From: invalid@invalid.invalid   
      
   Bud Frede  writes:   
   > Richard Kettlewell  writes:   
   >> So in terms of the Zawinski’s complaint post the situation has   
   >> deteriorated. Although I think the underlying issue is that he doesn’t   
   >> (or didn’t) believe in stable releases while Debian and its users do.   
   >   
   > I think the real issue is that problems with packaged software in a   
   > distro are often reported to the developer and not the packager or the   
   > distro itself as they should be. Jamie was annoyed that Debian users   
   > would report bugs to him that were only still extant in the very   
   > out-dated version of xscreensaver that Debian was shipping.   
      
   That seems like a problem with the users choosing to bypass the   
   distribution’s own bug reporting mechanisms.   
      
   19 months being “very out-dated” is JWZ’s viewpoint. It wasn’t the   
   distribution’s view and wasn’t (and still isn’t) the view of a large   
   chunk of end users, who (collectively) often complain about having to   
   upgrade.   
      
   Some software does legitimately have a very short upgrade cycle, even   
   disregarding security issues. For example, the timezone database, where   
   civil authorities have a tendency to make rather short-notice   
   changes. I’m not convinced that a screensaver should really fall into   
   this category. But that’s another question.   
      
   > I'm sure that there are other developers that experience the same thing,   
   > but are perhaps not as vocal about it as JWZ.   
      
   Yes, it happens all the time. If you publish software (for free or   
   otherwise) people will keep running it longer than you’d like, and they   
   will send low-quality support requests and bug reports to whatever   
   bug-reporting or support channel you advertize. Just part of life as a   
   software developer; you need to find a strategy for managing it which   
   works for you.   
      
   In the case at hand, JWZ was trying to manage it with grumpy code   
   comments and blog posts; it doesn’t sound like it was very effective.   
      
   Long ago I was in a conversation with someone about whether a particular   
   distribution’s bug tracking system should be made easier to use. It   
   wasn’t particularly user-friendly by the standards of the time, so it   
   seemed like a reasonable suggestion. But his view was that the   
   distribution had plenty of bug reports already, in fact far more than it   
   could address with the resources it had. Spending effort making it   
   easier to report bugs would help precisely nobody - the same effort   
   should be spent on something else (for example, fixing some bugs).   
      
   > Personally, I just assume that when I'm using a distro, I will report   
   > any issues to the distro's maintainers. The only times I would reach   
   > out to a developer whose software I use is if I want to make a   
   > donation in appreciation of their hard work, or perhaps if I had a   
   > suggestion for a new feature or something.   
      
   Yes, normally the distribution should be the first port of call. And   
   Debian’s tooling (e.g. the reportbug command) does try to make that   
   easy. But the distribution can’t force its users to behave in any   
   particular way.   
      
   --   
   https://www.greenend.org.uk/rjk/   
      
   --- SoupGate-DOS v1.05   
    * Origin: you cannot sedate... all the things you hate (1:229/2)