[Date Prev][Date Next][Thread Prev][Thread Next][Author Index][Date Index][Thread Index]
Re: [zzdev] The problem of versioning cursors: solved in Chartreuse
- To: ZZ Development <zzdev@xxxxxxxxxx>
- Subject: Re: [zzdev] The problem of versioning cursors: solved in Chartreuse
- From: "B. Fallenstein" <b.fallenstein@xxxxxx>
- Date: Fri, 19 Oct 2001 20:13:21 +0200
- References: <3BCFF107.CC2C76F8@xxxxxx> <20011019144605.B11539@xxxxxxxxxxxxxx>
Tuomas Lukka wrote:
> > In Chartreuse (GZigZag 0.8.0), we could potentially use a nice quick
> > hack to get around this problem, when we have a cleaner separation of
> > the primitive slices into clientspace and userspace and window state.
> > Then, we can simply forget old window states (expunge the blocks) and
> > yet version every change of the userspace. -- Probably not something we
> > want to do before the release of 0.8.0, but I think an idea worth considering.
> Definitely. I wouldn't even consider it a hack because different people
> working on the same data should be able to use diffent views without
> causing conflicts.
Oh, separating the clientspace and userspace is something I consider
absolutely neccessary (and separating clientspace [bindings etc.] and
windowspace [windows and cursors] is something I do consider useful).
The hacky part is to use it for dropping *the right* old versions,
because the problem is more complex than just data vs. view settings.
But as hacks go, I think it's a nicely working one, and we're far, far
from grasping how to really solve the problem behind it. After all, we
cannot do everything ;)