[Bf-funboard] File Browser proposal

Matt Ebb bf-funboard@blender.org
Tue, 2 Dec 2003 13:45:23 +1100


On 2 Dec 2003, at 4:24 AM, William Reynish wrote:

> Hi Matt. I'm sorry if I didn't credit your great work.
> I'm a true fan of all your ideas, and I love browsing
> the feature requests reading all your questions. You
> raise some very interesting questions there! I have
> updated my proposal with a link to your version, and
> reviewed my writing to not take any credtit from you.
> This was not intentional.

That's ok. I don't mean to be an arsehole, I just thought it was 
important to point out. It is good to see people thinking about it, 
that's what that tracker topic was aimed at doing :)

> About the file browser, I will review it today. You
> mention the OSX file browser - I'll just check it on
> the Mac. I forgot to implement your "new directory"
> button too. Also, I will look at some ideas and
> corrections posted by a few people and see if I can
> come up with something.

Actually, it seems it might be a better idea to take a step back first, 
and do some research, rather than working by trial and error. Like in 
that colour picker topic, it would be a good idea to go and analyse a 
bunch of different file selectors, look at what functionality 
(controls, hotkeys, etc) they contain, the advantages and disadvantages 
of how they are laid out, and also analysing just what functionality 
would be suited for Blender, and what would be superfluous and 
unnecessary. For example, It would probably be quite unnecessary and 
bloated for Blender to contain (like in Windows XP) Icon, Thumbnail, 
Tiled, List, Details, Filmstrip views. It's also good to keep in mind 
other file operations, like appending, importing, exporting, saving 
renders, etc. since the same thing would probably be re-used..

I also recall that the GNOME unix desktop project are in the process of 
re-designing their file selector - does anyone have any links to the 
discussions that have taken place surrounding that? We could learn 
something from them and avoid duplicating effort. Anyway, I think it 
would be much easier to make something more concrete, once we have some 
sort of a specification of what's good, bad, needed, or not.

What do you think? I'm finally getting some free time back again, so I 
can give a run down of the OS X one in the tracker.

cheers

matt