AppDB: Rating / Patching

[James Mckenzie]

>On Tue, Jan 6, 2009 at 6:48 AM, Rosanne DiMesio <dimesio at earthlink.net> wrote:
>>
>>> Now, the story changes if the patch is conforming and has been accepted
>>> by AJ and is pending the next development release.
>>>
>> Then the next development release can get the gold, but previous ones still shouldn't. AppDB test ratings are tied to specific releases, and intended to tell normal users how different versions of Wine will work with their app. Patching Wine is not something normal users can or want to do, and allowing ratings based on patched versions of Wine is misleading, even if the patch does eventually make it in to a later release.
>
>It sounds like the problem is that the version string in appdb isn't
>descriptive enough.  It's perfectly reasonable to wonder if a given
>program can be made to work with a patched version of wine, and wonder
>how well it will work.  It's also reasonable to wonder how it will
>work with a vanilla version.  Both types of reports are useful to have
>in the appdb.  Having a version "x.x.x (patched)" available to
>reporters would allow both types of reports to be clearly separated.
>
No, the appdb should not be touched.  Rosanne said it correctly, ordinary users are NOT going to take the time to build Wine, nor should they.   We can put in the bug report that the patch works and whether or not it has been submitted.  Sometimes a patch is to rough or a real hack that breaks other programs, but with refinement is acceptable and will be incorporated into Wine.  The appdb needs to stay as clean as it can.  Of course, you can always add a bug report to the appdb entry, add comments and let users decide what they want to do.  Rating a rogue patched Wine as Gold is very misleading.  We need to keep ratings to what is available for the ordinary, unknowing user (read nOOb.)

James McKenzie