caml-list - the Caml user's mailing list
 help / color / mirror / Atom feed
From: Pascal Cuoq <Pascal.Cuoq@inria.fr>
To: ohl@hep.tu-darmstadt.de, caml-list@inria.fr
Subject: Re: Preferred GUI Toolkit for O'Caml 3?
Date: Tue, 22 Feb 2000 18:05:41 +0100	[thread overview]
Message-ID: <20000222180541.16173@pauillac.inria.fr> (raw)
In-Reply-To: <14509.20226.13985.265187@heplix4.ikp.physik.tu-darmstadt.de>; from Thorsten Ohl on Fri, Feb 18, 2000 at 02:54:10PM +0100

>    -  mlgtk      [ (+) Gtk+ is nicer than Tk (and no-Tcl is nicer than Tcl)
>                    (-) is it ported to everything but the kitchen sink? ]
> 
>    -  lablgtk    [ same a mlgtk, plus
>                    (-) still beta
>                    (+) appears(!) to have the slickest API ]

Since Jacques defines "beta" as "the API might change later",
I should point out that mlgtk has had its share of these changes in
the past, and some more may be required in the future.

Personally, I am still working on mlgtk although there are very competent
people working on lablgtk because:

1) it is fun.  It looks good when it works, and when you're tired
   of watching it look good and work well, you change your gtk theme 
   and it feels like you just wrote another toolkit interface, which
   looks good too.

2) some people are actually using it.

Sven, who maintains mlgtk too, told me once that he himself actually 
used it, so that's one more reason why mlgtk should last for a while.

This said, at the time I was recruited to work on mlgtk I had no 
opinion about labels and optional arguments, and now I do: they are
definitely the way to go for a widget library.

There are other differences between the two libraries, because
they were written completely independently.  mlgtk is less object-oriented
in the sense that it relies on module abstraction to be essentially
type-safe from the outside.  I think that lablgtk is completely
type safe, but it has more complicated object hierarchies (if I understood
correctly).

Pascal Brisset, who wrote the interface between mlgtk and native threads,
seemed to be saying that threads were better in mlgtk too :) but that
may have changed...

Pascal



      parent reply	other threads:[~2000-02-22 17:54 UTC|newest]

Thread overview: 15+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2000-02-18 13:54 Thorsten Ohl
2000-02-21 17:22 ` Benjamin C. Pierce
2000-02-22 11:41   ` William Chesters
2000-02-23  7:36     ` Alan Schmitt
2000-02-23  9:24     ` Sven LUTHER
2000-02-24 10:24       ` Fabrice Le Fessant
2000-02-25 16:22         ` skaller
2000-02-22 13:41   ` skaller
2000-02-22 14:22     ` Benjamin C. Pierce
2000-02-22  1:46 ` Jacques Garrigue
2000-02-22 13:39   ` skaller
2000-02-22 20:13     ` Gerd Stolpmann
2000-02-22  8:17 ` Sven LUTHER
2000-02-22 11:04 ` jean-marc alliot
2000-02-22 17:05 ` Pascal Cuoq [this message]

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=20000222180541.16173@pauillac.inria.fr \
    --to=pascal.cuoq@inria.fr \
    --cc=caml-list@inria.fr \
    --cc=ohl@hep.tu-darmstadt.de \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).