Benoit St-Pierre wrote:
Hi!
> Not wanting to enter yet another language war, the question still begs
> to be asked : what limitations would justify to depart from Tcl/Tk ?
I see no real limitations.
The main drawback of Tcl/Tk is of course that you've a pretty small
developers community with an ancient language like that (IMHO). But
well, I'm not sure that we'd get much more developers for a chess
application in <place your favourite "modern" language here>.
> The two reasons so far are not really convincing. User-friendliness
> is unrelated to the language behind an application; "being not nice"
> is a bit fuzzy, to say the least.
I think the OT refers to Scid looking like, well, Tk looks. This is old
Motif style in most parts. To blend it better into your desktop you'd
also have to work in your .Xdefaults. (But: how many people know of it's
very existence, even more about how to use it? Besides that, Scid has
quite a bunch of hardcoded values inside so configuration by .Xdefaults
is limited a bit as well.) Beyond that: Tcl/Tk just does not respect
modern desktops preferences/styles. If that ever happens it will look
nice as well.
Anyway, "looking not nice" well, personally, I feel that e.g. the whole
KDE looks "not nice", but that's really a matter of personal taste.
(Most people will not like my NeXTStep look either.)
> That said, my main concern would be to justify Tcl/Tk, so that we
> could answer once and for all questions like "Why is Scid not
> developed in Java/Python/Gtk/Fortran etc?".
Java, Python, Gtk: were all not availalbe when Scid started.
Fortran: You don't want to do any GUI beyond basic CLI reading input
files in Fortran. (I can tell, I did quite a substential part of my work
in Fortran. A really cool language. For FORmula TRANslation.)
> Tcl/Tk seems like a
> reasonable choice : it has drawbacks, but they are compensated by its
> advantages.
Point it: it's there, everything is done in it. Alternative is to redo
everything. And as I said: its easier and much more efficient to learn
Tcl/Tk and work on Scid than redo everything from scratch. As Michael
pointed out: it's not only the GUI but you'd have to redo large parts of
actual functionality.
Coming to that: most people will not be aware of it, but if you have a
look at some banks e.g. most of the money transfer is handled by ...
Cobol or even older stuff. Or look at the usual nuclear power plant.
Most of it is done in concrete electronics. No Java, no PC no funny
stuff, just old but working things.
> We would need to have Pascal's position on this issue. I am not sure
> we already had that conversation. The only reason I am fussing about
> that is that I am starting to build a FAQ.
I think Pascals position is clear, and he loves Tcl/Tk. I'm more of the
"well if it has to be..." fraction that lives with it for pragmatic
reasons. (I do a lot of the coding in Perl and then translates the logic
to Tcl.)
--
Kind regards, / War is Peace.
| Freedom is Slavery.
Alexander Wagner | Ignorance is Strength.
|
| Theory : G. Orwell, "1984"
/ In practice: USA, since 2001
------------------------------------------------------------------------------
The Planet: dedicated and managed hosting, cloud storage, colocation
Stay online with enterprise data centers and the best network in the business
Choose flexible plans and management services without long-term contracts
Personal 24x7 support from experience hosting pros just a phone call away.
http://p.sf.net/sfu/theplanet-com
_______________________________________________
Scid-users mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/scid-users
