Go Qt, perhaps?
Go Qt, perhaps?
Posted Sep 28, 2010 18:00 UTC (Tue) by smokeing (guest, #53685)Parent article: Michael Meeks talks about LibreOffice and the Document Foundation
Mozilla holds on to XUL for a reason I think, it's *the* language everything renderable is described in in Firefox, but what's the point to have its own toolkit for an office suite?
(Was about to post this in /. as well, but there we go: http://developers.slashdot.org/comments.pl?sid=1801996&...).
And secondly, there is so much dross among the heap of rubble the feature set of OOo has become. Fair enough, users will want a picture placed on a page in Writer, but really, who needs a *movie* in what is, by definition, an essentially printable document? I understand those features had been accumulated per Sun market people telling programmers to catch up with a similar --questionable in its own right-- additions to MS Office, but now what's the pressing need to maintain it?  The entire Web-oriented Writer should also be ripped out and cast into eternal fire.
      Posted Sep 28, 2010 18:20 UTC (Tue)
                               by nicooo (guest, #69134)
                              [Link] (3 responses)
       
     
    
      Posted Sep 29, 2010 0:24 UTC (Wed)
                               by smokeing (guest, #53685)
                              [Link] (2 responses)
       
     
    
      Posted Sep 29, 2010 1:31 UTC (Wed)
                               by rgmoore (✭ supporter ✭, #75)
                              [Link] 
       
     
      Posted Sep 29, 2010 6:39 UTC (Wed)
                               by spaetz (guest, #32870)
                              [Link] 
       
     
      Posted Sep 28, 2010 18:49 UTC (Tue)
                               by oever (guest, #987)
                              [Link] 
       
     
      Posted Sep 28, 2010 19:06 UTC (Tue)
                               by elanthis (guest, #6227)
                              [Link] (3 responses)
       
That said, converting an entire codebase to a new toolkit is a truly massive undertaking.  It's also one that's relatively difficult to do piecemeal.  Without a really solid reason to do so -- which entails some use case where the existing toolkit simply doesn't work -- I don't foresee a toolkit change. 
     
    
      Posted Sep 29, 2010 1:09 UTC (Wed)
                               by smokeing (guest, #53685)
                              [Link] (2 responses)
       
No arguing it may be, in the short term, as unrewarding a task as it appears impossible. Ultimately, though, UI designers^w^w even the dreariest clerks will start to laugh at the 1990-ish look of OOo. Even the most conservative toolkit, GTK+, makes passes at OpenGL-enabled widgets (Clutter project), which I do believe will become as useful as a well-tuned compiz in lieu of your default wm. 
And, if undertaken in earnest, porting it to Qt will help separate the many processing and rendering layers the editable stuff goes through in OOo. Heck, I remember, around 2004, I had OOo 1.x installed on a Mac, and saw it bring along the entire X server to run on! 
     
    
      Posted Sep 29, 2010 6:48 UTC (Wed)
                               by eru (subscriber, #2753)
                              [Link] 
       
OOo has usability problems, but the lack of shine and glitter in widgets is not one of them... I'm desperately hoping the "new management" will not get sidetracked with eye candy, and instead will address the more serious problems first. Eg. try to do something about the ridiculous resource usage, which is one place where using a standard toolkit could actually help, provided the same toolkit (and version!) is used by other programs running on the same machine. This would be the case with QT in a KDE environment, or GTK on Gnome.
      
           
     
      Posted Sep 29, 2010 21:10 UTC (Wed)
                               by daniel (guest, #3181)
                              [Link] 
       
     
      Posted Sep 29, 2010 8:19 UTC (Wed)
                               by ibisum (guest, #59406)
                              [Link] 
       
     
    Go Qt, perhaps?
      
Go Qt, perhaps?
      
      I'd assume that it's a byproduct of their embedding scheme.  If the embedding code is shared between applications, then by default any application can embed any file type allowed by any other application.  When they let presentations embed movies, spreadsheets got the same capability automatically.  It's easier to leave embedding up to the users than to deliberately disable specific types for specific applications.  If nobody wants to embed movies in their spreadsheets, then the ability to do so will wind up being an unused feature.
      
          Go Qt, perhaps?
      Go Qt, perhaps?
      
Go Qt, perhaps?
      
Go Qt, perhaps?
      
Go Qt, perhaps?
      
No, not all toolkits are born equal, and while yes, gtkmm might be up for the task (inkscape uses it, to everyone's satisfaction I believe), it would be natural to use something natively written in C++. And that is Qt. Look at Scribus.
      even the dreariest clerks will start to laugh at the 1990-ish look of OOo. Even the most conservative toolkit, GTK+, makes passes at OpenGL-enabled widgets
Go Qt, perhaps?
      
      And, if undertaken in earnest, porting it to Qt will help separate the many processing and rendering layers the editable stuff goes through in OOo. Heck, I remember, around 2004, I had OOo 1.x installed on a Mac, and saw it bring along the entire X server to run on!
Go Qt, perhaps?
      
If an interested observer were to demonstrate a QT port patch that at least brings up part of the suite in not too ugly a way (a proof of concept) then I would be not at all surprised to see our new Libreoffice stewards react in favor of at least opening up an experimental branch.
      
          Go Qt, perhaps?
      
 
           