."

Sent from my iPhone

Post by Philip Monk
@thinks planets should be homesteads and moons should be steeds.

Post by Curtis Yarvin
dilnem-nodfun suggested "stead," which I thought was interesting. Any opinions in the peanut gallery?
Base is definitely "plot," it's final.
Sent from my iPhone

Post by Philip Monk
As long as we don't call it PC^2, because OS/2 was traumatic enough.

Post by Anton Dyudin
*went to a PCC for a semester once*

Post by Cameron Higgins
is PCC out? That's my favorite so far.

Post by Steve Dee
Fwiw I don't mind pserver.

Post by Anton Dyudin
And when in a decade that inevitably gets corrupted, PC3 (Personal
Closed-Cloud Computer)

Post by Galen Wolfe-Pauly
I think it's just unlikely that we can prefigure the name of the
product
category. I sincerely hope that it does get named out of necessity, but
we
may not be the ones to do it.
I'm pretty sure the only reason I'm not utterly opposed to pserver is
because at least it's better than memex.

Post by Curtis Yarvin
Before "pserver" we were appropriating the old Vannevar Bush word,
"memex"...
Sent from my iPhone
I recommend everyone who dislikes "pserver" to come up with an
alternative. Two hard things, and whatnot.

Post by Raymond Pasco
What *is* the correct approach to security?
(pserver totally sucks, especially if you can't pronounce the 'p'...)
Yours,
r
--
You received this message because you are subscribed to the Google
Groups
"urbit" group.
To unsubscribe from this group and stop receiving emails from it,
send an
For more options, visit https://groups.google.com/d/optout.

--
You received this message because you are subscribed to the Google
Groups
"urbit" group.
To unsubscribe from this group and stop receiving emails from it, send
an
For more options, visit https://groups.google.com/d/optout.
--
You received this message because you are subscribed to the Google
Groups
"urbit" group.
To unsubscribe from this group and stop receiving emails from it, send
an
For more options, visit https://groups.google.com/d/optout.

Nik Black

2015-09-20 18:08:50 UTC

I finally got around to reading through the whole thing and commenting
on confusing bits. It's entirely possible that I've just misunderstood
some things, but even in that case it might be worth tweaking the
wording or something. I *do* know the answer to some of these (eg. I
know that @ is used for odours), but I'm pretending I'm completely
ignorant of urbit, instead of mostly ignorant. All the comments are in
*roughly* the order they show up in the paper, everything should be
ctrl+f-able though.

I also think I found a bug, unrelated to the whitepaper itself. If I
click on one of the headers (not in the ToC) the tab's process pegs to
100% CPU usage and locks up (I can scroll, but can't select text or
ctrl+f). It looks like when I click on the header, it sets the anchor
in the URL to #-the-header (instead of #the-header), and when the
anchor in the URL doesn't exist on the page chrome becomes incredibly
unhappy.

(E): There should probably be more explanation, most of these probably
only need a couple words
(M): means probably a mistake, these should be simple fixes unless
I've just grossly misunderstood a sentence or something
(S): is a suggestion of some sort

(Style) Why is the whole second paragraph in the "A simpler OS"
section a parenthetical?
Also the fourth one.
Found another in "Why not a pserver built on JS or JVM?". It starts
with "It's worth mentioning..."

(E) Are you using prefix equality when you're talking about the
operating function? Is "= (T (V I)" equivalent to "T = V I"? If so,
you should probably either mention that, or switch to infix equality.
If not, you should definitely explain the notation.

(M) "Its job is to compile I2, the Hoon compiler as a Nock formula,
with I3, the Hoon compiler as source" Shouldn't it be compiling I3
with I2?

(E) What does the 1.234.456.789 notation mean? Is . just your
thousands separator? (Probably doesn't matter too much when you're
showing the Nock for the boot components, but definitely relevant when
you're talking about the types.

(S) I recall in the old docs you had some pseudo-assembly language for
Nock. If that's still in use, it might be neat to use it when you're
showing the nock for the boot stages.

(E) What is a fragment? You probably want to talk about fragment
address stuff when you're introducing nouns.

(E) What does ++thing mean?

(E) What is an arm? You use the term but don't give any sort of definition

(S) A more realistic example of a wire would be nice, [%foo %bar %baz
~] doesn't give me anything more than the text before it. (Perhaps
show an HTTP request wire?)

(E) What does the ~ at the end of that wire mean? You haven't used it
or mentioned it before.

(E) "The data in the wire is opaque to Urbit" Why is this the case,
and also how is the opaqueness enforced?

(M) "à la" not "a la", but this might just be because I know some French.

(E) "pull the original calling vane from the top wire" I thought wires
were opaque (except in that Unix interface case)

(E) Wait... What's a vase? I get that it's a [type noun] cell, but how
do you use that to simplify the duct/wire stuff?

(E) What is Urbit as a noun vs Urbit as a proper noun?

(E) Aha! Now I know that ~ is 0. But wait, wasn't that %$? Or is %$
more like null?

(E) What does the @ mean in [@da $+(path (unit (unit value)))]? This
is the *only* '@' in the whitepaper that isn't in the list of symbols

(M?) "The product of the ++scry gate is (unit (unit value))" Shouldn't
this be [unit [unit value]]?

(Technical question) "the current time (which is guaranteed to change
between events)" Is this *actually* true, or is there some stupid
convoluted case where this could bite you?

(??) "The vanes are separately documented" I take it this hasn't been
released yet. If it has, link?

(Style) "opaque atom mapped to it. This opaque ++bone, plus the
channel itself," -> "opaque atom mapped to it, a ++bone. The ++bone,
plus the channel itself,"

(M) You missed an "are": "local and remote causes *ARE* treated
exactly the same way"

(??) "For packet geeks only" Please please please give me a whole
section on the packety things :P

(M) "For instance, git where creates" should be "For instance, where
git creates"

(M) "Once, when code compiled" should be "One, when code compiled"

(M?) At the end of the %dill section. "Every system needs its". Is
that meant to be a clever "Every system needs its timers", or a
mistake?

Post by Curtis Yarvin
Moons would be steads, of course. All plots above 2^32 would just be
*pronounced* "steed..."
Sent from my iPhone
@thinks planets should be homesteads and moons should be steeds.

Post by Anton Dyudin
*went to a PCC for a semester once*

Post by Cameron Higgins
is PCC out? That's my favorite so far.

Post by Steve Dee
Fwiw I don't mind pserver.

Post by Anton Dyudin
And when in a decade that inevitably gets corrupted, PC3 (Personal
Closed-Cloud Computer)

Post by Galen Wolfe-Pauly
I think it's just unlikely that we can prefigure the name of the product
category. I sincerely hope that it does get named out of necessity, but
we
may not be the ones to do it.
I'm pretty sure the only reason I'm not utterly opposed to pserver is
because at least it's better than memex.

Post by Curtis Yarvin
Before "pserver" we were appropriating the old Vannevar Bush word,
"memex"...
Sent from my iPhone
I recommend everyone who dislikes "pserver" to come up with an
alternative. Two hard things, and whatnot.

Post by Raymond Pasco
What *is* the correct approach to security?
(pserver totally sucks, especially if you can't pronounce the 'p'...)
Yours,
r
--
You received this message because you are subscribed to the Google
Groups
"urbit" group.
To unsubscribe from this group and stop receiving emails from it,
send an
For more options, visit https://groups.google.com/d/optout.

--
You received this message because you are subscribed to the Google Groups
"urbit" group.
To unsubscribe from this group and stop receiving emails from it, send
an
For more options, visit https://groups.google.com/d/optout.
--
You received this message because you are subscribed to the Google Groups
"urbit" group.
To unsubscribe from this group and stop receiving emails from it, send
an
For more options, visit https://groups.google.com/d/optout.

Curtis Yarvin

2015-09-20 19:18:20 UTC

Nik, incredibly great detailed comments, I'll work through them for the next draft. Thanks!

Sent from my iPhone

Post by Nik Black
I finally got around to reading through the whole thing and commenting
on confusing bits. It's entirely possible that I've just misunderstood
some things, but even in that case it might be worth tweaking the
wording or something. I *do* know the answer to some of these (eg. I
ignorant of urbit, instead of mostly ignorant. All the comments are in
*roughly* the order they show up in the paper, everything should be
ctrl+f-able though.
I also think I found a bug, unrelated to the whitepaper itself. If I
click on one of the headers (not in the ToC) the tab's process pegs to
100% CPU usage and locks up (I can scroll, but can't select text or
ctrl+f). It looks like when I click on the header, it sets the anchor
in the URL to #-the-header (instead of #the-header), and when the
anchor in the URL doesn't exist on the page chrome becomes incredibly
unhappy.
(E): There should probably be more explanation, most of these probably
only need a couple words
(M): means probably a mistake, these should be simple fixes unless
I've just grossly misunderstood a sentence or something
(S): is a suggestion of some sort
(Style) Why is the whole second paragraph in the "A simpler OS"
section a parenthetical?
Also the fourth one.
Found another in "Why not a pserver built on JS or JVM?". It starts
with "It's worth mentioning..."
(E) Are you using prefix equality when you're talking about the
operating function? Is "= (T (V I)" equivalent to "T = V I"? If so,
you should probably either mention that, or switch to infix equality.
If not, you should definitely explain the notation.
(M) "Its job is to compile I2, the Hoon compiler as a Nock formula,
with I3, the Hoon compiler as source" Shouldn't it be compiling I3
with I2?
(E) What does the 1.234.456.789 notation mean? Is . just your
thousands separator? (Probably doesn't matter too much when you're
showing the Nock for the boot components, but definitely relevant when
you're talking about the types.
(S) I recall in the old docs you had some pseudo-assembly language for
Nock. If that's still in use, it might be neat to use it when you're
showing the nock for the boot stages.
(E) What is a fragment? You probably want to talk about fragment
address stuff when you're introducing nouns.
(E) What does ++thing mean?
(E) What is an arm? You use the term but don't give any sort of definition
(S) A more realistic example of a wire would be nice, [%foo %bar %baz
~] doesn't give me anything more than the text before it. (Perhaps
show an HTTP request wire?)
(E) What does the ~ at the end of that wire mean? You haven't used it
or mentioned it before.
(E) "The data in the wire is opaque to Urbit" Why is this the case,
and also how is the opaqueness enforced?
(M) "à la" not "a la", but this might just be because I know some French.
(E) "pull the original calling vane from the top wire" I thought wires
were opaque (except in that Unix interface case)
(E) Wait... What's a vase? I get that it's a [type noun] cell, but how
do you use that to simplify the duct/wire stuff?
(E) What is Urbit as a noun vs Urbit as a proper noun?
(E) Aha! Now I know that ~ is 0. But wait, wasn't that %$? Or is %$
more like null?
(M?) "The product of the ++scry gate is (unit (unit value))" Shouldn't
this be [unit [unit value]]?
(Technical question) "the current time (which is guaranteed to change
between events)" Is this *actually* true, or is there some stupid
convoluted case where this could bite you?
(??) "The vanes are separately documented" I take it this hasn't been
released yet. If it has, link?
(Style) "opaque atom mapped to it. This opaque ++bone, plus the
channel itself," -> "opaque atom mapped to it, a ++bone. The ++bone,
plus the channel itself,"
(M) You missed an "are": "local and remote causes *ARE* treated
exactly the same way"
(??) "For packet geeks only" Please please please give me a whole
section on the packety things :P
(M) "For instance, git where creates" should be "For instance, where
git creates"
(M) "Once, when code compiled" should be "One, when code compiled"
(M?) At the end of the %dill section. "Every system needs its". Is
that meant to be a clever "Every system needs its timers", or a
mistake?

Post by Anton Dyudin
*went to a PCC for a semester once*

Post by Cameron Higgins
is PCC out? That's my favorite so far.

Post by Steve Dee
Fwiw I don't mind pserver.

Post by Anton Dyudin
And when in a decade that inevitably gets corrupted, PC3 (Personal
Closed-Cloud Computer)

Post by Galen Wolfe-Pauly
I think it's just unlikely that we can prefigure the name of the product
category. I sincerely hope that it does get named out of necessity, but
we
may not be the ones to do it.
I'm pretty sure the only reason I'm not utterly opposed to pserver is
because at least it's better than memex.

Post by Curtis Yarvin
Before "pserver" we were appropriating the old Vannevar Bush word,
"memex"...
Sent from my iPhone
I recommend everyone who dislikes "pserver" to come up with an
alternative. Two hard things, and whatnot.

Post by Raymond Pasco
What *is* the correct approach to security?
(pserver totally sucks, especially if you can't pronounce the 'p'...)
Yours,
r
--
You received this message because you are subscribed to the Google
Groups
"urbit" group.
To unsubscribe from this group and stop receiving emails from it,
send an
For more options, visit https://groups.google.com/d/optout.

--
You received this message because you are subscribed to the Google Groups
"urbit" group.
To unsubscribe from this group and stop receiving emails from it, send
an
For more options, visit https://groups.google.com/d/optout.
--
You received this message because you are subscribed to the Google Groups
"urbit" group.
To unsubscribe from this group and stop receiving emails from it, send
an
For more options, visit https://groups.google.com/d/optout.

Philip Monk

2015-09-21 01:11:49 UTC

Nik, your comments are great. I have also noticed the tree bug that kills
chrome tabs. Not sure if it's related, but I've also seen the entire table
of contents get stuffed into the anchor, which caused problems.

Most of your questions should be answered by clarifying in the whitepaper,
but here's a couple of answers for you:

Regarding the current time, we do indeed guarantee that it will change
(monotonically) between events. This is one of those obvious design
choices that unix failed to make.

I've written some amount of partially-out-of-date doc for some of the vanes
[0], including fairly detailed commentary on the ames code [1], which is
basically in line with the current ames, though there are some relatively
minor changes in Curtis's testames branch. Incidentally, for some people
I'd guess that those commentaries would be a good way to learn hoon. The
clay commentary is probably the most out of date, but there's a Clay
architecture doc [2] that's basically up-to-date, and it's better anyway.

[0] http://happut-fopnys.urbit.org/home/tree/pub/doc/arvo
[1] http://happut-fopnys.urbit.org/home/tree/pub/doc/arvo/ames/commentary
[2]
https://github.com/philipcmonk/urbit/blob/9bc578e786985ec0ad73b53f6033c0be0d6505ee/urb/zod/pub/doc/arvo/clay/architecture.md

Post by Curtis Yarvin
Nik, incredibly great detailed comments, I'll work through them for the
next draft. Thanks!
Sent from my iPhone

definition

Post by Nik Black
(S) A more realistic example of a wire would be nice, [%foo %bar %baz
~] doesn't give me anything more than the text before it. (Perhaps
show an HTTP request wire?)
(E) What does the ~ at the end of that wire mean? You haven't used it
or mentioned it before.
(E) "The data in the wire is opaque to Urbit" Why is this the case,
and also how is the opaqueness enforced?
(M) "Ã la" not "a la", but this might just be because I know some French.
(E) "pull the original calling vane from the top wire" I thought wires
were opaque (except in that Unix interface case)
(E) Wait... What's a vase? I get that it's a [type noun] cell, but how
do you use that to simplify the duct/wire stuff?
(E) What is Urbit as a noun vs Urbit as a proper noun?
(E) Aha! Now I know that ~ is 0. But wait, wasn't that %$? Or is %$
more like null?
(M?) "The product of the ++scry gate is (unit (unit value))" Shouldn't
this be [unit [unit value]]?
(Technical question) "the current time (which is guaranteed to change
between events)" Is this *actually* true, or is there some stupid
convoluted case where this could bite you?
(??) "The vanes are separately documented" I take it this hasn't been
released yet. If it has, link?
(Style) "opaque atom mapped to it. This opaque ++bone, plus the
channel itself," -> "opaque atom mapped to it, a ++bone. The ++bone,
plus the channel itself,"
(M) You missed an "are": "local and remote causes *ARE* treated
exactly the same way"
(??) "For packet geeks only" Please please please give me a whole
section on the packety things :P
(M) "For instance, git where creates" should be "For instance, where
git creates"
(M) "Once, when code compiled" should be "One, when code compiled"
(M?) At the end of the %dill section. "Every system needs its". Is
that meant to be a clever "Every system needs its timers", or a
mistake?

Post by Anton Dyudin
*went to a PCC for a semester once*

Post by Cameron Higgins
is PCC out? That's my favorite so far.

Post by Steve Dee
Fwiw I don't mind pserver.

Post by Anton Dyudin
And when in a decade that inevitably gets corrupted, PC3 (Personal
Closed-Cloud Computer)

it's

Post by Nik Black
similar to PC but it isn't quite. (And it's not pserver :P).

pserver

Post by Galen Wolfe-Pauly
is
because at least it's better than memex.

Post by Curtis Yarvin
Before "pserver" we were appropriating the old Vannevar Bush word,
"memex"...
Sent from my iPhone
I recommend everyone who dislikes "pserver" to come up with an
alternative. Two hard things, and whatnot.

Post by Raymond Pasco
What *is* the correct approach to security?
(pserver totally sucks, especially if you can't pronounce the
'p'...)
Yours,
r
--
You received this message because you are subscribed to the Google
Groups
"urbit" group.
To unsubscribe from this group and stop receiving emails from it,
send an
To post to this group, send email to
For more options, visit https://groups.google.com/d/optout.

--
You received this message because you are subscribed to the

Google

Post by Galen Wolfe-Pauly
Groups
"urbit" group.
To unsubscribe from this group and stop receiving emails from it, send
an
For more options, visit https://groups.google.com/d/optout.
--
You received this message because you are subscribed to the

Google

Post by Galen Wolfe-Pauly
Groups
"urbit" group.
To unsubscribe from this group and stop receiving emails from it, send
an
For more options, visit https://groups.google.com/d/optout.

--
You received this message because you are subscribed to the Google Groups
"urbit" group.
To unsubscribe from this group and stop receiving emails from it,

send

Post by Steve Dee
an
For more options, visit https://groups.google.com/d/optout.

Mark Justman

2015-09-21 07:12:56 UTC

A lurker vote for PCC as the upgraded term for pserver.

Post by Philip Monk
Nik, your comments are great. I have also noticed the tree bug that kills
chrome tabs. Not sure if it's related, but I've also seen the entire table
of contents get stuffed into the anchor, which caused problems.
Most of your questions should be answered by clarifying in the whitepaper,
Regarding the current time, we do indeed guarantee that it will change
(monotonically) between events. This is one of those obvious design
choices that unix failed to make.
I've written some amount of partially-out-of-date doc for some of the
vanes [0], including fairly detailed commentary on the ames code [1], which
is basically in line with the current ames, though there are some
relatively minor changes in Curtis's testames branch. Incidentally, for
some people I'd guess that those commentaries would be a good way to learn
hoon. The clay commentary is probably the most out of date, but there's a
Clay architecture doc [2] that's basically up-to-date, and it's better
anyway.
[0] http://happut-fopnys.urbit.org/home/tree/pub/doc/arvo
[1] http://happut-fopnys.urbit.org/home/tree/pub/doc/arvo/ames/commentary
[2]
https://github.com/philipcmonk/urbit/blob/9bc578e786985ec0ad73b53f6033c0be0d6505ee/urb/zod/pub/doc/arvo/clay/architecture.md

Post by Curtis Yarvin
Nik, incredibly great detailed comments, I'll work through them for the
next draft. Thanks!
Sent from my iPhone

definition

French.

Post by Nik Black
(E) "pull the original calling vane from the top wire" I thought wires
were opaque (except in that Unix interface case)
(E) Wait... What's a vase? I get that it's a [type noun] cell, but how
do you use that to simplify the duct/wire stuff?
(E) What is Urbit as a noun vs Urbit as a proper noun?
(E) Aha! Now I know that ~ is 0. But wait, wasn't that %$? Or is %$
more like null?
(M?) "The product of the ++scry gate is (unit (unit value))" Shouldn't
this be [unit [unit value]]?
(Technical question) "the current time (which is guaranteed to change
between events)" Is this *actually* true, or is there some stupid
convoluted case where this could bite you?
(??) "The vanes are separately documented" I take it this hasn't been
released yet. If it has, link?
(Style) "opaque atom mapped to it. This opaque ++bone, plus the
channel itself," -> "opaque atom mapped to it, a ++bone. The ++bone,
plus the channel itself,"
(M) You missed an "are": "local and remote causes *ARE* treated
exactly the same way"
(??) "For packet geeks only" Please please please give me a whole
section on the packety things :P
(M) "For instance, git where creates" should be "For instance, where
git creates"
(M) "Once, when code compiled" should be "One, when code compiled"
(M?) At the end of the %dill section. "Every system needs its". Is
that meant to be a clever "Every system needs its timers", or a
mistake?

Post by Anton Dyudin
*went to a PCC for a semester once*
On Friday, September 18, 2015, Cameron Higgins <

Post by Cameron Higgins
is PCC out? That's my favorite so far.

Post by Steve Dee
Fwiw I don't mind pserver.

Post by Anton Dyudin
And when in a decade that inevitably gets corrupted, PC3 (Personal
Closed-Cloud Computer)

Post by Nik Black
While you can't say for certain what the category will be called, you
definitely have some influence since you're defining it right

now.

Post by Nik Black
"Personal Cloud Computer (PCC)" Seems like it would work well,

it's

Post by Nik Black
similar to PC but it isn't quite. (And it's not pserver :P).
On Fri, Sep 18, 2015 at 4:59 PM, Galen Wolfe-Pauly <

pserver

Post by Galen Wolfe-Pauly
is
because at least it's better than memex.

Post by Curtis Yarvin
Before "pserver" we were appropriating the old Vannevar Bush word,
"memex"...
Sent from my iPhone
I recommend everyone who dislikes "pserver" to come up with an
alternative. Two hard things, and whatnot.

Post by Raymond Pasco
What *is* the correct approach to security?
(pserver totally sucks, especially if you can't pronounce the
'p'...)
Yours,
r
--
You received this message because you are subscribed to the
Google
Groups
"urbit" group.
To unsubscribe from this group and stop receiving emails from it,
send an
To post to this group, send email to
For more options, visit https://groups.google.com/d/optout.

--
You received this message because you are subscribed to the

Google

Post by Galen Wolfe-Pauly
Groups
"urbit" group.
To unsubscribe from this group and stop receiving emails from

it,

Post by Galen Wolfe-Pauly
send
an

Post by Galen Wolfe-Pauly
For more options, visit https://groups.google.com/d/optout.
--
You received this message because you are subscribed to the

Google

Post by Galen Wolfe-Pauly
Groups
"urbit" group.
To unsubscribe from this group and stop receiving emails from

it,

Post by Galen Wolfe-Pauly
send
an

Post by Galen Wolfe-Pauly
For more options, visit https://groups.google.com/d/optout.

--
You received this message because you are subscribed to the

Google

Post by Nik Black
Groups
"urbit" group.
To unsubscribe from this group and stop receiving emails from it,
send an
For more options, visit https://groups.google.com/d/optout.

--
You received this message because you are subscribed to the Google Groups
"urbit" group.
To unsubscribe from this group and stop receiving emails from it,

send

Post by Steve Dee
an
For more options, visit https://groups.google.com/d/optout.

Anton Dyudin

2015-09-21 17:59:17 UTC

Vases definitely don't simplify duct stuff, the way zuse is currently set
up, all of the move types could just be checked statically by the vanes,
operating untyped at the arvo layer. Presumably there are reasons other
than "this used to be otherwise/ should be possible to revert to".

Post by Mark Justman
A lurker vote for PCC as the upgraded term for pserver.

Post by Philip Monk
Nik, your comments are great. I have also noticed the tree bug that
kills chrome tabs. Not sure if it's related, but I've also seen the entire
table of contents get stuffed into the anchor, which caused problems.
Most of your questions should be answered by clarifying in the
Regarding the current time, we do indeed guarantee that it will change
(monotonically) between events. This is one of those obvious design
choices that unix failed to make.
I've written some amount of partially-out-of-date doc for some of the
vanes [0], including fairly detailed commentary on the ames code [1], which
is basically in line with the current ames, though there are some
relatively minor changes in Curtis's testames branch. Incidentally, for
some people I'd guess that those commentaries would be a good way to learn
hoon. The clay commentary is probably the most out of date, but there's a
Clay architecture doc [2] that's basically up-to-date, and it's better
anyway.
[0] http://happut-fopnys.urbit.org/home/tree/pub/doc/arvo
[1] http://happut-fopnys.urbit.org/home/tree/pub/doc/arvo/ames/commentary
[2]
https://github.com/philipcmonk/urbit/blob/9bc578e786985ec0ad73b53f6033c0be0d6505ee/urb/zod/pub/doc/arvo/clay/architecture.md

Post by Curtis Yarvin
Nik, incredibly great detailed comments, I'll work through them for the
next draft. Thanks!
Sent from my iPhone

definition

French.

Post by Curtis Yarvin
dilnem-nodfun suggested "stead," which I thought was interesting.

Any

Post by Curtis Yarvin
opinions in the peanut gallery?
Base is definitely "plot," it's final.
Sent from my iPhone
As long as we don't call it PC^2, because OS/2 was traumatic enough.

Post by Anton Dyudin
*went to a PCC for a semester once*
On Friday, September 18, 2015, Cameron Higgins <

Post by Cameron Higgins
is PCC out? That's my favorite so far.

Post by Steve Dee
Fwiw I don't mind pserver.

Post by Anton Dyudin
And when in a decade that inevitably gets corrupted, PC3

(Personal

Post by Anton Dyudin
Closed-Cloud Computer)

Post by Nik Black
While you can't say for certain what the category will be

called,

Post by Nik Black
you
definitely have some influence since you're defining it right

now.

Post by Nik Black
"Personal Cloud Computer (PCC)" Seems like it would work well,

it's

Post by Nik Black
similar to PC but it isn't quite. (And it's not pserver :P).
On Fri, Sep 18, 2015 at 4:59 PM, Galen Wolfe-Pauly <

Post by Galen Wolfe-Pauly
I think it's just unlikely that we can prefigure the name of

the

Post by Galen Wolfe-Pauly
product
category. I sincerely hope that it does get named out of
necessity, but
we
may not be the ones to do it.
I'm pretty sure the only reason I'm not utterly opposed to

pserver

Post by Galen Wolfe-Pauly
is
because at least it's better than memex.

Post by Curtis Yarvin
Before "pserver" we were appropriating the old Vannevar Bush
word,
"memex"...
Sent from my iPhone
On Sep 18, 2015, at 12:49 PM, Philip Monk <
I recommend everyone who dislikes "pserver" to come up with an
alternative. Two hard things, and whatnot.

Post by Raymond Pasco
What *is* the correct approach to security?
(pserver totally sucks, especially if you can't pronounce the
'p'...)
Yours,
r
--
You received this message because you are subscribed to the
Google
Groups
"urbit" group.
To unsubscribe from this group and stop receiving emails from
it,
send an

Post by Raymond Pasco
To post to this group, send email to
For more options, visit https://groups.google.com/d/optout.

--
You received this message because you are subscribed to the

Google

Post by Galen Wolfe-Pauly
Groups
"urbit" group.
To unsubscribe from this group and stop receiving emails from

it,

Post by Galen Wolfe-Pauly
send
an

Post by Galen Wolfe-Pauly
To post to this group, send email to
For more options, visit https://groups.google.com/d/optout.
--
You received this message because you are subscribed to the

Google

Post by Galen Wolfe-Pauly
Groups
"urbit" group.
To unsubscribe from this group and stop receiving emails from

it,

Post by Galen Wolfe-Pauly
send
an

Post by Galen Wolfe-Pauly
To post to this group, send email to
For more options, visit https://groups.google.com/d/optout.

--
You received this message because you are subscribed to the

Google

Post by Nik Black
Groups
"urbit" group.
To unsubscribe from this group and stop receiving emails from

it,

Post by Nik Black
send an

Post by Nik Black
For more options, visit https://groups.google.com/d/optout.

--
You received this message because you are subscribed to the

Google

Post by Anton Dyudin
Groups
"urbit" group.
To unsubscribe from this group and stop receiving emails from it,
send an

Post by Anton Dyudin
For more options, visit https://groups.google.com/d/optout.

--
You received this message because you are subscribed to the Google Groups
"urbit" group.
To unsubscribe from this group and stop receiving emails from it,

send

Post by Steve Dee
an

Post by Steve Dee
For more options, visit https://groups.google.com/d/optout.

Philip Monk

2015-09-21 21:19:03 UTC

I believe the whitepaper anchor bug is fixed now. Can anyone else verify?

Post by Anton Dyudin
Vases definitely don't simplify duct stuff, the way zuse is currently set
up, all of the move types could just be checked statically by the vanes,
operating untyped at the arvo layer. Presumably there are reasons other
than "this used to be otherwise/ should be possible to revert to".

Post by Mark Justman
A lurker vote for PCC as the upgraded term for pserver.

Post by Philip Monk
Nik, your comments are great. I have also noticed the tree bug that
kills chrome tabs. Not sure if it's related, but I've also seen the entire
table of contents get stuffed into the anchor, which caused problems.
Most of your questions should be answered by clarifying in the
Regarding the current time, we do indeed guarantee that it will change
(monotonically) between events. This is one of those obvious design
choices that unix failed to make.
I've written some amount of partially-out-of-date doc for some of the
vanes [0], including fairly detailed commentary on the ames code [1], which
is basically in line with the current ames, though there are some
relatively minor changes in Curtis's testames branch. Incidentally, for
some people I'd guess that those commentaries would be a good way to learn
hoon. The clay commentary is probably the most out of date, but there's a
Clay architecture doc [2] that's basically up-to-date, and it's better
anyway.
[0] http://happut-fopnys.urbit.org/home/tree/pub/doc/arvo
[1]
http://happut-fopnys.urbit.org/home/tree/pub/doc/arvo/ames/commentary
[2]
https://github.com/philipcmonk/urbit/blob/9bc578e786985ec0ad73b53f6033c0be0d6505ee/urb/zod/pub/doc/arvo/clay/architecture.md

Post by Curtis Yarvin
Nik, incredibly great detailed comments, I'll work through them for the
next draft. Thanks!
Sent from my iPhone

definition

French.

Post by Curtis Yarvin
dilnem-nodfun suggested "stead," which I thought was interesting.

Any

Post by Curtis Yarvin
opinions in the peanut gallery?
Base is definitely "plot," it's final.
Sent from my iPhone
As long as we don't call it PC^2, because OS/2 was traumatic enough.

Post by Anton Dyudin
*went to a PCC for a semester once*
On Friday, September 18, 2015, Cameron Higgins <

Post by Cameron Higgins
is PCC out? That's my favorite so far.

Post by Steve Dee
Fwiw I don't mind pserver.

Post by Anton Dyudin
And when in a decade that inevitably gets corrupted, PC3

(Personal

Post by Anton Dyudin
Closed-Cloud Computer)

Post by Nik Black
While you can't say for certain what the category will be

called,

Post by Nik Black
you
definitely have some influence since you're defining it right

now.

Post by Nik Black
"Personal Cloud Computer (PCC)" Seems like it would work well,

it's

Post by Nik Black
similar to PC but it isn't quite. (And it's not pserver :P).
On Fri, Sep 18, 2015 at 4:59 PM, Galen Wolfe-Pauly <

Post by Galen Wolfe-Pauly
I think it's just unlikely that we can prefigure the name of

the

Post by Galen Wolfe-Pauly
product
category. I sincerely hope that it does get named out of
necessity, but
we
may not be the ones to do it.
I'm pretty sure the only reason I'm not utterly opposed to

pserver

Post by Galen Wolfe-Pauly
is
because at least it's better than memex.
On Fri, Sep 18, 2015 at 1:52 PM, Curtis Yarvin <

Post by Curtis Yarvin
alternative. Two hard things, and whatnot.

Post by Raymond Pasco
What *is* the correct approach to security?
(pserver totally sucks, especially if you can't pronounce

the

Post by Raymond Pasco
'p'...)
Yours,
r
--
You received this message because you are subscribed to the
Google
Groups
"urbit" group.
To unsubscribe from this group and stop receiving emails

from

Post by Raymond Pasco
it,
send an
To post to this group, send email to
For more options, visit https://groups.google.com/d/optout.

--
You received this message because you are subscribed to the

Google

Post by Galen Wolfe-Pauly
Groups
"urbit" group.
To unsubscribe from this group and stop receiving emails from

it,

Post by Galen Wolfe-Pauly
send
an
To post to this group, send email to
For more options, visit https://groups.google.com/d/optout.
--
You received this message because you are subscribed to the

Google

Post by Galen Wolfe-Pauly
Groups
"urbit" group.
To unsubscribe from this group and stop receiving emails from

it,

Post by Galen Wolfe-Pauly
send
an
To post to this group, send email to
For more options, visit https://groups.google.com/d/optout.

--
You received this message because you are subscribed to the

Google

Post by Nik Black
Groups
"urbit" group.
To unsubscribe from this group and stop receiving emails from

it,

Post by Nik Black
send an
To post to this group, send email to
For more options, visit https://groups.google.com/d/optout.

--
You received this message because you are subscribed to the

Google

Post by Anton Dyudin
Groups
"urbit" group.
To unsubscribe from this group and stop receiving emails from

it,

Post by Anton Dyudin
send an

Post by Anton Dyudin
For more options, visit https://groups.google.com/d/optout.

--
You received this message because you are subscribed to the

Google

Post by Steve Dee
Groups
"urbit" group.
To unsubscribe from this group and stop receiving emails from

it, send

Post by Steve Dee
an
For more options, visit https://groups.google.com/d/optout.

--
You received this message because you are subscribed to the Google
Groups "urbit" group.
To unsubscribe from this group and stop receiving emails from it, send
For more options, visit https://groups.google.com/d/optout.

Cameron Higgins

2015-09-22 00:02:37 UTC

Yeah, anchors seem fine now.

Post by Philip Monk
I believe the whitepaper anchor bug is fixed now. Can anyone else verify?

Post by Anton Dyudin
Vases definitely don't simplify duct stuff, the way zuse is currently set
up, all of the move types could just be checked statically by the vanes,
operating untyped at the arvo layer. Presumably there are reasons other than
"this used to be otherwise/ should be possible to revert to".

Post by Mark Justman
A lurker vote for PCC as the upgraded term for pserver.

Post by Philip Monk
Nik, your comments are great. I have also noticed the tree bug that
kills chrome tabs. Not sure if it's related, but I've also seen the entire
table of contents get stuffed into the anchor, which caused problems.
Most of your questions should be answered by clarifying in the
Regarding the current time, we do indeed guarantee that it will change
(monotonically) between events. This is one of those obvious design choices
that unix failed to make.
I've written some amount of partially-out-of-date doc for some of the
vanes [0], including fairly detailed commentary on the ames code [1], which
is basically in line with the current ames, though there are some relatively
minor changes in Curtis's testames branch. Incidentally, for some people
I'd guess that those commentaries would be a good way to learn hoon. The
clay commentary is probably the most out of date, but there's a Clay
architecture doc [2] that's basically up-to-date, and it's better anyway.
[0] http://happut-fopnys.urbit.org/home/tree/pub/doc/arvo
[1]
http://happut-fopnys.urbit.org/home/tree/pub/doc/arvo/ames/commentary
[2]
https://github.com/philipcmonk/urbit/blob/9bc578e786985ec0ad73b53f6033c0be0d6505ee/urb/zod/pub/doc/arvo/clay/architecture.md

Post by Curtis Yarvin
Nik, incredibly great detailed comments, I'll work through them for the
next draft. Thanks!
Sent from my iPhone

Post by Nik Black
I finally got around to reading through the whole thing and commenting
on confusing bits. It's entirely possible that I've just
misunderstood
some things, but even in that case it might be worth tweaking the
wording or something. I *do* know the answer to some of these (eg. I
ignorant of urbit, instead of mostly ignorant. All the comments are in
*roughly* the order they show up in the paper, everything should be
ctrl+f-able though.
I also think I found a bug, unrelated to the whitepaper itself. If I
click on one of the headers (not in the ToC) the tab's process pegs to
100% CPU usage and locks up (I can scroll, but can't select text or
ctrl+f). It looks like when I click on the header, it sets the anchor
in the URL to #-the-header (instead of #the-header), and when the
anchor in the URL doesn't exist on the page chrome becomes incredibly
unhappy.
(E): There should probably be more explanation, most of these probably
only need a couple words
(M): means probably a mistake, these should be simple fixes unless
I've just grossly misunderstood a sentence or something
(S): is a suggestion of some sort
(Style) Why is the whole second paragraph in the "A simpler OS"
section a parenthetical?
Also the fourth one.
Found another in "Why not a pserver built on JS or JVM?". It starts
with "It's worth mentioning..."
(E) Are you using prefix equality when you're talking about the
operating function? Is "= (T (V I)" equivalent to "T = V I"? If so,
you should probably either mention that, or switch to infix equality.
If not, you should definitely explain the notation.
(M) "Its job is to compile I2, the Hoon compiler as a Nock formula,
with I3, the Hoon compiler as source" Shouldn't it be compiling I3
with I2?
(E) What does the 1.234.456.789 notation mean? Is . just your
thousands separator? (Probably doesn't matter too much when you're
showing the Nock for the boot components, but definitely relevant when
you're talking about the types.
(S) I recall in the old docs you had some pseudo-assembly language for
Nock. If that's still in use, it might be neat to use it when you're
showing the nock for the boot stages.
(E) What is a fragment? You probably want to talk about fragment
address stuff when you're introducing nouns.
(E) What does ++thing mean?
(E) What is an arm? You use the term but don't give any sort of definition
(S) A more realistic example of a wire would be nice, [%foo %bar %baz
~] doesn't give me anything more than the text before it. (Perhaps
show an HTTP request wire?)
(E) What does the ~ at the end of that wire mean? You haven't used it
or mentioned it before.
(E) "The data in the wire is opaque to Urbit" Why is this the case,
and also how is the opaqueness enforced?
(M) "à la" not "a la", but this might just be because I know some French.
(E) "pull the original calling vane from the top wire" I thought wires
were opaque (except in that Unix interface case)
(E) Wait... What's a vase? I get that it's a [type noun] cell, but how
do you use that to simplify the duct/wire stuff?
(E) What is Urbit as a noun vs Urbit as a proper noun?
(E) Aha! Now I know that ~ is 0. But wait, wasn't that %$? Or is %$
more like null?
(M?) "The product of the ++scry gate is (unit (unit value))" Shouldn't
this be [unit [unit value]]?
(Technical question) "the current time (which is guaranteed to change
between events)" Is this *actually* true, or is there some stupid
convoluted case where this could bite you?
(??) "The vanes are separately documented" I take it this hasn't been
released yet. If it has, link?
(Style) "opaque atom mapped to it. This opaque ++bone, plus the
channel itself," -> "opaque atom mapped to it, a ++bone. The ++bone,
plus the channel itself,"
(M) You missed an "are": "local and remote causes *ARE* treated
exactly the same way"
(??) "For packet geeks only" Please please please give me a whole
section on the packety things :P
(M) "For instance, git where creates" should be "For instance, where
git creates"
(M) "Once, when code compiled" should be "One, when code compiled"
(M?) At the end of the %dill section. "Every system needs its". Is
that meant to be a clever "Every system needs its timers", or a
mistake?

Post by Curtis Yarvin
dilnem-nodfun suggested "stead," which I thought was interesting.
Any
opinions in the peanut gallery?
Base is definitely "plot," it's final.
Sent from my iPhone
As long as we don't call it PC^2, because OS/2 was traumatic enough.

Post by Anton Dyudin
*went to a PCC for a semester once*
On Friday, September 18, 2015, Cameron Higgins

Post by Cameron Higgins
is PCC out? That's my favorite so far.

Post by Steve Dee
Fwiw I don't mind pserver.

Post by Anton Dyudin
And when in a decade that inevitably gets corrupted, PC3 (Personal
Closed-Cloud Computer)

Post by Nik Black
While you can't say for certain what the category will be
called,
you
definitely have some influence since you're defining it right
now.
"Personal Cloud Computer (PCC)" Seems like it would work well,
it's
similar to PC but it isn't quite. (And it's not pserver :P).
On Fri, Sep 18, 2015 at 4:59 PM, Galen Wolfe-Pauly

Post by Galen Wolfe-Pauly
I think it's just unlikely that we can prefigure the name of
the
product
category. I sincerely hope that it does get named out of
necessity, but
we
may not be the ones to do it.
On Sep 18, 2015, at 1:53 PM, Philip Monk
I'm pretty sure the only reason I'm not utterly opposed to
pserver
is
because at least it's better than memex.
On Fri, Sep 18, 2015 at 1:52 PM, Curtis Yarvin

Post by Curtis Yarvin
Before "pserver" we were appropriating the old Vannevar Bush
word,
"memex"...
Sent from my iPhone
On Sep 18, 2015, at 12:49 PM, Philip Monk
I recommend everyone who dislikes "pserver" to come up with
an
alternative. Two hard things, and whatnot.

Post by Raymond Pasco
What *is* the correct approach to security?
(pserver totally sucks, especially if you can't pronounce
the
'p'...)
Yours,
r
--
You received this message because you are subscribed to the
Google
Groups
"urbit" group.
To unsubscribe from this group and stop receiving emails
from
it,
send an
To post to this group, send email to
For more options, visit https://groups.google.com/d/optout.

--
You received this message because you are subscribed to the
Google
Groups
"urbit" group.
To unsubscribe from this group and stop receiving emails from
it,
send
an
To post to this group, send email to
For more options, visit https://groups.google.com/d/optout.
--
You received this message because you are subscribed to the
Google
Groups
"urbit" group.
To unsubscribe from this group and stop receiving emails from
it,
send
an
To post to this group, send email to
For more options, visit https://groups.google.com/d/optout.

--
You received this message because you are subscribed to the Google
Groups "urbit" group.
To unsubscribe from this group and stop receiving emails from it, send
For more options, visit https://groups.google.com/d/optout.

Cameron Higgins

2015-09-22 00:14:23 UTC

Correction, it all seems to work except for the ones containing
special sections, such as the subsection for the tour of vanes. %a
%ames, for example, doesn't work. It seems any in that style is still
broken. The ones without special formatting seem fine though.

Post by Cameron Higgins
Yeah, anchors seem fine now.

Post by Philip Monk
I believe the whitepaper anchor bug is fixed now. Can anyone else verify?

Post by Anton Dyudin
Vases definitely don't simplify duct stuff, the way zuse is currently set
up, all of the move types could just be checked statically by the vanes,
operating untyped at the arvo layer. Presumably there are reasons other than
"this used to be otherwise/ should be possible to revert to".

Post by Mark Justman
A lurker vote for PCC as the upgraded term for pserver.

Post by Philip Monk
Nik, your comments are great. I have also noticed the tree bug that
kills chrome tabs. Not sure if it's related, but I've also seen the entire
table of contents get stuffed into the anchor, which caused problems.
Most of your questions should be answered by clarifying in the
Regarding the current time, we do indeed guarantee that it will change
(monotonically) between events. This is one of those obvious design choices
that unix failed to make.
I've written some amount of partially-out-of-date doc for some of the
vanes [0], including fairly detailed commentary on the ames code [1], which
is basically in line with the current ames, though there are some relatively
minor changes in Curtis's testames branch. Incidentally, for some people
I'd guess that those commentaries would be a good way to learn hoon. The
clay commentary is probably the most out of date, but there's a Clay
architecture doc [2] that's basically up-to-date, and it's better anyway.
[0] http://happut-fopnys.urbit.org/home/tree/pub/doc/arvo
[1]
http://happut-fopnys.urbit.org/home/tree/pub/doc/arvo/ames/commentary
[2]
https://github.com/philipcmonk/urbit/blob/9bc578e786985ec0ad73b53f6033c0be0d6505ee/urb/zod/pub/doc/arvo/clay/architecture.md

Post by Curtis Yarvin
Nik, incredibly great detailed comments, I'll work through them for the
next draft. Thanks!
Sent from my iPhone

Post by Curtis Yarvin
dilnem-nodfun suggested "stead," which I thought was interesting.
Any
opinions in the peanut gallery?
Base is definitely "plot," it's final.
Sent from my iPhone
As long as we don't call it PC^2, because OS/2 was traumatic enough.

Post by Anton Dyudin
*went to a PCC for a semester once*
On Friday, September 18, 2015, Cameron Higgins

Post by Cameron Higgins
is PCC out? That's my favorite so far.

Post by Steve Dee
Fwiw I don't mind pserver.

Post by Anton Dyudin
And when in a decade that inevitably gets corrupted, PC3
(Personal
Closed-Cloud Computer)

Post by Nik Black
While you can't say for certain what the category will be
called,
you
definitely have some influence since you're defining it right
now.
"Personal Cloud Computer (PCC)" Seems like it would work well,
it's
similar to PC but it isn't quite. (And it's not pserver :P).
On Fri, Sep 18, 2015 at 4:59 PM, Galen Wolfe-Pauly

Post by Galen Wolfe-Pauly
I think it's just unlikely that we can prefigure the name of
the
product
category. I sincerely hope that it does get named out of
necessity, but
we
may not be the ones to do it.
On Sep 18, 2015, at 1:53 PM, Philip Monk
I'm pretty sure the only reason I'm not utterly opposed to
pserver
is
because at least it's better than memex.
On Fri, Sep 18, 2015 at 1:52 PM, Curtis Yarvin

Post by Raymond Pasco
What *is* the correct approach to security?
(pserver totally sucks, especially if you can't pronounce
the
'p'...)
Yours,
r
--
You received this message because you are subscribed to the
Google
Groups
"urbit" group.
To unsubscribe from this group and stop receiving emails
from
it,
send an
To post to this group, send email to
For more options, visit https://groups.google.com/d/optout.

--
You received this message because you are subscribed to the Google
Groups "urbit" group.
To unsubscribe from this group and stop receiving emails from it, send
For more options, visit https://groups.google.com/d/optout.

Alex Krupp

2015-09-18 01:59:04 UTC

The abstract is a list of implementation details, not benefits. I would
delete it.

What makes social software successful isn't the hardware or the
software, but the new forms of social interaction that it enables. The
architecture details are interesting, but imho this would be stronger if
they came after the sociology.

Alex

Post by Galen Wolfe-Pauly
Psst.
http://happut-fopnys.urbit.org/home/tree/pub/release-0
Weâd love to hear feedback from the list before posting anywhere else.
--
You received this message because you are subscribed to the Google Groups "urbit" group.
To unsubscribe from this group and stop receiving emails from it, send
For more options, visit https://groups.google.com/d/optout.

--
Alex Krupp
Cell: (607) 351 2671

Subscribe to my blog: http://alexkrupp.typepad.com/
Follow me on twitter: @alexkrupp
My homepage: www.alexkrupp.com

Curtis Yarvin

2015-09-18 02:09:42 UTC

We'll post the introductory essay by itself, without the abstract, on Medium.

The abstract starts with the tech stack to warn/promise the reader that there's code under all this verbiage. Some people might suggest the opposite deletion, I think... but I feel it's important to have the combo in one place.

Sent from my iPhone

The abstract is a list of implementation details, not benefits. I would delete it.
What makes social software successful isn't the hardware or the software, but the new forms of social interaction that it enables. The architecture details are interesting, but imho this would be stronger if they came after the sociology.
Alex

Post by Galen Wolfe-Pauly
Psst.
http://happut-fopnys.urbit.org/home/tree/pub/release-0
Weâd love to hear feedback from the list before posting anywhere else.
--
You received this message because you are subscribed to the Google Groups "urbit" group.
For more options, visit https://groups.google.com/d/optout.

--
Alex Krupp
Cell: (607) 351 2671
Subscribe to my blog: http://alexkrupp.typepad.com/
My homepage: www.alexkrupp.com
--
You received this message because you are subscribed to the Google Groups "urbit" group.
For more options, visit https://groups.google.com/d/optout.

Galen Wolfe-Pauly

2015-09-18 02:33:29 UTC

Not so convinced about Medium.

I think many people just skim the abstract, so I’m not sure it’s a good idea to delete it. Maybe it could be improved by being more specific about how it’s "a different user experience”.

Post by Curtis Yarvin
We'll post the introductory essay by itself, without the abstract, on Medium.
The abstract starts with the tech stack to warn/promise the reader that there's code under all this verbiage. Some people might suggest the opposite deletion, I think... but I feel it's important to have the combo in one place.
Sent from my iPhone

The abstract is a list of implementation details, not benefits. I would delete it.
What makes social software successful isn't the hardware or the software, but the new forms of social interaction that it enables. The architecture details are interesting, but imho this would be stronger if they came after the sociology.
Alex

Post by Galen Wolfe-Pauly
Psst.
http://happut-fopnys.urbit.org/home/tree/pub/release-0
We’d love to hear feedback from the list before posting anywhere else.
--
You received this message because you are subscribed to the Google Groups "urbit" group.
For more options, visit https://groups.google.com/d/optout.

--
Alex Krupp
Cell: (607) 351 2671
http://alexkrupp.typepad.com/
www.alexkrupp.com
--
You received this message because you are subscribed to the Google Groups "urbit" group.
For more options, visit https://groups.google.com/d/optout.

--
You received this message because you are subscribed to the Google Groups "urbit" group.
For more options, visit https://groups.google.com/d/optout.

c***@gmail.com

2015-09-18 03:27:34 UTC

I second the sentiment about Medium...why not just host the introduction on
the landing page, above the whitepaper+video links?

Also, Andrew, "iff" ususally stands for "if and only if", so it might not
be a typo. Relevant XKCD: https://xkcd.com/1033/

On Thursday, September 17, 2015 at 10:33:33 PM UTC-4, Galen Wolfe-Pauly

Post by Galen Wolfe-Pauly
Not so convinced about Medium.
I think many people just skim the abstract, so Iâm not sure itâs a good
idea to delete it. Maybe it could be improved by being more specific about
how itâs "a different user experienceâ.

Post by Curtis Yarvin
We'll post the introductory essay by itself, without the abstract, on

Medium.

Post by Curtis Yarvin
The abstract starts with the tech stack to warn/promise the reader that

there's code under all this verbiage. Some people might suggest the
opposite deletion, I think... but I feel it's important to have the combo
in one place.

Post by Curtis Yarvin
Sent from my iPhone

Post by Alex Krupp
The abstract is a list of implementation details, not benefits. I would

delete it.

Post by Alex Krupp
What makes social software successful isn't the hardware or the

software, but the new forms of social interaction that it enables. The
architecture details are interesting, but imho this would be stronger if
they came after the sociology.

Post by Alex Krupp
Alex

Groups "urbit" group.

Post by Alex Krupp

Post by Galen Wolfe-Pauly
To unsubscribe from this group and stop receiving emails from it, send

<javascript:>.

Post by Alex Krupp

Post by Galen Wolfe-Pauly
For more options, visit https://groups.google.com/d/optout.

--
Alex Krupp
Cell: (607) 351 2671
http://alexkrupp.typepad.com/
www.alexkrupp.com
--
You received this message because you are subscribed to the Google

Groups "urbit" group.

Post by Alex Krupp
To unsubscribe from this group and stop receiving emails from it, send

<javascript:>.

Post by Alex Krupp
For more options, visit https://groups.google.com/d/optout.

--
You received this message because you are subscribed to the Google

Groups "urbit" group.

Post by Curtis Yarvin
To unsubscribe from this group and stop receiving emails from it, send

<javascript:>.

Post by Curtis Yarvin
For more options, visit https://groups.google.com/d/optout.

Steve Dee

2015-09-18 14:56:03 UTC

Great work! I really enjoyed reading that.

âAnd if not now, when? Will there ever be a clean-slate redesign of '70s
system software? A significant problem demands an ambitious solution.
Intuitively, clean-slate computing and the personal cloud feel like a fit.
Let's see if we can make the details work. We may never get another chance
to do it right.â

That paragraph, and especially that last sentence, perfectly expresses the
sentiment that motivated me to work on Urbit.

âA simple typed functional languageâ under âConcrete requirementsâ reads
really hand-wavy to me. The third paragraph in particular â âA pure,
higher-order, âŠâ â doesn't convince me at all of why the things it lists
are desirable. Obviously I know they're desirable from having written some
Hoon, but that exposition doesn't reach my model of the whitepaper's
audience. I'm not sure what the fix is â maybe it's just really hard to
explain Hoon short of spending a couple weeks learning Hoon â but two
things that come to mind are to either paper over it and go really abstract
and high-level in that paragraph, or to move or restate some of the âUI for
programmersâ stuff under âSyntax, textâ there.

Post by Galen Wolfe-Pauly
Psst.
http://happut-fopnys.urbit.org/home/tree/pub/release-0
Weâd love to hear feedback from the list before posting anywhere else.
--
You received this message because you are subscribed to the Google Groups "urbit" group.
To unsubscribe from this group and stop receiving emails from it, send an
For more options, visit https://groups.google.com/d/optout.

tenari

2015-09-18 16:57:42 UTC

super cool. works towards a solution to the problem of helping people even
understand what urbit is, let alone, why it is.

On Thursday, September 17, 2015 at 3:28:57 PM UTC-5, Galen Wolfe-Pauly

Post by Galen Wolfe-Pauly
Psst.
http://happut-fopnys.urbit.org/home/tree/pub/release-0
Weâd love to hear feedback from the list before posting anywhere else.

Kacper Cieśla

2015-09-25 18:18:52 UTC

Being gifted with spectacularly bad memory I decided to provide some
feedback to the whitepaper from a complete noob perspective.

Overall it's a nice comprehensive guide that I missed when I've heard about
Urbit the first time. I'll focus on what's wrong and not the things I like
(eg those little metaphors here and there), assuming that this will be more
helpful.

In general, I'm not sure I would made it through it, if not for the fact
that I was already familiar with the thing. I think at some point I could
be overwhelmed with amount of new information and associations that reader
has to make while reading it for the first time.

I think it would be nice to provide some general view of the whole system
(relations between nock, hoon, arvo vanes and more or less how it works),
before talking about cubes and forks. I know it's all connected and the
paper starts from the lowest level and nicely goes up, but it's hard to
get somebody interested in a great swiss watch starting with lecture about
shapes of the gears inside.

IMHO all names that are not used later in the paper could be dropped, new
reader has enough of them to get familiar with.

Quotes below are just search phrases to locate comments, not a context.

There are some comments referring to I0, I1, I2... I vaguely remember these
as hoon source code sections, but I don't think it's stated anywhere in the
paper e.g. "we end up in a main sequence of uniform events (starting at I5
)"

"no input not input to the pserver itself," - some typo?

"Of course every computer is deterministic at the CPU level" - I'm not even
sure that still holds true nowadays ;)

There was something about OS and dummy me wondered for a good while what
the "OP systems" are. Maybe it will help other dummies if there's some
'(OP)' next to the previous 'orthogonal persistence'.

'we can implement the JS or JVM specifications in Hoon' - that's the first
time Hoon is mentioned, having spend time on JS & JVM maybe it deserves a
few sentences of explicit introduction.

"I is a list of nouns - where a list is either 0 or a cell [item list]" -
so just a noun then?

The concept of [a b c] (even ignoring ~ that got there somewhere in the
paper) as list and being expressed as [a [b c]] was not explicitly stated
outside Nock spec I think. Maybe there could be a bit more talking how it
all runs on nock and about the ball of mud. It's all there in the paper,
but somewhere between the lines. I think it was much more explicit in the
original blog post and I liked it that way. Also I remember an info that
urbit can't call outside world but outside world can call urbit that got me
thinking in the original blog post. It's very easy to infer it from the
paper but reader has enough in his head happening already.

I was wondering if (V I) could be presented in a notation where there
wouldn't be need to reverse order later to [subject formula]

"the Arvo source. The main event sequence begins" - there was no word about
Arvo before this, reader has no idea what Arvo is yet.

"'atom' with text odor" - and no idea what odor is here

"(The syntax %string means a cord" - this %string jumps in here out of
nowhere, it's not within listed types, so it may raise confusion.

Examples of Hoon around here would get me confused. We're jumping right
into guts, perhaps it would be easier on the reader to provide a examples
how different types are getting translated into numbers like it was in some
previous doc, or some ++dec, or something else which shows more how Nock,
Hoon works rather than how it is implemented.

I don't think that telling user that %ta is ASCII and %da is date helps
reader at this point. Maybe it could with some enumerated list of chosen
types and examples of usage. But here it's just yet another confusing
string. At this point in paper she doesn't know yet anything about our
specific naming.

"'Each arm uses the whole core as its subject" - no idea for reader what an
arm is here

"nor a map of names to formulas a vtable" - not sure, but vtable was
supposed to be in parenthesis?

Hoon twigs "autocons." - typo, dot inside quote

"For twigs with a constant number of components" - maybe example with 3
elements would be better, very unsure with my memory

"I4 : Arvo" - starting with sentence what Arvo is could be nice

"++load is fragment 54 of the core battery, producing a gate whose formal
argument at fragment 6" - I'm don't know what these numbers are.

"the request with an action %this" - example of a name that I don't think
is necessary to mention in the paper, at least not while still trying to
describe how things work and not annotating source code

"a duct" - as in other places first occurrence so in italic I think?

I'd love to see more general talking how vanes communicate and how things
go together and less code annotations.

Before I remembered about clay, I thought %c was just a typo after %x and
%y (that is, that these were some examples, because [%foo %bar ~], and that
it should be %z). It could be just me. Definitely some feedback from a
person who actually hears about urbit for the first time will be very
valuable.

"contains the base and desk" - desk italic?

"head of the spur" - spur undefined

"While the Hoon type system works very well, it does not fill the niche
that MIME types fill on the Internets" - but it's still a type, right? It
seems to sound a bit like we use something else instead here.

"There are two ford request modes, %c and %r - cooked and raw."... -
another example of where I think names are not necessary. Even if you want
a whitepaper to be something where he gets back after reading it once and
it helps him understand how Hoon works, it would probably be better to
clearly separate sections describing how it works from those explaining
what is what under the hood.

"When a message fails to validate, the packet is dropped silently", whole
paragraph, there is some pretty much the same explanation earlier in the
paper (boith upgrades and backoff)

"all it sees are messages and subscriptions, just like it would receive
from another Urbit app." - few words about security model would be nice,
there's mention of %gall stopping apps from doing really weird stuff, but
what about how app distinguishes msgs that are local or not if it really
wants to (you may remember me playing with it some time ago)

"Every system needs its" - it?

"secret, it must register to receive an expiration notice." - random
thought, if it just drops packets, A and B are talking, then B subscription
breaks for some reason and B didn't get a chance to receive expiration
notice, but it's ignoring all packets now, how does this get fixed?

I assume that Hoon at this point is something less likely to change than
the communication protocol, but some more stuff about packets flying around
could be nice.

"Google Wave" - sigh, RIP

I like being upgraded from a destroyer to a whole planet. PCC sounds good,
pserver is OK for me too, although my brain won't accept that the P is
silent. Maybe PS.

So there's a lot of good stuff, but I'm not sure if it's organized in the
optimal way. Maybe to keep short attention spans like me reading it would
be better to move phonemic stuff and ships hierarchy before minting stuff.
It surely does great job at describing use case which was not clear in docs
earlier. But hmm.. some very subtle subjective feeling... like JS and JVM
are not really good for this, and IOS is problematic too, so we need
programming language, and OS, and blah.. It's kinda feels like wow, we're
fucked. In the original blog post my mind was more like - you know like
when you're coding and looking for bug, and diving through all these
abstraction layers to finally find it and whole compatibilieties and not
really knowing what inside.. we do BOOM and start from scratch. Nock. In
your face.

Anyway, I gave you pretty much everything that appeared in my head so I'm
sorry if it's pretty raw or chaotic. Really cool to see urbit getting a
solid whitepaper. If you are thinking about some more general introduction
I think all info should be in whitepaper too. Keep up the good work.

cheers,
todsef-nathes

(names stayed, right? I thought they were about to be regenerated or
something)

On Thursday, September 17, 2015 at 10:28:57 PM UTC+2, Galen Wolfe-Pauly

Post by Galen Wolfe-Pauly
Psst.
http://happut-fopnys.urbit.org/home/tree/pub/release-0
<http://www.google.com/url?q=http%3A%2F%2Fhapput-fopnys.urbit.org%2Fhome%2Ftree%2Fpub%2Frelease-0&sa=D&sntz=1&usg=AFQjCNGyjlsDvmMibiCa3SMJgb1s3N6hXA>
Weâd love to hear feedback from the list before posting anywhere else.

Curtis Yarvin

2015-09-25 22:58:23 UTC

Thanks, Kacper! Yeah, if you look at urbit.org today, there's a new
version. We ditched "pserver" and fixed a bunch of stuff.

Yes, the names are still scheduled to be replaced but have yet to be.

Ultimately I feel this whitepaper is too much of a monster -- I think it
needs to be split up, both for reading and just for managing it. On the
other hand, it is nice to have in one place...

Post by Kacper CieÅla
Being gifted with spectacularly bad memory I decided to provide some
feedback to the whitepaper from a complete noob perspective.
Overall it's a nice comprehensive guide that I missed when I've heard
about Urbit the first time. I'll focus on what's wrong and not the things I
like (eg those little metaphors here and there), assuming that this will be
more helpful.
In general, I'm not sure I would made it through it, if not for the fact
that I was already familiar with the thing. I think at some point I could
be overwhelmed with amount of new information and associations that reader
has to make while reading it for the first time.
I think it would be nice to provide some general view of the whole system
(relations between nock, hoon, arvo vanes and more or less how it works),
before talking about cubes and forks. I know it's all connected and the
paper starts from the lowest level and nicely goes up, but it's hard to
get somebody interested in a great swiss watch starting with lecture about
shapes of the gears inside.
IMHO all names that are not used later in the paper could be dropped, new
reader has enough of them to get familiar with.
Quotes below are just search phrases to locate comments, not a context.
There are some comments referring to I0, I1, I2... I vaguely remember
these as hoon source code sections, but I don't think it's stated anywhere
in the paper e.g. "we end up in a main sequence of uniform events (starting
at I5 )"
"no input not input to the pserver itself," - some typo?
"Of course every computer is deterministic at the CPU level" - I'm not
even sure that still holds true nowadays ;)
There was something about OS and dummy me wondered for a good while what
the "OP systems" are. Maybe it will help other dummies if there's some
'(OP)' next to the previous 'orthogonal persistence'.
'we can implement the JS or JVM specifications in Hoon' - that's the first
time Hoon is mentioned, having spend time on JS & JVM maybe it deserves a
few sentences of explicit introduction.
"I is a list of nouns - where a list is either 0 or a cell [item list]" -
so just a noun then?
The concept of [a b c] (even ignoring ~ that got there somewhere in the
paper) as list and being expressed as [a [b c]] was not explicitly stated
outside Nock spec I think. Maybe there could be a bit more talking how it
all runs on nock and about the ball of mud. It's all there in the paper,
but somewhere between the lines. I think it was much more explicit in the
original blog post and I liked it that way. Also I remember an info that
urbit can't call outside world but outside world can call urbit that got me
thinking in the original blog post. It's very easy to infer it from the
paper but reader has enough in his head happening already.
I was wondering if (V I) could be presented in a notation where there
wouldn't be need to reverse order later to [subject formula]
"the Arvo source. The main event sequence begins" - there was no word
about Arvo before this, reader has no idea what Arvo is yet.
"'atom' with text odor" - and no idea what odor is here
"(The syntax %string means a cord" - this %string jumps in here out of
nowhere, it's not within listed types, so it may raise confusion.
Examples of Hoon around here would get me confused. We're jumping right
into guts, perhaps it would be easier on the reader to provide a examples
how different types are getting translated into numbers like it was in some
previous doc, or some ++dec, or something else which shows more how Nock,
Hoon works rather than how it is implemented.
I don't think that telling user that %ta is ASCII and %da is date helps
reader at this point. Maybe it could with some enumerated list of chosen
types and examples of usage. But here it's just yet another confusing
string. At this point in paper she doesn't know yet anything about our
specific naming.
"'Each arm uses the whole core as its subject" - no idea for reader what
an arm is here
"nor a map of names to formulas a vtable" - not sure, but vtable was
supposed to be in parenthesis?
Hoon twigs "autocons." - typo, dot inside quote
"For twigs with a constant number of components" - maybe example with 3
elements would be better, very unsure with my memory
"I4 : Arvo" - starting with sentence what Arvo is could be nice
"++load is fragment 54 of the core battery, producing a gate whose formal
argument at fragment 6" - I'm don't know what these numbers are.
"the request with an action %this" - example of a name that I don't think
is necessary to mention in the paper, at least not while still trying to
describe how things work and not annotating source code
"a duct" - as in other places first occurrence so in italic I think?
I'd love to see more general talking how vanes communicate and how things
go together and less code annotations.
Before I remembered about clay, I thought %c was just a typo after %x and
%y (that is, that these were some examples, because [%foo %bar ~], and that
it should be %z). It could be just me. Definitely some feedback from a
person who actually hears about urbit for the first time will be very
valuable.
"contains the base and desk" - desk italic?
"head of the spur" - spur undefined
"While the Hoon type system works very well, it does not fill the niche
that MIME types fill on the Internets" - but it's still a type, right? It
seems to sound a bit like we use something else instead here.
"There are two ford request modes, %c and %r - cooked and raw."... -
another example of where I think names are not necessary. Even if you want
a whitepaper to be something where he gets back after reading it once and
it helps him understand how Hoon works, it would probably be better to
clearly separate sections describing how it works from those explaining
what is what under the hood.
"When a message fails to validate, the packet is dropped silently", whole
paragraph, there is some pretty much the same explanation earlier in the
paper (boith upgrades and backoff)
"all it sees are messages and subscriptions, just like it would receive
from another Urbit app." - few words about security model would be nice,
there's mention of %gall stopping apps from doing really weird stuff, but
what about how app distinguishes msgs that are local or not if it really
wants to (you may remember me playing with it some time ago)
"Every system needs its" - it?
"secret, it must register to receive an expiration notice." - random
thought, if it just drops packets, A and B are talking, then B subscription
breaks for some reason and B didn't get a chance to receive expiration
notice, but it's ignoring all packets now, how does this get fixed?
I assume that Hoon at this point is something less likely to change than
the communication protocol, but some more stuff about packets flying around
could be nice.
"Google Wave" - sigh, RIP
I like being upgraded from a destroyer to a whole planet. PCC sounds good,
pserver is OK for me too, although my brain won't accept that the P is
silent. Maybe PS.
So there's a lot of good stuff, but I'm not sure if it's organized in the
optimal way. Maybe to keep short attention spans like me reading it would
be better to move phonemic stuff and ships hierarchy before minting stuff.
It surely does great job at describing use case which was not clear in docs
earlier. But hmm.. some very subtle subjective feeling... like JS and JVM
are not really good for this, and IOS is problematic too, so we need
programming language, and OS, and blah.. It's kinda feels like wow, we're
fucked. In the original blog post my mind was more like - you know like
when you're coding and looking for bug, and diving through all these
abstraction layers to finally find it and whole compatibilieties and not
really knowing what inside.. we do BOOM and start from scratch. Nock. In
your face.
Anyway, I gave you pretty much everything that appeared in my head so I'm
sorry if it's pretty raw or chaotic. Really cool to see urbit getting a
solid whitepaper. If you are thinking about some more general introduction
I think all info should be in whitepaper too. Keep up the good work.
cheers,
todsef-nathes
(names stayed, right? I thought they were about to be regenerated or
something)
On Thursday, September 17, 2015 at 10:28:57 PM UTC+2, Galen Wolfe-Pauly

William Yznaga

2015-09-30 16:23:03 UTC

Concerning the whitepaper, I feel similarly to Kacper CieÅla. However, it would appear that I have much less of an acquaintance with the Urbit-related concepts introduced in the paper or referenced by it, but I would like to improve my understanding. I actually haven't yet finished reading through it, mostly because I'm unsure about something. Is an understanding of Nock (the VM, or language, or spec, or whichever term is preferred), necessary to really understand, in an integrated sort of way, the rest of the document's "Definition" section? I think I understand the proposals and arguments and such that comprise the opening of the document before the Nock spec, and I'm wondering if, to continue or finish reading, I need to understand Nock as well (relatively well, that is) as I understand the writing that comes before it.

Curtis Yarvin

2015-09-30 18:34:54 UTC

To be slightly coarse, the whitepaper is like the massive stool you release
after a week of constipation. It may not be perfectly formed, but it's
entirely substantial and extremely dense, and it had to get out there.

You definitely don't need to understand Nock as anything more a black box
marked "Turing-complete noun interpreter" or "functional assembly language"
to understand the rest of the paper. But different people have different
learning styles. For some it's fun to actually sit down and implement
Nock, then write decrement in it. For others, who cares...

Post by William Yznaga
Concerning the whitepaper, I feel similarly to Kacper CieÅla. However, it
would appear that I have much less of an acquaintance with the
Urbit-related concepts introduced in the paper or referenced by it, but I
would like to improve my understanding. I actually haven't yet finished
reading through it, mostly because I'm unsure about something. Is an
understanding of Nock (the VM, or language, or spec, or whichever term is
preferred), necessary to really understand, in an integrated sort of way,
the rest of the document's "Definition" section? I think I understand the
proposals and arguments and such that comprise the opening of the document
before the Nock spec, and I'm wondering if, to continue or finish reading,
I need to understand Nock as well (relatively well, that is) as I
understand the writing that comes before it.
--
You received this message because you are subscribed to the Google Groups "urbit" group.
To unsubscribe from this group and stop receiving emails from it, send an
For more options, visit https://groups.google.com/d/optout.

J LeBlanc

2015-09-30 18:49:55 UTC