Re: removing spatial viewer from GWorkspace

[Riccardo Mottola]

Hi,
> > It was added after a discussion on the mailing list long ago, it was a popular 
> > opinion.
> >      
> This is usually a good reason for keeping something, but then...
>
>    
it is indeed, but many people didn't like it back then either. I am 
investigating if opinions changed. Some features are a bit a "hype", 
they come with a trend, they fade silently again.
> I never use it
>    
> Neither is this - there is lots of stuff in GNUstep that I don't use, but try not to break just for the sake of it (or, at least, which I apologise to Richard about when I do break it).
>    
I am trying to understand if there are other users of it and if, how many.
> > and it adds code in GWorkspace that needs to be kept from bitrotting.
> >      
> Bit rot generally only happens when you make large changes to GWorkspace.  My 
> understanding is that you are currently just engaged in bug fixing and making 
> GWorkspace into something less of an unmaintainable mess, so bit rot should not 
> be a big issue.
>    
Currently yes, version 0.8.8 which I intend to release soon shall be 
more a maintenance release of bug fixes, build issues, crashes and 
general code clean-up. However, while looking a th the code base, it is 
legit to look in the future.
> The correct time to ask this question is when you've made, or about to make, 
> changes that will make the spacial mode difficult to maintain.  If there is 
> only one user of a feature, but it takes very little effort to maintain, then 
> it's a bad idea to delete it - all that we do is alienate users for no gain.
>    
Assessing the usage is my current goal. Currently the code has a wide 
degree of duplication, which enhances the effort of maintenance. If it 
is deemed a truly essential feature a way of factoring out common code 
about the two modes and thus enhancing maintenance needs to be found.

Thom's answer is so rough and impulsive that maybe it is not even 
trustworthy that he is really using it!


Cheers,
  Riccardo