<html>
<head>
<style><!--
.hmmessage P
{
margin:0px;
padding:0px
}
body.hmmessage
{
font-size: 10pt;
font-family:Verdana
}
--></style>
</head>
<body class='hmmessage'>
> > Trestle is easy to write for, but it really is butt ugly.<BR><BR>
<BR>
I should point that making Trestle not look ugly is perhaps<BR>
a much smaller task than wrapping any other library.<BR>
<BR>
<BR>
Randy's Win32 changes to Trestle definitely make it look better for example.<BR>
<BR>
<BR>
It was previously asserted that X Trestle shall not look like Win32.<BR>
However, X Trestle doesn't look like anything except itself, right?<BR>
So I figure Win32 is among the good choices.<BR>
Besides, given that there is no "standard X look", even if X Trestle<BR>
does look like *something*, Win32 would still be preferable.<BR>
(This would remove the forking we have where only one fork<BR>
or another has a bug, X vs. Win32..really need to merge/refactor<BR>
that code, no matter the decision and the resulting look..)<BR>
<BR>
- Jay<BR><BR> <BR>
<HR id=stopSpelling>
From: jay.krell@cornell.edu<BR>To: dragisha@m3w.org; peter.mckinna@gmail.com<BR>CC: m3devel@elegosoft.com<BR>Subject: RE: [M3devel] Modernising m3-ui?<BR>Date: Fri, 22 Jan 2010 12:05:10 +0000<BR><BR>
<STYLE>
.ExternalClass .ecxhmmessage P
{padding:0px;}
.ExternalClass body.ecxhmmessage
{font-size:10pt;font-family:Verdana;}
</STYLE>
> Linux desktop successful <BR> <BR>Haha. Military intelligence.<BR> <BR> <BR> > it's diversity of possible solutions.<BR> <BR>Chaos! Lack of interop. Duplicated effort.<BR>Granted, Darwinism/Capitalism, but very wasteful.<BR> <BR><BR> > Main problem with diversity of solutions is multi-platform nature of <BR> > Modula-3 - solutions not multi-platform are not likely to be accepted... <BR><BR> <BR>I'm sure we can win here, if we do anything.<BR> (I'm not volunteering. The idea of using Swig is good.)<BR>Just wrap Qt or FLTK or Tk or wxWidgets or such.<BR>Qt is my preferred.<BR> It seems to have the most active resources spent on it and<BR> is in the best condition already.<BR>GTk on Windows doesn't seem to get enough attention.<BR>Tk has a seemingly good track record in that<BR> people have successfully used it already from several<BR> languages: Python, Perl, Tcl (blech!), C.<BR>wxWidgets also has Python bindings at least.<BR> <BR> <BR>Things only get less clear if you worry about non Win32/64/XWindows<BR>platforms, like OS/2, Mac9, non-X Mac, Carbon/Cocoa, Win16, MS-DOS (framebuffer+mouse+keyboard), etc.<BR>And Trestle doesn't support any of those anyway, and they are all<BR>"obsolete" except non-X Mac (see also: iPhone!)<BR>I'm surprised I haven't see that Qt supports iPhone.<BR>If a library does support Carbon/Cocoa that is a point in favor imho.<BR> <BR><BR> - Jay<BR><BR> <BR>> From: dragisha@m3w.org<BR>> To: peter.mckinna@gmail.com<BR>> Date: Thu, 21 Jan 2010 23:27:12 +0100<BR>> CC: m3devel@elegosoft.com<BR>> Subject: Re: [M3devel] Modernising m3-ui?<BR>> <BR>> Modernizing of GUI is not so big a problem - lots of C(++) libraries<BR>> around.... And after wrapping _LOTS OF_ C libraries I can tell one thing<BR>> - there's nothing like a manual touch! And I've wrapped pthreads, mysql,<BR>> sqlite, gtk, gtk2, zip, bzip2, zaptel, libpri .....and many more.<BR>> <BR>> Modula-3 philosophy (at least it looks to me sometimes) is to think<BR>> about Modula-3 legacy libraries/project like it's something carved in<BR>> stone... To be kosher we must use right (legacy of course and<BR>> all-platforms of coursier :)) libraries for every job.... Nice idea, esp<BR>> when we whink longer term maintenance... But what made projects like,<BR>> for example, Linux desktop successful is not single solution path - it's<BR>> diversity of possible solutions.<BR>> <BR>> Main problem with diversity of solutions is multi-platform nature of<BR>> Modula-3 - solutions not multi-platform are not likely to be accepted...<BR>> And while it's relatively easy to wrte your Modula-3 code multiplatfom,<BR>> taking care of C(++) libraries can be real pain....<BR>> <BR>> Thus said - I can tell you one thing - JUST DO IT :). I don't see a<BR>> problem if my projects work only on Linux - once I have incentive to go<BR>> through a hassle to make it work on Windows, I'll synchronize. But it's<BR>> important to to many things with Modula-3 - more will come a lot easier.<BR>> <BR>> On Fri, 2010-01-22 at 06:33 +1100, Peter McKinna wrote:<BR>> > I've been looking at this area a bit. I have just completed the<BR>> > interface to GLUT which should be ready to commit in a few weeks as<BR>> > soon as i get a few more examples tested. This gives you the<BR>> > conventional opengl/X linking. Its taken a while to get my head around<BR>> > swig which seems a better way to feed into the C world. I have also<BR>> > nearly completed a new interface for mysql giving it a safe M3<BR>> > interface and gives the complete mysql.h api. I was thinking of using<BR>> > swig for the gtk bindings but not sure how well the c++ mappings are<BR>> > handled <BR>> > <BR>> > regards Peter <BR>> > <BR>> > <BR>> > On Thu, Jan 21, 2010 at 11:50 AM, Chris <Highjinks@gmx.com> wrote:<BR>> > It looks like most of the libs(Trestle, X11R4, OpenGL, etc...)<BR>> > in m3-ui haven't been touched or updated since the early 90s.<BR>> > <BR>> > Seems like modernising it might attract some more developers.<BR>> > i.e. Bring the X interface up to date(X11R6) and support<BR>> > things like DRM, XRandr, etc... And update OpenGL for NURBS,<BR>> > VBOs, etc...<BR>> > Trestle is easy to write for, but it really is butt ugly.<BR>> > <BR>> > Is anyone else looking at this area of the system?<BR>> > <BR>> > <BR>> > --<BR>> > Chris <Highjinks@gmx.com><BR>> > <BR>> -- <BR>> Dragiša Durić <dragisha@m3w.org><BR>> <BR> </body>
</html>