proposal: separate lists for ghc-cvs commits and ghc-dev chatter
hey all, It seems to me (and i've certainly heard other people echo this sentiment) that: ghc dev chatting gets buried in the huge volume of commit + build report emails, and that creates (perhaps) another barrier to involvement in ghc dev at the hobbyist (rather than part time/full time ) scale? So my question for the community (and of course current ghc devs ) 1) do others agree that theres value in separating the two? 2) would this just be another use of the ghc-users list or would it be worth having a ghc-dev list? 3) most importantly, would the folks actively involved in ghc dev be willing/able to do so? That said, it does seem like the majority of the interesting ghc-dev chatter is on Trac issues, which is a good thing, though theres lots of interesting wee nuggets buried on the ghc-cvs list intermittently I hope this question makes sense for y'all! -Carter
Carter Schonwald wrote:
hey all, It seems to me (and i've certainly heard other people echo this sentiment) that: ghc dev chatting gets buried in the huge volume of commit + build report emails, and that creates (perhaps) another barrier to involvement in ghc dev at the hobbyist (rather than part time/full time ) scale?
So my question for the community (and of course current ghc devs )
1) do others agree that theres value in separating the two?
2) would this just be another use of the ghc-users list or would it be worth having a ghc-dev list?
3) most importantly, would the folks actively involved in ghc dev be willing/able to do so?
+1 This is a good idea, one that I have solved at my end using Procmail. So even though I have solved it for me, I still approve of this suggestion since it seems silly to push this to the receiver's end when it could so easily be solved by having separate lists. Erik -- ---------------------------------------------------------------------- Erik de Castro Lopo http://www.mega-nerd.com/
Somebody claiming to be Erik de Castro Lopo wrote:
This is a good idea, one that I have solved at my end using Procmail.
I was planning to do this soon (as I get deeper into GHC hacking because of my port, but don't want the high-email-volume of the comitt and build emails). Procmail is awesome. -- Stephen Paul Weber, @singpolyma See http://singpolyma.net for how I prefer to be contacted edition right joseph
I think we already mostly have this. The separation (remembering something I read off Trac I think,) was something like: * cvs-ghc mailing list: Prospective patches, automatically generated build/commit emails, etc. * glasgow-haskell-users mailing list: Developers and users of GHC discussing issues of design, process, bugs or fielding questions et cetera Doing some counting: looking at my email (which is gmail,) I have the last 100 most recent email *threads* open from cvs-ghc. The 100th dates to Nov 30th. I count about ~7 instances of non-build-or-commit-generated discussions in this timeframe (that is, ~7 email threads where the content wasn't just a build email, but a follow up, or it was actually it's own thread.) Almost all of these instances are related to patches or patches that were committed and some talk happened about them. One (maybe two) were about build/test failures. In contrast, for glasgow-haskell-users, for the past 100 threads: the 100th thread dates to July 12th. However, the structure of all these conversations is far, far more rich: just about every single thread contains far more discussion, almost 100% of them containing at least 2-3 emails. Some go as far as 20-70 emails deep, and many have several handfuls of responses. This is just for the past few months but I've been subscribed to both for years, so I'd say that sounds about right.
From this, I would infer that most of the interesting discussion does, in fact, happen on glasgow-haskell-users. There are perhaps fewer topics at a less regular pace, but it's really here. In contrast, cvs-ghc is mostly autogenerated stuff regular developers etc care about, with the sprinkling of discussion, almost all of it for patches or failures it seems.
The thing is, I would think that most people who are trying to get
into routinely hacking on GHC would want to be subscribed to both
anyway. Personally I think commits are the most valuable asset I can
see at a glance. But Johan has discussed having Jenkins performance
regressions sent out in light of the recent performance Tsar work, and
(some of) the bots do report success nightly here. So for people who
aren't submitting a benign patch or doing one-off-work, you are
probably at least going to want that stuff on your radar. But even
just looking at the work others are doing can give you ideas, or give
you familiarity with something (it's even exciting to read patches
sometimes!)
Overall I think the current separation is actually pretty good. Most
of the real meaty questions are asked here, and the developers
routinely sit on cvs-ghc to talk about patches or build failures and
the like. However, I do agree that sometimes you can miss out on
interesting discussion that happens there. Perhaps the rules could be
changed so that:
* cvs-ghc is only for automatically-generated content like build
emails, regressions, and commits.
* glasgow-haskell-users is for everything else, including patch review
(which also happens on trac) or discussing failures/regressions.
I don't know how attached I am to the current scheme, but perhaps this
sounds better to some, and I thought I could offer .02c having been
listening here for a while. :)
On Wed, Dec 5, 2012 at 8:31 PM, Carter Schonwald
hey all, It seems to me (and i've certainly heard other people echo this sentiment) that: ghc dev chatting gets buried in the huge volume of commit + build report emails, and that creates (perhaps) another barrier to involvement in ghc dev at the hobbyist (rather than part time/full time ) scale?
So my question for the community (and of course current ghc devs )
1) do others agree that theres value in separating the two?
2) would this just be another use of the ghc-users list or would it be worth having a ghc-dev list?
3) most importantly, would the folks actively involved in ghc dev be willing/able to do so?
That said, it does seem like the majority of the interesting ghc-dev chatter is on Trac issues, which is a good thing, though theres lots of interesting wee nuggets buried on the ghc-cvs list intermittently
I hope this question makes sense for y'all! -Carter
_______________________________________________ Glasgow-haskell-users mailing list Glasgow-haskell-users@haskell.org http://www.haskell.org/mailman/listinfo/glasgow-haskell-users
-- Regards, Austin
My own understanding is this:
A GHC *user* is someone who uses GHC, but doesn't care how it is implemented.
A GHC *developer* is someone who wants to work on GHC itself in some way.
The current mailing lists:
* glasgow-haskell-users: for anything that a GHC *user* cares about
* glasgow-haskell-bugs: same, but with a focus on bug reporting
* cvs-ghc: for GHC *developers*
I don't think we want to confuse users with developers. If we flood users with dev-related conversations they'll get fed up.
I don't see a very useful distinction between glasgow-haskell-users and glasgow-haskell-bugs. The distinction was very important before we had a bug tracker, but it doesn't seem useful now.
I can see a perhaps-useful distinction between two *developer* lists
(A) human email about implementation aspects of GHC
(B) machine-generated email from buildbots etc
I rather think that (A) could usefully include Trac ticket creation and Git commit messages, since both are really human-generated. So that would leave only buildbot logs on (B).
So I would be content to
* Abolish glasgow-haskell-bugs in favour of glasgow-haskell-users
* Split out cvs-ghc into two in some way; details to be agreed.
But for me the issue is not a pressing one.
Simon
| -----Original Message-----
| From: glasgow-haskell-users-bounces@haskell.org [mailto:glasgow-haskell-
| users-bounces@haskell.org] On Behalf Of Austin Seipp
| Sent: 06 December 2012 03:21
| To: Carter Schonwald
| Cc: GHC Users List
| Subject: Re: proposal: separate lists for ghc-cvs commits and ghc-dev
| chatter
|
| I think we already mostly have this. The separation (remembering
| something I read off Trac I think,) was something like:
|
| * cvs-ghc mailing list: Prospective patches, automatically generated
| build/commit emails, etc.
| * glasgow-haskell-users mailing list: Developers and users of GHC
| discussing issues of design, process, bugs or fielding questions et
| cetera
|
| Doing some counting: looking at my email (which is gmail,) I have the
| last 100 most recent email *threads* open from cvs-ghc. The 100th dates
| to Nov 30th. I count about ~7 instances of non-build-or-commit-generated
| discussions in this timeframe (that is,
| ~7 email threads where the content wasn't just a build email, but a
| follow up, or it was actually it's own thread.) Almost all of these
| instances are related to patches or patches that were committed and some
| talk happened about them. One (maybe two) were about build/test
| failures.
|
| In contrast, for glasgow-haskell-users, for the past 100 threads: the
| 100th thread dates to July 12th. However, the structure of all these
| conversations is far, far more rich: just about every single thread
| contains far more discussion, almost 100% of them containing at least
| 2-3 emails. Some go as far as 20-70 emails deep, and many have several
| handfuls of responses. This is just for the past few months but I've
| been subscribed to both for years, so I'd say that sounds about right.
|
| From this, I would infer that most of the interesting discussion does,
| in fact, happen on glasgow-haskell-users. There are perhaps fewer topics
| at a less regular pace, but it's really here. In contrast, cvs-ghc is
| mostly autogenerated stuff regular developers etc care about, with the
| sprinkling of discussion, almost all of it for patches or failures it
| seems.
|
| The thing is, I would think that most people who are trying to get into
| routinely hacking on GHC would want to be subscribed to both anyway.
| Personally I think commits are the most valuable asset I can see at a
| glance. But Johan has discussed having Jenkins performance regressions
| sent out in light of the recent performance Tsar work, and (some of) the
| bots do report success nightly here. So for people who aren't submitting
| a benign patch or doing one-off-work, you are probably at least going to
| want that stuff on your radar. But even just looking at the work others
| are doing can give you ideas, or give you familiarity with something
| (it's even exciting to read patches
| sometimes!)
|
| Overall I think the current separation is actually pretty good. Most of
| the real meaty questions are asked here, and the developers routinely
| sit on cvs-ghc to talk about patches or build failures and the like.
| However, I do agree that sometimes you can miss out on interesting
| discussion that happens there. Perhaps the rules could be changed so
| that:
|
| * cvs-ghc is only for automatically-generated content like build emails,
| regressions, and commits.
| * glasgow-haskell-users is for everything else, including patch review
| (which also happens on trac) or discussing failures/regressions.
|
| I don't know how attached I am to the current scheme, but perhaps this
| sounds better to some, and I thought I could offer .02c having been
| listening here for a while. :)
|
| On Wed, Dec 5, 2012 at 8:31 PM, Carter Schonwald
|
On Thu, Dec 6, 2012 at 1:29 PM, Simon Peyton-Jones wrote:
My own understanding is this:
A GHC *user* is someone who uses GHC, but doesn't care how it is implemented. A GHC *developer* is someone who wants to work on GHC itself in some way.
The current mailing lists:
* glasgow-haskell-users: for anything that a GHC *user* cares about * glasgow-haskell-bugs: same, but with a focus on bug reporting * cvs-ghc: for GHC *developers*
I don't think we want to confuse users with developers. If we flood users with dev-related conversations they'll get fed up.
I don't see a very useful distinction between glasgow-haskell-users and glasgow-haskell-bugs. The distinction was very important before we had a bug tracker, but it doesn't seem useful now.
I can see a perhaps-useful distinction between two *developer* lists (A) human email about implementation aspects of GHC (B) machine-generated email from buildbots etc
I rather think that (A) could usefully include Trac ticket creation and Git commit messages, since both are really human-generated.
I think the last two things (tickets and commit messages) should be separate from a mailing that is intended for (email-only) discussion. The content may be human-generated, but: (1) The number of messages is overwhelming. Alternatively stated, if you consider each ticket or commit message a different thread (which many email clients do), the number of different threads is large. (2) The commit messages do not all lead to conversations, and most of the discussion on tickets takes place on Trac with every message duplicated to the list. Consequently, any email-only discussion threads on the mailing list can easily get lost among all the other threads.
So that would leave only buildbot logs on (B).
So I would be content to * Abolish glasgow-haskell-bugs in favour of glasgow-haskell-users * Split out cvs-ghc into two in some way; details to be agreed.
But for me the issue is not a pressing one.
I identify the following different needs: (1) User email discussion (2) Developer email discussion (3) Buildbot reports (4) Trac reports (5) Commit messages Users will be interested in (1). Developers or followers of GHC development are probably interested in (1) and (2) but not necessarily (3 - 5). Maintainers or "serious" developers are probably interested in some combination of the above, including (1) and (2). People who track bugs (in general or for fun) would be interested in (4). Since Trac tickets have a CC option, many people probably don't need to subscribe to this list. Developers who police development would be interested in (3 - 5). By having a separate mailing list for each of (1) and (2), the email-only discussion does not get overwhelmed by the traffic of (3 - 5). And a separate mailing for each of (3 - 5) would allow the interested parties to subscribe to only what they want. Also, if you want to respond to an email on (3 - 5), then you are probably already subscribed to (2), so you can CC (2) in your response, thus bringing more developers into the conversation. (Personally, I've been using filters in Gmail to mark-read stuff that I'm not interested in, but I would be even happier to have the mailing lists further segmented.) Regards, Sean
* Sean Leather
I think the last two things (tickets and commit messages) should be separate from a mailing that is intended for (email-only) discussion. The content may be human-generated, but:
(1) The number of messages is overwhelming. Alternatively stated, if you consider each ticket or commit message a different thread (which many email clients do), the number of different threads is large. (2) The commit messages do not all lead to conversations, and most of the discussion on tickets takes place on Trac with every message duplicated to the list.
+1. I'd like to follow GHC development discussions, but getting all the commits is too much. Roman
On Thu, Dec 06, 2012 at 06:25:49PM +0200, Roman Cheplyaka wrote:
+1. I'd like to follow GHC development discussions, but getting all the commits is too much.
I'm surprised by this, FWIW. I think skimming the commits is a good way to get an idea of what's going on, while discussions between developers tend to be focussed on particular obscure points (e.g. discussing correctness of a murky corner in the intersection between 2 new type system extensions, or discussing the way PIC is handled on OSX/PowerPC) which I wouldn't have thought were of much interest to any party not involved in the discussion and familiar with the details. Anyway, I'm not really too fussed about what mailing lists we have. I'll just subscribe to them all anyway :-) Thanks Ian
I have the same opinion as Ian; I'm happy with the current setup, but if
things
change I'll just subscribe to everything anyway.
Cheers,
Pedro
On Thu, Dec 6, 2012 at 4:55 PM, Ian Lynagh
On Thu, Dec 06, 2012 at 06:25:49PM +0200, Roman Cheplyaka wrote:
+1. I'd like to follow GHC development discussions, but getting all the commits is too much.
I'm surprised by this, FWIW. I think skimming the commits is a good way to get an idea of what's going on, while discussions between developers tend to be focussed on particular obscure points (e.g. discussing correctness of a murky corner in the intersection between 2 new type system extensions, or discussing the way PIC is handled on OSX/PowerPC) which I wouldn't have thought were of much interest to any party not involved in the discussion and familiar with the details.
Anyway, I'm not really too fussed about what mailing lists we have. I'll just subscribe to them all anyway :-)
Thanks Ian
_______________________________________________ Glasgow-haskell-users mailing list Glasgow-haskell-users@haskell.org http://www.haskell.org/mailman/listinfo/glasgow-haskell-users
On Thu, Dec 6, 2012 at 5:55 PM, Ian Lynagh wrote:
On Thu, Dec 06, 2012 at 06:25:49PM +0200, Roman Cheplyaka wrote:
+1. I'd like to follow GHC development discussions, but getting all the commits is too much.
I'm surprised by this, FWIW. I think skimming the commits is a good way to get an idea of what's going on,
Indeed, it can be. But I can't keep up with the huge number of commits everyday, and I'm not interested in the large majority of them. while discussions between developers
tend to be focussed on particular obscure points (e.g. discussing correctness of a murky corner in the intersection between 2 new type system extensions, or discussing the way PIC is handled on OSX/PowerPC) which I wouldn't have thought were of much interest to any party not involved in the discussion and familiar with the details.
I find these things (both examples, actually) more interesting. I'm more likely to follow such threads than I am to skim the commits. Anyway, I'm not really too fussed about what mailing lists we have. I'll
just subscribe to them all anyway :-)
I'm not too fussed about it, either, but I wouldn't mind a restructuring. I might even subscribe to them all as well, but then it will be easier to follow the commits at a different pace from the rest (without having to create a new filter, which I've been so far too lazy to do). Regards, Sean
I'd also like to follow discussions, but unsubscribed from cvs-ghc due to
the volume.
Maybe the reply-to field on cvs-ghc emails could be set to
glasgow-haskell-bugs? (or whatever the agreed upon dev list is called)
Then if someone replies to a bot/commit message on cvs-ghc, it
automatically goes to the place where active discussion happens, and those
of us that don't want so many emails won't have to dig through cvs-ghc to
find the interesting bits.
Andrew
On Thu, Dec 6, 2012 at 11:48 AM, Sean Leather
On Thu, Dec 6, 2012 at 5:55 PM, Ian Lynagh wrote:
On Thu, Dec 06, 2012 at 06:25:49PM +0200, Roman Cheplyaka wrote:
+1. I'd like to follow GHC development discussions, but getting all the commits is too much.
I'm surprised by this, FWIW. I think skimming the commits is a good way to get an idea of what's going on,
Indeed, it can be. But I can't keep up with the huge number of commits everyday, and I'm not interested in the large majority of them.
while discussions between developers
tend to be focussed on particular obscure points (e.g. discussing correctness of a murky corner in the intersection between 2 new type system extensions, or discussing the way PIC is handled on OSX/PowerPC) which I wouldn't have thought were of much interest to any party not involved in the discussion and familiar with the details.
I find these things (both examples, actually) more interesting. I'm more likely to follow such threads than I am to skim the commits.
Anyway, I'm not really too fussed about what mailing lists we have. I'll
just subscribe to them all anyway :-)
I'm not too fussed about it, either, but I wouldn't mind a restructuring. I might even subscribe to them all as well, but then it will be easier to follow the commits at a different pace from the rest (without having to create a new filter, which I've been so far too lazy to do).
Regards, Sean
_______________________________________________ Glasgow-haskell-users mailing list Glasgow-haskell-users@haskell.org http://www.haskell.org/mailman/listinfo/glasgow-haskell-users
On 06/12/12 13:23, Sean Leather wrote:
On Thu, Dec 6, 2012 at 1:29 PM, Simon Peyton-Jones wrote:
My own understanding is this:
A GHC *user* is someone who uses GHC, but doesn't care how it is implemented. A GHC *developer* is someone who wants to work on GHC itself in some way.
The current mailing lists:
* glasgow-haskell-users: for anything that a GHC *user* cares about * glasgow-haskell-bugs: same, but with a focus on bug reporting * cvs-ghc: for GHC *developers*
I don't think we want to confuse users with developers. If we flood users with dev-related conversations they'll get fed up.
I don't see a very useful distinction between glasgow-haskell-users and glasgow-haskell-bugs. The distinction was very important before we had a bug tracker, but it doesn't seem useful now.
I can see a perhaps-useful distinction between two *developer* lists (A) human email about implementation aspects of GHC (B) machine-generated email from buildbots etc
I rather think that (A) could usefully include Trac ticket creation and Git commit messages, since both are really human-generated.
I think the last two things (tickets and commit messages) should be separate from a mailing that is intended for (email-only) discussion. The content may be human-generated, but:
(1) The number of messages is overwhelming. Alternatively stated, if you consider each ticket or commit message a different thread (which many email clients do), the number of different threads is large. (2) The commit messages do not all lead to conversations, and most of the discussion on tickets takes place on Trac with every message duplicated to the list.
Consequently, any email-only discussion threads on the mailing list can easily get lost among all the other threads.
So that would leave only buildbot logs on (B).
So I would be content to * Abolish glasgow-haskell-bugs in favour of glasgow-haskell-users * Split out cvs-ghc into two in some way; details to be agreed.
But for me the issue is not a pressing one.
I identify the following different needs:
(1) User email discussion (2) Developer email discussion (3) Buildbot reports (4) Trac reports (5) Commit messages
Sounds good to me. I like the idea of separating out the buildbot reports too, because there tends to be little signal in those (I typically only look at one per day, just to check whether there's anything really broken). Although that problem could be solved a different way, by having the build server emit a single email with a good summary once per day. Cheers, Simon
On Thu, Dec 06, 2012 at 12:29:01PM +0000, Simon Peyton-Jones wrote:
My own understanding is this:
A GHC *user* is someone who uses GHC, but doesn't care how it is implemented. A GHC *developer* is someone who wants to work on GHC itself in some way.
The current mailing lists:
* glasgow-haskell-users: for anything that a GHC *user* cares about * glasgow-haskell-bugs: same, but with a focus on bug reporting
I see glasgow-haskell-bugs as being mainly for developers, who want to see what bugs are coming in. It's true that we do give e-mailing it as a (less preferred) way for users to submit a bug on http://hackage.haskell.org/trac/ghc/wiki/ReportABug but I wonder if we shouldn't change that. It's rare that we get a bug report e-mailed, and normally we ultimately end up creating a trac ticket for it anyway. I'm sure that people who really want to submit a bug report and for whatever reason can't use trac will e-mail it somewhere sensible. Thanks Ian
Hello,
I also sometimes wish that GHC developer discussions happened in a single
place, separate from the commits and ticket lists (at present they are
sometimes on ghc-users and sometimes on cvs-ghc). For what it's worth, I
don't follow the commit and build-bot messages closely, instead I use
various git tools to catch up on what's going on with the development
(mostly I use gitk, git-gub, and our own git-web instance on
darcs.haskell.org). I do find the trac bug messages useful though, both
as a GHC user and a GHC developer.
-Iavor
On Thu, Dec 6, 2012 at 9:04 AM, Ian Lynagh
On Thu, Dec 06, 2012 at 12:29:01PM +0000, Simon Peyton-Jones wrote:
My own understanding is this:
A GHC *user* is someone who uses GHC, but doesn't care how it is implemented. A GHC *developer* is someone who wants to work on GHC itself in some way.
The current mailing lists:
* glasgow-haskell-users: for anything that a GHC *user* cares about * glasgow-haskell-bugs: same, but with a focus on bug reporting
I see glasgow-haskell-bugs as being mainly for developers, who want to see what bugs are coming in.
It's true that we do give e-mailing it as a (less preferred) way for users to submit a bug on http://hackage.haskell.org/trac/ghc/wiki/ReportABug but I wonder if we shouldn't change that. It's rare that we get a bug report e-mailed, and normally we ultimately end up creating a trac ticket for it anyway. I'm sure that people who really want to submit a bug report and for whatever reason can't use trac will e-mail it somewhere sensible.
Thanks Ian
_______________________________________________ Glasgow-haskell-users mailing list Glasgow-haskell-users@haskell.org http://www.haskell.org/mailman/listinfo/glasgow-haskell-users
On 06/12/12 17:04, Ian Lynagh wrote:
On Thu, Dec 06, 2012 at 12:29:01PM +0000, Simon Peyton-Jones wrote:
My own understanding is this:
A GHC *user* is someone who uses GHC, but doesn't care how it is implemented. A GHC *developer* is someone who wants to work on GHC itself in some way.
The current mailing lists:
* glasgow-haskell-users: for anything that a GHC *user* cares about * glasgow-haskell-bugs: same, but with a focus on bug reporting
I see glasgow-haskell-bugs as being mainly for developers, who want to see what bugs are coming in.
It's true that we do give e-mailing it as a (less preferred) way for users to submit a bug on http://hackage.haskell.org/trac/ghc/wiki/ReportABug but I wonder if we shouldn't change that. It's rare that we get a bug report e-mailed, and normally we ultimately end up creating a trac ticket for it anyway. I'm sure that people who really want to submit a bug report and for whatever reason can't use trac will e-mail it somewhere sensible.
+1. ghc-bugs used to be for user-generated bug reports, but now it is almost exclusively Trac-generated emails. I don't think anything is gained by suggesting that people email bug reports any more. Cheers, Simon
On Thu, Dec 06, 2012 at 09:15:06PM +0000, Simon Marlow wrote:
On 06/12/12 17:04, Ian Lynagh wrote:
It's true that we do give e-mailing it as a (less preferred) way for users to submit a bug on http://hackage.haskell.org/trac/ghc/wiki/ReportABug but I wonder if we shouldn't change that. It's rare that we get a bug report e-mailed, and normally we ultimately end up creating a trac ticket for it anyway. I'm sure that people who really want to submit a bug report and for whatever reason can't use trac will e-mail it somewhere sensible.
+1. ghc-bugs used to be for user-generated bug reports, but now it is almost exclusively Trac-generated emails. I don't think anything is gained by suggesting that people email bug reports any more.
I've removed that option from ReportABug. Thanks Ian
participants (12)
-
Andrew Farmer -
Austin Seipp -
Carter Schonwald -
Erik de Castro Lopo -
Ian Lynagh -
Iavor Diatchki -
José Pedro Magalhães -
Roman Cheplyaka -
Sean Leather -
Simon Marlow -
Simon Peyton-Jones -
Stephen Paul Weber