The signature I (i.e., '''Lars Hellström''') uses on this Wiki. ---- I'm a Ph.D. in mathematics. These days, I quite a lot use Tcl as a language for rapid prototyping of research software. (BTW, the core part of one of these little programs was subsequently ported to C and set to work on some ''really'' large problems. We haven't even been able to run the largest yet, because it turned out SGI [http://www.sgi.com/] had limited memory allocation to 256 gigabytes per process and this particular problem needed 430, but it seems they have fixed that now, so right now we're just waiting for computer time. Also noteworthy: it's still a Tcl script that computes the basic tables needed by the core part of the program!) Also, I'm quite much into TeX. ---- Nice to see another Lars...--[LarsOlson] ---- Lars, I wanted to thank you for your comments on my page [User Interface Design for Tcl/Tk]. I've completely rewritten and redesigned it. I realized I'd bitten off more than I could chew, and was also duplicating work. As I continued to browse the Wiki, I realized that what I ''really'' wanted was a set of links to all the pages people had created about how to use Tk to make effective GUIs. I'm hoping the new page is more useful. I am planning to post a revised version of my example for the evolution of an interface on a separate page, this time using the appropriate geometry manager! :P [EKB] ---- [ABU] 21 jun 2005 I've read your notes about [Deep copy and persistence of a BWidget's Tree], but I dont understand what you mean with your last phrase : ''... you may want to think about whether it could be useful for a no-Tk interpreter to manipulate dried-trees.'' My guess is that you mean adding persistence and deep-copy capabilities to '''abstract''' trees (i.e pure data- structures). In this case the '''struct::tree''' package (see [struct]) already provides these capabilities (though implemented with a very different approach). Did I guess it ? [Lars H]: It depends. If dried-trees are '''struct::tree'''s then that would probably do the trick, but if not then you may have some work to do. Consider the following situation: Process A (perhaps a cron-job or whatever) generates tree-shaped data and places them in some file. Then every once in a while you run process B where you can view (edit, etc.) these trees. Now, it A could generate dried-trees then all B would have to do is a simple '''graft''' operation, is it not? [ABU] Lars, generally speaking, if A and B use independent internal representations for their trees, they should agree on a common data-exchange format. XML could be the solution (although I'm not enthusiastic about XML). If A and B are '''tcl-applications''', both using struct::tree for their internal representation, they could use 'serialized trees' for data-exchange. If A and B are '''tk-applications''', both using BWidget::Tree, they could use 'dried-trees' for data-exchange ---- [NEM] 13 Sept 2005 Lars, I've just received a notification that an email I sent you has not been delivered (but will retry for 5 days), with a problem trying to resolve the hostname of my email address (some authentication scheme?). Thought I'd let you know via the wiki in case it is a general email or DNS failure at your end. Cheers, Neil. ---- [Category Person]