Re: Open RFS lacking (further) response
On Wed, 3 Nov 2010, Charles Plessy wrote: Le Mon, Nov 01, 2010 at 11:00:23PM -0400, Asheesh Laroia a écrit : On Sun, 31 Oct 2010, Charles Plessy wrote: RFS has something similar in concept to pull requests. How about including in the RFS emails a link to the most relevant diff in the web interface of the VCS where the package is stored ? This will help to have the changes seen by many eyes. I think that is a great idea, so long as it's an "updated package" RFS. Do you see a way to make the idea work for all RFSs? (Honest question; trying to get a sense of what you mean.) Obviously, it will not work if the package is not a VCS, so it does not apply to all RFS. But we can encourage it by showing examples, for instance in the RFS template. From URLs like the one below, people can easily figure out the correct one for their package: Diff between revisions 5120 and 5399 of the imagej package: http://svn.debian.org/wsvn/debian-med/?op=comp&compare[]=%2ftrunk%2fpackages%2fimagej%2ftr...@5120&compare[]=%2ftrunk%2fpackages%2fimagej%2ftr...@5399 Diff between the tags debian/0.1.8-1 and debian/0.1.9-1 of the samtools package: http://git.debian.org/?p=debian-med/samtools.git;a=commitdiff;h=debian/0.1.9-1;hp=debian/0.1.8-1 That makes a lot of sense! I see what you mean. -- Asheesh. -- ROMEO: Courage, man; the hurt cannot be much. MERCUTIO: No, 'tis not so deep as a well, nor so wide as a church-door; but 'tis enough, 'twill serve.
Re: Open RFS lacking (further) response
Le Mon, Nov 01, 2010 at 11:00:23PM -0400, Asheesh Laroia a écrit : > On Sun, 31 Oct 2010, Charles Plessy wrote: >> >> RFS has something similar in concept to pull requests. How about including in >> the RFS emails a link to the most relevant diff in the web interface of the >> VCS >> where the package is stored ? This will help to have the changes seen by many >> eyes. > > I think that is a great idea, so long as it's an "updated package" RFS. > > Do you see a way to make the idea work for all RFSs? (Honest question; > trying to get a sense of what you mean.) Obviously, it will not work if the package is not a VCS, so it does not apply to all RFS. But we can encourage it by showing examples, for instance in the RFS template. From URLs like the one below, people can easily figure out the correct one for their package: Diff between revisions 5120 and 5399 of the imagej package: http://svn.debian.org/wsvn/debian-med/?op=comp&compare[]=%2ftrunk%2fpackages%2fimagej%2ftr...@5120&compare[]=%2ftrunk%2fpackages%2fimagej%2ftr...@5399 Diff between the tags debian/0.1.8-1 and debian/0.1.9-1 of the samtools package: http://git.debian.org/?p=debian-med/samtools.git;a=commitdiff;h=debian/0.1.9-1;hp=debian/0.1.8-1 Have a nice day, -- Charles Plessy Tsurumi, Kanagawa, Japan -- To UNSUBSCRIBE, email to debian-mentors-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org Archive: http://lists.debian.org/20101102152142.gb32...@merveille.plessy.net
Re: Open RFS lacking (further) response
On Sat, 30 Oct 2010, Niels Thykier wrote: On 2010-10-30 08:59, Michael Tautschnig wrote: [...] One technical question, however, remains: Could we have some list of packages that remain to be reviewed? Just telling people "please review some package" is pretty awkward... Best regards, Michael Asheesh: Sounds like you got a feature request for Debexpo! (-: Currently we got a list of "unanswered" list emails at [1]; though it has a couple of issues (not limited to it not updating regularly). It also creates a number of false-positives (e.g. "nanoblogger-extra" was also handled in the reply to the "nanoblogger" email, but detecting this is non-trivial). I can try to extend this script to be more suited for this purpose. So I have no idea why this script isn't running properly. /me looks into it Oh, hey, I fixed it. It'll run every night and update now. The problem was that my cron job called "make -q" not "make --quiet". I think that, as far as false positives: we should just reply to all the messages. For better or for worse, people will look at mailing list archives and get a feel for what our community is like. (That's my opinion, anyway.) It's clear to me that the source code to this cron job should be somewhere public. Let's see... Okay, published: http://gitorious.org/debian-mentors-mailing-list-summary/debian-mentors-mailing-list-summary Anyway -- that's all for that from me for now! -- Asheesh. -- A tall, dark stranger will have more fun than you. -- To UNSUBSCRIBE, email to debian-mentors-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org Archive: http://lists.debian.org/alpine.deb.2.00.1011012215310.7...@rose.makesad.us
Re: Open RFS lacking (further) response
On Sun, 31 Oct 2010, Charles Plessy wrote: Le Sat, Oct 30, 2010 at 09:59:45PM +0200, Michael Tautschnig a écrit : I think this is a superb step in the right direction; while clearly the list on mentors.debian.net, as suggested in another email in this thread, is the more comprehensive list of packages in need of sponsorship, this web page is a proper list of RFS needing attention. Hi all, I just read a post on Planet Debian about pull requests: http://ondrejcertik.blogspot.com/2010/10/git-has-won.html http://help.github.com/pull-requests/ RFS has something similar in concept to pull requests. How about including in the RFS emails a link to the most relevant diff in the web interface of the VCS where the package is stored ? This will help to have the changes seen by many eyes. I think that is a great idea, so long as it's an "updated package" RFS. Do you see a way to make the idea work for all RFSs? (Honest question; trying to get a sense of what you mean.) -- Asheesh. -- Q: What is the sound of one cat napping? A: Mu.
Re: Open RFS lacking (further) response
On Sat, Oct 30, 2010 at 11:49:25AM +0200, Niels Thykier wrote:
> Personally I believe that Noel's statement is relates to upstream,
> who have their own packaging + their own little repository as
> compared to upstreams like you, who also work on the Debian side.
[...]
Which was more or less my point as well...
I have my own packaging and I publish it in my own repository too.
However, I spent and continue to spend a disproportionate amount of
time making sure my packaging is correct for the Debian project as a
whole (to a great extent because Debian is also my preferred
operating system and the primary reference platform on which I
develop and test my software). I do that by getting my packaging
peer-reviewed on the mentors list, having an official DD upload my
packages once he's comfortable with the level of effort I've
applied, keeping up with Debian's BTS, participating in discussions
here and on debian-devel, et cetera.
In contrast, the packaging I've seen provided by upstreams who don't
actually put forth the effort to get involved to that degree are
pretty much always non-compliant with Debian standards... many FTBFS
on sid (if they even publish source packages at all), most do not
have proper Debian.changelogs with BTS-relevant closes fields, lots
are ignorant of the FHS or more Debian-specific policy directives,
few run them through QA filters like lintian or test their builds in
clean chroots... I won't even bother to enumerate the more
scary/egregious/brain-dead packaging no-nos I've seen from some.
I expect it would be less work for a seasoned DD or DM to simply
disregard the upstream-provided packaging and redo it from scratch
than try to hunt through it for those kinds of land mines. If it
were my software and I weren't submitting my own packaging work
directly into Debian, I wouldn't take it personally if the package
maintainer scrapped my unofficial packaging work and replaced it
with his own.
> On a related note, while an experienced DD (or non-DD for that
> matter) may make a "better" debian package most of them would
> still be an inferior maintainer. All bug reports are handled by
> one who knows the code and the latency of forwarding upstream bugs
> is ... 0!
A lot of that has to do more generally with the responsiveness of
the upstream developers. I personally take bug reports from any
distribution's users of my software seriously, whether or not they
come through the package maintainers. Though I also work a day job
where I constantly get bombarded by harebrained requests from
clueless users, so I'm used to handling some of the worst sorts of
end-user interaction with a smile and a nod (and the open-source
community is more competent and much easier to deal with than my
day-job customers anyway).
Faster bug triage and awareness of upstream software's development
process aside, I still expect DDs who constantly deal with
maintenance of a broad variety of packages have a more useful
perspective on distribution-wide integration which I lack.
> Personally I think you should have an extra look at the DM role.
> You may not be ready for it just yet, but I think aiming for it
> would be an improvement for you. :)
[...]
It's been on my to do list to look into, but I haven't yet gone out
of my way to pinopint any of my local DDs... no doubt there are many
here near RTP where I live, even though Raleigh is more of a Red Hat
town. I figured I'd revisit the DM process once I found an
opportunity to get a DD signature or two on my key. Also, lack of
some official status hasn't yet been an obstacle for contributing to
Debian in a variety of other ways, so I haven't seen it as a
particularly immediate need. On the other hand, given that I've been
participating in the project in an unofficial capacity for over a
decade, maybe it's time.
Thanks for the suggestion and kind words!
--
{ IRL(Jeremy_Stanley); WWW(http://fungi.yuggoth.org/); PGP(43495829);
WHOIS(STANL3-ARIN); SMTP(fu...@yuggoth.org); FINGER(fu...@yuggoth.org);
MUD(kin...@katarsis.mudpy.org:6669); IRC(fu...@irc.yuggoth.org#ccl);
ICQ(114362511); YAHOO(crawlingchaoslabs); AIM(dreadazathoth); }
--
To UNSUBSCRIBE, email to debian-mentors-requ...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Archive: http://lists.debian.org/20101031171757.gg8...@yuggoth.org
Re: Open RFS lacking (further) response
> Perhaps (hopefully) this initiative will change the culture of this > list. I'll certainly try to remember to point out when people are > ignoring non-DDs input. > Me too, probably I'll even refrain from sponsoring should I note such behavior. Best regards, Michael pgpWQgPPBsI8a.pgp Description: PGP signature
Re: Open RFS lacking (further) response
Perhaps (hopefully) this initiative will change the culture of this list. I'll certainly try to remember to point out when people are ignoring non-DDs input. -- bye, pabs http://wiki.debian.org/PaulWise -- To UNSUBSCRIBE, email to debian-mentors-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org Archive: http://lists.debian.org/aanlktimu2l6dot_db3jsbohxpk1ka-scif07gur8y...@mail.gmail.com
Re: Open RFS lacking (further) response
Hello, On Fri, Oct 29, 2010 at 3:22 AM, Michael Tautschnig wrote: > Unfortunately, proposals such as the collaborative package review by Charles > Plessy [2] haven't caught any attention. In particular, this is a call to > people > not being DD (yet): Everyone could review packages, but it just seems people > only keep posting even more RFS. I'm a non-DD that has done reviews on mentors and IRC. I can report that there is significant pushback whenever someone hears you are a non-DD. Understandably, the sponsees don't know if you know if they can trust what you are saying, and since you don't have power to help them upload, they don't even have to respond to your comments - only to a DDs comments. A few times I have suggested something which the sponsee disregarded (or questioned) to have a DD come and say the same thing. My point is that non-DD reviews might not be as helpful as we think, if the sponsees are going to wait for a DD to respond anyways. It was a good exercise (for me) to go through other people's packages - but I understand that they are probably not going to trust something I say and a DD will have to repeat my work. Regards, Scott -- To UNSUBSCRIBE, email to debian-mentors-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org Archive: http://lists.debian.org/aanlktim2h6r-luhw0k2oqdbbngmpob9r3ig1j3q0w...@mail.gmail.com
Re: Open RFS lacking (further) response
Le Sat, Oct 30, 2010 at 09:59:45PM +0200, Michael Tautschnig a écrit : > > I think this is a superb step in the right direction; while clearly the list > on > mentors.debian.net, as suggested in another email in this thread, is the more > comprehensive list of packages in need of sponsorship, this web page is a > proper > list of RFS needing attention. Hi all, I just read a post on Planet Debian about pull requests: http://ondrejcertik.blogspot.com/2010/10/git-has-won.html http://help.github.com/pull-requests/ RFS has something similar in concept to pull requests. How about including in the RFS emails a link to the most relevant diff in the web interface of the VCS where the package is stored ? This will help to have the changes seen by many eyes. Cheers, -- Charles Plessy Tsurumi, Kanagawa, Japan -- To UNSUBSCRIBE, email to debian-mentors-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org Archive: http://lists.debian.org/20101031083042.gb11...@merveille.plessy.net
Re: Open RFS lacking (further) response
[...] > > I liked the idea of requesting reviews from people whose packages will get > sponsored; but I'd just ask for doing that voluntarily. I probably should have > taken the chance to ask people for doing so yesterday, maybe I'll even send > those people another email. > [...] Done as [1]. Should anybody else like this idea, feel free to take my take, improve it, etc. If you have ideas to improve it, please let me know as well. Best, Michael [1] http://lists.debian.org/debian-mentors/2010/10/msg00478.html pgprC6iEDy3XJ.pgp Description: PGP signature
Re: Open RFS lacking (further) response
> On 2010-10-30 08:59, Michael Tautschnig wrote: > > [...] > > > > One technical question, however, remains: Could we have some list of > > packages > > that remain to be reviewed? Just telling people "please review some > > package" is > > pretty awkward... > > > > Best regards, > > Michael > > > > Asheesh: Sounds like you got a feature request for Debexpo! > > Currently we got a list of "unanswered" list emails at [1]; though it > has a couple of issues (not limited to it not updating regularly). It > also creates a number of false-positives (e.g. "nanoblogger-extra" was > also handled in the reply to the "nanoblogger" email, but detecting this > is non-trivial). > [...] I think this is a superb step in the right direction; while clearly the list on mentors.debian.net, as suggested in another email in this thread, is the more comprehensive list of packages in need of sponsorship, this web page is a proper list of RFS needing attention. I have no idea about the technical details of this script, but the following would seem like interesting improvements: - Have it run daily, if possible. As you said, its pretty much out-of-date already. - Could the list of open RFS be checked against http://ftp-master.debian.org/new.html and maybe, possibly using UDD, against the list of packages already in the archives? If the version can be extracted from the RFS, then a check against UDD could help cleanup such nanoblogger-extra cases. Best regards, Michael pgpwv4lPx2hdc.pgp Description: PGP signature
Re: Open RFS lacking (further) response
-BEGIN PGP SIGNED MESSAGE- Hash: SHA256 On 2010-10-30 02:16, The Fungi wrote: > On Fri, Oct 29, 2010 at 10:44:37PM +0100, Noel David Torres Taño wrote: > [...] >> Is it always an upstream package worst than a repackaged package? > [...] > >>From what I've seen, in most cases, yes. > > I am one of the guilty "one pet package" upstream authors who gets > my Debian work uploaded by a gracious sponsor. [...] > Hi Personally I believe that Noel's statement is relates to upstream, who have their own packaging + their own little repository as compared to upstreams like you, who also work on the Debian side. > Even with a "special interest in Debian" I am fairly certain my > *one* package would be improved if repackaged by an experienced > Debian Developer. Mmm... if such an improvement exists, it is likely a non-issue (or not worth the time) since your sponsor has not mentioned it. On a related note, while an experienced DD (or non-DD for that matter) may make a "better" debian package most of them would still be an inferior maintainer. All bug reports are handled by one who knows the code and the latency of forwarding upstream bugs is ... 0! > On several occasions I've gotten close to ... > even applying for a DM/DD role > in the project, but I'm mindful of the limited spare time I have in > my life and careful not to commit to responsibilities beyond my > present capacity. Personally I think you should have an extra look at the DM role. You may not be ready for it just yet, but I think aiming for it would be an improvement for you. :) As I see it, the DM role does not come with any extra responsibilities for you other than the annual ping, if you are already double checking your own package before sending it to your sponsor. It is not like becoming a DM forces you to become a DD later[1] or anything. Sure, you also have to explicitly agree to abide by some rules and guidelines, but you are probably following all these already. On the plus side, you would be able to upload your pet package at will. This saves you an email to your (at that time ex-)sponsor with every upload (at the price of one email per year). But of course, you should contact your sponsor about it - he/she knows you, your package and your skills better than I do. :) ~Niels [1] "Individuals may apply to become a Debian Maintainer without being in the n-m queue, or having any intention of joining the n-m queue." http://www.debian.org/vote/2007/vote_003 -BEGIN PGP SIGNATURE- Version: GnuPG v1.4.10 (GNU/Linux) Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org/ iQIcBAEBCAAGBQJMy+okAAoJEAVLu599gGRCesQP/1FF0vpnom/N8xpjV9T8x617 Mtb8SXK1W8CJ+guomx8tRYuSTU2O/IPguJAMRtG2vm+taBW7szI0SLKC3QYMLFik /xxqZOk16EQPBtNCybYEJ49QkLQ/9oIqjZjApZryNJsojeSTsfoHBDNMbORngrRU LxF9gK643Trd/wgmnwDCNYA9jA34xzQvVUI3LHT+oosI7kwf5p21vNFjuTBhdbxn tGz82ovQOWenuYisZiMalMjr7bJHrdqNdMZo4iF9bBs5Kv+85yMMWTO7yn2KQm6w LrxdxSuYAmF5eN121T7ThcPyr0J0KCeDnaq9gXLLz+dyNiLAZv645MyhEzeH4JRc /BxoHe+m9LZe3QzykCIDEVXqOFGJIhucXbRLWcB0TfrqrYLC8At04TL/MFTy+JtL bvMmRQ47A5bY3TIAuKGhTyu0kBJLsj3fcL0VyktrOxSNVQ0maQyccZLvk7HzLkcr I61y0htmasO4esYW2fMuiLOsC69GEjILYu6sc/0vP1JSL0PXCFb2r5o5wZmay0NK NeC47ePFKpYesF6qHtY/+t/ACaWkYke559LegI6GFU+HYB5t4uKYAGhwale+ADLH t6de3B16HB2vR+cTVIAMECktcATvlLmoS3iyyRU5jP79e/pf3xXkGIZgOLbaXHRF 7zUV4pyGSD6IelTN6jx/ =SITe -END PGP SIGNATURE- -- To UNSUBSCRIBE, email to debian-mentors-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org Archive: http://lists.debian.org/4ccbea25.5050...@thykier.net
Re: Open RFS lacking (further) response
Quoting "Michael Tautschnig" : Hi Michael, I liked the idea of requesting reviews from people whose packages will get sponsored; but I'd just ask for doing that voluntarily. I probably should have taken the chance to ask people for doing so yesterday, maybe I'll even send those people another email. One technical question, however, remains: Could we have some list of packages that remain to be reviewed? Just telling people "please review some package" is pretty awkward... It seems to me that this technical question is automatically sorted by itself, i.e. any package listed at sponsor-pkglist [1] with 'Needs a sponsor' != 'No, Thanks' would make a good candidate for volunteer reviewing. [1] http://mentors.debian.net/cgi-bin/sponsor-pkglist -- To UNSUBSCRIBE, email to debian-mentors-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org Archive: http://lists.debian.org/20101030113354.38021rtsa2swz...@webmail.spnet.net
Re: Open RFS lacking (further) response
-BEGIN PGP SIGNED MESSAGE- Hash: SHA256 On 2010-10-30 08:59, Michael Tautschnig wrote: > [...] > > One technical question, however, remains: Could we have some list of packages > that remain to be reviewed? Just telling people "please review some package" > is > pretty awkward... > > Best regards, > Michael > Asheesh: Sounds like you got a feature request for Debexpo! Currently we got a list of "unanswered" list emails at [1]; though it has a couple of issues (not limited to it not updating regularly). It also creates a number of false-positives (e.g. "nanoblogger-extra" was also handled in the reply to the "nanoblogger" email, but detecting this is non-trivial). I can try to extend this script to be more suited for this purpose. ~Niels [1] http://rose.makesad.us/~paulproteus/four-days/ -BEGIN PGP SIGNATURE- Version: GnuPG v1.4.10 (GNU/Linux) Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org/ iQIcBAEBCAAGBQJMy9gMAAoJEAVLu599gGRC758P/RdzE+Alo25HtZBfyfr8ilUJ TOYPn44A5IeI1ohlSi+8pJEGuAFjliD2pvXDmoPWokgERAaowsE/SYzlCoqCsZxL zYMUnhnYLUBfdOrcRgsMNPeibiNJr/ckxq+/HZ1E705jSfNQy8SCBBWtrKSVj8ba i2Yr5qZwc+z6QAaOTbHwmJDjgZiCerhVqsISaVgTNGY8CeiSz6S9FziklN1yKa0A 18YpRiTxaZOy9TyMQfU4iosPjQ0n5A41bAZux88mfG2I0ANzFTOa1aHspcEhswqU J2PRVhSa7mopFWYvzwasr/oN7lQLvLcb8JAt4NHQqpJawoxb139Aq7NFrY7WZ4ja aISZV8EMnzxgdyZUkCJpeHPxuI+3z/OR6Yj23IyzZRxxAjBMgYor50Uqdv8iza/d YAv8+vTHlYc6Peern12mCtgJCiM56e4Rfy7f/N8TKd5334DvtaNXgNCCBnNxw5fH 6/UH5ZF0SmIHJHwwWZQjQk44ebiAMPqYWUgWhgy4dtJ9CxmVAn9Ufpn7Ed5Xv9j2 mSqAJeRLwIsa1ehMNLF8LaXdiLEvFXlSCSrpT4klK+UyjO9tHf1pbqUnxCVX7LU0 fmVzjzg4NdKJJVr7SIwoA6x+k+R395QR1JZMRn8FiTznNg+PlRYYB53Ng72e9Vqe 98Vu6m+NBHSQlMIvHh98 =ItMW -END PGP SIGNATURE- -- To UNSUBSCRIBE, email to debian-mentors-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org Archive: http://lists.debian.org/4ccbd80d.5010...@thykier.net
Re: Open RFS lacking (further) response
Hello, On Friday 29 October 2010 at 20:13:21, Noel David Torres Taño wrote: > On Viernes 29 Octubre 2010 08:34:25 Paul Wise escribió: [...] > > I think I'll also do a workshop about reviewing packages at the > > MiniDebConf in Vietnam. Does anyone think that an online workshop (IRC > > or similar) about package review would be helpful? > > Yes, I think it will be, but I do not know if it will be popular (timing > issues). Maybe several of them, or even periodic bug squashing^H^H^H package > reviewing parties? Of course, round the clock, for attendants (packagers and > mentors) be able to participate from all places in the world. This month in > European morning time, next month in American morning time, next month in > Japan morning time, last month in Middle East morning time, and start again. I think it is a wonderful idea. Maybe there are people (like me) who want to help Debian in general and don't have interest in a concrete package, but want to contribute to Debian on the whole. If this workshop is done (and I'm not sleeping or working), be sure I'll be there :-) Cheers. -- Mònica signature.asc Description: This is a digitally signed message part.
Re: Open RFS lacking (further) response
[...] > > > > Hmm, how can we encourage non-DDs to review packages. > > > > I'm thinking it might work to offer sponsorship in exchange for > > reviews. So, a DD reviews a package from maintainer. If there are no > > blockers to the upload, the DD says, "looks good, review someone elses > > package and I'll upload it". > > I see a small problem with this, and it is the following: imagine that > somebody packages a useful application. He has no special interest in Debian, > but only in that application, and it is an application we consider worth > enough to have in Debian. > [...] I liked the idea of requesting reviews from people whose packages will get sponsored; but I'd just ask for doing that voluntarily. I probably should have taken the chance to ask people for doing so yesterday, maybe I'll even send those people another email. One technical question, however, remains: Could we have some list of packages that remain to be reviewed? Just telling people "please review some package" is pretty awkward... Best regards, Michael pgphcgNyI8ZVV.pgp Description: PGP signature
Re: Open RFS lacking (further) response
Noel David Torres Taño writes: > On Viernes 29 Octubre 2010 22:05:40 Ben Finney escribió: > > Noel David Torres Taño writes: > > > That person creates the Debian package (as he creates packages for > > > other distros) and puts it here. It is a good package of a good > > > app, and he will have continuous interest in time for the package > > > (that is, he will maintain the package in shape). > > > > Do you have a concrete example of such a package that is especially > > well-maintained by someone who “has no special interest in Debian”? > > No, I havn't. I know there are a lot of programs out there which have > a debian dir or even finished deb packages, and are later repackaged > by their Debian maintainers. But must that be always that way? Is it > always an upstream package worst than a repackaged package? Until there's a concrete example of such a package, I don't think we should let that get in the way of an informal policy that we both agree is beneficial for the examples we do have to hand. -- \“The problem with television is that the people must sit and | `\keep their eyes glued on a screen: the average American family | _o__) hasn't time for it.” —_The New York Times_, 1939 | Ben Finney pgpUKZqVdlJ1a.pgp Description: PGP signature
Re: Open RFS lacking (further) response
On Fri, Oct 29, 2010 at 10:44:37PM +0100, Noel David Torres Taño wrote:
[...]
> Is it always an upstream package worst than a repackaged package?
[...]
>From what I've seen, in most cases, yes.
I am one of the guilty "one pet package" upstream authors who gets
my Debian work uploaded by a gracious sponsor. I also have a
"special interest" in Debian as a whole, since it's been my primary
operating system as far back as the bo release and I run it on
roughly 10 distinct hardware architectures (including hundreds of
servers at my place of employment).
I spent substantial time learning the proper way to package
applications for Debian and lurked on this list for over a year
before requesting sponsorship. Debian policy and standards are a
moving target. Paying attention to debian-devel-announce and
(preferably) debian-devel are important for me to proactively stay
abrest of issues and transitions which could affect my package.
Keeping up with Debian's bug tracker is crucial as well.
I try to contribute in other ways, participating in ML discussions
within my field of expertise, diligently filing detailed bugs (and
when possible providing patches) any time I encounter an
undocumented issue, and advocating Debian within my technical
community. On several occasions I've gotten close to submitting
packages of other upstreams' work or even applying for a DM/DD role
in the project, but I'm mindful of the limited spare time I have in
my life and careful not to commit to responsibilities beyond my
present capacity.
Even with a "special interest in Debian" I am fairly certain my
*one* package would be improved if repackaged by an experienced
Debian Developer.
--
{ IRL(Jeremy_Stanley); WWW(http://fungi.yuggoth.org/); PGP(43495829);
WHOIS(STANL3-ARIN); SMTP(fu...@yuggoth.org); FINGER(fu...@yuggoth.org);
MUD(kin...@katarsis.mudpy.org:6669); IRC(fu...@irc.yuggoth.org#ccl);
ICQ(114362511); YAHOO(crawlingchaoslabs); AIM(dreadazathoth); }
--
To UNSUBSCRIBE, email to debian-mentors-requ...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Archive: http://lists.debian.org/20101030001619.gb8...@yuggoth.org
Re: Open RFS lacking (further) response
On Viernes 29 Octubre 2010 22:05:40 Ben Finney escribió: > Noel David Torres Taño writes: > > I see a small problem with this, and it is the following: imagine that > > somebody packages a useful application. He has no special interest in > > Debian, but only in that application, and it is an application we > > consider worth enough to have in Debian. > > I see that scenario as problematic in itself. I would much rather delay > the inclusion of that package until it gains a maintainer who > demonstrates interest in Debian as a whole, than have it maintained by > someone who “has no special interest in Debian”. > > > That person creates the Debian package (as he creates packages for > > other distros) and puts it here. It is a good package of a good app, > > and he will have continuous interest in time for the package (that is, > > he will maintain the package in shape). > > Do you have a concrete example of such a package that is especially > well-maintained by someone who “has no special interest in Debian”? No, I havn't. I know there are a lot of programs out there which have a debian dir or even finished deb packages, and are later repackaged by their Debian maintainers. But must that be always that way? Is it always an upstream package worst than a repackaged package? > > > Will we say him we will not upload his work just because he does not > > review other package he has no interest at all in? > > My opinion: we have an excess of packages, and a dearth of maintainers > who have interest in improving Debian as a whole. > > I would much rather address the latter by preferring maintainers who > demonstrate an interest in Debian as a whole, not just their pet > package. Yes, I agree. I was just trying to raise an academic question, but it is a possible case. What if inclusion of that particular package is a benefit for Debian as a whole? To throw an (imaginary) example, think on a CUPS equivalent for sound: Common Unix Sound System which really unificates all the mess we have with OSS, ALSA, aRTS, Pulse, phonon and Jack. Let it be the work of one team on one University, people who do not care on Debian (nor any other single distro) but do quite necessary software and quite good packaging. Just trying to get to the corners of the proposal, you know Noel er Envite signature.asc Description: This is a digitally signed message part.
Re: Open RFS lacking (further) response
Noel David Torres Taño writes: > I see a small problem with this, and it is the following: imagine that > somebody packages a useful application. He has no special interest in > Debian, but only in that application, and it is an application we > consider worth enough to have in Debian. I see that scenario as problematic in itself. I would much rather delay the inclusion of that package until it gains a maintainer who demonstrates interest in Debian as a whole, than have it maintained by someone who “has no special interest in Debian”. > That person creates the Debian package (as he creates packages for > other distros) and puts it here. It is a good package of a good app, > and he will have continuous interest in time for the package (that is, > he will maintain the package in shape). Do you have a concrete example of such a package that is especially well-maintained by someone who “has no special interest in Debian”? > Will we say him we will not upload his work just because he does not > review other package he has no interest at all in? My opinion: we have an excess of packages, and a dearth of maintainers who have interest in improving Debian as a whole. I would much rather address the latter by preferring maintainers who demonstrate an interest in Debian as a whole, not just their pet package. -- \ “By instructing students how to learn, unlearn, and relearn, a | `\ powerful new dimension can be added to education.” —Alvin | _o__)Toffler, _Future Shock_, 1970 | Ben Finney -- To UNSUBSCRIBE, email to debian-mentors-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org Archive: http://lists.debian.org/8739ron1zv@benfinney.id.au
Re: Open RFS lacking (further) response
On Viernes 29 Octubre 2010 08:34:25 Paul Wise escribió: > On Fri, Oct 29, 2010 at 3:22 PM, Michael Tautschnig wrote: > > So this was the first time I dragged together such a list, and most > > probably it will already be the last time. It just takes way too much > > time and feels pretty much like a Sisyphus task as people just keep > > sending RFS with very few non-DDs doing reviews (there are noteworthy > > exceptions - [13] and possibly others). > > Hmm, how can we encourage non-DDs to review packages. > > I'm thinking it might work to offer sponsorship in exchange for > reviews. So, a DD reviews a package from maintainer. If there are no > blockers to the upload, the DD says, "looks good, review someone elses > package and I'll upload it". I see a small problem with this, and it is the following: imagine that somebody packages a useful application. He has no special interest in Debian, but only in that application, and it is an application we consider worth enough to have in Debian. That person creates the Debian package (as he creates packages for other distros) and puts it here. It is a good package of a good app, and he will have continuous interest in time for the package (that is, he will maintain the package in shape). Will we say him we will not upload his work just because he does not review other package he has no interest at all in? > > I think I'll also do a workshop about reviewing packages at the > MiniDebConf in Vietnam. Does anyone think that an online workshop (IRC > or similar) about package review would be helpful? Yes, I think it will be, but I do not know if it will be popular (timing issues). Maybe several of them, or even periodic bug squashing^H^H^H package reviewing parties? Of course, round the clock, for attendants (packagers and mentors) be able to participate from all places in the world. This month in European morning time, next month in American morning time, next month in Japan morning time, last month in Middle East morning time, and start again. Just thinking about Noel er envite -- To UNSUBSCRIBE, email to debian-mentors-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org Archive: http://lists.debian.org/201010291913.22239.env...@rolamasao.org
Re: Open RFS lacking (further) response
On 10/29/2010 03:34 PM, Paul Wise wrote: > I think I'll also do a workshop about reviewing packages at the > MiniDebConf in Vietnam. Does anyone think that an online workshop (IRC > or similar) about package review would be helpful? > Paul, I'll try to bring my DVD cam, so we can film all workshops and put videos online. That might help too! Looking forward to be in Vietnam, Thomas -- To UNSUBSCRIBE, email to debian-mentors-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org Archive: http://lists.debian.org/4cca89e6.4050...@debian.org
Re: Open RFS lacking (further) response
On Fri, Oct 29, 2010 at 3:22 PM, Michael Tautschnig wrote: > So this was the first time I dragged together such a list, and most probably > it > will already be the last time. It just takes way too much time and feels > pretty > much like a Sisyphus task as people just keep sending RFS with very few > non-DDs > doing reviews (there are noteworthy exceptions - [13] and possibly others). Hmm, how can we encourage non-DDs to review packages. I'm thinking it might work to offer sponsorship in exchange for reviews. So, a DD reviews a package from maintainer. If there are no blockers to the upload, the DD says, "looks good, review someone elses package and I'll upload it". I think I'll also do a workshop about reviewing packages at the MiniDebConf in Vietnam. Does anyone think that an online workshop (IRC or similar) about package review would be helpful? -- bye, pabs http://wiki.debian.org/PaulWise -- To UNSUBSCRIBE, email to debian-mentors-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org Archive: http://lists.debian.org/aanlktin=_qykzec+bytpn_nzuhm5a91jriunow_nq...@mail.gmail.com
Open RFS lacking (further) response
Hi all, Following Asheesh's proposal [1] we should try to reply to a post to debian-mentors within four days. I felt there was an increased willingness to kindly reply to RFS following this proposal, but it seems to have decreased again. Myself, I'd be willing to keep up this kind service to people posting RFS, but I just lack experience in many areas, and hence feel unable to take all the packages listed below. Not that time would permit to take all those packages either, but a lack of experience is my main point. Unfortunately, proposals such as the collaborative package review by Charles Plessy [2] haven't caught any attention. In particular, this is a call to people not being DD (yet): Everyone could review packages, but it just seems people only keep posting even more RFS. Below I'll list several RFS that, to the best of my knowledge, are still open, were sent after the four-days-proposal, and are older than four days. I'll add further information where deemed useful, and I'll also indicate packages that I intend to review myself. The date is the date of last action on debian-mentors: - 2010-10-12 [3]: disco - A distributed computing framework Niels Thykier and David Bremner provided feedback; this package involves lots of python code. - 2010-10-12 [4]: crystalhd - Crystal HD Video Decoder Submitter of RFS should probably talk to kernel folks. - 2010-10-18 [5]: sciteproj - Project manager for usage with the SciTE editor I'll respond to that one. - 2010-10-22 [6]: minidlna - lightweight DLNA/UPnP-AV server targeted at embedded systems I'll try to look at it. - 2010-10-22 [7]: pgfouine - PostgreSQL log analyzer Matthew Palmer had reviewed that one. - 2010-10-22 [8]: downloadstatusbar: Iceweasel/Firefox addon that provides an improved download status I'll talk to the submitter. - 2010-10-22 [9]: seeks - Extensible websearch proxy & meta-search engine with personalized I have no idea about web stuff. - 2010-10-23 [10]: morse - 'Morse Classic' is a morse-code training program for aspiring rad New upstream release, package has been sponsored once already. - 2010-10-24 [11]: ovito - visualization and analysis tool for atomistic simulation data I'll try to look at it. - 2010-10-24 [12]: 9menu - Creates X menus from the shell Updated package. So this was the first time I dragged together such a list, and most probably it will already be the last time. It just takes way too much time and feels pretty much like a Sisyphus task as people just keep sending RFS with very few non-DDs doing reviews (there are noteworthy exceptions - [13] and possibly others). Best regards, Michael [1] http://lists.debian.org/debian-mentors/2010/10/msg5.html [2] http://wiki.debian.org/PackageReview [3] http://lists.debian.org/debian-mentors/2010/10/msg00225.html [4] http://lists.debian.org/debian-mentors/2010/10/msg00247.html [5] http://lists.debian.org/debian-mentors/2010/10/msg00283.html [6] http://lists.debian.org/debian-mentors/2010/10/msg00347.html [7] http://lists.debian.org/debian-mentors/2010/10/msg00349.html [8] http://lists.debian.org/debian-mentors/2010/10/msg00351.html [9] http://lists.debian.org/debian-mentors/2010/10/msg00352.html [10] http://lists.debian.org/debian-mentors/2010/10/msg00354.html [11] http://lists.debian.org/debian-mentors/2010/10/msg00371.html [12] http://lists.debian.org/debian-mentors/2010/10/msg00372.html [13] http://lists.debian.org/debian-mentors/2010/10/msg00297.html pgpByNhHCDXuS.pgp Description: PGP signature
