[Date Prev][Date Next][Thread Prev][Thread Next][Author Index][Date Index][Thread Index]
Re: [zzdev] Re: [zzdev] Raster-specific keybindings
- To: Benjamin Fallenstein <b.fallenstein@xxxxxx>
- Subject: Re: [zzdev] Re: [zzdev] Raster-specific keybindings
- From: Tuomas Lukka <lukka@xxxxxxxxxxx>
- Date: Sat, 12 Aug 2000 13:58:49 +0300 (EETDST)
- Cc: zzdev@xxxxxxxxxx
- In-reply-to: <399434B9.67379360@xxxxxx>
> I haven't been able to do a consistent draft on versioning. All
> approaches I would have liked proved to be mathematically impossible...
> I still hope that there is some good way for local versioning in ZZ, and
> that it'll be found. Looking forward to the results of your meeting with
> Ted (not only in respect to versioning).
Well, we need to collect all the aspects of the problem together, and
start seeing which are logically incompatible. Then it becomes a
(NP-complete) 3-sat problem ;) ;) ;)
> In the spec there's something about selectors to complement paths: is
> this implemented yet? That could be helpful in order to generalize the
> level raster -- it could even be used for Ted's genealogy stuff,
> although this would require some more serious changes which I'm unable
> to do right now. But maybe over the next few weeks...
Not yet, but will soon be, as it's needed for mail raster.
Actually, I just had a really strange idea: why not make selectors be
simply inverse paths... that would make a lot of sense in some ways.
So when the way from the child to the parent is step here, headcell here,
step there, then the selector would be the inverse (automatically: the
same path definition in the structure would be used): step -there,
all cells poswards from here, step -here.
It's a nice idea but I don't know whether it'll work.