RFD: comp.lang.go - LAST CALL FOR COMMENTS (revised)

[Usenet Big-8 Management Board]

[This is a revised version of the Final RFD/Last Call for Comments,
to address concerns raised regarding the possibility of future changes
in moderation status. The wording in question has been removed at the
request of the proponent. The posting date for the first Final RFD in
the Change History was also incorrect and has been amended.]

REQUEST FOR DISCUSSION (RFD)
unmoderated group comp.lang.go

This is a formal Request for Discussion (RFD) for the creation of the
unmoderated newsgroup comp.lang.go

NEWSGROUPS LINE:

comp.lang.go The Go programming language (Golang)

RATIONALE:

On Usenet at the moment there are only computer programming languages
that were around in the 80's and 90's. To be able to bring more people
and colleagues to Usenet we need updated groups where people will find
it interesting to log in on a daily basis.

CHARTER:

comp.lang.go will be a unmoderated group where people interested in the
Go programming language can share ideas and topics, help each other, and
keep up with interesting projects.

PROCEDURE:

Please refer to the newsgroup creation policies listed here:

http://www.big-8.org/wiki/How_to_Create_a_New_Big-8_Newsgroup

All discussion of active proposals should be posted to
news.groups.proposals.

To this end, the followup header of this RFD has been set to
news.groups.proposals.

The final comment period lasts for five (5) days from the
time that this RFD is posted.

DISTRIBUTION:

This document has been posted to the following newsgroups:

news.announce.newgroups
news.groups.proposals
comp.lang.misc

PROPONENT:

Name: ReK2
Email: re...@hispagatos.org
Gemini: gemini://rek2.hispagatos.org

CHANGE HISTORY:

2023-07-03 1st RFD
2023-09-22 Final RFD / Last Call for Comments
2023-09-25 Final RFD / Last Call for Comments (revised)

[rek2 hispagatos]

On 2023-09-25, Usenet Big-8 Management Board <bo...@big-8.org> wrote:
> [This is a revised version of the Final RFD/Last Call for Comments,
> to address concerns raised regarding the possibility of future changes
> in moderation status. The wording in question has been removed at the
> request of the proponent. The posting date for the first Final RFD in
> the Change History was also incorrect and has been amended.]
>
> REQUEST FOR DISCUSSION (RFD)
> unmoderated group comp.lang.go
>

Thanks for the revision.

Happy Hacking
ReK2

--
- {gemini,https}://{,rek2.}hispagatos.org - mastodon: @re...@hispagatos.space
- [https|gemini]://2600.Madrid - https://hispagatos.space/@rek2
- https://keyoxide.org/A31C7CE19D9C58084EA42BA26C0B0D11E9303EC5

[Mima-sama]

On 9/26/2023 00:23, Usenet Big-8 Management Board wrote:
> [This is a revised version of the Final RFD/Last Call for Comments,
> to address concerns raised regarding the possibility of future changes
> in moderation status. The wording in question has been removed at the
> request of the proponent. The posting date for the first Final RFD in
> the Change History was also incorrect and has been amended.]
>
> REQUEST FOR DISCUSSION (RFD)
> unmoderated group comp.lang.go
>
> This is a formal Request for Discussion (RFD) for the creation of the
> unmoderated newsgroup comp.lang.go

I've been holding this back to see if the proponent and its supporters
would actually prove that this newsgroup deserves to be created, but
alas, no. Not even a single post about Golang or its projects in
comp.lang.misc. So count me in those who oppose the creation of
comp.lang.go, for the same reason as ahk in news.groups. Nobody is
seriously talking about it which means the topic does not need its own
place in the Big-8 for easier discussion. How can you make discussion
easier if there's no discussion in the first place?

If the proponent wants their own newsgroup then they can newgroup one at
alt.*. If they want to stick to Big-8 then they should prove that it is
being discussed regularly throughout Usenet and that it's causing
problems therefore necessitating the creation of its own group. Like
posting to comp.lang.misc. I'm pretty sure you could also somehow use
comp.unix.shell too if the Go program is a command line tool and you
need some help piping it somewhere or integrating it to a shell script.

The point is to spread around multiple relevant newsgroups to see if
someone knowledgeable of the topic you're discussing about can follow
you up. And then if it becomes a problem (like suddenly a lot of people
are talking about Go in multiple newsgroups and it's getting annoying or
difficult to keep track of them), that's when you propose a newsgroup to
collect them all into one place. That's how Usenet has been working for
decades and I don't see any good reason to stray from it when again,
alt.* exists if they don't want to go through that process.

I've also posted about this RFD in the fediverse:
https://makai.chaotic.ninja/notes/9k2cotdoow

--
Mima
Reincarnated Legendary Evil Spirit of Complete Darkness

[Spiros Bousbouras]

Once upon a time
Spiros Bousbouras <spi...@gmail.com> wrote:

> On Wed, 27 Sep 2023 00:17:55 CST
> Mima-sama <m...@masa.ma> wrote:
> > I've been holding this back to see if the proponent and its supporters
> > would actually prove that this newsgroup deserves to be created, but
> > alas, no. Not even a single post about Golang or its projects in
> > comp.lang.misc.
>

> There does exist
> From: Vasco Costa <vasco...@invalid.invalid>
> Newsgroups: comp.lang.misc
> Subject: [GO] Who uses Go and what do you like/dislike about it?
> Date: Sun, 10 Apr 2022 14:30:08 -0000 (UTC)
> Organization: Aioe.org NNTP Server
> Message-ID: <t2uplg$k4l$1...@gioia.aioe.org>
>
> (or http://al.howardknight.net/?ID=166327300700) and that thread got a few
> replies. Nothing since then I think.

Apologies , I should have checked before posting. There is also

From: rek2 hispagatos <re...@hispagatos.org.invalid>
Newsgroups: comp.lang.misc,hispagatos.talk,es.comp.hackers,alt.2600
Subject: mpd to matrix - please review
Followup-To: hispagatos.talk
Date: Tue, 26 Sep 2023 15:34:55 -0000 (UTC)
Message-ID: <ueutmv$1l1ds$2...@matrix.hispagatos.org>

.But I'm puzzled by the Followup-To: .Presumably , if a comp.lang.go
existed , then any comments about the code should also appear on the
same group.

[Spiros Bousbouras]

On Wed, 27 Sep 2023 00:17:55 CST
Mima-sama <m...@masa.ma> wrote:

> I've been holding this back to see if the proponent and its supporters
> would actually prove that this newsgroup deserves to be created, but
> alas, no. Not even a single post about Golang or its projects in
> comp.lang.misc.

There does exist
From: Vasco Costa <vasco...@invalid.invalid>
Newsgroups: comp.lang.misc
Subject: [GO] Who uses Go and what do you like/dislike about it?
Date: Sun, 10 Apr 2022 14:30:08 -0000 (UTC)
Organization: Aioe.org NNTP Server
Message-ID: <t2uplg$k4l$1...@gioia.aioe.org>

(or http://al.howardknight.net/?ID=166327300700) and that thread got a few
replies. Nothing since then I think.

> So count me in those who oppose the creation of
> comp.lang.go, for the same reason as ahk in news.groups. Nobody is
> seriously talking about it which means the topic does not need its own
> place in the Big-8 for easier discussion. How can you make discussion
> easier if there's no discussion in the first place?

The idea is that the lack of a specific group leads to lack of discussion.
Personally I think that the idea "Build it and perhaps they will come" should
be tested every now and again but not too often. Earlier posts mentioned the
possibility of also creating a comp.lang.rust group soon. I would be
opposed to that now because I haven't seen discussion of Rust on various
technical newsgroups. If the idea is to be tested , it's enough to test it
with one subject and since Go was proposed first then the idea may as well be
tested with a comp.lang.go group. But if a comp.lang.go gets created and
attracts some discussion then at some point in the future it would be
reasonable to also discuss about a comp.lang.rust group.

> I'm pretty sure you could also somehow use
> comp.unix.shell too if the Go program is a command line tool and you
> need some help piping it somewhere or integrating it to a shell script.

The output of any command line programme written in any language can be piped
to a shell script and any programme can be called from a shell script.
comp.unix.shell is not for discussion of every programming language under
the sun.

> The point is to spread around multiple relevant newsgroups to see if
> someone knowledgeable of the topic you're discussing about can follow
> you up.

Which relevant newsgroups ? The only one I know which would be relevant is
comp.lang.misc .Anyone who wants to discuss Go now can start a thread on
comp.lang.misc .The fact that noone has done so (apart from the example I
gave above) doesn't bode well for a comp.lang.go group but , as I've said ,
I consider it acceptable to test every now and again the idea that people
prefer a specific newsgroup as opposed to a generic one.

--
vlaho.ninja/prog

[Mima-sama]

On 9/27/2023 23:22, Spiros Bousbouras wrote:
> There does exist

Ok, I will concede to that as well as the yesterday article you
mentioned in a later follow-up. I haven't thought of using a news server
with a longer retention time. But even then the fact that it took like
what, one and a half year before another article related to Go (and that
recent article is just too little too late anyway, and it doesn't have
good follow-up potential IMO, but I'm digressing now) just proves that
pretty much nobody is discussing about or interested in Go. Which is a
reason why I oppose the creation of this newsgroup.

> The idea is that the lack of a specific group leads to lack of discussion.

And I disagree with that idea. If that's true then how come Usenet has
thrived for many years before with users seemingly not minding to post
on broader newsgroups fully knowing that a specific newsgroup is not
ready yet?

It's people's laziness in posting something on-topic and making the
effort to invite their friends to Usenet that leads to lack of
discussion. The "lack of a specific group" is just an excuse as proven
by the many unused specific newsgroups in Big-8 waiting to be posted on.

> Personally I think that the idea "Build it and perhaps they will come" should
> be tested every now and again but not too often.

It has already been tested many times in the past, as ahk said
news.groups (<ueomm5$18gkb$1...@dont-email.me>). They failed.

Now you might say "but what about comp.infosystems.gemini". But I argue
it's an exception to the rule. (And it seems it couldn't keep up with
the 10 posts/day guideline over 90 days anyway which is used to gauge
whether a topic's discussion is self-sufficient enough to be split off
into its own newsgroup.)

> Earlier posts mentioned the
> possibility of also creating a comp.lang.rust group soon. I would be
> opposed to that now because I haven't seen discussion of Rust on various
> technical newsgroups.

I'm opposed to it too, but for the different reason that is that there
exists already one at alt.comp.lang.rust. There's no point in creating a
newsgroup in Big-8 for a topic whose home already exists in alt.*, as
pointed out by ahk again in the same article I mentioned by ID.

> The output of any command line programme written in any language can be piped
> to a shell script and any programme can be called from a shell script.
> comp.unix.shell is not for discussion of every programming language under
> the sun.

I'm not saying every CLI program written in Go should go there. It's
just an idea that can be a valid option to take in some cases if you use
a bit of creativity. For example, what if there's a Go program a user is
dependent on but is badly designed that it's errors are outputting to
stdout instead of stderr for some reason? User has little knowledge
about Go but is comfortable with modifying the source code with help
from someone more experienced in Golang and shell. comp.unix.shell would
be perfectly fine to (cross)post to for that scenario.

Or what about if a user is looking for a CLI HTML parser written
specifically in Go, so they could use it with their shell script? Again,
perfectly fine to (cross)post to comp.unix.shell.

> Which relevant newsgroups ?

Idk, do a keyword search of "program" in your client? There's no way you
can't find a single one that can be related to programming in Go. You're
developing a game written in Go? Post to alt.games.programming! Need
help porting a Golang program to Plan 9? Crosspost to comp.os.plan9!

> I consider it acceptable to test every now and again the idea that people
> prefer a specific newsgroup as opposed to a generic one.

If people prefer a specific newsgroup without having to prove that
Usenet is interested in their topic, they're always free to newgroup at
alt.*, as I said earlier. No RFDs required.

[Russ Allbery]

Mima-sama <m...@masa.ma> writes:

> And I disagree with that idea. If that's true then how come Usenet has
> thrived for many years before with users seemingly not minding to post
> on broader newsgroups fully knowing that a specific newsgroup is not
> ready yet?

Sometimes it works and sometimes it doesn't. The amount of crossover of
interest is a major factor.

For example, most people who play one type of Rogue-like game are at least
moderately interested in other Rogue-like games, so it was fairly easy to
build interest for a new game on rec.games.roguelike.misc and see if it
got critical mass. Or, similarly, most people who were fans of one genre
of comic books at least dipped their toes in other genres of comic books,
so rec.arts.comics.misc worked reasonably well.

On the other hand, despite being interested in programming languages and
even new programming languages, comp.lang.misc has never seemed like a
reasonable newsgroup to follow in my time on Usenet. The scope is just
way too broad for me and turns into a weird grab bag of miscellaneous
stuff, 95% of which is irrelevant to anything I'm interested in. It's the
same reason why I read rec.arts.sf.written back in the day but have zero
interest in reading rec.arts.books, whose topic scope is in theory the
entirety of written human knowledge and entertainment.

This is all in response to your comment about what Usenet was like when it
was thriving. It's a way different place now, so I don't know if that
older experience applies. But I would not immediately reject the theory
that a more specific group could get traffic that is never going to show
up on a catch-all group. People's tolerance for searching through random
stuff they're not interested in to find a few posts they are interested in
varies widely.

> It's people's laziness in posting something on-topic and making the
> effort to invite their friends to Usenet that leads to lack of
> discussion.

I'm not sure it's a very useful statement to make that some method of
newsgroup organization would work if people weren't lazy. Posting to and
reading Usenet isn't a job. No one is under any obligation to do work.
Of course people are lazy about their casual entertainment!

I'm not going to do a bunch of work to find topics that are interesting to
me on Usenet, let alone try to start that discussion myself. I'm going to
try a little bit, in proportion to how much I think I'm going to get out
of it, and then I'm going to go use some other medium entirely that
requires less effort on my part.

If Usenet wants to stick around in a world in which there are way
easier-to-use and less-obscure discussion forums, it's going to need to be
at least a little welcoming. More focused topic groups may or may not be
a way to be more welcoming; I don't know! But it's at least worth
considering.

> I'm opposed to it too, but for the different reason that is that there
> exists already one at alt.comp.lang.rust. There's no point in creating a
> newsgroup in Big-8 for a topic whose home already exists in alt.*, as
> pointed out by ahk again in the same article I mentioned by ID.

By this rule, we would never have created any Big Eight newsgroups at all.
alt.* predates all of the groups we're discussing on this newsgroup and,
due to the nature of alt.* newsgroup creation, there was essentially
always already an alt.* group.

--
Russ Allbery (ea...@eyrie.org) <https://www.eyrie.org/~eagle/>

[Julien ÉLIE]

Hi Russ,

> For example, most people who play one type of Rogue-like game are at least
> moderately interested in other Rogue-like games, so it was fairly easy to
> build interest for a new game on rec.games.roguelike.misc and see if it
> got critical mass. Or, similarly, most people who were fans of one genre
> of comic books at least dipped their toes in other genres of comic books,
> so rec.arts.comics.misc worked reasonably well.

Indeed, and in our recent systemic review of the fr.* groups, the
usefulness of some specific newsgroups was mentioned.
For instance the necessity to keep separate newsgroups for mail clients
(fr.comp.mail) and mail servers (fr.comp.mail.serveurs) as some people
are only interested in one of them. Massification to increase the
number of posts was not desired.

> On the other hand, despite being interested in programming languages and
> even new programming languages, comp.lang.misc has never seemed like a
> reasonable newsgroup to follow in my time on Usenet.

Same thing for fr.comp.lang.general. Nobody basically writes there, but
there are people reading it and willing to respond!

FWIW, one person asked there a question about Rust last year:
https://groups.google.com/g/fr.comp.lang.general/c/9EW2R2bntio

So this language can be discussed on Usenet, if need be to give a proof...

> I would not immediately reject the theory
> that a more specific group could get traffic that is never going to show
> up on a catch-all group. People's tolerance for searching through random
> stuff they're not interested in to find a few posts they are interested in
> varies widely.

Yes, I totally agree.

> I'm not going to do a bunch of work to find topics that are interesting to
> me on Usenet, let alone try to start that discussion myself. I'm going to
> try a little bit, in proportion to how much I think I'm going to get out
> of it, and then I'm going to go use some other medium entirely that
> requires less effort on my part.

Same observation here. People look for the easiest way to receive a
quick and wise response. When they see an empty newsgroup, they'll go
away most of the time (same thing as when they look at a web forum where
the last message dates back to 5 years ago, they just search another web
forum and won't loose time writing there).

> If Usenet wants to stick around in a world in which there are way
> easier-to-use and less-obscure discussion forums, it's going to need to be
> at least a little welcoming. More focused topic groups may or may not be
> a way to be more welcoming; I don't know! But it's at least worth
> considering.

That's exactly why we decided to go ahead in the fr.* hierarchy, making
it less stale. Naturally not everybody agrees (like for the Big-8) but
it's worth considering and trying.

Anyway, doing nothing won't do much good to attract people looking for a
specific subject. If the subject is not in a newsgroup name or its
description, they'll probably look elsewhere.

If someone asks for the creation of a reasonable newsgroup and the only
response is there won't be the audience, use the .misc group, that's not
very welcoming. There may be no experts in the .misc group for that
subject, and he will be told to ask in Reddit or another medium.

We decided in fr.* to be more agile and flexible about newsgroup
creations (and removals). Let's give it a real try!

--
Julien ÉLIE

« Affirmanti incumbit probatio. »

[Mima-sama]

On 9/28/2023 02:16, Russ Allbery wrote:
> If Usenet wants to stick around in a world in which there are way
> easier-to-use and less-obscure discussion forums, it's going to need to be
> at least a little welcoming. More focused topic groups may or may not be
> a way to be more welcoming; I don't know! But it's at least worth
> considering.

Call me a gatekeeper if you want, but I don't think Big-8 needs to be
like the web forums where you can easily create topic groups just
because you said you want one. It doesn't need to be Reddit or Lemmy or
kbin. If a user thinks those two or others would be more useful to them
than Usenet, I have no qualms with it. More power to them! Because at
the end of the day, all of these forums are just tools. One just uses
the right tool for the job.

I never see Usenet as competition to the web forums. It never was and
never should've been considered that way. It's like Gopher to the web;
they can both coexist. I don't see anyone calling for radically changing
Gopher to make it easier just like what seems to be being done here
(which is to turn Big-8 to what's essentially just another alt.*).

Creating a newsgroup in the Big-8 is a big (heh) deal. You're
essentially signalling to many news servers who gave their trust to the
board that this topic exists and that this topic is regularly being
discussed in Usenet, and that the newsgroup created will have a
self-sustaining discussion for many years to come. If Big-8 keeps
breaking the trust of these news admins with this idea of "more
welcoming", we're going to see fragmentation and therefore an even worse
Usenet than it already is because plenty of news admins decide that
nothing coming out of the board is making sense anymore. I don't want
that to happen. It's actually a miracle (AFAICS) that the majority of
news admins haven't revolted against the board when they let skirv
create those newsgroups redundant of alt.* which never got really used.
"It's obvious!" Yeah it's obvious that you shouldn't be in charge of
Big-8 anymore.

To quote from the Big-8 wiki's Content and Format of a Request for
Discussion (RFD):

> One of the goals of the Big-8 Management Board is to create groups
> that are well-used. Proponents may look at other newsgroups, web
> sites, mailing lists, or forums to present evidence that many people
> are interested in the topic. "Traffic Analysis" is a catch-all title
> for evidence that supports the conclusion that the topic is popular
> and that there is therefore some likelihood that the group will be
> well-used.
>
> **It is in the best interests of the proposed group** to get as many
> people as possible interested in the proposal during the discussion of
> the RFD. The requirement that *prospective supporters show themselves
> to be familiar with Usenet* means that the new group may have a good
> nucleus of posters if and when it is created. A newsgroup with no news
> is no fun. The more people whom proponents persuade to show support
> for the proposal by making Usenet posts during the discussion, *the
> better it is for the newsgroup itself* if and when it is created. A
> proponent may not simply assume that the creation of a newsgroup will
> attract traffic from existing alternatives to Usenet.

As long as the mission of board includes creating *well-used*
newsgroups, and as long as *traffic analysis* remains a valid tool to
determine whether a newsgroup deserves to be created, I will continue to
oppose these RFDs which have no proof of having a self-sustaining
discussion. The board will have to reform its process (maybe even
abolish the RFD altogether!) and seek for itself what its mission really
is in this 2020s and beyond if they want me to stop opposing. I'm not
pressuring them to do so (come on, it's just a single dissenting opinion
from me). But I guess they can take it as a suggestion if they wish.

Again, if the proponents and future proponents don't like this kind of
Big-8 they're better off asking in alt.config and having a news
administrator create it for them. They shouldn't see alt.* as something
to avoid. It's there for people who think they can't do the
self-sustaining discussion guideline that Big-8 obligates them to prove.
There's nothing wrong with admitting you want to skip the RFD, me thinks
people should try it from time to time!

> By this rule, we would never have created any Big Eight newsgroups at all.
> alt.* predates all of the groups we're discussing on this newsgroup and,
> due to the nature of alt.* newsgroup creation, there was essentially
> always already an alt.* group.

I mean an alt.* group already exists *with discussion* on it. Sorry for
not making it clear. If an alt.* exists and people are talking there,
there isn't really any point to making a Big-8 to it in the future. It's
simply annoying and unnecessarily disruptive to the people reading and
posting there. It's not like the name chosen for that alt.* group is
wrong or bad either. alt.comp.lang.rust is a perfectly fine name; one
should be able to find it easily from traversing each hierarchy and
subhierarchy. And even then, just use a keyword search in your news client?

[Russ Allbery]

Mima-sama <m...@masa.ma> writes:

> Creating a newsgroup in the Big-8 is a big (heh) deal. You're
> essentially signalling to many news servers who gave their trust to the
> board that this topic exists and that this topic is regularly being
> discussed in Usenet, and that the newsgroup created will have a
> self-sustaining discussion for many years to come.

If you're at all in the mood to listen to someone who has spent nearly
thirty years at this point involved in Usenet newsgroup creation (good
lord), I would try to convince you that this really isn't true. It's been
part of the conversation about creating new newsgroups since forever, but
the cost of a new newsgroup is mostly miniscule. It's a few inodes, and
if it's empty, it takes essentially zero resources. The biggest cost of a
new newsgroup is that it may make it marginally harder for people to find
newsgroups. That's about it.

We've been treating creating new Big Eight newsgroups like it's a big deal
for decades and decades now. When that involved splitting an active group
and trying to tell people to move a bunch of discussions, that was
justified. But for new topics that aren't really being discussed
anywhere, I think this concept has caused a lot of frustration and
annoyance and no real benefits. I'm sorry that I didn't manage to get rid
of it when I was running Big-8 newsgroup creation.

This is just my opinion, and I neither work on nor want to work on this
stuff any more. I did my time and other people can make all the decisions
now with my blessing. But making a new newsgroup is only marginally more
of a big deal than running mkdir on a few computers.

> If Big-8 keeps breaking the trust of these news admins with this idea of
> "more welcoming", we're going to see fragmentation and therefore an even
> worse Usenet than it already is because plenty of news admins decide
> that nothing coming out of the board is making sense anymore.

I can say with about 95% certainty that almost none of those admins care.
News admins really do not care about this kind of thing nearly as much as
people think they do. And when they do care, it's often about really
random and unpredictable things that were never part of the discussion.

I think you're worried about some sort of slippery slope problem, but I've
been watching this discussion about this group slowly creep along for...
a really long time now. Maybe there's some slippery slope towards
excessive newsgroup creation, but you couldn't see it from this process
with a telescope.

>> By this rule, we would never have created any Big Eight newsgroups at
>> all. alt.* predates all of the groups we're discussing on this
>> newsgroup and, due to the nature of alt.* newsgroup creation, there was
>> essentially always already an alt.* group.

> I mean an alt.* group already exists *with discussion* on it. Sorry for
> not making it clear. If an alt.* exists and people are talking there,
> there isn't really any point to making a Big-8 to it in the future.

Once again, this is a very old argument that has been roundly rejected
throughout the entire history of the Big-8. Maybe you think we've always
been wrong to reject that argument! But we've created groups that
duplicate trafficed alt.* groups for the entire time I've been involved in
Usenet newsgroup creation. A few people would always complain about it,
but it was never considered a reason not to proceed.

[Marco Moock]

Am 27.09.2023 schrieb Julien ÉLIE <iul...@nom-de-mon-site.com.invalid>:

> We decided in fr.* to be more agile and flexible about newsgroup
> creations (and removals). Let's give it a real try!

dana did the same for de.*. I can see no disadvantage.

[Spiros Bousbouras]

On Wed, 27 Sep 2023 22:01:51 CST
Russ Allbery <ea...@eyrie.org> wrote:

> Mima-sama <m...@masa.ma> writes:
> I think you're worried about some sort of slippery slope problem, but I've
> been watching this discussion about this group slowly creep along for...
> a really long time now.

It started with
Date: Wed, 30 Mar 2022 07:46:29 CST
From: rek2 hispagatos <re...@hispagatos.org.invalid>
Newsgroups: news.groups.proposals
Subject: News group for the GO programming language pls
Message-ID: <t20mgj$tjt$1...@dont-email.me>

or http://al.howardknight.net/?ID=165913333800 .

> Maybe there's some slippery slope towards
> excessive newsgroup creation, but you couldn't see it from this process
> with a telescope.

I'm worried too so , if a comp.lang.go gets created , I wouldn't say
just go ahead and create other "reasonable" groups but leave it several
months to see what happens with comp.lang.go .

> > I mean an alt.* group already exists *with discussion* on it. Sorry for
> > not making it clear. If an alt.* exists and people are talking there,
> > there isn't really any point to making a Big-8 to it in the future.
>
> Once again, this is a very old argument that has been roundly rejected
> throughout the entire history of the Big-8. Maybe you think we've always
> been wrong to reject that argument! But we've created groups that
> duplicate trafficed alt.* groups for the entire time I've been involved in
> Usenet newsgroup creation. A few people would always complain about it,
> but it was never considered a reason not to proceed.

What's the motivation for duplicating under big-8 an alt.* group with on
topic discussion ?

[Theo]

Russ Allbery <ea...@eyrie.org> wrote:
> On the other hand, despite being interested in programming languages and
> even new programming languages, comp.lang.misc has never seemed like a
> reasonable newsgroup to follow in my time on Usenet. The scope is just
> way too broad for me and turns into a weird grab bag of miscellaneous
> stuff, 95% of which is irrelevant to anything I'm interested in. It's the
> same reason why I read rec.arts.sf.written back in the day but have zero
> interest in reading rec.arts.books, whose topic scope is in theory the
> entirety of written human knowledge and entertainment.
>
> This is all in response to your comment about what Usenet was like when it
> was thriving. It's a way different place now, so I don't know if that
> older experience applies. But I would not immediately reject the theory
> that a more specific group could get traffic that is never going to show
> up on a catch-all group. People's tolerance for searching through random
> stuff they're not interested in to find a few posts they are interested in
> varies widely.

For a specific example, comp.sys.raspberry-pi is moderately busy and doing
fine. Before it was created you might argue that the majority was on topic
for comp.os.linux.* - but the discussion wasn't there. Having a specific
group acted as a nucleus of the discussion. What happened is people have
subscribed who were previously subscribed to other overlapping groups (eg
uk.d-i-y, uk.comp.os.linux are a couple I read where there's a crossover of
people), but they weren't discussing Raspberry Pi stuff in those because the
pool of contributors was too weak to have meaningful discussions.

In essence, you need a certain concentration of interest to have a
discussion: you need some people able to answer the questions, and only once
there is that concentration will others reckon it's worth their time asking
them.

With Usenet in its current state many groups are too thin of people to be
viable, especially for 'misc' groups about disparate topics which dilutes
the answerer-pool even more. Having focused 'concentrated' groups is one
way to make them viable.

Theo

[Spiros Bousbouras]

On Wed, 27 Sep 2023 12:16:58 CST
Russ Allbery <ea...@eyrie.org> wrote:
> Mima-sama <m...@masa.ma> writes:
>
> > And I disagree with that idea. If that's true then how come Usenet has
> > thrived for many years before with users seemingly not minding to post
> > on broader newsgroups fully knowing that a specific newsgroup is not
> > ready yet?
>
> Sometimes it works and sometimes it doesn't. The amount of crossover of
> interest is a major factor.

[...]

> On the other hand, despite being interested in programming languages and
> even new programming languages, comp.lang.misc has never seemed like a
> reasonable newsgroup to follow in my time on Usenet. The scope is just
> way too broad for me and turns into a weird grab bag of miscellaneous
> stuff, 95% of which is irrelevant to anything I'm interested in.

I have been reading comp.lang.misc on and off. The only threads I see
are about design of new computer languages or how certain design features
ought to work. I've never seen it operate as a place to ask question about
preexisting language X when a comp.lang.X does not already exist. The
only exception is a couple of dicussions on Go whose motivation was at least
partially to bolster the case for a comp.lang.go group.

> This is all in response to your comment about what Usenet was like when it
> was thriving. It's a way different place now, so I don't know if that
> older experience applies. But I would not immediately reject the theory
> that a more specific group could get traffic that is never going to show
> up on a catch-all group. People's tolerance for searching through random
> stuff they're not interested in to find a few posts they are interested in
> varies widely.
>
> > It's people's laziness in posting something on-topic and making the
> > effort to invite their friends to Usenet that leads to lack of
> > discussion.
>
> I'm not sure it's a very useful statement to make that some method of
> newsgroup organization would work if people weren't lazy. Posting to and
> reading Usenet isn't a job. No one is under any obligation to do work.
> Of course people are lazy about their casual entertainment!

Actually on technical newsgroups people treat posting and reading very
seriously and clearly some posts have put a lot of work in them. But clearly
this is work people find intellectually satisfying and take pride in and
also do it because they know that other knowledgeable people read the group.

But that's a completely thing to just go through newgroups one is not
familiar with hoping to get an answer to a specific question. There is no
satisfaction in that and one can only hope to get lucky. Not wanting to do
that is not laziness , it is just preserving one's time for more important
stuff.

> I'm not going to do a bunch of work to find topics that are interesting to
> me on Usenet, let alone try to start that discussion myself. I'm going to
> try a little bit, in proportion to how much I think I'm going to get out
> of it, and then I'm going to go use some other medium entirely that
> requires less effort on my part.
>
> If Usenet wants to stick around in a world in which there are way
> easier-to-use and less-obscure discussion forums, it's going to need to be
> at least a little welcoming. More focused topic groups may or may not be
> a way to be more welcoming; I don't know! But it's at least worth
> considering.

I wouldn't use the word "welcoming" because it's too "warm fuzzy feelings"
for my taste. I would say "practical".

[Spiros Bousbouras]

On Wed, 27 Sep 2023 10:56:14 CST
Mima-sama <m...@masa.ma> wrote:
> On 9/27/2023 23:22, Spiros Bousbouras wrote:
> It's people's laziness in posting something on-topic and making the
> effort to invite their friends to Usenet that leads to lack of
> discussion. The "lack of a specific group" is just an excuse as proven
> by the many unused specific newsgroups in Big-8 waiting to be posted on.

I don't think it's laziness , at least not for technical groups. As I've said
in <duPfEzRa...@bongo-ra.co> , (some) people put a lot of work in
(some of) their posts. What I think happens with many groups is that only
experts remain reading them so they don't have any questions to ask or , if
they do , they can find the answers on their own. I'm sure this is the case
with comp.programming for example because , on the rare occasion that
anything approaching topical gets asked , it will receive many good replies.

This scenario I consider most likely with comp.lang.go too (if it gets
created) namely that the only people regularly visiting the group will be
either those who are already too knowledgeable to ask questions or those who
only have a casual interest in the language like myself. But it's a plausible
scenario that the occasional relative beginner in the language will turn up
and ask questions.

Regarding inviting one's friends , perhaps they don't have friends who do
online discussions , I certainly don't. I do have on my CV under "Hobies/
Interests" that I post on usenet.

> > Personally I think that the idea "Build it and perhaps they will come" should
> > be tested every now and again but not too often.
>
> It has already been tested many times in the past, as ahk said
> news.groups (<ueomm5$18gkb$1...@dont-email.me>). They failed.

I don't see in <ueomm5$18gkb$1...@dont-email.me> any comment on the subject.

> Now you might say "but what about comp.infosystems.gemini". But I argue
> it's an exception to the rule. (And it seems it couldn't keep up with
> the 10 posts/day guideline over 90 days anyway which is used to gauge
> whether a topic's discussion is self-sufficient enough to be split off
> into its own newsgroup.)

10 posts/day is totally unrealistic , I doubt that any comp.lang* group
achieves that on the average. Perhaps the number was realistic a long time
in the past when usenet was a lot more popular but not anymore.

> > Earlier posts mentioned the
> > possibility of also creating a comp.lang.rust group soon. I would be
> > opposed to that now because I haven't seen discussion of Rust on various
> > technical newsgroups.
>
> I'm opposed to it too, but for the different reason that is that there
> exists already one at alt.comp.lang.rust. There's no point in creating a
> newsgroup in Big-8 for a topic whose home already exists in alt.*, as
> pointed out by ahk again in the same article I mentioned by ID.

Ok , in that case I agree with you. By the way , <ueomm5$18gkb$1...@dont-email.me>
does not mention alt.comp.lang.rust .

> > The output of any command line programme written in any language can be piped
> > to a shell script and any programme can be called from a shell script.
> > comp.unix.shell is not for discussion of every programming language under
> > the sun.
>
> I'm not saying every CLI program written in Go should go there. It's
> just an idea that can be a valid option to take in some cases if you use
> a bit of creativity. For example, what if there's a Go program a user is
> dependent on but is badly designed that it's errors are outputting to
> stdout instead of stderr for some reason? User has little knowledge
> about Go but is comfortable with modifying the source code with help
> from someone more experienced in Golang and shell. comp.unix.shell would
> be perfectly fine to (cross)post to for that scenario.

If the question is about modifying Go source code then it doesn't belong on
comp.unix.shell ; if it's about doing redirection from the shell then the
language the programme is written in is irrelevant. So I can't imagine a
realistic scenario where a question about Go would be appropriate for
comp.unix.shell .Whether any of the regulars on that group is knowledgeable
about Go I don't know.

I note also that with enough "creativity" anything's possible. For example ,
many programmers seem to be interested in SF so one can mark their post as OT
and ask on rec.arts.sf.written .But one is more likely to get useful answers
by the most knowledgeable people if one comes across as moderately clueful
and having done a basic amount of preliminary search.

> Or what about if a user is looking for a CLI HTML parser written
> specifically in Go, so they could use it with their shell script? Again,
> perfectly fine to (cross)post to comp.unix.shell.

No. If one is asking for code in Go , doing whatever then comp.unix.shell is
not an appropriate place and is extremely unlikely to get a useful reply.

> > Which relevant newsgroups ?
>
> Idk, do a keyword search of "program" in your client? There's no way you
> can't find a single one that can be related to programming in Go. You're
> developing a game written in Go? Post to alt.games.programming! Need
> help porting a Golang program to Plan 9? Crosspost to comp.os.plan9!

In my experience , questions on comp.lang* groups tend to be like this :

- One is asking about an ambiguity in the relevant standard (if the language
has one).

- One has a piece of code which compiles with error(s) or does not produce
correct results and they're asking for help how to fix it. With such
questions the suggestion is always "post a minimal working example which
reproduces the error". With such a minimal working example , the wider code
in which the problematic code is embedded is irrelevant. It doesn't matter
if the wider programme is about a computer game or numerical simulation or
accounting software. But I'm not really familiar with
alt.games.programming . Perhaps people there welcome any question about any
language as long as the code has something to do with a computer game.

- How do I achieve task X in language Y ? What books or tutorials are there ?
I need help to solve some homework exercise.

None of the above questions belong to alt.games.programming or
comp.os.plan9 .For all I know , if one is lucky , they might get useful
replies in any of the above groups but this is under the general idea that
anything's possible. It is possible they would get a useful reply on
comp.lang.misc .But it's not reasonable to ask people to post to various
vaguely relevant (to Go) groups hoping they would get lucky and it's not
laziness if one won't bother to do so but instead seek an online discussion
medium other than usenet.

--
vlaho.ninja/prog

[Russ Allbery]

Spiros Bousbouras <spi...@gmail.com> writes:

> What's the motivation for duplicating under big-8 an alt.* group with on
> topic discussion ?

The general belief that Big Eight groups are more widely propagated, since
a lot of sites won't carry alt.* or at least won't carry new alt.* groups.

This may be wrong! As I said, I'm out of the newsgroup creation game, and
am not trying to defend any of these arguments. But that's been the
approach for many years. It used to even be semi-formal: people would try
out their idea in an alt.* group and then create a Big Eight group if it
caught on.