Re: Windows support (was: Pissed Off)

> Given this latest feedback, I think I should've gone with my old idea of
> dropping the windows support altogether and simply continue the player
> as a Linux only player. 
> 
> Hey Ed, wanna drop Windows support? 

Dropping windows support has a couple major advantages I'll mention.

1.  We can fix the config/make system that some package maintainers have 
complained about and it makes the tree easier to deal with by reducing 
it by more than half the number of directories and code.

2.  We dont have to deal with bugs from a port that nobody wants to keep 
  up to date with the more developed unix version.


For the most part as far as i can see the code for these two OS's are 
separate.  We've basically got two programs sharing some similar files 
but for the most part are entirely separate.  That is our main problem 
with why the windows port works so much less than the unix port.  The 
lack of shared code.  The gui and ui is entirely separate and that's a 
big problem with maintainance.


A couple disadvantages with dropping windows...
1. We leave windows and it's userbase.  Whether that's good or bad i'm 
not sure but i've always felt being able to run on as many OS's as 
possible is best if you can do it with ease.

2.  A lot of what is left of the interest in zinf comes from emusic.  A 
lot of emusic users are windows users.  By leaving the windows OS we 
leave that "advertising" and with that may come less interest and less 
interest means any programmers out there wont be compelled to want to be 
a zinf developer.


Now there are some obvious faults with the disadvantages.  I've never 
even been to emusic's website so that cuts some big holes in the "no 
advertising" to windows argument.  Also, despite being visible to the 
huge windows  community, we've gotten hardly any programming interest 
from it.  Also, developing on windows to make the project compile is a 
pain in the ass because we either have to deal with cygwin or VC and the 
very nature of windows makes this kind of project have to have a large 
quantity of duplicated functionality and thus separate code.




> EMusic will roll out their download manager soon, and I've changed the
> handled mime type and removed the calls to the Jukebox that handles
> pass-offs of unhandled RMP types. That should improve the stability of
> the window version, I think. I'll put that out as version 2.2.1 in the
> next week or so, and then we can wipe the windows code from the
> codebase. 
> 
> Whaddya think?
> 



it's not practical to keep a port that requires so much work but nobody 
from that platform wants to do it. The next version of zinf has always 
been to cut dead weight from zinf, to remove the code that doesn't work 
in favor of trying to hack it to make it work.  While i dont think there 
will never be a windows port of zinf, i think it's best to remove the 
port we do have now and if people want one work can be started on a 
clean slate and done correctly.

So to sum it up, I say remove the windows port.  Remove the bug reports 
  related to it. We then remove the bulk of the problems with zinf that 
are reported and can get to the real problems with zinf that have to be 
dealt with before the next release. That and things being more managable 
  may lead to less programmers being scared away from the project.



-------------------------------------------------------
This sf.net email is sponsored by:ThinkGeek
Welcome to geek heaven.
http://thinkgeek.com/sf