RFD: comp.lang.haskell
Jeffrey M. Vinocur
unmoderated group comp.lang.haskell
This is a formal Request For Discussion (RFD) for the creation of the
unmoderated Usenet newsgroup, comp.lang.haskell.
NEWSGROUPS LINE: comp.lang.haskell
comp.lang.haskell Haskell concepts, implementation, and use.
RATIONALE: comp.lang.haskell
Discussion of Haskell is currently fragmented through Usenet. It is the
only major functional programming language without a dedicated Big 8
group, and thus has become the topic of a substantial portion of the
traffic in comp.lang.functional. The main purposes of this group,
therefore, are to separate out this traffic into a dedicated group, and
to make it easier for new users to locate Haskell-related discussion.
This group has been the subject of periodic discussion for several
years, most notably in 2001 when a straw poll showed general support but
not enough interest, numerically, to warrant proposal under the old Big
8 group creation system (see <9j6ufp$fil$3...@marduk.litech.org> and the
accompanying thread for details).
TRAFFIC ANALYSIS:
This analysis is limited by the Google Groups interface. However, in
the past year, at least 280 distinct threads (a fifth of the total
number of threads for that time) in comp.lang.functional include
discussion of Haskell. The number of posts per thread is unknown, as is
the number of relevant posts that do not contain the word "haskell" (a
not infrequent occurrence, as the distinct syntax is enough to make
Haskell code easily recognizable without specific identification).
There is additionally some traffic scattered through other parts
of the comp.lang.* hierarchy as well as in alt.comp.lang.haskell.
It is unknown what fraction of the active mailing list community
responsible for generating the traffic in fa.haskell would be
interested in using a Big 8 newsgroups instead or in addition.
CHARTER:
The comp.lang.haskell unmoderated newsgroup is for the discussion of the
programming language Haskell, including but not limited to:
* conceptual and mathematical foundations
* interpreter and compiler development
* libraries
* learning Haskell
* programming in Haskell
* implementation-specific issues
* comparison to other programming languages (within reason)
Cryptographic signatures and small source code snippets are permitted,
but spam and other binary posts are prohibited.
Students seeking help on school work should be handled on a case-by-case
basis. In general, thoughtful questions by students showing a true
desire to learn may be appropriate for discussion, but blatant attempts
to have newsgroup participants complete assignments are likely to be
ignored at best, and possibly reported to the relevant institution.
PROCEDURE:
For more information on the newsgroup creation process, please see:
http://www.big-8.org/dokuwiki/doku.php?id=policies:creation
Those who wish to influence the development of this RFD and its final
resolution should subscribe to news.groups and participate in the
relevant threads in that newsgroup. This is both a courtesy to groups
in which discussion of creating a new group is off-topic as well as the
best method of making sure that one's comments or criticisms are heard.
All discussion of active proposals should be posted to news.groups.
To this end, the followup header of this RFD has been set to
news.groups.
If desired by the readership of closely affected groups, the discussion
may be crossposted to those groups, but care must be taken to ensure
that all discussion appears in news.groups as well.
If you would like to read or post in the proposed newsgroup, please make
a comment to that effect in this thread; the proponent will keep a list
of such positive posts with the relevant message ID (e.g., Barney Fife,
<4JGdnb60fsMzHA7Z...@sysmatrix.net>). This positive
feedback for the proposal may constitute good evidence that the group
will be well-used if it is created.
DISTRIBUTION:
This document has been posted to the following newsgroups:
news.announce.newgroups
news.groups
comp.lang.functional
alt.comp.lang.haskell
The proponent will also post pointers to:
http://www.haskell.org/haskellwiki/Mailing_lists
PROPONENT:
Jeffrey M. Vinocur <je...@litech.org>
CHANGE HISTORY:
2006-10-16 1st RFD
Martin X. Moleski, SJ
> REQUEST FOR DISCUSSION (RFD)
> unmoderated group comp.lang.haskell
This looks like a no-brainer RFD to me.
I think this is the kind of newsgroup Usenet was
originally created for. (Don't everybody flame
me all at once, OK?)
If I had my druthers, we'd just create it without
a whole lot of discussion.
The only value I see of discussing it is that it
might help you to promote the group--something
that I'm sure you're as interested in as I am.
A newsgroup with no news is no fun.
Then again, I don't expect high-volume traffic
in technical groups. A few excellent questions
and answers every year might build up a repository
of information that haskell programmers will
find useful.
Marty
--
Member of the Big-8 Management Board (B8MB).
See http://www.big-8.org for more information.
Brian Mailman
> Discussion of Haskell is currently fragmented through Usenet. It is
> the only major functional programming language without a dedicated
> Big 8 group, and thus has become the topic of a substantial portion
> of the traffic in comp.lang.functional. The main purposes of this
> group, therefore, are to separate out this traffic into a dedicated
^consolidate
> group, and to make it easier for new users to locate Haskell-related
^facilitate finding Haskell-related discussion.
> ... but blatant attempts to have newsgroup participants complete
> assignments are likely to be ignored at best, and possibly reported
> to the relevant institution.
The former is fine, but the latter seems like overkill, especially in an
unmoderated newsgroup, since it can't be guaranted such reportage will
occur (and such an attempt won't be ignored, it'll be commented on).
B/
Mark T.B. Carroll
(snip)
> If you would like to read or post in the proposed newsgroup, please make
> a comment to that effect in this thread; the proponent will keep a list
> of such positive posts with the relevant message ID (e.g., Barney Fife,
> <4JGdnb60fsMzHA7Z...@sysmatrix.net>).
I would like to read and post in comp.lang.haskell. I have hoped for
such a group for quite some time.
-- Mark
Aatu Koskensilta
> Discussion of Haskell is currently fragmented through Usenet. It is the
> only major functional programming language without a dedicated Big 8
> group, and thus has become the topic of a substantial portion of the
> traffic in comp.lang.functional.
It seems most serious discussion about Haskell takes place in
comp.lang.functional. The discussion thus does not seem to be
"fragmented through Usenet". comp.lang.functional is not a very
high-volume group, so why the need for a specific Haskell group? I'm not
saying that I'm opposed to such a group - I certainly would read it! - I
just don't see why Haskell can't be discussed in comp.lang.functional as
it has to this day. Is there some evidence that newcomers have hard time
locating Haskell discussion in USENET? (Most probably use the mailing
list in any case?).
Anyhow, if a group dedicated to Haskell is created, I'll be one of the
first subscribers.
--
Aatu Koskensilta (aatu.kos...@xortec.fi)
"Wovon man nicht sprechen kann, daruber muss man schweigen"
- Ludwig Wittgenstein, Tractatus Logico-Philosophicus
2Rowdy
by the entity known as Aatu Koskensilta, that requests spam to be sent
to <aatu.kos...@xortec.fi> and I became inspired,
> Anyhow, if a group dedicated to Haskell is created, I'll be one of
> the first subscribers.
There is an empty alt.* version of the proposed newsgroup.
--
d:J0han; Certifiable me
http://www.aacity.net Citroen Newsgroup
Click -[ HERE ]- to continue.....
Aatu Koskensilta
> I was reading <news:e2WYg.7662$wn7....@reader1.news.jippii.net>, made
> by the entity known as Aatu Koskensilta, that requests spam to be sent
> to <aatu.kos...@xortec.fi> and I became inspired,
>
>> Anyhow, if a group dedicated to Haskell is created, I'll be one of
>> the first subscribers.
>
> There is an empty alt.* version of the proposed newsgroup.
So it appears. Great! I can subscribe to that, then.
Perhaps the good people of the Haskell mailing lists should be consulted
on whether they have any intention of moving to USENET? If not, I'm more
and more inclined to think comp.lang.functional suffices just fine.
Jeffrey M. Vinocur
Brian Mailman <bmai...@sfo.invalid> wrote:
>Jeffrey M. Vinocur wrote:
>
>> [...]
> ^consolidate
> ^facilitate finding Haskell-related discussion.
Both excellent, thanks.
>> ... but blatant attempts to have newsgroup participants complete
>> assignments are likely to be ignored at best, and possibly reported
>> to the relevant institution.
>
>The former is fine, but the latter seems like overkill, especially in an
> unmoderated newsgroup, since it can't be guaranted such reportage will
>occur (and such an attempt won't be ignored, it'll be commented on).
I was trying to simply report current practice in the comp.lang.*
groups that I read, which seems to be that sometimes the posts
are completely ignored, sometimes everyone makes fun of the user,
sometimes they try to gently guide the user in the right
direction, and sometimes they track down what university it's
coming from and report or threaten to email the professor.
I tried to use descriptive rather than prescriptive language in
the paragraph you quoted, because of course you're right, we have
no guarantee about what will happen.
Other people want to chime in here? Should I be briefer?
--
Jeffrey M. Vinocur
je...@litech.org
Jeffrey M. Vinocur
Aatu Koskensilta <aatu.kos...@xortec.fi> wrote:
>
>Perhaps the good people of the Haskell mailing lists should be consulted
>on whether they have any intention of moving to USENET? If not, I'm more
>and more inclined to think comp.lang.functional suffices just fine.
Well, I think you're right that we're on the border in terms of
traffic to justify a separate group vs just leaving it in
comp.lang.functional. And under the old system, where strict
traffic criteria and expected readership numbers were required,
we never would have gotten this group created. But it has come
up for discussion multiple times in the past, which suggests that
at least some people are interested.
And we truly have no idea how many people try to find a place for
Haskell discussion and get lost because there is no eponymous
group.
By no means do I think this group is critically necessary and
likely to revolutionize the online Haskell community. But I
think that, under the new Big 8 system, there is a very low cost
to creating this group, and so if it has the potential for a
small-to-moderate benefit to some users, and no real harm, then
we should go for it.
2Rowdy
known as Jeffrey M. Vinocur, that requests spam to be sent to
<je...@litech.org> and I became inspired,
The paragraph about students is not enforceable and I think it
shouldn't be in a Charter. If anywhere perhaps in a FAQ (though I
don't like FAQ for newsgroups).
Suppose the newsgroup becomes a success and in a few years from now
another type of misuse is noticed, are you going to request a change
in Charter to address that type of misuse?
Posting behavior can not be caught in a Charter, abuse can.
--
d:J0han; Certifiable me
http://www.aacity.net Citroen Newsgroup
I drank of the purifying Nirang, the sterile urine of the spotless white bull of which there is but one in ten thousand. I am the wearer of The Secret Girdle that once belonged to DE.
2Rowdy
<je...@litech.org> and I became inspired,
> By no means do I think this group is critically necessary and
> likely to revolutionize the online Haskell community. But I
> think that, under the new Big 8 system, there is a very low cost
> to creating this group, and so if it has the potential for a
> small-to-moderate benefit to some users, and no real harm, then
> we should go for it.
Though I don't favor copying newsgroups from alt to beight in this
case there is a rationale for it.
Some academic servers don't add alt groups (shame on them) and the
subject probably has interest from the academic world. The failure in
alt doesn't guarantee failure in beight, it hints towards it.
--
d:J0han; Certifiable me
http://www.aacity.net Citroen Newsgroup
This posting is optimised for reading with Microsoft Outlook Express
Aatu Koskensilta
> By no means do I think this group is critically necessary and
> likely to revolutionize the online Haskell community. But I
> think that, under the new Big 8 system, there is a very low cost
> to creating this group, and so if it has the potential for a
> small-to-moderate benefit to some users, and no real harm, then
> we should go for it.
If I have understood correctly, now that there is a removal procedure -
or at least an rmgrouping procedure - in place the board feels that
creating a group that fails to eventually attract sufficient amount of
traffic to count as "live" is not such a horrible tragedy. So yes, they
might very well choose to create the Haskell group, and I'm not opposed
to that, just mildly skeptical about the success of the group.
Do you think the results of the straw poll taken during the previous RFD
are still reasonably accurate? If not, perhaps you could just ask people
in e.g. haskell-cafe what they think of the idea?
Another thought occurred to me. As a reader of comp.lang.functional I'm
not sure I *want* to see the Haskell traffic to go away. Now that I
think of it, I like comp.lang.functional the way it is, with its happy
motley of languages, happily living in midsts of more abstract
discussions...
Steve Bonine
> Another thought occurred to me. As a reader of comp.lang.functional I'm
> not sure I *want* to see the Haskell traffic to go away. Now that I
> think of it, I like comp.lang.functional the way it is, with its happy
> motley of languages, happily living in midsts of more abstract
> discussions...
Nothing is ever simple. When I saw this proposal, my first impression
was that there was no reason not to do it. Now I see two potential
downsides . . . the harm to comp.lang.fuctional that Aatu mentions, and
the possibility that fragmenting the existing Haskell discussion on the
mailing list will result in two discussions, neither of which is as
useful as the current one already is. Unfortunately, these same issues
exist with the creation of any new newsgroup.
In this specific case, it would be nice to know how robust the discusion
is on the current mailing list, and how many readers of
comp.lang.functional share Aatu's feeling that having the discussion
melding with others is a useful consequence.
Martin X. Moleski, SJ
...
>>> ... but blatant attempts to have newsgroup participants complete
>>> assignments are likely to be ignored at best, and possibly reported
>>> to the relevant institution.
> ... I tried to use descriptive rather than prescriptive language in
>the paragraph you quoted, because of course you're right, we have
>no guarantee about what will happen.
>Other people want to chime in here? Should I be briefer?
Put the warning in a FAQ.
Q: Will you help me with my homework?
A: No.
Q: Why not?
A: ...
Martin X. Moleski, SJ
> ... I
>think that, under the new Big 8 system, there is a very low cost
>to creating this group, and so if it has the potential for a
>small-to-moderate benefit to some users, and no real harm, then
>we should go for it.
AOL.
Wayne Brown
>
> I was trying to simply report current practice in the comp.lang.*
> groups that I read, which seems to be that sometimes the posts
> are completely ignored, sometimes everyone makes fun of the user,
> sometimes they try to gently guide the user in the right
> direction, and sometimes they track down what university it's
> coming from and report or threaten to email the professor.
Those sound just like the reactions that occur in sci.math when a student
asks for answers to homework problems without showing evidence of having
attempted to solve them first.
--
Wayne Brown <fwb...@bellsouth.net> (HPCC #1104)
Þæs ofereode, ðisses swa mæg. ("That passed away, this also can.")
"Deor," from the Exeter Book (folios 100r-100v)
Mark T.B. Carroll
(snip)
> Perhaps the good people of the Haskell mailing lists should be consulted
> on whether they have any intention of moving to USENET?
I should have mentioned as someone hoping for comp.lang.haskell that I'm
subscribed to a few of the Haskell mailing lists.
> If not, I'm more and more inclined to think comp.lang.functional
> suffices just fine.
Partly why I'd be happy to see a particularly-Haskell group is that some
days I just don't care how OCaml or whatever does things, and then we
can perhaps escape most of the interminable functional-vs-OO threads and
whatever. comp.lang.functional gets a fair bit of "which language? which
paradigm?" discussion that's not so personally relevant once you've made
your choice and are now trying to live with it.
Not that the current situation isn't tolerable, mind, given decent
newsreading software.
-- Mark
nobr...@gmail.com
>
> AOL.
LOL.
Just one side note - if this proposal does go forward (and I can't
support it because I think it will cause *more* fragmentation, not
less) and if it's decided to gateway with the haskell mailing list(s),
*do not use current versions of Mailman* for that. Mailman has a bug
in that it changes the Message-Id when it injects into news articles
sent originally to the mailing list. This ruins the reading experience
on *both* the list and the group for threaded readers.
Martin X. Moleski, SJ
> ... Just one side note - if this proposal does go forward (and I can't
This question is above my pay grade: Any chance of getting the bug fixed?
Is it a simple problem or hard?
In other words, is it a feature deliberately chosen by
Mailman that the authors won't remove?
Jeffrey M. Vinocur
Martin X. Moleski, SJ <mol...@canisius.edu> wrote:
>On 17 Oct 2006 08:32:12 -0700, nobr...@gmail.com wrote in
><1161099132.1...@h48g2000cwc.googlegroups.com>:
>
>>*do not use current versions of Mailman* for that. Mailman has a bug
>>in that it changes the Message-Id when it injects into news articles
>>sent originally to the mailing list. This ruins the reading experience
>>on *both* the list and the group for threaded readers.
>
>This question is above my pay grade: Any chance of getting the bug fixed?
>
>Is it a simple problem or hard?
It turns out that gatewaying between mail and news is fairly
tricky, because the standards are very different. News has
strict limitations on Message-ID format, and relies on those
Message-IDs in the References header for threading. So any naive
approach (e.g. simply passing articles in both directions and not
doing anything much about Message-IDs) is doomed to have much
breakage. I'm not sure what the state-of-the-art is in solutions
to this problem, but it's not trivial.
On the other hand, the haskell-cafe mailing list is apparently
being gatewayed into fa.haskell with some degree of success; I
don't know if this is a bidirectional gateway.
Jeffrey M. Vinocur
2Rowdy <Harry...@gmail.com> wrote:
>
>The paragraph about students is not enforceable and I think it
>shouldn't be in a Charter.
Well, in an unmoderated group, nothing is enforceable.
All we can really do is set out the intent of the newsgroup, and
I think the intent is that it not be a "homework help" forum, but
that of course well-meaning questions from students putting in a
good effort will be treated reasonably.
Jeffrey M. Vinocur
Aatu Koskensilta <aatu.kos...@xortec.fi> wrote:
>
>Do you think the results of the straw poll taken during the previous RFD
>are still reasonably accurate? If not, perhaps you could just ask people
>in e.g. haskell-cafe what they think of the idea?
I have no reason to suspect anything drastic has changed; I think
the readers of comp.lang.* would be content to consolidate their
Haskell-related traffic into a separate group, we might attract a
small amount of new activity from the mailing list, and we
probably would make it easier for new posters to find the group
(we know that some do find comp.lang.functional since they ask
Haskell questions there, but we have no idea how many give up
when the can't find a specific Haskell group).
I do think it's a good idea to contact the mailing lists.
>Another thought occurred to me. As a reader of comp.lang.functional I'm
>not sure I *want* to see the Haskell traffic to go away. Now that I
>think of it, I like comp.lang.functional the way it is, with its happy
>motley of languages, happily living in midsts of more abstract
>discussions...
Personally I'm unswayed by this argument. The current state of
comp.lang.functional is purely an artifact of which languages
were invented before Usenet declined enough to make new group
creation difficult. So lisp and scheme and ML have their own
groups, but Haskell doesn't. Do you see any reason that Haskell
should be distinguished in this way?
Jeffrey M. Vinocur
Steve Bonine <s...@pobox.com> wrote:
>
>In this specific case, it would be nice to know how robust the discusion
>is on the current mailing list,
I'm not a subscriber, but from outward inspection, I suspect it's
extremely robust.
>and how many readers of comp.lang.functional share Aatu's
>feeling that having the discussion melding with others is a
>useful consequence.
Well, the readers of comp.lang.functional are invited by the RFD
to participate in this discussion, and we'll have to see.
I'm skeptical that this is going to upset some delicate balance
that exists in comp.lang.functional, personally. I think the
users of functional languages (Haskell and others) that are
interested in meta-discussions and such will continue to
participate in comp.lang.functional, and then people who
specifically want to start such discussions will continue to
crosspost to several related groups and have no trouble getting
participants.
Jeffrey M. Vinocur
>Aatu Koskensilta <aatu.kos...@xortec.fi> writes:
>(snip)
>> Perhaps the good people of the Haskell mailing lists should be consulted
>> on whether they have any intention of moving to USENET?
>
>I should have mentioned as someone hoping for comp.lang.haskell that I'm
>subscribed to a few of the Haskell mailing lists.
Well then, could we impose on you to briefly inform them of this
discussion, and invite them (particularly those that would
participate in a Usenet-based Haskell forum) to share their
opinions?
I should make clear here that I have no intention to create
comp.lang.haskell as a gateway to/from any Haskell mailing list,
unless perhaps there should be overwhelming demand for that from
the members of one of the mailing lists.
Martin X. Moleski, SJ
> ... I do think it's a good idea to contact the mailing lists.
Agreed.
When and if you do, you might invite them to post in this
or a subsequent thread.
Then, if you could collect the feedback (as I've been
doing in the so-called "straw poll" for the n.g.p.
RFD, it might bolster your case that this is a viable
group.
We don't have any other polling system in place
at present.
Martin X. Moleski, SJ
>>>[...] and if it's decided to gateway with the haskell mailing list(s),
>>>*do not use current versions of Mailman* for that. Mailman has a bug
>>>in that it changes the Message-Id when it injects into news articles
>>>sent originally to the mailing list. This ruins the reading experience
>>>on *both* the list and the group for threaded readers.
...
>It turns out that gatewaying between mail and news is fairly
>tricky, because the standards are very different. News has
>strict limitations on Message-ID format, and relies on those
>Message-IDs in the References header for threading. So any naive
>approach (e.g. simply passing articles in both directions and not
>doing anything much about Message-IDs) is doomed to have much
>breakage. I'm not sure what the state-of-the-art is in solutions
>to this problem, but it's not trivial.
>On the other hand, the haskell-cafe mailing list is apparently
>being gatewayed into fa.haskell with some degree of success; I
>don't know if this is a bidirectional gateway.
FWIW, there is a service that advertises free uni- and
bi-directional gatewaying of newsgroups:
http://www.newsadmin.com/bit/gatewayhosts.asp
Dunno if it works or how they do it. I just came
across the page once upon a time and Mark Dodel
put it on our NSP page for future reference.
Mark T.B. Carroll
(snip)
> Well then, could we impose on you to briefly inform them of this
> discussion, and invite them (particularly those that would
> participate in a Usenet-based Haskell forum) to share their
> opinions?
Certainly, yes. I have now posted to the main Haskell mailing list,
so let's see what happens.
> I should make clear here that I have no intention to create
> comp.lang.haskell as a gateway to/from any Haskell mailing list,
> unless perhaps there should be overwhelming demand for that from
> the members of one of the mailing lists.
Sounds good to me.
-- Mark
Tim Skirvin
[mailman breaks message-ids on Usenet posting]
> This question is above my pay grade: Any chance of getting the bug fixed?
Good question. I haven't actually pursued the mailman team with
this, but that's in large part because I don't like the installation
process and don't want to get involved in the debugging.
Russ maintains a news gateway software package (News::Gateway)
that does a better job of this. I think the time required to get mailman
fixed would be better spent on setting up a decent gateway based on Russ'
software somewhere.
- Tim Skirvin (sk...@big-8.org)
--
http://www.big-8.org/ Big-8 Management Board
http://www.killfile.org/~tskirvin/ Skirv's Homepage <FISH>< <*>
Doug Freyburger
> Jeffrey M. Vinocur wrote:
>
> > unmoderated group comp.lang.haskell
>
> This looks like a no-brainer RFD to me.
Agreed. Minimum times at each step of the process?
> I think this is the kind of newsgroup Usenet was
> originally created for. (Don't everybody flame
> me all at once, OK?)
So I waited a day until -
> If I had my druthers, we'd just create it without
> a whole lot of discussion.
Having your druthers would undermine the system being
build. Far better to go through the current version of the
process with minimum timings at each step.
If the B8BM wants to discuss a policy for slam-dunk
proposals ...
Brian Mailman
> I was reading <news:e2WYg.7662$wn7....@reader1.news.jippii.net>, made
> by the entity known as Aatu Koskensilta, that requests spam to be sent
> to <aatu.kos...@xortec.fi> and I became inspired,
>
>> Anyhow, if a group dedicated to Haskell is created, I'll be one of
>> the first subscribers.
>
> There is an empty alt.* version of the proposed newsgroup.
alt.comp.lang.haskell is not on Supernews.
B/
Brian Mailman
> In article <12j7ei4...@news.supernews.com>,
> Brian Mailman <bmai...@sfo.invalid> wrote:
>>Jeffrey M. Vinocur wrote:
>>
>>> [...]
>> ^consolidate
>> ^facilitate finding Haskell-related discussion.
>
> Both excellent, thanks.
My pleasure.
B/
Brian Mailman
> If I have understood correctly, now that there is a removal procedure -
> or at least an rmgrouping procedure
http://groupsearch.aacity.net/cgi-bin/prop.pl?action=search&group=rec.pets.cats&sorting=group
B/
Brian Mailman
Someone decided they didn't like news readers and tacked on a Mailman (a
Lyris product, no relation) interface on a group I read (the ethics of
doing so without consultation with the group is an exercise left for
another time). I don't know what the fix was, if there was any, but for
a while when someone from the mailing side replied to my messages, it
sent a duplicate--one to me and one to the group. At the time, my ISP
wasn't letting post with .invalid unless I had an valid Reply To: so I
was caught in it. Several times I received an email that I replied to
and I found out I had to spend the time and energy reposting when I read
the group.
B/
>
nobr...@gmail.com
> This question is above my pay grade: Any chance of getting the bug fixed?
>
> Is it a simple problem or hard?
>
> In other words, is it a feature deliberately chosen by
> Mailman that the authors won't remove?
>
It is filed in the sourceforge bugtracker for mailman (not by me), and
the report includes a (very hacky) fix, if I remember right.
Ok, found it, it's here:
http://sourceforge.net/tracker/index.php?func=detail&aid=1256272&group_id=103&atid=100103
But, it seems the team has switched to a new bugtracker (JIRA?) so this
report may be in a black hole.
2Rowdy
entity known as Brian Mailman, that requests spam to be sent to
<bmai...@sfo.invalid> and I became inspired,
You don't miss much. All I see is one message, this RFD.
--
d:J0han; Certifiable me
No, I am not demented,
just gray haired
Aaron Denney
There is also gmane <http://www.gmane.org/> that provides an excellent
usenet interface to many mailing lists, and a decent web front-end as
well. The main Haskell lists are among those it provides. If all you
want is a usenet interface to the same set of people it does the job.
Personally, I'd like more Haskell discussions with usenet denizens, so
I'm all in favor of the creation of comp.lang.haskell.
--
Aaron Denney
-><-
Antti-Juhani Kaijanaho
> If you would like to read or post in the proposed newsgroup, please make
> a comment to that effect in this thread; the proponent will keep a list
> of such positive posts with the relevant message ID (e.g., Barney Fife,
><4JGdnb60fsMzHA7Z...@sysmatrix.net>). This positive
> feedback for the proposal may constitute good evidence that the group
> will be well-used if it is created.
I will read the group should it be created. I will also post in it if
interesting threads appear. I doubt I will be creating threads myself :)
(I have read comp.lang.functional occasionally - but not in recent
months -, and I follow, and occasionally post in, the Haskell mailing
list.)
Martin X. Moleski, SJ
> ... There is also gmane <http://www.gmane.org/> that provides an excellent
>usenet interface to many mailing lists, and a decent web front-end as
>well. The main Haskell lists are among those it provides. If all you
>want is a usenet interface to the same set of people it does the job.
Thanks, Aaron.
I've started drafting a page on mail-to-news gateways:
http://www.big-8.org/dokuwiki/doku.php?id=faqs:mail-to-news
>Personally, I'd like more Haskell discussions with usenet denizens, so
>I'm all in favor of the creation of comp.lang.haskell.
Thanks for the feedback.
Adam H. Kerman
>Jeffrey M. Vinocur <je...@litech.org> wrote in <11610163...@isc.org>:
>> REQUEST FOR DISCUSSION (RFD)
>> unmoderated group comp.lang.haskell
>This looks like a no-brainer RFD to me.
>I think this is the kind of newsgroup Usenet was
>originally created for. (Don't everybody flame
>me all at once, OK?)
It's an alt group newgrouped nearly four years ago. Our proponent should
request that alt.comp.lang.haskell be created on his system, and post to it.
>If I had my druthers, we'd just create it without
>a whole lot of discussion.
Great. Do you have a hit list of comp groups in alt you would duplicate
without discussion?
>The only value I see of discussing it is that it
>might help you to promote the group--something
>that I'm sure you're as interested in as I am.
>
>A newsgroup with no news is no fun.
>
>Then again, I don't expect high-volume traffic
>in technical groups. A few excellent questions
>and answers every year might build up a repository
>of information that haskell programmers will
>find useful.
Martin X. Moleski, SJ
<Pine.LNX.4.63.06...@qbbshf.puvarg.pbz>:
>>If I had my druthers, we'd just create it without
>>a whole lot of discussion.
>Great. Do you have a hit list of comp groups in alt you would duplicate
>without discussion?
Nope.
I don't look at alt.
It's in its own universe.
Dave Sill
> If the B8BM wants to discuss a policy for slam-dunk
> proposals ...
We've already got one. Proponent submits the slam-dunk RFD to the
board and requests a vote without public pre-vote debate. The only
opportunity for public feedback prior to issuing the newgroup is
during the post-vote waiting period.
-Dave
Adam H. Kerman
>On Wed, 18 Oct 2006 11:33:34 -0500, "Adam H. Kerman" <a...@chinet.com> wrote:
>>>If I had my druthers, we'd just create it without
>>>a whole lot of discussion.
>>Great. Do you have a hit list of comp groups in alt you would duplicate
>>without discussion?
>Nope.
>I don't look at alt.
>It's in its own universe.
Way to sum up B8MBy's problems.
Tim Skirvin
> It's an alt group newgrouped nearly four years ago. Our proponent should
> request that alt.comp.lang.haskell be created on his system, and post to it.
alt.comp.lang.haskell is a fairly empty group (according to
Google), and not very well propagated (carried by 16 servers, according to
the http://groupsearch.aacity.net/ search engine). Based on this, I
expect that it would be far easier and more effective for the proponent to
create and publicize a new group in comp.* than it would be to try to
publicize and popularize the alt.* group.
Adam H. Kerman
>Jeffrey M. Vinocur wrote:
>>By no means do I think this group is critically necessary and
>>likely to revolutionize the online Haskell community. But I
>>think that, under the new Big 8 system, there is a very low cost
>>to creating this group, and so if it has the potential for a
>>small-to-moderate benefit to some users, and no real harm, then
>>we should go for it.
>If I have understood correctly, now that there is a removal procedure - or
>at least an rmgrouping procedure - in place the board feels that creating a
>group that fails to eventually attract sufficient amount of traffic to
>count as "live" is not such a horrible tragedy. So yes, they might very
>well choose to create the Haskell group, and I'm not opposed to that, just
>mildly skeptical about the success of the group.
>Do you think the results of the straw poll taken during the previous RFD
>are still reasonably accurate? If not, perhaps you could just ask people in
>e.g. haskell-cafe what they think of the idea?
>Another thought occurred to me. As a reader of comp.lang.functional I'm not
>sure I *want* to see the Haskell traffic to go away. Now that I think of
>it, I like comp.lang.functional the way it is, with its happy motley of
>languages, happily living in midsts of more abstract discussions...
This is exactly the reason why I've argued for years that splits should be
discussed in the affected newsgroups FIRST, and not in config newsgroups,
something the alt group proponent didn't do.
Knee-jerk "let's just newgroup it and damn the consequences!" isn't helpful.
Consideration should be given for those discussing the topic and what their
preferences are.
If the gentleman didn't happen to post in news.groups, the Board would have
been unaware of reasonable objections as it doesn't feel it's its place to
seek out information on how the topic is discussed.
Adam H. Kerman
>Aatu Koskensilta <aatu.kos...@xortec.fi> wrote:
>>Do you think the results of the straw poll taken during the previous RFD
>>are still reasonably accurate? If not, perhaps you could just ask people
>>in e.g. haskell-cafe what they think of the idea?
>I have no reason to suspect anything drastic has changed; I think
>the readers of comp.lang.* would be content to consolidate their
>Haskell-related traffic into a separate group, we might attract a
>small amount of new activity from the mailing list, and we
>probably would make it easier for new posters to find the group
>(we know that some do find comp.lang.functional since they ask
>Haskell questions there, but we have no idea how many give up
>when the can't find a specific Haskell group).
Could you ask, please?
>I do think it's a good idea to contact the mailing lists.
Don't count on their support, unless there is a widespread dissatisfaction
with their operation or unless they are already using Usenet.
>>Another thought occurred to me. As a reader of comp.lang.functional I'm
>>not sure I *want* to see the Haskell traffic to go away. Now that I
>>think of it, I like comp.lang.functional the way it is, with its happy
>>motley of languages, happily living in midsts of more abstract
>>discussions...
>Personally I'm unswayed by this argument. The current state of
>comp.lang.functional is purely an artifact of which languages
>were invented before Usenet declined enough to make new group
>creation difficult. So lisp and scheme and ML have their own
>groups, but Haskell doesn't. Do you see any reason that Haskell
>should be distinguished in this way?
Topics don't "deserve" newsgroups. It really depends on the nature of the
discussion. In this case, for whatever reason, people are using a newsgroup
for a broad topic, likely because it's well established.
And, no, numerous computer language newsgroups for newer-than-Usenet
languages have been started over the years.
Aratzio
<mol...@canisius.edu> transparently proposed:
>On Wed, 18 Oct 2006 11:33:34 -0500, "Adam H. Kerman" <a...@chinet.com> wrote in
><Pine.LNX.4.63.06...@qbbshf.puvarg.pbz>:
>
>>>If I had my druthers, we'd just create it without
>>>a whole lot of discussion.
>
>>Great. Do you have a hit list of comp groups in alt you would duplicate
>>without discussion?
>
>Nope.
>
>I don't look at alt.
>
>It's in its own universe.
>
If only the irony had been intentional.
--
Does the name Pavlov ring a bell?
Adam H. Kerman
>"Adam H. Kerman" <a...@chinet.com> writes:
>> It's an alt group newgrouped nearly four years ago. Our proponent should
>> request that alt.comp.lang.haskell be created on his system, and post to it.
> alt.comp.lang.haskell is a fairly empty group (according to
>Google), and not very well propagated (carried by 16 servers, according to
>the http://groupsearch.aacity.net/ search engine). Based on this, I
>expect that it would be far easier and more effective for the proponent to
>create and publicize a new group in comp.* than it would be to try to
>publicize and popularize the alt.* group.
Sure, if one advises the proponent to ignore the wishes of those posting on
the topic in the functional language group, you know, just about the entire
audience for the topic on Usenet. And if one advises the proponent that
there's no obligation to publicize a Big 8 group, exactly the same as
publicizing an alt group, despite what Marty thinks.
There's that "far easier" thing again. None of this is easy. There are only
people to consider and choices to be made and the difficulty of getting the
proponent to consider that those discussing the topic deserve consideration.
Jeffrey M. Vinocur
Adam H. Kerman <a...@chinet.com> wrote:
>At 4:37pm -0000, 10/17/06, Jeffrey M. Vinocur <je...@litech.org> wrote:
>
>
>Could you ask, please?
I'm not quite sure what you're looking for here. I get the
impression that there's something, besides having the RFD
crossposted to comp.lang.functional, that you want me to do to
involve the readers of comp.lang.functional in this discussion.
But I don't have a good idea of how you think I should do that,
or why you think it's necessary.
I'm also not sure where all of your hostility comes from. I
certainly do not appreciate having your emotions regarding the
B8MB directed onto me and my proposal for a pleasant and entirely
unremarkable comp.* group.
>>I do think it's a good idea to contact the mailing lists.
>
>Don't count on their support, unless there is a widespread dissatisfaction
>with their operation or unless they are already using Usenet.
I wasn't planning on it.
>Topics don't "deserve" newsgroups. It really depends on the nature of the
>discussion. In this case, for whatever reason, people are using a newsgroup
>for a broad topic, likely because it's well established.
Well sure...but some of us have been suggesting that splitting
off a newsgroup for the narrower topic would be valuable. I'm
not sure why you think the existence of discussion in a broader
group should be taken as anything but in favor of the new group.
>And, no, numerous computer language newsgroups for newer-than-Usenet
>languages have been started over the years.
Whatever point you're attempting to make here escapes me. If
you're trying to say that Haskell shouldn't have a group because
it doesn't yet and other new languages do, that strikes me as
some circular reasoning that would suggest we should never create
new groups.
Jeffrey M. Vinocur
>those discussing the topic deserve consideration.
Again, I'm not sure where this reaction comes from, or why you
think I'm considering anyone -but- the readers of the newsgroups
in question.
My goal is purely to facilitate discussion of Haskell and other
functional language by the provision of an appropriate set of
forums for such discussion to take place, and if you are sensing
any other motivation you may wish to reanalyze your perceptions.
Adam H. Kerman
>Adam H. Kerman <a...@chinet.com> wrote:
>>At 4:37pm -0000, 10/17/06, Jeffrey M. Vinocur <je...@litech.org> wrote:
>>>I have no reason to suspect anything drastic has changed [...]
>>Could you ask, please?
>I'm not quite sure what you're looking for here. I get the
>impression that there's something, besides having the RFD
>crossposted to comp.lang.functional, that you want me to do to
>involve the readers of comp.lang.functional in this discussion.
According to this thread, discussion of the topic generally takes place in
the functional language group. That makes this a split, because just about
the entire Usenet audience for the topic is in the one group. So you work
things out with those posting on the topic, and take their input on whether
they think the split is a good idea and if the proposed group is workable
and if any of them think there's sufficient discussion for a separate group.
>But I don't have a good idea of how you think I should do that,
>or why you think it's necessary.
>
>I'm also not sure where all of your hostility comes from. I
>certainly do not appreciate having your emotions regarding the
>B8MB directed onto me and my proposal for a pleasant and entirely
>unremarkable comp.* group.
I don't think you've been given good advice.
Now, you can take anyone's advice or leave it. Our opinion is irrelevant.
The only opinions that are important are those of the people currently
discussing the topic on Usenet. Work it out with them. I don't think the
details of splits can be successfully worked out anywhere but in the
affected newsgroup.
>>Topics don't "deserve" newsgroups. It really depends on the nature of the
>>discussion. In this case, for whatever reason, people are using a newsgroup
>>for a broad topic, likely because it's well established.
>Well sure...but some of us have been suggesting that splitting
>off a newsgroup for the narrower topic would be valuable. I'm
>not sure why you think the existence of discussion in a broader
>group should be taken as anything but in favor of the new group.
I'm going by the discussion in this thread. You have opposition, and you
need to work things out. It's important in this type of proposal.
>>And, no, numerous computer language newsgroups for newer-than-Usenet
>>languages have been started over the years.
>Whatever point you're attempting to make here escapes me.
You made a very odd comment that the only computer language newsgroups on
Usenet were those for languages in existence when Usenet began.
>If you're trying to say that Haskell shouldn't have a group because it
>doesn't yet and other new languages do, that strikes me as some circular
>reasoning that would suggest we should never create new groups.
Whether 200 other computer languages have Usenet newsgroups is irrelevant in
the immediate instance. It depends on the nature of the discussion.
As several people have pointed out, there is a newsgroup, though little
used. Ask that it be created on your server.
Adam H. Kerman
>Adam H. Kerman <a...@chinet.com> wrote:
>>[...] the difficulty of getting the proponent to consider that
>>those discussing the topic deserve consideration.
>Again, I'm not sure where this reaction comes from, or why you
>think I'm considering anyone -but- the readers of the newsgroups
>in question.
Ask THEM.
>My goal is purely to facilitate discussion of Haskell and other
>functional language by the provision of an appropriate set of
>forums for such discussion to take place, and if you are sensing
>any other motivation you may wish to reanalyze your perceptions.
What if, after serious consideration of the needs of those posting on the
topic, you find you don't have enough support for the split? It may not be
the answer you'll want to hear. B8MBys will likely newgroup it anyway, if
you choose to proceed.
What if a lot of people like discussing the topic where it is currently
discussed? Are they not already facilitated?
Martin X. Moleski, SJ
<Pine.LNX.4.63.06...@qbbshf.puvarg.pbz>:
> ... Sure, if one advises the proponent to ignore the wishes of those posting on
>the topic in the functional language group, you know, just about the entire
>audience for the topic on Usenet. And if one advises the proponent that
>there's no obligation to publicize a Big 8 group, exactly the same as
>publicizing an alt group, despite what Marty thinks. ...
What I think is:
1. Some members of a generic group always want to keep
the topic area in the generic group. I don't give a
lot of weight to such complaints. If they like the
new topic, they can subscribe to both newsgroups.
If they just want to hold the folks interested in
the topic hostage to pump up the stats for the
generic group, then I'd like to see the captives
set free.
2. Some members of a generic group want to see a
more focused newsgroup (hence, the RFD in this
case).
3. A newsgroup with no news is no fun. Therefore, it
behooves the proponent to adverise the group as
widely as possible during the RFD process and,
if the group is created, afterward as well.
4. A proponent ought to do what can be done to
show support for the RFD by inviting people to
post in the related thread in news.groups. At
present, we have no other method of polling.
5. As in alt, some NSPs only add newsgroups
requested by their customers. The proponents
should do what they can to assist in the
process of increasing propagation.
Steve Bonine
> My goal is purely to facilitate discussion of Haskell and other
> functional language by the provision of an appropriate set of
> forums for such discussion to take place, and if you are sensing
> any other motivation you may wish to reanalyze your perceptions.
I think that what some people are saying is that more is not always
better when it comes to "an appropriate set of forums."
What you're proposing is to split some discussion of Haskell out of
comp.lang.functional. So just for the sake of discussion, let's say
that there are 100 people who participate in comp.lang.functional. How
many of these will the new newsgroup attract?
Let's say it attracts 10. Now, instead of 100 potential contributors to
a thread, you have 10. In many cases, those are the same 10 who would
have been interested if the thread had been in comp.lang.functional, but
sometimes people who are interested in other languages might deign to
post to a Haskell question.
Let's say it attracts 90. Now what happens to the discussion in
comp.lang.functional?
These are worst-case scenarios and I don't think are likely in the real
world. A more realistic outcome would be somewhere in between with a
lot of people reading both groups and would indeed achieve your goal "to
facilitate discussion of Haskell." I am just trying to point out that
there are possible negative ramifications to creating a new group.
Here's my specific advice to satisfy the naysayers here in news.groups.
Even though the RFD has been posted in comp.lang.functional, post a
short article there and try to start a thread in the newsgroup that
gauges how people feel about the new group. People are more willing to
express their opinion when the followup is not set to news.groups. When
you get a few opinions from the membership of the group you can bring
that information back to news.groups.
The cost of this exercise is small, and you might garner some
interesting data. If no one follows up to your article, that in itself
is an interesting fact.
Thank you for taking time to sponsor this proposal. The environment in
news.groups can seem hostile at times. For what it's worth, I do not
include myself in the "naysayer" category mentiioned above, but I do
think that it's appropriate to make at least some effort to discover the
opinion of current readers. Posting the RFD in the affected group
really should be enough, but it's not unreasonable to take one more step.
Aatu Koskensilta
> I'm skeptical that this is going to upset some delicate balance
> that exists in comp.lang.functional, personally. I think the
> users of functional languages (Haskell and others) that are
> interested in meta-discussions and such will continue to
> participate in comp.lang.functional, and then people who
> specifically want to start such discussions will continue to
> crosspost to several related groups and have no trouble getting
> participants.
Sure. As said, I am not opposed to the creation of a Haskell group, and
wished to just offer some random thoughts that occurred to me.
Personally, I come from a proof theory and type theory background, and
am more interested in general discussion than in the specifics of this
or that language, even though I do dabble in Haskell and other
functional languages now and then. I suspect the expected userbase of
the proposed group to consist mainly of actual Haskell users rather than
theory-oriented people like myself - though of course there is
considerable overlap between those two groups!
--
Aatu Koskensilta (aatu.kos...@xortec.fi)
"Wovon man nicht sprechen kann, daruber muss man schweigen"
- Ludwig Wittgenstein, Tractatus Logico-Philosophicus
Jim Logajan
> unmoderated group comp.lang.haskell
> DISTRIBUTION:
>
> This document has been posted to the following newsgroups:
>
> news.announce.newgroups
> news.groups
> comp.lang.functional
> alt.comp.lang.haskell
Just FYI: The group alt.comp.lang.haskell is not carried by my NSP,
Supernews, but a newgroup is archived at ftp.isc.org (a condition that
Supernews seems to require) so for consistency I asked them to carry it.
Adam H. Kerman
>"Adam H. Kerman" <a...@chinet.com> wrote:
>> ... Sure, if one advises the proponent to ignore the wishes of those posting on
>>the topic in the functional language group, you know, just about the entire
>>audience for the topic on Usenet. And if one advises the proponent that
>>there's no obligation to publicize a Big 8 group, exactly the same as
>>publicizing an alt group, despite what Marty thinks. ...
>What I think is:
>1. Some members of a generic group always want to keep
>the topic area in the generic group. I don't give a
>lot of weight to such complaints.
Why is it your place to tell them that they are posting incorrectly?
>If they like the
>new topic, they can subscribe to both newsgroups.
>If they just want to hold the folks interested in
>the topic hostage to pump up the stats for the
>generic group, then I'd like to see the captives
>set free.
Wow. This is truly offensive.
Tim Skirvin
>>> It's an alt group newgrouped nearly four years ago. Our proponent should
>>> request that alt.comp.lang.haskell be created on his system, and post to it.
>> alt.comp.lang.haskell is a fairly empty group (according to
>> Google), and not very well propagated (carried by 16 servers, according to
>> the http://groupsearch.aacity.net/ search engine). Based on this, I
>> expect that it would be far easier and more effective for the proponent to
>> create and publicize a new group in comp.* than it would be to try to
>> publicize and popularize the alt.* group.
> Sure, if one advises the proponent to ignore the wishes of those posting on
> the topic in the functional language group, you know, just about the entire
> audience for the topic on Usenet.
What does that have to do with what I wrote, with what you wrote,
or with the situation at hand?
comp.lang.functional has been contacted, and the feedback in this
thread seems generally positive. I see little evidence that this is a
particularly controversial split.
But to clarify: do you think that this proposal is a 'hostile
split' of comp.lang.functional? How about alt.comp.lang.haskell?
> And if one advises the proponent that there's no obligation to publicize
> a Big 8 group,
To my knowledge, no such advice has been offered to the proponent by
Board members; and based on the proponent's background, I am confident
that he wouldn't believe it if somebody else suggested it.
> There's that "far easier" thing again. None of this is easy.
Plastic surgery is far easier than brain surgery. That doesn't
mean that either of them is easy.
Dave Sill
> Plastic surgery is far easier than brain surgery.
Is it really? Is it harder to do a lobotomy than to perform a complex
facial reconstruction involving bone grafts and prosthetics?
-Dave
Adam H. Kerman
>"Adam H. Kerman" <a...@chinet.com> writes:
>>>>It's an alt group newgrouped nearly four years ago. Our proponent should
>>>>request that alt.comp.lang.haskell be created on his system, and post to it.
>>>alt.comp.lang.haskell is a fairly empty group (according to Google), and
>>>not very well propagated (carried by 16 servers, according to the
>>>http://groupsearch.aacity.net/ search engine). Based on this, I expect
>>>that it would be far easier and more effective for the proponent to
>>>create and publicize a new group in comp.* than it would be to try to
>>>publicize and popularize the alt.* group.
>>Sure, if one advises the proponent to ignore the wishes of those posting
>>on the topic in the functional language group, you know, just about the
>>entire audience for the topic on Usenet.
>What does that have to do with what I wrote, with what you wrote, or with
>the situation at hand?
>comp.lang.functional has been contacted, and the feedback in this thread
>seems generally positive. I see little evidence that this is a
>particularly controversial split.
People have been quite pleasant about it all, proving that non-hostile
discussion can take place in news.groups. Still, important concerns have
been raised. Shouldn't they be addressed? In fact, when people raise
concerns without vitriol, shouldn't the Board take them a whole lot more
seriously than, say, when insults and accusations are bandied about?
>But to clarify: do you think that this proposal is a 'hostile split' of
>comp.lang.functional?
That's where discussion is taking place. That's where details of the split
should be worked out, not here in news.groups. Do those discussing the
language wish to move their discussion to another group? Is there enough
discussion for the new group to be a success? Would enough discussion remain
in the old group on other topics that it remains viable?
I wish the proponent would ask. My opinion is irrelevant.
>How about alt.comp.lang.haskell?
Sigh. Proposal was made in alt.config after newgroup in bad syntax lacking a
charter was sent, usually the sign of a hopeless proponent. Proponent was
asked if he was aware of the mail-gated newsgroup. fa.haskell. He said he
was. He was not heard from again.
I have noted the existence of the haskell newsgroup. Proponent made no
effort. It's not a success, but it is named correctly, so it could be used
for its intended purpose. It's an option to consider, even if it is
rejected.
Does anyone take fa.* groups and have a good idea which ones are active and
still gated?
>>And if one advises the proponent that there's no obligation to publicize a
>>Big 8 group,
>To my knowledge, no such advice has been offered to the proponent by Board
>members; and based on the proponent's background, I am confident that he
>wouldn't believe it if somebody else suggested it.
Ok. We've had this unsuccessful discussion before of what should be expected
of proponents.
>>There's that "far easier" thing again. None of this is easy.
>Plastic surgery is far easier than brain surgery. That doesn't mean that
>either of them is easy.
If someone got the idea in his head that it was from an off-hand comment of
"far easier", should not one watch his language?
Kjetil Torgrim Homme
>
> Sigh. Proposal was made in alt.config after newgroup in bad syntax
> lacking a charter was sent, usually the sign of a hopeless
> proponent. Proponent was asked if he was aware of the mail-gated
> newsgroup. fa.haskell. He said he was. He was not heard from
> again.
>
>
> Does anyone take fa.* groups and have a good idea which ones are
> active and still gated?
I run the gateway at fa.*, including fa.haskell. I'm sorry to say I
haven't put much effort into maintaining subscriptions the last few
years, and as you allude to, many of the groups in the hierarchy are
dead. Gmane now does a much better job, so ...
all the fa.* groups are one-way, btw.
--
Kjetil T.
Adam H. Kerman
>Adam H. Kerman:
>>Does anyone take fa.* groups and have a good idea which ones are active
>>and still gated?
>I run the gateway at fa.*, including fa.haskell. I'm sorry to say I
>haven't put much effort into maintaining subscriptions the last few years,
>and as you allude to, many of the groups in the hierarchy are dead.
Are you saying that the gated mailing lists themselves are dead? That's too
bad.
>Gmane now does a much better job, so ...
>all the fa.* groups are one-way, btw.
I didn't know that. Was this always the case, or a deliberate attempt to
keep Usenet spam out of Mail?
Jim Riley
wrote:
>At 5:56am -0000, 10/19/06, Jeffrey M. Vinocur <je...@litech.org> wrote:
>>Adam H. Kerman <a...@chinet.com> wrote:
>>My goal is purely to facilitate discussion of Haskell and other
>>functional language by the provision of an appropriate set of
>>forums for such discussion to take place, and if you are sensing
>>any other motivation you may wish to reanalyze your perceptions.
>
>What if, after serious consideration of the needs of those posting on the
>topic, you find you don't have enough support for the split?
Why do you characterize this as a split?
If people are only interested in Haskell as a way of exploring the
concept of functional languages, then your position might make sense.
But if there are persons who are interested in using Haskell, then
they may not find comp.lang.haskell, or find it difficult to use for
their purpose. If they want to know how to do X in Haskell, they may
get an answer comparing several languages, or spring off on a tangent
about why they want to do something.
Those who are interested in both practical and conceptual aspects of
Haskell can participate in both groups.
--
Jim Riley
Kjetil Torgrim Homme
>
> Kjetil Torgrim Homme <kjet...@kaksi.ifi.uio.no> wrote:
> > I run the gateway at fa.*, including fa.haskell. I'm sorry to
> > say I haven't put much effort into maintaining subscriptions the
> > last few years, and as you allude to, many of the groups in the
> > hierarchy are dead.
>
> Are you saying that the gated mailing lists themselves are dead?
> That's too bad.
a little of both, I think. (I know fa.haskell is alive, though.)
> > all the fa.* groups are one-way, btw.
>
> I didn't know that. Was this always the case, or a deliberate
> attempt to keep Usenet spam out of Mail?
our gateway has always been one-way.
a little background: as you probably know, the FA name is from before
the Great Renaming (it stands for From ARPAnet). I don't know how it
worked back then. some years later, Anders Ellefsrud, my predecessor
as news admin here at ifi.uio.no, thought that it was wasteful that
our students subscribed to mailing lists individually, and set up a
gateway to save bandwidth and to get automatic expiration. he chose
to reuse the name fa.* which was no longer in use on Usenet. this was
a local hierarchy at first, but some neighbours thought it was useful
and wanted feeds, and in 1998 it started leaking to the Usenet at
large. this was under my watch, but I thought "hey, indexing at
Google is actually pretty neat", so I didn't try to find the leak.
--
Kjetil T.
Adam H. Kerman
>On Thu, 19 Oct 2006 02:17:19 -0500, "Adam H. Kerman" <a...@chinet.com> wrote:
>>At 5:56am -0000, 10/19/06, Jeffrey M. Vinocur <je...@litech.org> wrote:
>>>My goal is purely to facilitate discussion of Haskell and other
>>>functional language by the provision of an appropriate set of
>>>forums for such discussion to take place, and if you are sensing
>>>any other motivation you may wish to reanalyze your perceptions.
>>What if, after serious consideration of the needs of those posting on the
>>topic, you find you don't have enough support for the split?
>Why do you characterize this as a split?
Why wouldn't I? We have been informed that discussion of Haskell generally
takes place in the functional language group.
>If people are only interested in Haskell as a way of exploring the
>concept of functional languages, then your position might make sense.
>But if there are persons who are interested in using Haskell, then
>they may not find comp.lang.haskell, or find it difficult to use for
>their purpose. If they want to know how to do X in Haskell, they may
>get an answer comparing several languages, or spring off on a tangent
>about why they want to do something.
Those are interesting predictions of how future conversations in unmoderated
newsgroups might take place, completely unenforceable of course. I'm not
going to ask the proponent to perform a justification based on concept
versus practice (not that I asked for justification at all, just a
discussion with current posters about exactly what they want), for I don't
predict that's how the discussion would break down between the two groups,
if both were reasonably well used after newgrouping. There's no reason not
to believe that discussion in the narrower group wouldn't drift into
comparisons with other languages.
Did you read the functional language group, discovering a significant amount
of conceptual discussion?
>Those who are interested in both practical and conceptual aspects of
>Haskell can participate in both groups.
Your comment about discussion of practical Haskell is an argument in favor
of the status quo.
Aatu Koskensilta
> But if there are persons who are interested in using Haskell, then
> they may not find comp.lang.haskell, or find it difficult to use for
> their purpose. If they want to know how to do X in Haskell, they may
> get an answer comparing several languages, or spring off on a tangent
> about why they want to do something.
They may, of course, but in my experience questions about how to do this
or that in Haskell receive helpful answers explaining how to do this or
that in Haskell. There might be tangential discussion about why one
should do this or that that way in Haskell while in ML another approach
would be preferred, or why some solution that might appear obvious to
someone coming from an imperative background is highly non-idiomatic in
Haskell, or why there is no need to do this or that in Haskell in the
first place, since in purely functional settings some totally different
data structure is to be preferred, or ...
In any programming language group questions about how to do this or that
are going to occasionally spring off on a tangent about why the poster
wans to do this or that. Often such questions are helpful, as posters
not infrequently try to solve a specific problem in a particular way
that might be unsuitable or needlessly complicated. Sometimes they do
know what they're doing, and have good reasons for wanting to do
something that is usually not recommended; depending on the
interlocutors the discussion might then become heated as the original
poster asks others to just answer his question, while others demand he
tells why he wants to do what he wants to do and so forth, or it might
happen that everyone gets alone in the most commendable manner and
useful information and advice is given. We see all this in any language
group, and there seems to be no reason to suppose the proposed
comp.lang.haskell group would be an exception.
Finally, to repeat myself one more time: I'm not opposed to the creation
of the Haskell group if the Haskell people really see a need for it.
I'll certainly read the group if it is created, but I personally am
perfectly happy with Haskell discussion taking place in
comp.lang.functional (and the mailing lists). It is also my impression
that most of serious Haskell discussion takes place on the mailing
lists, and am interested in seeing how the people there feel about the
proposed group. The notice of the RFD hasn't exactly spurred a blizzard
of replies yet ...
Tim Skirvin
> Still, important concerns have been raised. Shouldn't they be addressed?
It looked to me like the concerns are being addressed. If any
specific point has been missed, why don't you specifically ask Jeff about
it?
>> But to clarify: do you think that this proposal is a 'hostile split' of
>> comp.lang.functional?
> That's where discussion is taking place.
Is that a 'yes'? I'll assume yes.
> That's where details of the split should be worked out, not here in
> news.groups. Do those discussing the language wish to move their
> discussion to another group? Is there enough discussion for the new
> group to be a success? Would enough discussion remain in the old group
> on other topics that it remains viable?
Go ahead and ask the proponent, preferably without being too
hostile about it.
> I have noted the existence of the haskell newsgroup. Proponent made no
> effort.
If you're still referring to alt.comp.lang.haskell, it was
included in the crosspost of the first RFD at the proponent's request.
If you're referring to fa.haskell, it doesn't appear to be a
bi-directional gateway, so posting there wouldn't do much good anyway.
>>> There's that "far easier" thing again. None of this is easy.
>> Plastic surgery is far easier than brain surgery. That doesn't mean that
>> either of them is easy.
> If someone got the idea in his head that it was from an off-hand comment of
> "far easier", should not one watch his language?
Mu. Do you have any evidence of your not-quite-assertion?
Adam H. Kerman
>"Adam H. Kerman" <a...@chinet.com> writes:
>>Still, important concerns have been raised. Shouldn't they be addressed?
>It looked to me like the concerns are being addressed. If any specific
>point has been missed, why don't you specifically ask Jeff about it?
It's a question to you.
>>>But to clarify: do you think that this proposal is a 'hostile split' of
>>>comp.lang.functional?
>>That's where discussion is taking place.
>Is that a 'yes'? I'll assume yes.
It's a split. I never once used the term "hostile split" in this discussion,
although I'm not entirely satisfied with the proponent's course of action,
this has been rather civilized.
>>That's where details of the split should be worked out, not here in
>>news.groups. Do those discussing the language wish to move their
>>discussion to another group? Is there enough discussion for the new group
>>to be a success? Would enough discussion remain in the old group on other
>>topics that it remains viable?
>Go ahead and ask the proponent, preferably without being too hostile about
>it.
I believe I have. At the moment, my concern is that several of these
concerns have been raised by people other than me. They haven't been
addressed yet. Will you see to it that they are addressed before this
proceeds any further?
>>>>There's that "far easier" thing again. None of this is easy.
>>>Plastic surgery is far easier than brain surgery. That doesn't mean that
>>>either of them is easy.
>>If someone got the idea in his head that it was from an off-hand comment of
>>"far easier", should not one watch his language?
>Mu. Do you have any evidence of your not-quite-assertion?
Someone else argued that plastic surgery isn't far easier than brain
surgery.
Mark T.B. Carroll
> That's where discussion is taking place. That's where details of the split
> should be worked out, not here in news.groups.
I can perhaps help a little here, having been on comp.lang.functional
and the main Haskell mailing lists for a while, and at times having
followed other groups like comp.lang.misc.
> Do those discussing the language wish to move their discussion to
> another group?
That, I don't know. I'm much more interested in the practical use of
Haskell threads rather than the discussion of FP in general with Haskell
and other languages as examples, so the separation would work well for
me, but I may be atypical as more a working programmer than an academic.
> Is there enough discussion for the new group to be a success?
I'm quite sure this would be so: the mailing lists get a lot of traffic
already, there are a fair few practical Haskell-only questions asked on
comp.lang.functional, and the Haskell user base is growing healthily
with new users, finding it through it now being taught in various
computer science courses, and through its use in Pugs, the fact that
it's now useful for real-world applications, etc.
> Would enough discussion remain in the old group on other topics that
> it remains viable?
Yes, there's no end of discussion of FP in general as FP languages are
compared, they borrow ideas from each other, they are compared with more
conventional approaches, new/minority FP languages are explored, etc.
(snip)
> Sigh. Proposal was made in alt.config after newgroup in bad syntax lacking a
> charter was sent, usually the sign of a hopeless proponent. Proponent was
> asked if he was aware of the mail-gated newsgroup. fa.haskell. He said he
> was. He was not heard from again.
(snip)
I didn't actually know about alt.comp.os.haskell before this thread.
It's not carried on any of the servers I use (those that carry alt.* at
all, anyway) and not mentioned in the regular Haskell community at all
that I've noticed. Given how few Google hits it gets, too (compared to,
say, alt.comp.lang.applescript) I'd be amazed if it actually propagated
well even if I got my local newsmasters to add it.
-- Mark
Brian Mailman
>
> I didn't actually know about alt.comp.os.haskell before this thread.
alt.comp.lang.haskell.
> It's not carried on any of the servers I use (those that carry alt.* at
> all, anyway) and not mentioned in the regular Haskell community at all
> that I've noticed. Given how few Google hits it gets, too (compared to,
> say, alt.comp.lang.applescript) I'd be amazed if it actually propagated
> well even if I got my local newsmasters to add it.
Apparently, the proponent wasn't particularly enthusiastic about
promoting it, since it's missing on what I would consider some major
servers:
Although it's recently been requested from Supernews (which gives some
dozens--at least--servers that SN outsources to).
If this were an alt.* group proposal, I'd say to promote it by starting
to crosspost ontopic between the *.functional group and aclh and put a
note in your .sig file to request the alt.* group from your sysadmin.
B/
Mark T.B. Carroll
> Mark T.B. Carroll wrote:
>>
>> I didn't actually know about alt.comp.os.haskell before this thread.
>
> alt.comp.lang.haskell.
(snip)
Doh, sorry, yes. I did search for the right thing, at least. (-:
-- Mark
Henrietta K Thomas
What he said.
Since I'm not a comp.* person, I've decided not to get too involved in
the arguments going on in this thread, except for the name of the group.
My understanding is that Haskell is currently discussed in
comp.lang.functional, and that some people are satisfied with that, but
that others would like to move it out into a separate group.
Your proposal is to create comp.lang.haskell, which makes a clean break
with the .functional group.
It might be better to create the new group as an extension of the old
rather than to make a clean break. Then you would have two groups:
comp.lang.functional
comp.lang.functional.haskell
and people could make their own choices.
Just a thought.....
--
Henrietta K. Thomas
Successful proponent for soc.support.vision-impaired.
Aatu Koskensilta
> It might be better to create the new group as an extension of the old
> rather than to make a clean break. Then you would have two groups:
>
> comp.lang.functional
> comp.lang.functional.haskell
>
> and people could make their own choices.
>
> Just a thought.....
comp.lang.functional.haskell doesn't really make much sense. We don't
have comp.lang.functional.ml, comp.lang.imperative.c,
comp.lang.imperative-oo.java or anything like that. If a Haskell group
is created it should be named comp.lang.haskell.
Jim Riley
wrote:
>At 6:25am -0000, 10/20/06, Jim Riley <jim...@pipeline.com> wrote:
>>Why do you characterize this as a split?
>
>Why wouldn't I? We have been informed that discussion of Haskell generally
>takes place in the functional language group.
We have been informed for lack of a specific place to discuss Haskell,
that some people use the group.
>>If people are only interested in Haskell as a way of exploring the
>>concept of functional languages, then your position might make sense.
>
>>But if there are persons who are interested in using Haskell, then
>>they may not find comp.lang.haskell, or find it difficult to use for
this should have been "comp.lang.functional"
>>their purpose. If they want to know how to do X in Haskell, they may
>>get an answer comparing several languages, or spring off on a tangent
>>about why they want to do something.
>
>Those are interesting predictions of how future conversations in unmoderated
>newsgroups might take place, completely unenforceable of course. I'm not
>going to ask the proponent to perform a justification based on concept
>versus practice (not that I asked for justification at all, just a
>discussion with current posters about exactly what they want), for I don't
>predict that's how the discussion would break down between the two groups,
>if both were reasonably well used after newgrouping. There's no reason not
>to believe that discussion in the narrower group wouldn't drift into
>comparisons with other languages.
Such discussion could be cross-posted. It is quite likely that some
persons will participate in both groups.
>Did you read the functional language group, discovering a significant amount
>of conceptual discussion?
I just scanned over the past year of comp.lang.functional. There were
many lengthly threads (10s and 100s of articles) that were conceptual
in nature. And then bunches of short threads (under 10 articles) that
were concrete questions, primarily about either Haskell or ML.
Depending on how one encountered the group, and how it was presented
in their newsreader, it might not be obvious that practical discussion
was appropriate.
Curiously, someone in June asked:
How hard is it to start new news groups?
Like
comp.lang.haskell
One response was:
I do find that comp.lang.haskell is conspicuous by its absence,
and included a pointer to the old group creation FAQs.
Jeffrey M. Vinocur (current proponent) noted that he had taken a straw
poll several years ago, and had concluded there was not enough support
to win a newgroup vote, but noted that the group creation proposal
system had changed. This apparently led to the current RFD.
So there you have a couple of people out of the blue, saying that it
sure would be nice to have a newsgroup dedicated to Haskell, how do we
create one?
>>Those who are interested in both practical and conceptual aspects of
>>Haskell can participate in both groups.
>
>Your comment about discussion of practical Haskell is an argument in favor
>of the status quo.
It would be, except I wrote comp.lang.haskell, where I mean
comp.lang.functional. That is, persons who had a particular interest
in practical Haskell, would be less likely to find
comp.lang.functional, and may find much of the discussion there not
particularly useful for their immediate needs, if it was too
conceptual in nature, or about another functional programming
language.
--
Jim Riley
Jim Riley
wrote:
>Your proposal is to create comp.lang.haskell, which makes a clean break
>with the .functional group.
>
>It might be better to create the new group as an extension of the old
>rather than to make a clean break. Then you would have two groups:
>
>comp.lang.functional
>comp.lang.functional.haskell
>
>and people could make their own choices.
There are already groups for specific functional languages that are
found directly under comp.lang.*.
The convention is:
comp.lang.<name-of-language>
You would be introducing a new naming scheme that does not fit dozens
of existing groups, and possibly even hides other languages. If
Haskell were listed under comp.lang.functional.haskell, then someone
might expect that a group for ML might be under
comp.lang.functional.ml. Since it doesn't they might presume that it
doesn't exist.
There are a very few groups that are not for particular languages.
comp.lang.functional
comp.lang.misc
comp.lang.visual
comp.object
--
Jim Riley
Adam H. Kerman
>On Fri, 20 Oct 2006 05:01:22 -0500, "Adam H. Kerman" <a...@chinet.com> wrote:
>>At 6:25am -0000, 10/20/06, Jim Riley <jim...@pipeline.com> wrote:
>>>Why do you characterize this as a split?
>>Why wouldn't I? We have been informed that discussion of Haskell generally
>>takes place in the functional language group.
>We have been informed for lack of a specific place to discuss Haskell,
>that some people use the group.
No, Jim, we haven't. We've been informed that those posting on Usenet on the
topic are satisfied with the current group. It's odd that an experience
Usenet user such as yourself is taking a position that newsgroups for
broader topics are not "specific places" for narrower topics. You're the
rmgroup queen. Seems that you have concluded that a great many of these
groups started years ago were too narrow and that several reorganizations
were useless.
>>>If people are only interested in Haskell as a way of exploring the
>>>concept of functional languages, then your position might make sense.
>>>But if there are persons who are interested in using Haskell, then
>>>they may not find comp.lang.haskell, or find it difficult to use for
>this should have been "comp.lang.functional"
>>>their purpose. If they want to know how to do X in Haskell, they may
>>>get an answer comparing several languages, or spring off on a tangent
>>>about why they want to do something.
>>Those are interesting predictions of how future conversations in unmoderated
>>newsgroups might take place, completely unenforceable of course. I'm not
>>going to ask the proponent to perform a justification based on concept
>>versus practice (not that I asked for justification at all, just a
>>discussion with current posters about exactly what they want), for I don't
>>predict that's how the discussion would break down between the two groups,
>>if both were reasonably well used after newgrouping. There's no reason not
>>to believe that discussion in the narrower group wouldn't drift into
>>comparisons with other languages.
>Such discussion could be cross-posted.
People may do any number of things. You can neither predict nor enforce what
they should do in unmoderated Usenet. If you feel that there's a need to
crosspost certain articles, then such articles do not justify the proposed
group. In my experience, when following up in a thread, people generally
don't change the newsgroups header. If the thread wasn't crossposted,
followups probably won't be crossposted. If the thread was crossposted and a
followup no longer addresses a topic of one or more of the groups on the
crosspost, it'll still be crossposted.
Topic drift is a Usenet feature. You're in the minority.
>It is quite likely that some persons will participate in both groups.
It's thought that when discussion of a topic is consolidated in one
newsgroup, there's more discussion and the discussion is more fruitful.
Splits are not warranted under the scenario that you have presented, in
which some of the discussion of the topic would still take place in the
existing newsgroup.
>>Did you read the functional language group, discovering a significant amount
>>of conceptual discussion?
>I just scanned over the past year of comp.lang.functional. There were
>many lengthly threads (10s and 100s of articles) that were conceptual
>in nature. And then bunches of short threads (under 10 articles) that
>were concrete questions, primarily about either Haskell or ML.
Well, then. If someone asked a question comparing Haskell to another
language, then they posted in the correct newsgroup anyway, even if the
narrower group already existed.
>Depending on how one encountered the group, and how it was presented
>in their newsreader, it might not be obvious that practical discussion
>was appropriate.
This is why one lurks in a newsgroup, to learn what is actually discussed.
>Curiously, someone in June asked:
>
> How hard is it to start new news groups?
>
> Like
>
> comp.lang.haskell
>One response was:
> I do find that comp.lang.haskell is conspicuous by its absence,
>and included a pointer to the old group creation FAQs.
>Jeffrey M. Vinocur (current proponent) noted that he had taken a straw
>poll several years ago, and had concluded there was not enough support
>to win a newgroup vote, but noted that the group creation proposal
>system had changed. This apparently led to the current RFD.
>So there you have a couple of people out of the blue, saying that it
>sure would be nice to have a newsgroup dedicated to Haskell, how do we
>create one?
Actually, it appears to have led to the alt group newgrouped by
urf...@optushome.com.au who posted a bit on Haskell in 2002 and 2003 and
not since. I take it back. He did make one attempt to publicize the group in
the functional language group.
Groups usually get started in this manner, with someone raising the issue in
an existing group where the narrow topic is discussed.
>>>Those who are interested in both practical and conceptual aspects of
>>>Haskell can participate in both groups.
>>Your comment about discussion of practical Haskell is an argument in favor
>>of the status quo.
>It would be, except I wrote comp.lang.haskell, where I mean
>comp.lang.functional.
I already noted that I understood your practical versus conceptual
distinction.
>That is, persons who had a particular interest in practical Haskell, would
>be less likely to find comp.lang.functional, and may find much of the
>discussion there not particularly useful for their immediate needs, if it
>was too conceptual in nature, or about another functional programming
>language.
Why on earth would you assume that people familiar with computer languages
are unable to search for the discussion that they want? Will you next argue
that if those looking for conceptional discussion of Haskell will be so
disappointed if it's not found in the proposed group that they won't read
the functional languages group?
In any event, your belief that the discussion will neatly split itself
between two groups as practical or conceptual isn't logical.
If you correctly predict that this will be an extraordinary newsgroup
without drift, that practical discussion won't drift into conceptual
discussion, then there will hardly be enough to discuss.
Henrietta K Thomas
<aatu.kos...@xortec.fi> wrote:
>Henrietta K Thomas wrote:
>> It might be better to create the new group as an extension of the old
>> rather than to make a clean break. Then you would have two groups:
>>
>> comp.lang.functional
>> comp.lang.functional.haskell
>>
>> and people could make their own choices.
>>
>> Just a thought.....
>
>comp.lang.functional.haskell doesn't really make much sense. We don't
>have comp.lang.functional.ml, comp.lang.imperative.c,
>comp.lang.imperative-oo.java or anything like that. If a Haskell group
>is created it should be named comp.lang.haskell.
Jim Riley apparently agrees with you, so I withdraw the suggestion.
Jeffrey M. Vinocur
Okay, I started a whole separate thread in comp.lang.functional
trying to satisfy this request of yours. We'll see if it changes
anything.
--
Jeffrey M. Vinocur
je...@litech.org
Jeffrey M. Vinocur
Adam, do you think we should rmgroup comp.lang.ml and
comp.lang.scheme, and ask those users to hold their discussions
in comp.lang.functional?
Jim Riley
wrote:
>At 6:42am -0000, 10/21/06, Jim Riley <jim...@pipeline.com> wrote:
>>On Fri, 20 Oct 2006 05:01:22 -0500, "Adam H. Kerman" <a...@chinet.com> wrote:
>>>At 6:25am -0000, 10/20/06, Jim Riley <jim...@pipeline.com> wrote:
>
>>>>Why do you characterize this as a split?
>
>>>Why wouldn't I? We have been informed that discussion of Haskell generally
>>>takes place in the functional language group.
>
>>We have been informed for lack of a specific place to discuss Haskell,
>>that some people use the group.
>
>No, Jim, we haven't. We've been informed that those posting on Usenet on the
>topic are satisfied with the current group.
I think you are misunderstanding the purpose and focus of
comp.lang.functional. You appear to be seeing it as a container for
the set of functional languages. But it is more conceptual in nature.
People discuss functional languages that _do_ have their own
comp.lang.* groups. But the discussion of those languages that occurs
in comp.lang.functional tends to be more conceptual or comparative.
That is not to say that no practical discussion slips through, but it
simply is not the focus of the group.
Since Haskell does not have its own group, people make use of what is
available. It's not really the wrong place. But at the same time it
is not really the right place.
> It's odd that an experience
>Usenet user such as yourself is taking a position that newsgroups for
>broader topics are not "specific places" for narrower topics.
I recognize that people will use whatever group that they can find
that is close to the topic they want to discuss. The discussion on
comp.unix.cray is not primarily about Unix, but Cray. The discussion
on comp.os.aos is not about AOS as much as it is Data General.
> You're the
>rmgroup queen. Seems that you have concluded that a great many of these
>groups started years ago were too narrow and that several reorganizations
>were useless.
Which groups in *particular* are *you* referring to? It is not
usually a good idea to make generalizations based on the null set.
Very few of the groups are the result of what would be truly
considered re-organizations. Some of the binaries and sources groups
may have been created as part of a split. Before there were binaries
groups, peoples sometimes would post binaries to discussion groups.
The groups that have been removed have been largely unused. In some
cases, people were wondering 20 years ago what the heck the group was
for. Unused and useless - lacking a use - are the same thing.
>>Such discussion could be cross-posted.
>
>People may do any number of things. You can neither predict nor enforce what
>they should do in unmoderated Usenet.
What people should do is not a prediction. What people will do can
certainly be predicted. It is certainly reasonable to cross-post
discussion that is more conceptual or comparative in nature between
comp.lang.haskell and compl.lang.functional. And further, I predict
that it will happen.
> If you feel that there's a need to
>crosspost certain articles, then such articles do not justify the proposed
>group.
The cross-posted articles would not provide justification for
comp.lang.haskell. It is the articles that would be more appropriate
for comp.lang.haskell than comp.lang.functional that form the
justification. Nonetheless, there are some articles that would
appropriately be cross-posted between the groups.
Let's say that there is a proposal for a moderated version of
misc.foobar, and the discussion is cross-posted between news.groups
and misc.foobar. Does that cross-posted discussion form part of the
justification for either misc.foobar or news.groups? NOOOO. It that
cross-posted discussion appropriate? MAYBE.
>In my experience, when following up in a thread, people generally
>don't change the newsgroups header. If the thread wasn't crossposted,
>followups probably won't be crossposted.
In this case, it would be likely be that someone in comp.lang.haskell
would notice that a discussion was moving more into the conceptual
realm and would be of interest to comp.lang.functional as well, and
add the cross-post. I would not be surprised if many people who
subscribe to comp.lang.haskell would also continue to subscribe to
comp.lang.functional.
>If the thread was crossposted and a
>followup no longer addresses a topic of one or more of the groups on the
>crosspost, it'll still be crossposted.
So? It may then last a little bit longer in comp.lang.haskell.
Threads in comp.lang.functional can be rather lengthy.
>Topic drift is a Usenet feature. You're in the minority.
Eh? The thread drift was from something that was in
comp.lang.haskell, that was more appropriate for comp.lang.functional.
Cross-posting supports that.
>>It is quite likely that some persons will participate in both groups.
>
>It's thought that when discussion of a topic is consolidated in one
>newsgroup, there's more discussion and the discussion is more fruitful.
Is is thought by _whom_? Cite your authority.
>Splits are not warranted under the scenario that you have presented, in
>which some of the discussion of the topic would still take place in the
>existing newsgroup.
It is not a split.
>>>Did you read the functional language group, discovering a significant amount
>>>of conceptual discussion?
>
>>I just scanned over the past year of comp.lang.functional. There were
>>many lengthly threads (10s and 100s of articles) that were conceptual
>>in nature. And then bunches of short threads (under 10 articles) that
>>were concrete questions, primarily about either Haskell or ML.
>
>Well, then. If someone asked a question comparing Haskell to another
>language, then they posted in the correct newsgroup anyway, even if the
>narrower group already existed.
Most of the discussion about Haskell was not comparative in nature.
There was some discussion comparing how different ideas were (or were
not) implemented in various functional languages, and examples in
Haskell were given. But to repeat - most of the discussion about
Haskell was not comparative in nature.
>>Depending on how one encountered the group, and how it was presented
>>in their newsreader, it might not be obvious that practical discussion
>>was appropriate.
>
>This is why one lurks in a newsgroup, to learn what is actually discussed.
Lurking may not help. The discussion in comp.lang.functional may be
too abstract and conceptual for it to be readily understood that
practical matters are sometimes discussed, It would be transparent if
it were in comp.lang.haskell.
>>Curiously, someone in June asked:
>>
>> How hard is it to start new news groups?
>>
>> Like
>>
>> comp.lang.haskell
>
>>One response was:
>
>> I do find that comp.lang.haskell is conspicuous by its absence,
>
>>and included a pointer to the old group creation FAQs.
>
>>Jeffrey M. Vinocur (current proponent) noted that he had taken a straw
>>poll several years ago, and had concluded there was not enough support
>>to win a newgroup vote, but noted that the group creation proposal
>>system had changed. This apparently led to the current RFD.
>
>>So there you have a couple of people out of the blue, saying that it
>>sure would be nice to have a newsgroup dedicated to Haskell, how do we
>>create one?
>
>Actually, it appears to have led to the alt group newgrouped by
>urf...@optushome.com.au who posted a bit on Haskell in 2002 and 2003 and
>not since. I take it back. He did make one attempt to publicize the group in
>the functional language group.
The person who made the comment in June 2006, was not aware of the
existence of the alt.* group. It was perfectly obvious to the person
that there _should_ be a comp.lang.haskell given the groups for many
other programming languages. But maybe that was because they hadn't
lurked in news.groups or alt.config for 10 years - and there common
sense had not been excised.
>Groups usually get started in this manner, with someone raising the issue in
>an existing group where the narrow topic is discussed.
I don't believe this is true at all. As often as not, people use
other groups as a pattern. For example, the RFD for
soc.culture.french says something like, "this would be like
soc.culture.nordic, except we wear beréts". They weren't discussing
French culture in soc.culture.misc or soc.culture.nordic and proposing
a split, but were using another group for a model for a topic they
wanted to discuss.
comp.lang.haskell would be based on the pattern of comp.lang.fortran,
and comp.lang.c, and comp.lang.cobol, etc., even if Haskell is never
mentioned in any of them, and they use a completely different approach
to problem solving.
>>That is, persons who had a particular interest in practical Haskell, would
>>be less likely to find comp.lang.functional, and may find much of the
>>discussion there not particularly useful for their immediate needs, if it
>>was too conceptual in nature, or about another functional programming
>>language.
>
>Why on earth would you assume that people familiar with computer languages
>are unable to search for the discussion that they want?
I believe that people who were familiar with most comp.lang.* groups
(which focus more on the practical) would recognize that
comp.lang.functional is not exactly what they would want in a Haskell
newsgroup.
> Will you next argue
>that if those looking for conceptional discussion of Haskell will be so
>disappointed if it's not found in the proposed group that they won't read
>the functional languages group?
What do you mean by conceptual discussion of Haskell?
>In any event, your belief that the discussion will neatly split itself
>between two groups as practical or conceptual isn't logical.
I believe that discussion of ML will neatly split itself. It won't
happen in comp.lang.haskell, except where someone is trying to figure
out how to do something in Haskell, that they know how to do in ML.
But they might also want to know how to do something in Haskell, that
they know how to do in C.
>If you correctly predict that this will be an extraordinary newsgroup
>without drift, that practical discussion won't drift into conceptual
>discussion, then there will hardly be enough to discuss.
Eh?
--
Jim Riley
Jonathan Kamens
>In article <Pine.LNX.4.63.06...@qbbshf.puvarg.pbz>,
>Adam H. Kerman <a...@chinet.com> wrote:
>>
>>Ask THEM.
>
>Okay, I started a whole separate thread in comp.lang.functional
>trying to satisfy this request of yours. We'll see if it changes
>anything.
You shouldn't be doing it "to satify this request of [Adam's]." You
should be doing it because it's a good idea and, as Adam has pointed
out, relevant to the question of whether the group should be created.
--
Help stop the genocide in Darfur!
http://www.genocideintervention.net/
Jeffrey M. Vinocur
Jonathan Kamens <j...@kamens.brookline.ma.us> wrote:
>je...@litech.org (Jeffrey M. Vinocur) writes:
>>In article <Pine.LNX.4.63.06...@qbbshf.puvarg.pbz>,
>>Adam H. Kerman <a...@chinet.com> wrote:
>>>
>>>Ask THEM.
>>
>>Okay, I started a whole separate thread in comp.lang.functional
>>trying to satisfy this request of yours. We'll see if it changes
>>anything.
>
>You shouldn't be doing it "to satify this request of [Adam's]." You
>should be doing it because it's a good idea and, as Adam has pointed
>out, relevant to the question of whether the group should be created.
Certainly, which is why the RFD is crossposted to
comp.lang.functional. But I'm not quite sure how much jumping up
and down and waving we should do to drag people into this
discussion.
Adam H. Kerman
>Adam H. Kerman <a...@chinet.com> wrote:
>>Ask THEM.
>Okay, I started a whole separate thread in comp.lang.functional
Thank you kindly.
>trying to satisfy this request of yours. We'll see if it changes anything.
My satisfaction is irrelevant. The point is that these guys are the entire
Usenet audience for the proposed newsgroup. If they don't support the group,
it will fail.
New groups need lots and lots of support.
Adam H. Kerman
>Adam, do you think we should rmgroup comp.lang.ml and
>comp.lang.scheme, and ask those users to hold their discussions
>in comp.lang.functional?
No, of course not. I do not believe in reorganizations, nor do I believe in
rmgroups, in this era in which all too few servers are running checkgroups.
Your (I hope) sarcasm aside, Jim Riley and I do not see eye to eye on neat
divisions of discussion among several related unmoderated newsgroups. If the
haskell group is started, and it gets both practical and conceptual
discussion, what's wrong with that? Various rmgroup RFDs by JR note that,
wonder of wonders, certain reorganizations attempting to herd cats utterly
failed. His true belief confuses me.
Ivan Jager
> Jeffrey M. Vinocur wrote:
>> Discussion of Haskell is currently fragmented through Usenet. It is the
>> only major functional programming language without a dedicated Big 8
>> group, and thus has become the topic of a substantial portion of the
>> traffic in comp.lang.functional.
>
> It seems most serious discussion about Haskell takes place in
> comp.lang.functional. The discussion thus does not seem to be "fragmented
> through Usenet". comp.lang.functional is not a very high-volume group, so why
> the need for a specific Haskell group? I'm not saying that I'm opposed to
> such a group - I certainly would read it! - I just don't see why Haskell
> can't be discussed in comp.lang.functional as it has to this day. Is there
> some evidence that newcomers have hard time locating Haskell discussion in
> USENET? (Most probably use the mailing list in any case?).
I had been subscribed to alt.comp.lang.haskell for a while, and it never
occured to me to subscribe to comp.lang.functional until I saw the post
about creating the comp.lang.haskell group, and it mentioned that most
Haskell discussion is on comp.lang.functional.
So, I would be in favor of creating the group.
Ivan
Adam H. Kerman
>I had been subscribed to alt.comp.lang.haskell for a while,
How much on-topic discussion is there? As the proponent did a rotten job,
how did you learn of the group?
mike....@gmail.com
> REQUEST FOR DISCUSSION (RFD)
> unmoderated group comp.lang.haskell
>
> This is a formal Request For Discussion (RFD) for the creation of the
> unmoderated Usenet newsgroup, comp.lang.haskell.
>
> NEWSGROUPS LINE: comp.lang.haskell
>
> comp.lang.haskell Haskell concepts, implementation, and use.
+1(*) Practical issues of haskell programming are an uneasy fit with
comp.lang.funtional, and Haskell has sufficient traction to make this
work.
(*) As an aside, many open-source communities use a scheme for issue
resolution that works as follows: Once an course of action is proposed,
people involved in the discussion can indicate the their opinion using
a numeric thingy. Note that this isn't a vote: the numbers are not
usually counted or summed. Instead it is used to get a more concrete
idea of what a group of people are thinking.
Typically there are four options:
+1: in favour
+0: not opposed and don't see potential harm, but not endorsing the
idea and don't necessarily se benefit
-0: not opposed to the idea's implementation, but am uneasy or have
reservations about the proposal
-1: opposed: the idea is harmful
Note that there are _no_ defined ways in which these numbers are used.
Some projects have a single dictatorial head from which a "-1" is a
definite veto. Most projects have members of higher standing (due to
major contribution or experience), and their opinion tends to be more
highly weighed.
It tends to be clear: seas of -1's tend to kill a proposal. Vast
quantities of +/-0's with few +/-1's tend to be ideas of marginal
value. Lots of +1's is also an easy case. Any other distribution
tends to be less clear, but it at least gives something from which a
decision can be based.
Also, changing your "number" is often a more concrete way of indicating
that someone has persuaded you to modify your position in an important
way. "I'm changing my +1 to -0, as you've convinced me that the
proposal isn't beneficial and might cause problems"
Of course, this meshes uneasily with soliciting the users of a group,
since they are not familiar with the scheme. But as long as this is
taken into consideration by the decision makers, this could at least be
used to sample the opinion of news.groupies (presumably the opinions of
groupies would change to reflect the feedback from affected groups as
well).
Take this proposal as an example. There are several people I'd
classify as +1 (Jeffrey, Jim, some cl.functional regs, myself), some +0
(Martin, HKT; mainly people that don't see problems but aren't as
familiar with the issues), and some people with caveats. It would be
useful for decision makers to know whether the caveats translated into
-0 opinions or -1 opinions. At least it would for me, were I
ultimately making the call.
cheers,
-Mike
Martin X. Moleski, SJ
<1161559726.3...@f16g2000cwb.googlegroups.com>:
...
>Typically there are four options:
>+1: in favour
>+0: not opposed and don't see potential harm, but not endorsing the
>idea and don't necessarily se benefit
>-0: not opposed to the idea's implementation, but am uneasy or have
>reservations about the proposal
>-1: opposed: the idea is harmful
...
>Take this proposal as an example. There are several people I'd
>classify as +1 (Jeffrey, Jim, some cl.functional regs, myself), some +0
>(Martin, HKT; mainly people that don't see problems but aren't as
>familiar with the issues), and some people with caveats. It would be
>useful for decision makers to know whether the caveats translated into
>-0 opinions or -1 opinions. At least it would for me, were I
>ultimately making the call.
The Apache Software Foundation actually counts the votes,
I believe:
<http://www.apache.org/foundation/voting.html>:
> Expressing Votes: +1, 0, -1, and Fractions
> ------------------------------------------
>
> The voting process in Apache may seem more than a little weird if you've
> never encountered it before. Votes are represented as numbers between -1
> and +1, with '-1' meaning 'no' and '+1' meaning 'yes.'
>
> The in-between values are indicative of how strongly the voting individual
> feels. Here are some examples of fractional votes and ways in which they
> might be intended and interpreted:
>
> * +0: 'I don't feel strongly about it, but I'm okey with this.'
> * -0: 'I won't get in the way, but I'd rather we didn't do this.'
> * -0.5: 'I don't like this idea, but I can't find any rational justification for my feelings.'
> * ++1: 'Wow! I like this! Let's do it!'
> * -0.9: 'I really don't like this, but I'm not going to stand in the way if everyone else wants to go ahead with it.'
> * +0.9: 'This is a cool idea and i like it, but I don't have time/the skills necessary to help out.'
>
> Votes should generally be permitted to run for at least 72 hours to
> provide an opportunity for all concerned persons to participate regardless
> of their geographic locations.
>
> Votes on Code Modification
> --------------------------
>
> For code-modification votes, +1 votes are in favour of the proposal, but -
> 1 votes are vetos and kill the proposal dead until all vetoers withdraw
> their -1 votes.
>
> Unless a vote has been declared as using lazy consensus, three +1 votes
> are required for a code-modification proposal to pass.
>
> Whole numbers are recommended for this type of vote, as the opinion being
> expressed is Boolean: 'I approve/do not approve of this change.'
I guess this is an open-ended Likert scale. Instead of just
the standard 5 possible responses, it lets people get as
fuzzy as they want to get. (-0.001: I want to throw
a grain of sand in the gears).
The standard Likert responses are:
1. Strongly disagree
2. Disagree
3. Neither agree nor disagree
4. Agree
5. Strongly agree
I'd rate myself at +0.25 on the applause-o-meter. I
do see benefit in the creation of the group, though
I wouldn't use it myself.
Marty
--
Member of the Big-8 Management Board (B8MB).
See http://www.big-8.org for more information.
mike....@gmail.com
> On 22 Oct 2006 16:28:46 -0700, "mike....@gmail.com" <mike....@gmail.com> wrote in
> <1161559726.3...@f16g2000cwb.googlegroups.com>:
> The Apache Software Foundation actually counts the votes,
> I believe:
>
> <http://www.apache.org/foundation/voting.html>:
<snip>
As a committer in a modest-sizes ASF project, I am indeed aware of
that. You can only count votes when you have a well-defined electorate
(as you do with ASF projects). There are many other examples of this
scheme being used outside of a formal counting framework.
> I guess this is an open-ended Likert scale. Instead of just
> the standard 5 possible responses, it lets people get as
> fuzzy as they want to get. (-0.001: I want to throw
> a grain of sand in the gears).
>
> The standard Likert responses are:
>
> 1. Strongly disagree
> 2. Disagree
> 3. Neither agree nor disagree
> 4. Agree
> 5. Strongly agree
I think there is a difference: in the +-1 +-0 scheme, you effectively
express an separate opinion on whether you think it is good idea, and
whether the idea should be effectuated. Usually these two things are
highly corollated, so there isn't a huge difference.
> I'd rate myself at +0.25 on the applause-o-meter. I
> do see benefit in the creation of the group, though
> I wouldn't use it myself.
People can and do submit any sort of value (a favourite is -1000 or
-infinity). But this is just playfully extending the metaphor--the
numbers are metaphorical, not concrete. There's not point in saying
"+.25", as everyone's idea of scaling these values is different, and it
invites false comparison (is someone that indicates -.25 exactly as
opposed to the proposal as you are for it?).
As I mentioned above, the useful axes are twofold: what is your opinion
of the idea, and should it be implemented. The opinion is conveyed in
the + or -, and whether your opinion is strong enough to recommend
either for/against action or not is conveyed in the 0 or 1.
-Mike
Mark T.B. Carroll
> Usenet audience for the proposed newsgroup. If they don't support the group,
> it will fail.
No, over time they're not: not only is the Haskell user base growing, so
there'll be more people looking for a place to talk about it, but it's
certainly conceivable that some of the users of the Haskell mailing
lists are using them more because they're Haskell-specific than because
they're not Usenet.
-- Mark
Adam H. Kerman
>"Adam H. Kerman" <a...@chinet.com> writes:
>>My satisfaction is irrelevant. The point is that these guys are the entire
>>Usenet audience for the proposed newsgroup. If they don't support the group,
>>it will fail.
>No, over time they're not: not only is the Haskell user base growing, so
>there'll be more people looking for a place to talk about it, but it's
>certainly conceivable that some of the users of the Haskell mailing
>lists are using them more because they're Haskell-specific than because
>they're not Usenet.
Why do you say that? Is there widespread dissatisfaction with the mailing
lists? If there were, you'd think they'd have used the nearly four year old
alt group at some point. Motivated people might have gotten that going; too
bad that wasn't the proponent.
Let's not make the same mistake with the second newsgroup, assuming that
people are waiting in the wings if only... Somebody has to find the audience
and motivate people to use the group, possibly after requestion creation of
the group on a server that doesn't run checkgroups.
Martin X. Moleski, SJ
<1161563868.6...@i3g2000cwc.googlegroups.com>:
> ... the useful axes are twofold: what is your opinion
>of the idea, and should it be implemented. The opinion is conveyed in
>the + or -, and whether your opinion is strong enough to recommend
>either for/against action or not is conveyed in the 0 or 1.
Ah, I didn't see that you were dealing with just
four quadrants defining all possible combinations
of answers to the two separate questions:
+1: for the idea,
it should be implemented
+0: for the idea,
it should not be implemented
on the strength of this opinion
-0: against the idea,
it should not fail to be implemented
on the strength of this opinion
-1: against the idea,
it should not be implemented
If people agree on what the responses mean,
I suppose they can start using them in discussions
here.
As has happened with the Apache Software Foundation,
I doubt that you can keep people from using decimals
to indicate nuances in their position. People are
like that.
Tim Skirvin
> Motivated people might have gotten that going; too bad that wasn't the
> proponent.
Adam, can you please stop gratuitously insulting the proponents?
- Tim Skirvin (sk...@big-8.org)
--
http://www.big-8.org/ Big-8 Management Board
http://www.killfile.org/~tskirvin/ Skirv's Homepage <FISH>< <*>
Mark T.B. Carroll
> Why do you say that? Is there widespread dissatisfaction with the mailing
> lists?
No, simply that the fact that the people are using the mailing lists and
not comp.lang.functional doesn't necessarily reflect that they're not
part of the prospective comp.lang.haskell audience, it might be more
that they're following more by topic than modality.
> If there were, you'd think they'd have used the nearly four year old
> alt group at some point.
(snip)
I've already pointed out how poorly propagated that group is. I don't
think that its quietness is any indication at all of the prospects for a
comp.lang.haskell. Certainly it wasn't on any of the servers I routinely
use. (Besides, Haskell was nowhere near as popular at the time that that
group was created, so it would have been harder to drum up an audience
anyway.)
-- Mark
Pekka Karjalainen
>If you would like to read or post in the proposed newsgroup, please make
>a comment to that effect in this thread; the proponent will keep a list
I would like to read and post in comp.lang.haskell. I'd prefer it to both
the Haskell café mailing list and comp.lang.functional, which I follow,
but which I haven't posted in.
Pekka Karjalainen
Adam H. Kerman
>"Adam H. Kerman" <a...@chinet.com> writes:
>>Motivated people might have gotten that going; too bad that wasn't the
>>proponent.
>Adam, can you please stop gratuitously insulting the proponents?
It's an observation, not the least bit gratuitous. Proponents of tens of
thousands of alt newgroups have done little or nothing to promote the
newsgroups they pretended to want. It's most unfortunate when this comes at
the expense of discussion of a serious subject like Haskell.
Nothing good can be said of that proponent. I'm not bothering to suggest
that it's worth salvaging alt.comp.lang.haskell in lieu of starting a Big 8
group.
Are we now playing a game whereby you belittle what I say for the hell of
it? Cute bit taking my remark out of context and implying that it's some
bizarre attack on the current proponent, who has nothing to do with the
original proponent.
The context you snipped:
"Is there widespread dissatisfaction with the mailing lists? If there were,
you'd think they'd have used the nearly four year old alt group at some
point. Motivated people might have gotten that going; too bad that wasn't
the proponent."
Tim Skirvin
>>> Motivated people might have gotten that going; too bad that wasn't the
>>> proponent.
>> Adam, can you please stop gratuitously insulting the proponents?
> It's an observation, not the least bit gratuitous. Proponents of tens of
> thousands of alt newgroups have done little or nothing to promote the
> newsgroups they pretended to want. It's most unfortunate when this comes at
> the expense of discussion of a serious subject like Haskell.
So you're attacking the proponent of alt.comp.lang.haskell, not
the proponent of comp.lang.haskell? If so, I apologize; but I wish you'd
make that more clear.
Jeffrey M. Vinocur
>
>>Adam, do you think we should rmgroup comp.lang.ml and
>>comp.lang.scheme, and ask those users to hold their discussions
>>in comp.lang.functional?
>
>No, of course not. I do not believe in reorganizations, nor do I believe in
>rmgroups, in this era in which all too few servers are running checkgroups.
Okay, that wasn't really the question I was trying to ask.
In a hypothetical parallel Usenet where we had
comp.lang.functional and did not yet have comp.lang.ml,
comp.lang.scheme, or comp.lang.haskell (so all ML, Scheme, and
Haskell discussion was taking place in comp.lang.functional as
the next best place), would you be interested in the creation of
any of the mentioned groups? Do you think those groups were a
mistake in the first place?
Should we just have misc.misc for all discussion and call it a
day, and never make any new groups at all?
Jeffrey M. Vinocur
Steve Bonine <s...@pobox.com> wrote:
>
> Even though the RFD has been posted in comp.lang.functional, post a
>short article there and try to start a thread in the newsgroup that
>gauges how people feel about the new group. People are more willing to
>express their opinion when the followup is not set to news.groups. When
>you get a few opinions from the membership of the group you can bring
>that information back to news.groups.
For what it's worth, the small amount of response elicited was
generally in favor, with one person skeptical about whether the
quantity of discussion warrented a separate group, and no real
objections.
Steve Bonine
Thank you. I feel that I learned something from your trouble.
Adam H. Kerman
>"Adam H. Kerman" <a...@chinet.com> writes:
>>>> Motivated people might have gotten that going; too bad that wasn't the
>>>> proponent.
>>> Adam, can you please stop gratuitously insulting the proponents?
>> It's an observation, not the least bit gratuitous. Proponents of tens of
>> thousands of alt newgroups have done little or nothing to promote the
>> newsgroups they pretended to want. It's most unfortunate when this comes at
>> the expense of discussion of a serious subject like Haskell.
> So you're attacking the proponent of alt.comp.lang.haskell, not
>the proponent of comp.lang.haskell? If so, I apologize; but I wish you'd
>make that more clear.
Stick a sock in it, will you Tim?