Peter Hildebrandt wrote:

Still I remember you changing the way without-c-dependency worked -- it might be that this was for debugging purposes or as a future extension, tho.

It was actually a Fundamental Repair. I had overloaded the compute stack with the responsibility of conveying dependency. Sounds reasonable until one decides to go without dependency but of course still does not want to get into an infinite cycle computing values. So I created a new special variable to represent the "depender", which one could bind to nil to avoid dependencies while still binding oneself onto the *call-stack* to support cycle detection.

I believe (!) you changed it from setfing to dynamically binding ...

Me? setf a special?! :)

well, all that is buried deep down there beneath a ton of thesis work ;-)

...

...

Andy, if you could send us a more detailed description of the problem, that might help trigger our recollection from the cells-gtk3 migration.

I think a runtime error from Cells complaining about a cycle that goes away when reverting to older Cells would be a smoking gun. :)

As a matter of fact, that's what I had in mind.

Btw, Chris Smith got back to me concerning the doc work and we mailed a couple times to and fro. He's now working on putting together a good (!) getting started for cells-gtk3 and maybe smoothen out the install process a bit. (I should apologize that I lured him over here into the cells-gtk camp, again leaving celtk and cello without a doc.

Just as well, I haven't much time to spare these days.

Maybe one day Frank will take it on ...)

Is that algebra software coming along? I remember you talking about a closed-user-group alpha a while ago on cll, but haven't heard anything since. Maybe I should go back and join that yahoo group (I believe it was).

Going good, so well I got a little greedy and went for another milestone before releasing. Wait for an announcement, it will be a new/different group (the name is kinda taken by other Algebra stuff). Thx for the interest.

kt