-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Je Mardo 15 Majo 2001 14:11, vi skribis:
> Don't you then run the risk of reducing the underlying toolkits' unique
> strengths to a lowest common denominator -- the same problem that you
> flamed AWT for earlier in your post ("... the limitations are many, but

Oh, definately.  This is my own personal hell... I like the *design* of AWT 
and what it gives you, but it does have serious limitations.  What I'm hoping 
is that the limitations are not *inherent* to the paradigm, and that they can 
be avoided with a better design.  For example, say we have two GUIs that 
support Buttons.  One GUI supports setting the background image of the 
Button, while the other does not.  Rather than not implement setting the 
background with the abstract API, which is what AWT did, we allow it, and 
ignore it on the platform that doesn't allow the feature.  This may be a 
dangerous path; the API could be huge and complex with features that may not 
be supported on the majority of platforms.  I don't know, but I would like 
to.  One thing I *do* know: the folks at Sun are smart, but they aren't 
omniscient, and AWT's failure doesn't imply that all abstract toolkits based 
on native toolkits must be.

> By the way, didn't we have this discussion on the newsgroup just a few
> months ago? This:
>
>     http://www.rubygarden.org/ruby?AbstractionLayer
>
> looks like the beginning of the thread but I don't remember where it went
> from there ;)

Perhaps; I'm new to Ruby.  I haven't seen anything in the RAA for this, and 
it seems that none of the GUI toolkit bindings are at all Ruby-esque.  In my 
mind, this means that it is still an issue.

Gxis,

=== SER   Deutsch|Esperanto|Francaise|Linux|Java|Aikido|Dirigibles|GPG
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.0.2 (GNU/Linux)
Comment: For info see http://www.gnupg.org

iD8DBQE7AaJEP0KxygnleI8RAj/WAJ9up0mVYNnhjB4LNi0g6hvcos21zwCbBgam
qKamVEV3qeNaduGb2Q4BdNE=
=IORo
-----END PGP SIGNATURE-----