Menu
▾
▴
pidgin-devel — DO NOT USE. See https://pidgin.im/cgi-bin/mailman/listinfo/devel
You can subscribe to this list here.
| 2002 |
Jan
|
Feb
|
Mar
|
Apr
|
May
|
Jun
(27) |
Jul
(25) |
Aug
(21) |
Sep
(136) |
Oct
(123) |
Nov
(87) |
Dec
(110) |
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| 2003 |
Jan
(87) |
Feb
(88) |
Mar
(81) |
Apr
(255) |
May
(73) |
Jun
(96) |
Jul
(131) |
Aug
(94) |
Sep
(148) |
Oct
(171) |
Nov
(166) |
Dec
(172) |
| 2004 |
Jan
(251) |
Feb
(140) |
Mar
(213) |
Apr
(298) |
May
(182) |
Jun
(185) |
Jul
(159) |
Aug
(376) |
Sep
(334) |
Oct
(256) |
Nov
(217) |
Dec
(189) |
| 2005 |
Jan
(186) |
Feb
(151) |
Mar
(199) |
Apr
(115) |
May
(203) |
Jun
(228) |
Jul
(116) |
Aug
(189) |
Sep
(136) |
Oct
(198) |
Nov
(249) |
Dec
(339) |
| 2006 |
Jan
(167) |
Feb
(185) |
Mar
(95) |
Apr
(133) |
May
(86) |
Jun
(156) |
Jul
(149) |
Aug
(170) |
Sep
(208) |
Oct
(151) |
Nov
(270) |
Dec
(148) |
| 2007 |
Jan
(240) |
Feb
(127) |
Mar
(150) |
Apr
(40) |
May
|
Jun
|
Jul
|
Aug
|
Sep
|
Oct
|
Nov
|
Dec
|
|
From: Ethan B. <ebl...@cs...> - 2003-04-02 21:22:11
|
Vincent Ho spake unto us the following wisdom: > CVS commit logs don't correlate changes between files. This is one of > the key drawbacks of CVS IMHO, that commits are done on a per-file basis > and it has no concept that any changes to multiple files are linked. > Without tags, there is no simple way to say "find this change I did to > multiple files", and view/revert it. http://www.cobite.com/cvsps It seems to choke on generating patchset *patches* for gaim, but I find it very useful for browsing the logs. > Hence why I prefer the ChangeLog as a log of discrete changes, and NEWS > for the bird's eye view of release changes. But I digress; it seems the > main reason logging on a commit-by-commit basis to the ChangeLog is a > pain is because you guys have made it so =3D) How so? The point that I and Rob both made was that we don't want to see that kind of volume of information in the ChangeLog. The current method has what we want in the ChangeLog. Perhaps it is painful for you... Ethan --=20 "Among the many misdeeds of the British rule in India, history will look upon the act of depriving a whole nation of arms, as the blackest." -- Mahatma Gandhi |
|
From: Vincent Ho <lo...@in...> - 2003-04-02 21:05:41
|
On Wed, Apr 02, 2003 at 10:52:09AM -0500, Ethan Blanton wrote:
> Morten Brix Pedersen spake unto us the following wisdom:
>
> > The ChangeLog is not meant to track changes between releases. Its
> > meant to track code changes. The NEWS file is used to track changes
> > betweeen releases.
>
> That is very much an opinion ... and one I am inclined to disagree
> with. Since we have CVS logs to track commit-by-commit changes
> (although the commit messages are not always what they could be), I
> don't see any need to duplicate that information. I like to see major
> changes since the last release when I look at a ChangeLog.
CVS commit logs don't correlate changes between files. This is one of
the key drawbacks of CVS IMHO, that commits are done on a per-file basis
and it has no concept that any changes to multiple files are linked.
Without tags, there is no simple way to say "find this change I did to
multiple files", and view/revert it.
Hence why I prefer the ChangeLog as a log of discrete changes, and NEWS
for the bird's eye view of release changes. But I digress; it seems the
main reason logging on a commit-by-commit basis to the ChangeLog is a
pain is because you guys have made it so =)
Ah well, no skin off my nose really.
Cheers,
Vince
--
Vincent Ho
lo...@in...
QOTD:
"Everything I am today I owe to people, whom it is now
too late to punish."
|
|
From: Luke S. <lsc...@us...> - 2003-04-02 20:38:42
|
On Wed, Apr 02, 2003 at 12:40:41PM -0500, Richard Hoskins wrote: > Ethan Blanton <ebl...@cs...> writes: > > > Interface change discussion is, of course, welcome on this list. The > > sort of email which doesn't get a very good response is "I don't like > > X, make it how it used to be", or "Is there an option to make X how it > > used to be?". A well thought out email presenting pros and cons might > > make some progress ... a knee-jerk reaction to change will not. Users > > should keep in mind that we, as gaim developers, use gaim every day as > > well. What might initially seem like the end of the world (come on > > people, it's a freaking IM client ... if it has something you don't > > like in CVS for a week, the world will NOT end) may in fact be > > something you decide you like if you just give it a chance for a few > > days. > > Fair enough, but I don't see a lot of knee jerk reactions here. I see > no discussion of any issues, as a matter of fact. yeah, gaim-devel is VERY low traffic, we created it more as a way to get stuff out when a majority of us developers aren't present on irc at a given time. latter the wiki proved a more useful way of doing that. this has continued because it has developed its own nitche. still, by the time _any_ issue hits this list, we've heard about it umpteen million times in #gaim and/or in the sf forums. > > > It's very off-putting to have every single design change second- and > > third-guessed simply because a few (not by any means a majority!) > > users are so afraid of change and progress that they cannot handle the > > fact that, e.g., the tabs on their conversations are not optional for > > a _few days_. The developers of gaim work very hard to ensure that > > they are, in fact, *improving* gaim. Sometimes change is good. > > Fair enough again. I'm sure the developers are working very hard, and > I appreciate their work. But where's this second and third guessing > going on? It's not on this list. If anything, the knee jerk > reactions have been on the other side of the fence. the knee jerk reactions happen mostly in irc, some in the forum. you are correct, this list is ussually more sane. > > > > Take the new buddy list, for another example. When it was first > > committed, people were like "oh, gah! it's so huuuuge! wah wah can i > > have an option?". I know of at least a few people, however, that > > decided (in the two or three days, it wasn't that long!) they liked > > the bigger buddy list with icons and kept it even after the option to > > reduce it was readded. > > Perhaps the discussion of the new buddy list was warranted, then. It > did some good. Just because some acted like they wanted the world, > doesn't mean that there wasn't room for discussion. You don't want > feedback from the users? Are you going to treat all questions as > complaints, and all concerns as whining? it depends. seriously. we _try_ to take serious comments, substantive comments, seriously. sometimes we are having a bad day. sometimes we've had so many complaints we react before we realize that your comment is the first meaningful one of the day. this isn't good, but its kinda just part of us being human. > > > I'm relatively new to this whole gaim development thing, so I came in > > just before 0.59 was frozen and 0.60 went under heavy rewrite. I have > > seen in this past few months *countless* times where some feature X > > gets ripped out for rewrite, or is only partially reimplemented (or > > whatever), and people come literally flocking to #gaim (or, less > > often, this mailing list) begging to have this feature > > readded/removed/whatever so they can have the behavior they are used > > to back. Within HOURS or MINUTES. It's awfully hard to develop, make > > forward progress, and keep these people happy all at the same time. > > Thus you get short answers like "no, shut up" both here and in #gaim > > and in the SF forums. If you're the 200th user to say "when are > > conversation tabs going to be optional?", you can expect a terse and > > perhaps somewhat unfriendly response. > > Don't do IRC, so I won't comment on #gaim, but I don't see that > behavior on this list. I don't know what has gone on in the past. > (Is this list archived anywhere?) this is archived, i'm not sure where sf keeps the archives, gaim-commits is not, as the cvs -z3 log IS the archive. > > > I think it is also reasonable to expect people who wish to track gaim > > CVS to keep at least somewhat on top of the changes going in. If > > something shows up that I didn't expect, I don't come crying to this > > list or IRC first thing ... I peruse the recent CVS logs to try and > > determine if it is intentional or a bug. If it appears to be > > intentional, I give it a few hours to shake itself out before I start > > announcing that gaim has ended my life and career and exploded my > > monitor. If it seems to be a bug, I attempt to figure out why or get > > a backtrace or take whatever action seems most appropriate. Is it > > unreasonable to expect this out of users? Not at all, if they're > > using gaim from CVS and not a released version. As a matter of fact, > > if a user comes to #gaim and says "gaim is dumping core, but I don't > > know how to get a backtrace like the instructions say, can someone > > help me?", I'm sure that user would get more or less immediate help. > > Gaim dumped core on me, or at least that is what STDERR said, along > with a request for a backtrace. I couldn't find the core file to get > said backtrace, so I posted to this list. I was told "There's a reason > this is CVS. The bugs will be fixed when we find fixes." for some VERY odd reason, most linux distros have things set to suppress core files. run ulimit -c unlimited in the terminal you start gaim from to re-enable them (you can replace unlimited with a size as well). you might want to put that in your bash profile or rc or whathaveyou. luke > > > -- > Lift me down, so I can make the Earth tremble. > --Bucky Katt > > > > ------------------------------------------------------- > This SF.net email is sponsored by: ValueWeb: > Dedicated Hosting for just $79/mo with 500 GB of bandwidth! > No other company gives more support or power for your dedicated server > http://click.atdmt.com/AFF/go/sdnxxaff00300020aff/direct/01/ > _______________________________________________ > Gaim-devel mailing list > Gai...@li... > https://lists.sourceforge.net/lists/listinfo/gaim-devel -- -This email is made of 100% recycled electrons. |
|
From: Richard H. <rm...@ap...> - 2003-04-02 19:45:52
|
Rob Flynn <ro...@fa...> writes:
> We don't mean to come off as sounding rude. Sometimes we've just
> had a bad day, are sick, or have been bombarded all day with the
> same question. I think that we (gaim) have one of the more
> approachable development teams in comparison to some of the other
> projects out there.
Isn't this an artifact of these questions being asked/answered on IM
and IRC? I realize the 'eat your own dogfood' logic in an IM client
using IM and IRC for discussion, but if questions were redirected to
the mailing list, everybody would see them once[1], and have an
opportunity to answer them once. Also, you could read/answer them at
your convenience.
I don't know what has been asked in your IM, and even if I logged on
to #gaim, I don't know what has been asked yesterday...
Just a suggestion.
{1] Well, <12 times anyway... :)
--
Lift me down, so I can make the Earth tremble.
--Bucky Katt
|
|
From: Rob F. <ro...@fa...> - 2003-04-02 19:23:17
|
CVS is the development sandbox, and > people who want to play in the sandbox need too understand that the > Tonka's might be not work sometimes. > Hahahahahahahaha!! I like that analogy. - R |
|
From: Rob F. <ro...@fa...> - 2003-04-02 19:21:18
|
> Because, contrary to how this thread might look, we *do* get plenty of
> useful feedback. There are stretches of days at a time where all of
> the feedback is redundant and hardly useful, but fortunately they
> aren't contiguous. :-)
>
> Ethan
Exactly.
I want to say that we DO appreciate ALL feedback from our users. You're
right that there isn't a lot of discussion and/or reporting of bugs and
such on the mailing list. I gets MANY IMs each day from people who usually
all say the same thing.
We don't mean to come off as sounding rude. Sometimes we've just had a bad
day, are sick, or have been bombarded all day with the same question. I
think that we (gaim) have one of the more approachable development teams in
comparison to some of the other projects out there.
We usually try to take the time to listen to what you've got to say.
Granted, a lot of times when I get an IM, I respond very shortly and just
sa "busy. email me." I don't do this to be rude, but I have a regular job
where I have to get things done, so I just send people to e-mail. I do,
however, reply to their message eventually.
We like the fact that we allow people to use our CVS, to help out with
development, and to talk to us and join in on discussions about it. That's
what open source is all about. That's what having a good project is all
about.
I think that after this week, everything will be cool and back to normal. I
think we're all just getting anxious. We're already behind our initial
schedule on releasing 0.60. We, as developers, want it released just as
much as you, as the users, do.
After that, we can kick back and have a nice {cold|warm} glass of
{beer|milk|soda} and enjoy the world again.
This has been a public service announcement by a rambling idiot. ;-)
Rob
|
|
From: Matthew K. <kel...@po...> - 2003-04-02 19:13:07
|
On Wed, 2003-04-02 at 12:56, Sean Egan wrote: > These people get annoyed when something breaks as they depend on CVS > Gaim to be perfect. Just as an aside, the reason that Samba-TNG forked from Samba is because there was SO much user-pressure on the Samba developers to always have a perfect CVS, that major changes were impossible. I, as a Gaim user, would much prefer CVS breaking, then Gaim forking- Or worse, developers feeling disenfranchised. CVS is the development sandbox, and people who want to play in the sandbox need too understand that the Tonka's might be not work sometimes. My $0.02. -- Matthew Keller Enterprise Systems Analyst Computing & Technology Services State University of New York @ Potsdam Potsdam, NY USA http://mattwork.potsdam.edu/ |
|
From: Ethan B. <ebl...@cs...> - 2003-04-02 19:05:20
|
Richard Hoskins spake unto us the following wisdom:
> Ethan Blanton <ebl...@cs...> writes:
> > As far as not finding a core file, I *personally* expect people who
> > wish to follow development versions of software to know how to
> > enable core dumpage ... (ulimit -c unlimited). Maybe I set the bar
> > too high. In either case, perhaps we should add some text to
> > gdb.php explaining how to cause your machine to dump cores if it
> > does not by default.
>=20
> OK.
> But, I *personally* expect not to be treated rudely. I was trying to
> help. There you go. Maybe I set the bar to high.
Hmm, my point was not that there was any justification to treating you
rudely, but I realize that what I wrote doesn't look very good as I
reread it. I apologize.
> You complain when your lusers just whine, and you're rude when a luser
> tries to help, but his help isn't up to your standards.
There is a problem here with becoming jaded ... if someone tries to
help but needs a little push, and it's a problem we haven't seen
before, it's easy to give that little push. If someone tries to help
with a problem we've seen a dozen times and needs a little push, it's
very easy to snap and say "stuff it". That's not the fault of the
person reporting the bug, of course ...
> Why don't you guys just close the CVS and this mailing list? (Serious
> question.) They don't seem to be doing anything but annoying you
> anyway.
Because, contrary to how this thread might look, we *do* get plenty of
useful feedback. There are stretches of days at a time where all of
the feedback is redundant and hardly useful, but fortunately they
aren't contiguous. :-)
Ethan
--=20
"Among the many misdeeds of the British rule in India, history will
look upon the act of depriving a whole nation of arms, as the
blackest."
-- Mahatma Gandhi
|
|
From: Richard H. <rm...@ap...> - 2003-04-02 18:47:36
|
Ethan Blanton <ebl...@cs...> writes:
> As far as not finding a core file, I *personally* expect people who
> wish to follow development versions of software to know how to
> enable core dumpage ... (ulimit -c unlimited). Maybe I set the bar
> too high. In either case, perhaps we should add some text to
> gdb.php explaining how to cause your machine to dump cores if it
> does not by default.
OK.
But, I *personally* expect not to be treated rudely. I was trying to
help. There you go. Maybe I set the bar to high.
You complain when your lusers just whine, and you're rude when a luser
tries to help, but his help isn't up to your standards.
Why don't you guys just close the CVS and this mailing list? (Serious
question.) They don't seem to be doing anything but annoying you
anyway.
--
Lift me down, so I can make the Earth tremble.
--Bucky Katt
|
|
From: Ethan B. <ebl...@cs...> - 2003-04-02 18:24:46
|
Richard Hoskins spake unto us the following wisdom:
> Fair enough, but I don't see a lot of knee jerk reactions here. I see
> no discussion of any issues, as a matter of fact.
This is true, the gaim-devel list tends to be more reasonable if you
look at it in a vacuum ... I have a tendency to correlate posts to
gaim-devel with comments in #gaim on IRC. Note that many of my
comments addressed both forums. It seems that you don't follow #gaim,
so I can see how parts of my message wouldn't have made much sense to
you.
> > Take the new buddy list, for another example. When it was first
> > committed, people were like "oh, gah! it's so huuuuge! wah wah can i
> > have an option?". I know of at least a few people, however, that
> > decided (in the two or three days, it wasn't that long!) they liked
> > the bigger buddy list with icons and kept it even after the option to
> > reduce it was readded.
>=20
> Perhaps the discussion of the new buddy list was warranted, then. It
> did some good. Just because some acted like they wanted the world,
> doesn't mean that there wasn't room for discussion. You don't want
> feedback from the users? Are you going to treat all questions as
> complaints, and all concerns as whining?
What good? Did you think the option for a smaller blist wasn't going
to go back in eventually? My point was that a few of the users who
cried the loudest decided they didn't even need the option when it
*did* go back in.
> Gaim dumped core on me, or at least that is what STDERR said, along
> with a request for a backtrace. I couldn't find the core file to get
> said backtrace, so I posted to this list. I was told "There's a reason
> this is CVS. The bugs will be fixed when we find fixes."
I do not recall this exact incident, but that was probably because it
was a well known bug which had been beaten to death in #gaim.
As far as not finding a core file, I *personally* expect people who
wish to follow development versions of software to know how to enable
core dumpage ... (ulimit -c unlimited). Maybe I set the bar too high.
In either case, perhaps we should add some text to gdb.php explaining
how to cause your machine to dump cores if it does not by default.
Ethan
--=20
"Among the many misdeeds of the British rule in India, history will
look upon the act of depriving a whole nation of arms, as the
blackest."
-- Mahatma Gandhi
|
|
From: Richard H. <rm...@ap...> - 2003-04-02 18:14:20
|
Ka-Hing Cheung <kh...@ja...> writes: > We want feedback from users, but not in the form of "This is bad and I > don't like it." Not saying this happened on this list, but it sure > happens a lot on #gaim. Sorry to hear that. > In this case your system probably set a size limit on core file to 0. > Try 'ulimit -c unlimited' then run gaim from the same terminal. After > you get a core see http://gaim.sf.net/gdb.php for how to get a bt. Thank you. It has stopped dumping core, however. At least for the time being. :) > Other than the bt, it is often useful to tell us what you were doing at > the time of segfault. Other information such as distro, gtk versions are > less helpful, but they don't hurt. For example, recently we have a > segfault-on-start problem with Gentoo users only. Will do. > May I also add that it's not very useful to report bugs that you > cannot reproduce? Obviously. > Also, a quick search on the sf tracker, this emailing list, and the > sf forum would help prevent multiple bug reports. If we get the same > bug report over and over again, it's easy to get annoyed by it. Understood. Thanks again. -- Lift me down, so I can make the Earth tremble. --Bucky Katt |
|
From: Ka-Hing C. <kh...@ja...> - 2003-04-02 18:02:35
|
On Wed, 2003-04-02 at 09:40, Richard Hoskins wrote: > Ethan Blanton <ebl...@cs...> writes: > Perhaps the discussion of the new buddy list was warranted, then. It > did some good. Just because some acted like they wanted the world, > doesn't mean that there wasn't room for discussion. You don't want > feedback from the users? Are you going to treat all questions as > complaints, and all concerns as whining? > We want feedback from users, but not in the form of "This is bad and I don't like it." Not saying this happened on this list, but it sure happens a lot on #gaim. > Gaim dumped core on me, or at least that is what STDERR said, along > with a request for a backtrace. I couldn't find the core file to get > said backtrace, so I posted to this list. I was told "There's a reason > this is CVS. The bugs will be fixed when we find fixes." In this case your system probably set a size limit on core file to 0. Try 'ulimit -c unlimited' then run gaim from the same terminal. After you get a core see http://gaim.sf.net/gdb.php for how to get a bt. Other than the bt, it is often useful to tell us what you were doing at the time of segfault. Other information such as distro, gtk versions are less helpful, but they don't hurt. For example, recently we have a segfault-on-start problem with Gentoo users only. May I also add that it's not very useful to report bugs that you cannot reproduce? Also, a quick search on the sf tracker, this emailing list, and the sf forum would help prevent multiple bug reports. If we get the same bug report over and over again, it's easy to get annoyed by it. -- Ka-Hing "A complex system that works is invariably found to have evolved from a simple system that worked." -- John Gall, _Systemantics_ |
|
From: Sean E. <sea...@bi...> - 2003-04-02 17:56:51
|
On Wed, 2003-04-02 at 12:19, Richard Hoskins wrote: > Asking a question does not constitute complaining, or expecting > anything, let alone a release-quality app. When you hear the same question from a hundred different people, the net affect is the same. My two cents: CVS is by definition incopmlete. Because we haven't had a major release in 9 months, too many people (read: people who don't actually care about development) are using CVS (a normal Gaim release has around 40,000 downloads from sourceforge. The latest stable has 8,000 downloads. The CVS nightlies take about 13GB/Week.) These people get annoyed when something breaks as they depend on CVS Gaim to be perfect. Everytime something gets partially implemented or temporarily removed, #gaim, this mailing list, our mailboxes and IM windows get filled with people "asking questions" (read: complaining). This is annoying. Very annyoing. 0.60 really will be released this week, and then we'll return to our normal 2-week release cycle. Then, people who don't care about Gaim development can use releases in which we promise everything works, and you can complain your little hearts out if it doesn't. -s. |
|
From: Richard H. <rm...@ap...> - 2003-04-02 17:41:02
|
Ethan Blanton <ebl...@cs...> writes:
> Interface change discussion is, of course, welcome on this list. The
> sort of email which doesn't get a very good response is "I don't like
> X, make it how it used to be", or "Is there an option to make X how it
> used to be?". A well thought out email presenting pros and cons might
> make some progress ... a knee-jerk reaction to change will not. Users
> should keep in mind that we, as gaim developers, use gaim every day as
> well. What might initially seem like the end of the world (come on
> people, it's a freaking IM client ... if it has something you don't
> like in CVS for a week, the world will NOT end) may in fact be
> something you decide you like if you just give it a chance for a few
> days.
Fair enough, but I don't see a lot of knee jerk reactions here. I see
no discussion of any issues, as a matter of fact.
> It's very off-putting to have every single design change second- and
> third-guessed simply because a few (not by any means a majority!)
> users are so afraid of change and progress that they cannot handle the
> fact that, e.g., the tabs on their conversations are not optional for
> a _few days_. The developers of gaim work very hard to ensure that
> they are, in fact, *improving* gaim. Sometimes change is good.
Fair enough again. I'm sure the developers are working very hard, and
I appreciate their work. But where's this second and third guessing
going on? It's not on this list. If anything, the knee jerk
reactions have been on the other side of the fence.
> Take the new buddy list, for another example. When it was first
> committed, people were like "oh, gah! it's so huuuuge! wah wah can i
> have an option?". I know of at least a few people, however, that
> decided (in the two or three days, it wasn't that long!) they liked
> the bigger buddy list with icons and kept it even after the option to
> reduce it was readded.
Perhaps the discussion of the new buddy list was warranted, then. It
did some good. Just because some acted like they wanted the world,
doesn't mean that there wasn't room for discussion. You don't want
feedback from the users? Are you going to treat all questions as
complaints, and all concerns as whining?
> I'm relatively new to this whole gaim development thing, so I came in
> just before 0.59 was frozen and 0.60 went under heavy rewrite. I have
> seen in this past few months *countless* times where some feature X
> gets ripped out for rewrite, or is only partially reimplemented (or
> whatever), and people come literally flocking to #gaim (or, less
> often, this mailing list) begging to have this feature
> readded/removed/whatever so they can have the behavior they are used
> to back. Within HOURS or MINUTES. It's awfully hard to develop, make
> forward progress, and keep these people happy all at the same time.
> Thus you get short answers like "no, shut up" both here and in #gaim
> and in the SF forums. If you're the 200th user to say "when are
> conversation tabs going to be optional?", you can expect a terse and
> perhaps somewhat unfriendly response.
Don't do IRC, so I won't comment on #gaim, but I don't see that
behavior on this list. I don't know what has gone on in the past.
(Is this list archived anywhere?)
> I think it is also reasonable to expect people who wish to track gaim
> CVS to keep at least somewhat on top of the changes going in. If
> something shows up that I didn't expect, I don't come crying to this
> list or IRC first thing ... I peruse the recent CVS logs to try and
> determine if it is intentional or a bug. If it appears to be
> intentional, I give it a few hours to shake itself out before I start
> announcing that gaim has ended my life and career and exploded my
> monitor. If it seems to be a bug, I attempt to figure out why or get
> a backtrace or take whatever action seems most appropriate. Is it
> unreasonable to expect this out of users? Not at all, if they're
> using gaim from CVS and not a released version. As a matter of fact,
> if a user comes to #gaim and says "gaim is dumping core, but I don't
> know how to get a backtrace like the instructions say, can someone
> help me?", I'm sure that user would get more or less immediate help.
Gaim dumped core on me, or at least that is what STDERR said, along
with a request for a backtrace. I couldn't find the core file to get
said backtrace, so I posted to this list. I was told "There's a reason
this is CVS. The bugs will be fixed when we find fixes."
--
Lift me down, so I can make the Earth tremble.
--Bucky Katt
|
|
From: Richard H. <rm...@ap...> - 2003-04-02 17:19:23
|
Rob Flynn <ro...@fa...> writes:
> My main problem comes from the fact that some expect what is in CVS
> to be release-quality. They don't want a half completed feature.
> If some UI is not fully implemented then they will complain. That
> is mostly what I was complaining about. The changes made yesterday
> are an exception being that it was April Fool's day.
Asking a question does not constitute complaining, or expecting
anything, let alone a release-quality app.
But you know that.
Hope you are feeling better.
--
Lift me down, so I can make the Earth tremble.
--Bucky Katt
|
|
From: Nathan W. <fac...@fa...> - 2003-04-02 17:16:52
|
On Wed, Apr 02, 2003 at 11:08:32AM -0500, Scott Henson wrote: > Maybe starting to archive it on the web somewhere would be nice. Im not > sure what the volume on it is, but I would rather not add another list > to my mailbox. Also if it was archived it would be much simpler to go > back and track the changes instead of having to keep every commit > message. =20 Once 0.60 releases (real soon now, really), people can go back to: a) using releases, and reading the changelog to find out what has changed b) using CVS, and subscribing to gaim-commits to watch what is changing. once we release, the only people who should be using cvs are those willing to keep up with the changes, or those willing to put up with it. hell, there's a digest for gaim-commits, it won't ruin your life to get a daily email about what we're doing. =20 If it _will_ ruin your life to get an extra email or two per day, you can always go read about stuff in the web interface to cvs that sourceforge is so nice to provide. If any of the above options are too much work for you, then you shouldn't be using cvs. You should be waiting for us to tell you "it's ok to use it now", which we say in the form of a release. Nathan --=20 Nathan Walp || fac...@fa... GPG Fingerprint: || http://faceprint.com/ 5509 6EF3 928B 2363 9B2B DA17 3E46 2CDC 492D DB7E |
|
From: Ethan B. <ebl...@cs...> - 2003-04-02 17:13:35
|
Richard Hoskins spake unto us the following wisdom:
> Perhaps interface changes can be discussed on this list as well. That
> doesn't seem to be happening, and when a user asks a question about
> them here, they get the stock "it's CVS. Things change. Deal with
> it. Read the Logs."
Interface change discussion is, of course, welcome on this list. The
sort of email which doesn't get a very good response is "I don't like
X, make it how it used to be", or "Is there an option to make X how it
used to be?". A well thought out email presenting pros and cons might
make some progress ... a knee-jerk reaction to change will not. Users
should keep in mind that we, as gaim developers, use gaim every day as
well. What might initially seem like the end of the world (come on
people, it's a freaking IM client ... if it has something you don't
like in CVS for a week, the world will NOT end) may in fact be
something you decide you like if you just give it a chance for a few
days.
It's very off-putting to have every single design change second- and
third-guessed simply because a few (not by any means a majority!)
users are so afraid of change and progress that they cannot handle the
fact that, e.g., the tabs on their conversations are not optional for
a _few days_. The developers of gaim work very hard to ensure that
they are, in fact, *improving* gaim. Sometimes change is good.
Take the new buddy list, for another example. When it was first
committed, people were like "oh, gah! it's so huuuuge! wah wah can i
have an option?". I know of at least a few people, however, that
decided (in the two or three days, it wasn't that long!) they liked
the bigger buddy list with icons and kept it even after the option to
reduce it was readded.
I'm relatively new to this whole gaim development thing, so I came in
just before 0.59 was frozen and 0.60 went under heavy rewrite. I have
seen in this past few months *countless* times where some feature X
gets ripped out for rewrite, or is only partially reimplemented (or
whatever), and people come literally flocking to #gaim (or, less
often, this mailing list) begging to have this feature
readded/removed/whatever so they can have the behavior they are used
to back. Within HOURS or MINUTES. It's awfully hard to develop, make
forward progress, and keep these people happy all at the same time.
Thus you get short answers like "no, shut up" both here and in #gaim
and in the SF forums. If you're the 200th user to say "when are
conversation tabs going to be optional?", you can expect a terse and
perhaps somewhat unfriendly response.
I think it is also reasonable to expect people who wish to track gaim
CVS to keep at least somewhat on top of the changes going in. If
something shows up that I didn't expect, I don't come crying to this
list or IRC first thing ... I peruse the recent CVS logs to try and
determine if it is intentional or a bug. If it appears to be
intentional, I give it a few hours to shake itself out before I start
announcing that gaim has ended my life and career and exploded my
monitor. If it seems to be a bug, I attempt to figure out why or get
a backtrace or take whatever action seems most appropriate. Is it
unreasonable to expect this out of users? Not at all, if they're
using gaim from CVS and not a released version. As a matter of fact,
if a user comes to #gaim and says "gaim is dumping core, but I don't
know how to get a backtrace like the instructions say, can someone
help me?", I'm sure that user would get more or less immediate help.
But this has become tangential to the original discussion of UI
changes, so I'll let it rest.
> Which, of course, are after the fact. (Still hoping the input on top
> is an April Fools change.)
Why? What's wrong with it? Would it be better if we moved the
toolbar from the bottom to the top, to bring it closer to the input
box? Do you understand why randomly slamming the change without
providing any useful input is pointless?
Ethan
--=20
"Among the many misdeeds of the British rule in India, history will
look upon the act of depriving a whole nation of arms, as the
blackest."
-- Mahatma Gandhi
|
|
From: Luke S. <lsc...@ra...> - 2003-04-02 17:05:44
|
On Wed, Apr 02, 2003 at 11:32:32AM -0500, Richard Hoskins wrote: > Scott Henson <deb...@si...> writes: > > > On Wed, 2003-04-02 at 10:52, Ethan Blanton wrote: > >> That is very much an opinion ... and one I am inclined to disagree > >> with. Since we have CVS logs to track commit-by-commit changes > >> (although the commit messages are not always what they could be), I > >> don't see any need to duplicate that information. I like to see major > >> changes since the last release when I look at a ChangeLog. > > Maybe starting to archive it on the web somewhere would be nice. Im not > > sure what the volume on it is, but I would rather not add another list > > to my mailbox. Also if it was archived it would be much simpler to go > > back and track the changes instead of having to keep every commit > > message. > > Perhaps interface changes can be discussed on this list as well. That > doesn't seem to be happening, and when a user asks a question about > them here, they get the stock "it's CVS. Things change. Deal with > it. Read the Logs." when someone complains about something not working, ie the buddy list not having the ability to edit, that is the responce, yes. basically, any time someone "reports" a bug that its impossible to use cvs w/out noticing, they will get that responce. any time someone says something in this list that was mentioned in the commit log as "this doesn't work yet, but here is some code for it" then they will also get that responce. gaim-commits has a very variable traffic, if you are worried about volume, i suggest the digest of it, which will generate an email 1-2 times a day. as Ethan and I have both said now, we are using the commit log to notate the commital of bug fixes, improvements, and anything else that either 1)already has a changelog entry or 2)doesn't warrent one (ie the gtk2-ification of yet another dialog doesn't need yet another changelog entry). gaim-commits mails you the commit log, the list of modified files, and the patch version of the commit. i'm not sure what all of that is in the digest, the logs themselves at very least are. interface changes themselves happen in a variety of media, which reflects the variety of ways we developers are in touch with each other. some of it happens on the list, more of it happens over irc. some of it happens in the wiki, some of happens in im conversations between a pair of developers. we aren't going to spend time duplicating this discussion across media just because the subset of users on gaim-devel isn't the same as the subset in #gaim isn't the same as the subset elsewhere, its too much work. luke > > Which, of course, are after the fact. (Still hoping the input on top > is an April Fools change.) > > -- > Lift me down, so I can make the Earth tremble. > --Bucky Katt > > > > ------------------------------------------------------- > This SF.net email is sponsored by: ValueWeb: > Dedicated Hosting for just $79/mo with 500 GB of bandwidth! > No other company gives more support or power for your dedicated server > http://click.atdmt.com/AFF/go/sdnxxaff00300020aff/direct/01/ > _______________________________________________ > Gaim-devel mailing list > Gai...@li... > https://lists.sourceforge.net/lists/listinfo/gaim-devel |
|
From: Rob F. <ro...@fa...> - 2003-04-02 16:53:20
|
My view on the matter is that the ChangeLog will contain a list of new features and other major changes. This is the way we've done it since the initial release, so I intend on continuing to do it the same way. I probably acted a little rude yesterday with my hasty response. However, I was sick and therefore a bit irritable. My main problem comes from the fact that some expect what is in CVS to be release-quality. They don't want a half completed feature. If some UI is not fully implemented then they will complain. That is mostly what I was complaining about. The changes made yesterday are an exception being that it was April Fool's day. Rob > Scott Henson <deb...@si...> writes: > > > Perhaps interface changes can be discussed on this list as well. That > doesn't seem to be happening, and when a user asks a question about > them here, they get the stock "it's CVS. Things change. Deal with it. > Read the Logs." > > Which, of course, are after the fact. (Still hoping the input on top > is an April Fools change.) > |
|
From: Richard H. <rm...@ap...> - 2003-04-02 16:32:52
|
Scott Henson <deb...@si...> writes:
> On Wed, 2003-04-02 at 10:52, Ethan Blanton wrote:
>> That is very much an opinion ... and one I am inclined to disagree
>> with. Since we have CVS logs to track commit-by-commit changes
>> (although the commit messages are not always what they could be), I
>> don't see any need to duplicate that information. I like to see major
>> changes since the last release when I look at a ChangeLog.
> Maybe starting to archive it on the web somewhere would be nice. Im not
> sure what the volume on it is, but I would rather not add another list
> to my mailbox. Also if it was archived it would be much simpler to go
> back and track the changes instead of having to keep every commit
> message.
Perhaps interface changes can be discussed on this list as well. That
doesn't seem to be happening, and when a user asks a question about
them here, they get the stock "it's CVS. Things change. Deal with
it. Read the Logs."
Which, of course, are after the fact. (Still hoping the input on top
is an April Fools change.)
--
Lift me down, so I can make the Earth tremble.
--Bucky Katt
|
|
From: Ethan B. <ebl...@cs...> - 2003-04-02 16:30:11
|
Scott Henson spake unto us the following wisdom:
> On Wed, 2003-04-02 at 10:52, Ethan Blanton wrote:
> > That is very much an opinion ... and one I am inclined to disagree
> > with. Since we have CVS logs to track commit-by-commit changes
> > (although the commit messages are not always what they could be), I
> > don't see any need to duplicate that information. I like to see major
> > changes since the last release when I look at a ChangeLog.
>
> Maybe starting to archive it on the web somewhere would be nice. Im not
> sure what the volume on it is, but I would rather not add another list
> to my mailbox. Also if it was archived it would be much simpler to go
> back and track the changes instead of having to keep every commit
> message. =20
cvs -z3 log ...
Ethan
--=20
"Among the many misdeeds of the British rule in India, history will
look upon the act of depriving a whole nation of arms, as the
blackest."
-- Mahatma Gandhi
|
|
From: Scott H. <deb...@si...> - 2003-04-02 16:09:47
|
On Wed, 2003-04-02 at 10:52, Ethan Blanton wrote: > That is very much an opinion ... and one I am inclined to disagree > with. Since we have CVS logs to track commit-by-commit changes > (although the commit messages are not always what they could be), I > don't see any need to duplicate that information. I like to see major > changes since the last release when I look at a ChangeLog. Maybe starting to archive it on the web somewhere would be nice. Im not sure what the volume on it is, but I would rather not add another list to my mailbox. Also if it was archived it would be much simpler to go back and track the changes instead of having to keep every commit message. -- Scott Henson <deb...@si...> |
|
From: Ethan B. <ebl...@cs...> - 2003-04-02 15:52:23
|
Morten Brix Pedersen spake unto us the following wisdom:
> * Luke Schierer <lsc...@us...> [2003-04-02 14:34:19]:
> > true, but many commits are fixing bugs that didn't exist in the previou=
s=20
> > version, and thus aren't _changes_. or fixing bugs in a feature that h=
as=20
> > been added since the last version, but didn't work perfectly the first =
time=20
> > it was committed, and noted in the changelog. its these, the majority o=
f=20
> > commits, that make the changelog painful on a commit-by-commit basis.
> > luke
>=20
> The ChangeLog is not meant to track changes between releases. Its meant
> to track code changes. The NEWS file is used to track changes betweeen
> releases.
That is very much an opinion ... and one I am inclined to disagree
with. Since we have CVS logs to track commit-by-commit changes
(although the commit messages are not always what they could be), I
don't see any need to duplicate that information. I like to see major
changes since the last release when I look at a ChangeLog.
Ethan
--=20
"Among the many misdeeds of the British rule in India, history will
look upon the act of depriving a whole nation of arms, as the
blackest."
-- Mahatma Gandhi
|
|
From: Morten B. P. <mo...@wt...> - 2003-04-02 14:56:54
|
* Luke Schierer <lsc...@us...> [2003-04-02 14:34:19]: > On Wed, Apr 02, 2003 at 08:33:19PM +1000, Vincent Ho wrote: > > On Tue, Apr 01, 2003 at 03:57:26PM -0500, Luke Schierer wrote: > > > > > the changelog is a pain to edit on a commit-by-commit basis. > > > > No more of a pain than to edit any other file. Just leave it open when > > you're editing other source files and it's no hassle to switch buffers > > and add a line. > > true, but many commits are fixing bugs that didn't exist in the previous > version, and thus aren't _changes_. or fixing bugs in a feature that has > been added since the last version, but didn't work perfectly the first time > it was committed, and noted in the changelog. its these, the majority of > commits, that make the changelog painful on a commit-by-commit basis. > luke The ChangeLog is not meant to track changes between releases. Its meant to track code changes. The NEWS file is used to track changes betweeen releases. - Morten. -- http://mbrix.dk/ |
|
From: Luke S. <lsc...@us...> - 2003-04-02 12:24:04
|
On Wed, Apr 02, 2003 at 08:33:19PM +1000, Vincent Ho wrote: > On Tue, Apr 01, 2003 at 03:57:26PM -0500, Luke Schierer wrote: > > > the changelog is a pain to edit on a commit-by-commit basis. > > No more of a pain than to edit any other file. Just leave it open when > you're editing other source files and it's no hassle to switch buffers > and add a line. true, but many commits are fixing bugs that didn't exist in the previous version, and thus aren't _changes_. or fixing bugs in a feature that has been added since the last version, but didn't work perfectly the first time it was committed, and noted in the changelog. its these, the majority of commits, that make the changelog painful on a commit-by-commit basis. luke > > > Cheers, > > Vince > > -- > Vincent Ho > lo...@in... > > Kitchen activity is highlighted. Butter up a friend. > > > ------------------------------------------------------- > This SF.net email is sponsored by: ValueWeb: > Dedicated Hosting for just $79/mo with 500 GB of bandwidth! > No other company gives more support or power for your dedicated server > http://click.atdmt.com/AFF/go/sdnxxaff00300020aff/direct/01/ > _______________________________________________ > Gaim-devel mailing list > Gai...@li... > https://lists.sourceforge.net/lists/listinfo/gaim-devel -- -This email is made of 100% recycled electrons. |
171 messages has been excluded from this view by a project administrator.
