Otto Kekäläinen <[email protected]> writes: > Hi! > >> >>>Do you (or anyone else) update the Uploaders field? In what way? >> >> >> >> I feel like I probably should in some way, but at the moment I basically >> >> don't unless I particularly feel inclined to take over primary >> >> maintenance of that particular package. > .. >> It seems to me that the Uploaders: fields doesn't really reflect any >> relevant information, or at least none that is kept up to date. >> >> Is there any reason we couldn't make this change to policy? > > I use the Uploaders field like described in the Policy, and add myself > as Uploader on packages where I am willing to take responsibility like > others describe above. Please don't change the meaning or remove the > Uploaders field. > > On the contrary, I would actually ask you to consider using it more > often. For example you have done all golang-go.crypto uploads in > (nearly) the past year, but you didn't add yourself as Uploader
Do Policy say anything which suggest that? What I find is this: Section 3.3 https://www.debian.org/doc/debian-policy/ch-binary.html#the-maintainer-of-a-package If the maintainer of the package is a team of people with a shared email address, the Uploaders control field must be present and must contain at least one human with their personal email address. See Uploaders for the syntax of that field. I think golang-go.crypto both follow the letter and spirit of that recommendation. Section 5.6.3 https://www.debian.org/doc/debian-policy/ch-controlfields.html#uploaders List of the names and email addresses of co-maintainers of the package, if any. If the package has other maintainers besides the one named in the Maintainer field, their names and email addresses should be listed here. The format of each entry is the same as that of the Maintainer field, and multiple entries must be comma separated. This is normally an optional field, but if the Maintainer control field names a group of people and a shared email address, the Uploaders field must be present and must contain at least one human with their personal email address. I think the letter of this is followed for golang-go.crypto too. If it follows the spirit of the text depends on what is meant by "co-maintainer". I suspect you have one interpretation in mind when you read it. This paragraph is the ONLY use of the word "co-maintainer" in the Debian Policy manual, from what I can tell. So what it is supposed to mean isn't terribly clear. What DO we want "co-maintainer" to mean? How do you define it? I wouldn't necessarily consider myself a co-maintainer of golang-go.crypto since I prefer to consider myself part of a team that maintains packages. And preferrably the team is "Debian" rather than "Debian Go Team", but that is mostly semantics. The few golang-go.crypto uploads I've done are mostly drive-by uploads because other packages needed functionality from newer golang-go.crypto. I have only contributed ~0.5% of the its debian/ history: jas@frallan:~/dpkg/golang-go.crypto$ git ls-files debian|xargs -n1 git blame|wc -l 31757 jas@frallan:~/dpkg/golang-go.crypto$ git ls-files debian|xargs -n1 git blame|grep 'Simon Josefsson'|wc -l 168 jas@frallan:~/dpkg/golang-go.crypto$ However I believe the entire concept of Maintainer/Uploader fields in Debian is a gigantic bike shed. It can be part of an explaination of some poor social behaviour historically (e.g., systemd, or almost anything involving any cross-package co-ordination). These problems aren't technical. Thus we can't solve them fully with technical improvements. As they are currently defined, I believe dropping Maintainer/Uploader fields from the debian policy manual is the simplest improvement. Or at least make them optional. I would agree that sometimes social problems can be mitigated by technical clarifications or changes, but I think that would involve some serious work and I don't see anyone working on that. /Simon
signature.asc
Description: PGP signature
