Hello, On Mon, Jan 02, 2006 at 02:00:59PM +0100, Andreas Barth wrote: > Hi, > > * Anand Kumria (wildfire@progsoc.uts.edu.au) [051230 02:20]: > > I think it'd more useful for users if the debian-volatile group were > > able to process requests (such as gtk-gnutella) by maintainers in a > > timely fashion. > > > > I'm pretty disappointed so far; despite being contacted via IRC a few > > days ago by Martin, who: > > what do you expect by this mail to happen? Annoy Martin? Increase our > frustration level? Actually get a response? I got a response from Martin (via IRC) and now yourself via email. > > - wasn't able to determine whether I was a Debian developer or > > not (I am). > > Well, I usually would expect that - if something worked wrong in your > opinion with the automatic running script - you directly contact the No output *at* *all* from any automatic (or otherwise) program has been received by me regarding my upload. With regards to the key, nothing in the IRC log indicates that this was a problem in a script. The log seems to indicate to me that Martin is attempting to manually verify my keyid using a gpg command. I gave him a corrected version and was asked to join #debian-volatile -- however it was 6am and sleep was decidedly more important to me. > > - hadn't actually bothered to look over the upload changelog I > > had uploaded > > This is not true. I call bullshit on this one. I'm loathe to quote IRC logs but I will in this case: (05:59:53) zobel: changelogs like these really help us and our users, as it clearly says what has changed. (06:00:05) wildfire: here is what my last entry has (06:00:09) wildfire: * New upstream release [... more cut and paste _from_ _the_ _upload_ ...] (06:00:39) wildfire: * Urgency set to medium since all current versions of gtk-gnutella have (06:00:39) wildfire: expired (06:00:44) zobel: yeah, i realy would like to see that for volatile! (06:00:50) wildfire: that is what I uploaded! > > You have maintainers who want to support their users but whom you are > > ignoring. > > This is not true. You send mails to the volatile contact address, and we > answered. I've sent many emails. I've got one answer from you and one from Martin. In particular I point you to: http://lists.debian.org/debian-volatile/2005/12/msg00020.html > There are massive reasons against gtk-gnutella as written in > e.g. http://lists.debian.org/debian-volatile/2005/12/msg00001.html - What? Have you actually read what you wrote? You "massive reasons" are: "You don't really expect that we can accpet such massive changes in volatile? Besides, BTW, the changelog is *really* broken" to which I've already responded. See: http://lists.debian.org/debian-volatile/2005/12/msg00006.html > Martin has been extra nice, and still wanted to work out a way to As have I. I've been extra patient as well. If you are busy, no problem, *tell someone*. Don't wait 30 (or more days) and then wonder why people get frustrated. > It is always easy to claim "from my maintainers point of view, doing > this and that will help users of my package". Are you disputing my claim that the package is now broken for existing users? Are you disputing my claim that upgrading to gtk-gnutella 0.96b, via volatile, would be of benefit to them? If so, I'd certainly love to hear an actual reason rather than rhetoric. > However, we as volatile > team have (also) to take care of the larger picture - long-term trust > and supportability etc. Indeed. You have to earn trust. Partly that means being responsive. Partly that also mean being upfront. As an example, if you (plural) are too busy -- no problems -- just let people know. Expectation management isn't hard or difficult, it really requires communication though. I understand why you feel you want to review changes prior to them being uploaded. From my outsiders perspective it has reduced the utility of volatile significantly though. > And it is not true that just because some > maintainer says "it works fine", everything works fine. Otherwise, there > wouldn't be e.g. any RC bugs in Debian sid. That really is a non-congruent argument. > > It seems to me your are wasting your time by ovrelooking the former for > > the latter as well as letting down whatever userbase volatile has. > > If you think the volatile team is making wrong decisions, you are free > to try to convince us Isn't that what I am trying to do? But in order for me to convince you of anything it would require a dialogue (at the very least) which certainly hasn't been the forté of the debian-volatile group recently. > or even ask the DPL on replacing the team (I'm not > going to argue we're not delegated, and even running volatile on private > hardware - if the DPL wants to use his constitutional right, we're going > to give him way) - While I appreciate the offer, I haven't totally given up on convincing either of you as yet. Even if it does come across as if I have. And my "escalation" wouldn't be to appeal to a higher authority straight away -- I'd be more inclined to have some kind of real-time chat about things via a phonecall. > however, please refrain from using killer arguments > like "letting down whatever userbase volatile has". I've already pointed out that gtk-gnutella has a large userbase but lets see what popcon says about its relative size versus what you already have in volatile: popcon/net whois jwhois gtk-gnutella rank 29 274 127 inst 8132 155 576 vote 2283 65 227 [note1] [note1]: whois in volatile is older than whois in stable popcon/mail spamassassin spamc postgrey rank 9 11 61 inst 3225 3130 188 vote 2160 1366 110 popcon/utils clamav-base clamav-daemon clamav-docs clamav-freshclam clamav-milter clamav-data rank 55 61 784 56 280 783 inst 1864 1174 312 1798 68 34 vote 1261 762 0 1255 48 0 popcon/libs popcon/libdevel libclamav1 libclamav-dev rank 175 396 inst 1897 36 vote 1345 16 Which indicates to me: - clamav-data isn't so useful to have in volatile, very few people actually use it (probably all using clamav-freshclam) - spamassassin, and clamav (software) are widely used - postgrey and jwhois have a very small userbase - gtk-gnutella's userbase is below spamassassin & clamav but above postgrey & jwhois; so it is a middle sized program for volatile. Anand -- `When any government, or any church for that matter, undertakes to say to its subjects, "This you may not read, this you must not see, this you are forbidden to know," the end result is tyranny and oppression no matter how holy the motives' -- Robert A Heinlein, "If this goes on --"
Attachment:
signature.asc
Description: Digital signature