On Sun, 14 Jan 2007 21:21:00 +0100, Thijs Kinkhorst wrote: > > He imposes a high quality standard when he sponsors a package. > > I'm not sure about this reference to a quality standard multiple people > in this thread are making. I did not question anything about the > quality, just about communications. > Hi Thijs, On the point of your email, At the end of last year I had some free time and spent it improving my packaging by getting involved with debian-mentors. As I am not a DD, I only helped with problems, and checked 20 or so packages for people. This is far below the number that Daniel has sponsored, and I don't claim that it makes me an expert on the subject. However I would like to contribute to the discussion. When I checked that humongous number of packages I was finding the same mistakes over and over again. Mistakes that are really easy to fix, and while they are not always highlighted clearly in the places that we look are repeatedly mentioned on the list. I felt that if people simply looked at one or two RFS threads on the list before they posted theirs they would be able to clean these things up. Picking three packages at random http://lists.debian.org/debian-mentors/2006/09/msg00178.html http://lists.debian.org/debian-mentors/2006/09/msg00419.html http://lists.debian.org/debian-mentors/2006/09/msg00662.html there are issues that come up over and over again (not to crticise those maintainers). All of this led me to be slightly disappointed that people didn't put as much effort in as they could when they were asking for time from other people. This detracted from my want to be verbose when explaining these things. As for the practical aspect, it gets boring to repeat yourself and say almost the same things over and over again (hehe!). In terms of solving this you could have preprepared answers ready for the common case (I liked to find a good reference on the problem and reference it, the how not to write a copyright file email being the best example), but that is likely to end up sounding cold as every response is the same. I don't know how any of this relates to any other sponsor on the list, and as I said this is from a short experience. Now, how to try and alleviate any problems that we have (if indeed a cold tone of voice is a problem, but I think it is symptomatic of a deeper problem). These are just some ideas of mine, and I am proposing them for discussion as there is absolutely nothing I can do to implement most of them. Please prefix everyone with "In my opinion" when you read it. For sponsors it would be good if they sponsored less packages, this way they will probably be more willing to work with each maintainer, as it will be less like a conveyor belt of packages. This is difficult without more sponsors, and there have been very few on the list for months now. For maintainers looking for sponsorship I think they should spend a little extra time checking the package, and reading about common mistakes, and certainly not think, "well, it looks OK, and my sponsor will find any problems". I think this currently happens with the second or third sponsorships as people wont make the same mistakes twice. I have seen a couple of people going from making these sorts of mistakes to being excellent packagers in no time. I also have a couple of ideas for increasing the visibility of resources for packagers, and for highlighting common problems and learning good practices. I am thinking about some, and working slowly on others, but as always I have more ideas than time. My premise for most of this email has been that it has been repetition that has sparked off this thread, but I also think there are other things going on here that I haven't highlighted, and don't have the quick fixes that I have mentioned. Thanks for reading, James -- James Westby -- GPG Key ID: B577FE13 -- http://jameswestby.net/ seccure key - (3+)k7|M*edCX/.A:n*N!>|&7U.L#9E)Tu)T0>AM - secp256r1/nistp256
Attachment:
signature.asc
Description: Digital signature