Quoting Bas Wijnen (2016-02-26 20:16:32) > On Fri, Feb 26, 2016 at 07:59:29PM +0100, Jonas Smedegaard wrote: >> Do we favor tracking the true upstreams when packaging for Debian? > > There was some discussion about this on the list recently, but this is > a question that didn't really come up, AFAIK. > > IMO, there are two things that matter here: > 1. We require source. If the "fake" upstream does not provide that, > it is certainly not adequate. IIUC, this is your situation (but I > didn't check your links). That is: minified js is not source, and > a project including it in its distribution is equivalent to a > compiled project including a static library. In both cases, the > code must be packaged from its source, and the bundled version must > be discarded. This was discussed, and AFAIK what I wrote here is > what most (but not all) people agreed with. Above is *not* the issue I ask about here: We may disagree on how we should interpret them, but rules _do_ exist in Policy. It is below that I ask about here: > 2. Needless forking is bad. There is no consent on what is "needless" > though. My point is that having multiple copies of a thing that are > all treated as source leads to problems. In Debian, we recognize > that and one effect of that is that we don't want bundled libraries > in packages. In the greater free software community, not everyone > sees it this way. Having this opinion in Debian, I think we should > use our influence to try to push upstreams the right way. That > means we should package real upstream if there are multiple sources > to choose from. Another reason for doing this is that future code > duplication in Debian is automatically prevented. In your example: > if someone needs the serverside version of the package, they would > package node-handlebars and then we have two versions of the code > in Debian. If the real upstream was used to begin with, that > problem would have been avoided. Right. That is the issue. Question I raise is how to deal with it? I agree with you that the real upstream should be used when possible - but is that just the personal opinion of two Debian developers which should not be imposed on others (read: at most file wishlist bugreports) or do we have rough consensus in Debian to make it into Policy, so that issues of that kind can be treated as more severe bugs? - Jonas P.S. (The concrete case had the additional oddity that the real upstream _was_ used to begin with but later abandoned as less convenient.) -- * Jonas Smedegaard - idealist & Internet-arkitekt * Tlf.: +45 40843136 Website: http://dr.jones.dk/ [x] quote me freely [ ] ask before reusing [ ] keep private
Attachment:
signature.asc
Description: signature