From: John Canning
Subject: CLIM, does anyone really use this beast??
Date:
Message-ID: <3801E674.BAC83465@_ou.edu>
I am working on a research project at the Univ. of Oklahoma; we are
developing a deductive database that uses Arc/INFO as a database
for representing facts in our rulebase.
Anyway, with that background info out of the way, I would like to
ask a question. Does anyone really use CLIM? I am having a lot of
trouble understanding it. I have made some abortive attempts to layout
a simple dialog that accepts data from the user. But CLIM's very
unconventional event system (or lack thereof) is causing me a lot of
grief. It's simply too dificult to grasp right away for someone like me
who is used to Java's AWT and MSVC++ MFC way of doing things.
In these windowing toolkits, control over appearance, size, and
position, and response to events is a lot more explicit.
Can someone here help me get a handle on this beast? (Or suggest
an alternative?) I am using Harlequin LISP 4.1 with CLIM 2.0.
Jonathan Moody
email: (remove #\_ ········@_ou.edu")
On Mon, 11 Oct 1999 08:30:28 -0500, John Canning <········@ou.edu> wrote:
> Can someone here help me get a handle on this beast? (Or suggest
> an alternative?) I am using Harlequin LISP 4.1 with CLIM 2.0.
If you're using a Harlequin Lisp (you don't say which one) have you tried
using CAPI?
__Jason
Jason Trenouth <·····@harlequin.com> writes:
> On Mon, 11 Oct 1999 08:30:28 -0500, John Canning <········@ou.edu> wrote:
>
> > Can someone here help me get a handle on this beast? (Or suggest
> > an alternative?) I am using Harlequin LISP 4.1 with CLIM 2.0.
>
> If you're using a Harlequin Lisp (you don't say which one) have you tried
> using CAPI?
Of course your program won't run on ACL then.
Cheers
--
Marco Antoniotti ===========================================
PARADES, Via San Pantaleo 66, I-00186 Rome, ITALY
tel. +39 - 06 68 10 03 17, fax. +39 - 06 68 80 79 26
http://www.parades.rm.cnr.it/~marcoxa
In article <·················@_ou.edu>, John Canning <········@ou.edu> wrote:
> Anyway, with that background info out of the way, I would like to
> ask a question. Does anyone really use CLIM? I am having a lot of
> trouble understanding it. I have made some abortive attempts to layout
> a simple dialog that accepts data from the user. But CLIM's very
> unconventional event system (or lack thereof) is causing me a lot of
> grief. It's simply too dificult to grasp right away for someone like me
> who is used to Java's AWT and MSVC++ MFC way of doing things.
> In these windowing toolkits, control over appearance, size, and
> position, and response to events is a lot more explicit.
There is a mailing list for CLIM: ····@bbn.com.
> Can someone here help me get a handle on this beast? (Or suggest
> an alternative?) I am using Harlequin LISP 4.1 with CLIM 2.0.
If you ask specific questions you might get answers here
or on the above mentioned mailing list.
From: Howard R. Stearns
Subject: Re: CLIM, does anyone really use this beast??
Date:
Message-ID: <3802239F.79D17EF1@elwood.com>
I agree that it is very hard to figure out how to get started in CLIM.
I am also not aware of a good basic tutorial. What I do know of is
listed at http://www.elwood.com/alu/table/references.htm#clim
This should really be answered by someone who has done a lot of CLIM
applications and taught CLIM to others. However, the lack of available
material has lead me to form my own approach, which I offer for whatever
its worth.
I did find it helpful to look at examples. I think most implementations
(and resources listed above) have the "address-book" demo, which gives
one a decent place to start.
I also find it helpful to consider that there's a lot of ways of doing
things in CLIM. There are also a lot of low level tools that allow fine
control over things. Try not to need them. Just concentrate on
1. the application-frame and the main window "panes" you need in that
frame.
2. the commands you need to support, but never mind how they get
invoked.
3. the objects that a user sees that get displayed in the panes.
You can then read in the User Guide or spec just the sections on
application-frames, commands, and presentation-types. This should
actually be enough to understand most of the example code you might
see.
John Canning wrote:
>
> I am working on a research project at the Univ. of Oklahoma; we are
> developing a deductive database that uses Arc/INFO as a database
> for representing facts in our rulebase.
>
> Anyway, with that background info out of the way, I would like to
> ask a question. Does anyone really use CLIM? I am having a lot of
> trouble understanding it. I have made some abortive attempts to layout
> a simple dialog that accepts data from the user. But CLIM's very
> unconventional event system (or lack thereof) is causing me a lot of
> grief. It's simply too dificult to grasp right away for someone like me
> who is used to Java's AWT and MSVC++ MFC way of doing things.
> In these windowing toolkits, control over appearance, size, and
> position, and response to events is a lot more explicit.
>
> Can someone here help me get a handle on this beast? (Or suggest
> an alternative?) I am using Harlequin LISP 4.1 with CLIM 2.0.
>
> Jonathan Moody
> email: (remove #\_ ········@_ou.edu")
John Canning <········@ou.edu> wrote in message
······················@_ou.edu...
> Anyway, with that background info out of the way, I would like to
> ask a question. Does anyone really use CLIM? I am having a lot of
> trouble understanding it. I have made some abortive attempts to layout
> a simple dialog that accepts data from the user. But CLIM's very
> unconventional event system (or lack thereof) is causing me a lot of
> grief. It's simply too dificult to grasp right away for someone like me
> who is used to Java's AWT and MSVC++ MFC way of doing things.
> In these windowing toolkits, control over appearance, size, and
> position, and response to events is a lot more explicit.
>
When I first came to Lisp I was expecting to find Tk and maybe Wx available.
It seems strange for the Lisp world to try and stick with a single language
standard and not get the benefit of being part of a wider community of
users. The benefit is a double one - Lisp gets better libraries, and it's
easier for new programmers to get useful work done in Lisp. (Perl and Python
would seem natural places for Lisp to draw strength from.)
I find this mono-language standard especially strange as CLIM - from my
perspective at least - seems rather under-powered. I know that my (much)
preferred vendor, Franz, doesn't offer an OpenGL CLIM component, for
instance. But I can that capability with Wx, and I think Tk (maybe depending
on platform.) And opensource makes it really easy to add missing
functionality to these toolkits.
Jonathan Coupe, XWare
On Mon, 11 Oct 1999 19:02:33 +0100, "Jonathan"
<········@meanwhile.freeserve.co.uk> wrote:
> When I first came to Lisp I was expecting to find Tk and maybe Wx available.
If by Lisp you mean a family of languages, those toolkits are already
available: Tk is supported by STk, Wx by DrScheme. Both of them are Scheme
systems.
Paolo
--
EncyCMUCLopedia * Extensive collection of CMU Common Lisp documentation
http://cvs2.cons.org:8000/cmucl/doc/EncyCMUCLopedia/
From: Robert Monfera
Subject: Re: CLIM, does anyone really use this beast??
Date:
Message-ID: <38033AA8.F38D967A@fisec.com>
Paolo Amoroso wrote:
> If by Lisp you mean a family of languages, those toolkits are already
> available: Tk is supported by STk, Wx by DrScheme. Both of them are
> Scheme systems.
I was looking at STk about 2 years ago, but at that time it crashed
pretty often, especially when using with Tk. Maybe it has changed
since?
Robert
>>>>> "Robert" == Robert Monfera <·······@fisec.com> writes:
Robert> Paolo Amoroso wrote:
>> If by Lisp you mean a family of languages, those toolkits are already
>> available: Tk is supported by STk, Wx by DrScheme. Both of them are
>> Scheme systems.
Robert> I was looking at STk about 2 years ago, but at that time it crashed
Robert> pretty often, especially when using with Tk. Maybe it has changed
Robert> since?
Odd. We've used stk for at least that long, and I've never had it
crash on me. Different uses/usage styles, I guess.
Ray
From: Robert Monfera
Subject: Re: CLIM, does anyone really use this beast??
Date:
Message-ID: <38036486.AAC0CBE9@fisec.com>
Thanks, good to know - I tested layouts quite a bit, and I ran it on
windows besides linux (the whole point is that you can write a
platform-independent GUI).
By the way, thanks for maintaining the SERIES package - I prefer it over
loop if possible.
Robert
Raymond Toy wrote:
>
> >>>>> "Robert" == Robert Monfera <·······@fisec.com> writes:
>
> Robert> Paolo Amoroso wrote:
> >> If by Lisp you mean a family of languages, those toolkits are already
> >> available: Tk is supported by STk, Wx by DrScheme. Both of them are
> >> Scheme systems.
>
> Robert> I was looking at STk about 2 years ago, but at that time it crashed
> Robert> pretty often, especially when using with Tk. Maybe it has changed
> Robert> since?
>
> Odd. We've used stk for at least that long, and I've never had it
> crash on me. Different uses/usage styles, I guess.
>
> Ray
In article <··············@rtp.ericsson.se>,
Raymond Toy <···@rtp.ericsson.se> writes:
>>>>>> "Robert" == Robert Monfera <·······@fisec.com> writes:
>
> Robert> Paolo Amoroso wrote:
> >> If by Lisp you mean a family of languages, those toolkits are already
> >> available: Tk is supported by STk, Wx by DrScheme. Both of them are
> >> Scheme systems.
>
> Robert> I was looking at STk about 2 years ago, but at that time it crashed
> Robert> pretty often, especially when using with Tk. Maybe it has changed
> Robert> since?
>
> Odd. We've used stk for at least that long, and I've never had it
> crash on me. Different uses/usage styles, I guess.
>
> Ray
Last time I tried it, about 18 months ago, it core dumped in some of the
demos.
Mike McDonald
·······@mikemac.com
On Tue, 12 Oct 1999 09:42:00 -0400, Robert Monfera <·······@fisec.com>
wrote:
> I was looking at STk about 2 years ago, but at that time it crashed
> pretty often, especially when using with Tk. Maybe it has changed
> since?
Here is part of a message posted last March to the STk mailing list:
"On another note, a big THANK YOU to Erick for all his work on STk.
Our company has just released a product which uses STk for the
user interface. In all, it has about 30,000 lines of STk source --
a testament to the power and quality of STk.
(For anyone who is interested, a description can be found at:
http://www.grammatech.com/products/codesurfer/codesurfer.html)"
It's just a data point, but it looks interesting.
Paolo
--
EncyCMUCLopedia * Extensive collection of CMU Common Lisp documentation
http://cvs2.cons.org:8000/cmucl/doc/EncyCMUCLopedia/
Paolo Amoroso <·······@mclink.it> wrote in message
·····················@news.mclink.it...
> On Mon, 11 Oct 1999 19:02:33 +0100, "Jonathan"
> <········@meanwhile.freeserve.co.uk> wrote:
>
> > When I first came to Lisp I was expecting to find Tk and maybe Wx
available.
>
> If by Lisp you mean a family of languages, those toolkits are already
> available: Tk is supported by STk, Wx by DrScheme. Both of them are Scheme
> systems.
>
>
> Paolo
> --
> EncyCMUCLopedia * Extensive collection of CMU Common Lisp documentation
> http://cvs2.cons.org:8000/cmucl/doc/EncyCMUCLopedia/
Unfortunately, I meant Common Lisp - I don't think Scheme would stretch to
the couple of roles we want to try Lisp in. Allegro seems to be our only
feasible choice.
Jonathan
Jonathan wrote:
>
> When I first came to Lisp I was expecting to find Tk and maybe Wx available.
> It seems strange for the Lisp world to try and stick with a single language
> standard and not get the benefit of being part of a wider community of
> users. The benefit is a double one - Lisp gets better libraries, and it's
> easier for new programmers to get useful work done in Lisp. (Perl and Python
> would seem natural places for Lisp to draw strength from.)
>
> I find this mono-language standard especially strange as CLIM - from my
> perspective at least - seems rather under-powered. I know that my (much)
> preferred vendor, Franz, doesn't offer an OpenGL CLIM component, for
> instance. But I can that capability with Wx, and I think Tk (maybe depending
> on platform.) And opensource makes it really easy to add missing
> functionality to these toolkits.
>
> Jonathan Coupe, XWare
You will always encounter difficulties, regardless of the GUI-building kit.
There seems to be no easy way around this sad fact.
When using graphic packages that are "foreign" to Lisp, you must create
functions to administer the graphic resources, since they are usually unknown to
Lisp's memory management. In short, you must marry two, essentially separate
systems.
You will also have to tie the "problem-oriented" model you have constructed, to
a visualization/interaction model, which is usually different. This is fairly
easy when the relations between the two models are simple, eg: menus, icons,
text, simple curves and surfaces.
Experience shows that designing a GUI beyond the "simple approach" is a complex
task. (Which is why most application programs running in MS Windows or X11 has
buttons and pull-down menus all over the screen.)
CLIM was meant for the design of user visualization/interaction models at a more
advanced level. This approach requires more of it's programmers, there is no way
around *that*. The interesting question is, I think, whether CLIM is too hard to
penetrate when designing a GUI of today's standard?
You mention OpenGL as a desirable component, and I agree. I used Garnet years
ago, for a simple simulation demo. It worked well, but was much too slow at
screen updates. But I would prefer to have OpenGL functionality truly integrated
in a CLIM library, not merely available as a foreign package, and I would like
it to take advantage of any graphics hardware that is present.
But this is not a language issue of course.
Lars Lundb�ck
Lars Lundback <·······@eralslk.ericsson.se> wrote in message
······················@eralslk.ericsson.se...
> CLIM was meant for the design of user visualization/interaction models at
a more
> advanced level. This approach requires more of it's programmers, there is
no way
> around *that*. The interesting question is, I think, whether CLIM is too
hard to
> penetrate when designing a GUI of today's standard?
I'm not quite sure what a "more advanced level" means in this context. (I'd
be interested to hear though.) Tk and Wx both offer everything I need for
GUI. But then my main coding takes place a long way from GUI - its at most a
tertiary isssue in my own work. I find the fast to use GUI toolkits useful
because they let me get back to doing interesting stuff fast. I'm willing to
give up some generality to do this.
> You mention OpenGL as a desirable component, and I agree. I used Garnet
years
> ago, for a simple simulation demo. It worked well, but was much too slow
at
> screen updates. But I would prefer to have OpenGL functionality truly
integrated
> in a CLIM library, not merely available as a foreign package, and I would
like
> it to take advantage of any graphics hardware that is present.
Taking advantage of hardware is pretty simple, under OpenGL or D3D - there
isn't much point to either of them without it imho. It's a real shame that
Allegro and Harlequin don't come with 3D libraries. They'd be cheap to add,
add real functionality to each, and maybe open up an extra market to Lisp.
Jonathan Coupe, XWare
Jonathan <········@meanwhile.freeserve.co.uk> wrote in message
·················@news7.svr.pol.co.uk...
> Taking advantage of hardware is pretty simple, under OpenGL or D3D - there
> isn't much point to either of them without it imho. It's a real shame that
> Allegro and Harlequin don't come with 3D libraries. They'd be cheap to
add,
> add real functionality to each, and maybe open up an extra market to Lisp.
>
> Jonathan Coupe, XWare
Harlequin has an OpenGL Interface that works very well in LWW.
Marc Battyani
In article <············@news7.svr.pol.co.uk>, "Jonathan" <········@meanwhile.freeserve.co.uk> wrote:
> Taking advantage of hardware is pretty simple, under OpenGL or D3D - there
> isn't much point to either of them without it imho. It's a real shame that
> Allegro and Harlequin don't come with 3D libraries. They'd be cheap to add,
Hmm, my version of Harlequin's LispWorks for Windows has an OpenGL
example.
Rainer Joswig <······@lavielle.com> wrote in message
····························@194.163.195.67...
> In article <············@news7.svr.pol.co.uk>, "Jonathan"
<········@meanwhile.freeserve.co.uk> wrote:
>
> > Taking advantage of hardware is pretty simple, under OpenGL or D3D -
there
> > isn't much point to either of them without it imho. It's a real shame
that
> > Allegro and Harlequin don't come with 3D libraries. They'd be cheap to
add,
>
> Hmm, my version of Harlequin's LispWorks for Windows has an OpenGL
> example.
It wouldn't surprise me. I found it almost impossible to get information out
of Harlequin when I called them. They neglected to answer repeated calls
about weeks - looking back this would have been just before the buyout. When
I finally got hold of someone, he later emailed me reversing some of the
"information" he gave me. It seems the OpenGL thing was an another mistake.
I'd love for Harlequin to be as good as Allegro - but the benchmarks seem
way behind (matters in my business.) And I ideally need efficient
multithreading support on Windows NT (and maybe other OS's too.) Pity:
Harlequin's cheaper and the license is so much more pleasant.
Jonathan
From: Robert Monfera
Subject: Re: CLIM, does anyone really use this beast??
Date:
Message-ID: <3805271A.404D7465@fisec.com>
Jonathan wrote:
> I'd love for Harlequin to be as good as Allegro - but the benchmarks seem
> way behind (matters in my business.)
What benchmarks have you obtained? (It seems that you do not have
Lispworks installed.)
> And I ideally need efficient
> multithreading support on Windows NT (and maybe other OS's too.)
Do those benchmarks cover multithreading efficiency?
> Pity:
> Harlequin's cheaper and the license is so much more pleasant.
I'm sure you could convince Harlequin to give you a higher price and a
more restrictive license :-)
Robert
Jonathan wrote:
>
> I'm not quite sure what a "more advanced level" means in this context. (I'd
> be interested to hear though.) Tk and Wx both offer everything I need for
> GUI. But then my main coding takes place a long way from GUI - its at most a
> tertiary isssue in my own work. I find the fast to use GUI toolkits useful
> because they let me get back to doing interesting stuff fast. I'm willing to
> give up some generality to do this.
>
I would say both yes and no. My needs at that time were somewhat similar, and
the GUI was not a main issue. The direct method, ie: to patch a foreign graphics
library onto Lisp and then go straight ahead, could be quick, especially if the
internal structures and components in your GUI need not be reused in other work,
or by other people. But Garnet did provide a functionality that saved me a lot
of tedious low-level work, and the first results were soon available for
feedback into the program design.
I guess that by "more advanced level", I mean a system that can do a better job
at catching the user's input and then interpret his intention, according to the
context. The environment in some of the Lisp Machines is one example. Another
could be one of the best 3D-modelling CAD systems of today. Some 25 years ago,
when I was investigating possibilities to implement such functionality in
application programs, mainly in the 3D area, we regarded Lisp as the most
promising candidate.
One introduction to this topic is available at
http://kogs-www.informatik.uni-hamburg.de/~moeller/uims-clim/clim-intro.html
which describes the situation. When trying to reach this "advanced level", it is
important to make the system sensitive to mouse movements as well, and to relate
these to the context. I think that a lot of extra effort is needed if a standard
GUI toolkit is chosen.
>
> Taking advantage of hardware is pretty simple, under OpenGL or D3D - there
> isn't much point to either of them without it imho. It's a real shame that
> Allegro and Harlequin don't come with 3D libraries. They'd be cheap to add,
> add real functionality to each, and maybe open up an extra market to Lisp.
>
It is not always so, since it depends on the particular library implementation.
I would like to add that the main advantage with OpenGL is the standard it
imposes, rather than it's speed - if any.
3D libraries for Lisp have been available for quite some time now, as have
Lisp-based GUI toolkits. But they enforce a mental model upon you, which is
different from the usual simple one. Perhaps the beast to conquer is deficient
introductory documents and unsatisfactory examples to learn from?
Regards, Lars
Lars Lundback <·······@eralslk.ericsson.se> wrote in message
·····················@eralslk.ericsson.se...
> Jonathan wrote:
> >
> > I'm not quite sure what a "more advanced level" means in this context.
(I'd
> > be interested to hear though.) Tk and Wx both offer everything I need
for
> > GUI. But then my main coding takes place a long way from GUI - its at
most a
> > tertiary isssue in my own work. I find the fast to use GUI toolkits
useful
> > because they let me get back to doing interesting stuff fast. I'm
willing to
> > give up some generality to do this.
> >
>
> I would say both yes and no. My needs at that time were somewhat similar,
and
> the GUI was not a main issue. The direct method, ie: to patch a foreign
graphics
> library onto Lisp and then go straight ahead, could be quick, especially
if the
> internal structures and components in your GUI need not be reused in other
work,
> or by other people. But Garnet did provide a functionality that saved me a
lot
> of tedious low-level work, and the first results were soon available for
> feedback into the program design.
>
> I guess that by "more advanced level", I mean a system that can do a
better job
> at catching the user's input and then interpret his intention, according
to the
> context. The environment in some of the Lisp Machines is one example.
Another
> could be one of the best 3D-modelling CAD systems of today. Some 25 years
ago,
> when I was investigating possibilities to implement such functionality in
> application programs, mainly in the 3D area, we regarded Lisp as the most
> promising candidate.
>
> One introduction to this topic is available at
>
http://kogs-www.informatik.uni-hamburg.de/~moeller/uims-clim/clim-intro.html
> which describes the situation. When trying to reach this "advanced level",
it is
> important to make the system sensitive to mouse movements as well, and to
relate
> these to the context. I think that a lot of extra effort is needed if a
standard
> GUI toolkit is chosen.
>
I'd say that's a damn good definition of "advanced" in GUI design. I'd love
to see a Lisp Machine someday. Does anyone know of any in the UK? With
owners who would't mind a visit?
> >
> > Taking advantage of hardware is pretty simple, under OpenGL or D3D -
there
> > isn't much point to either of them without it imho. It's a real shame
that
> > Allegro and Harlequin don't come with 3D libraries. They'd be cheap to
add,
> > add real functionality to each, and maybe open up an extra market to
Lisp.
> >
>
> It is not always so, since it depends on the particular library
implementation.
> I would like to add that the main advantage with OpenGL is the standard it
> imposes, rather than it's speed - if any.
>
Regarding speed, you mean compared to desktop graphics, or 3D card vendor's
own API?
I don't know any 3D card vendor who has a Lisp api. And a modern $200 OpenGL
or D3D card will run at speeds dozens of times beyond that you can reach
with desktop graphics if properly used - certainly for polygons fills. And
now you can lighting and geometry acceleartion in a consumer level card. In
a few weeks you'll probably be able to get gaussian filtering too.
> 3D libraries for Lisp have been available for quite some time now, as have
> Lisp-based GUI toolkits. But they enforce a mental model upon you, which
is
> different from the usual simple one. Perhaps the beast to conquer is
deficient
> introductory documents and unsatisfactory examples to learn from?
>
> Regards, Lars
I agree. But the toolkits that do that are the Performer level ones. I just
want to be able to initialize any OpenGL compliant card from a (fast,
multithreaded, well-supported) version of Lisp, and then have a function to
hook into for ploygon drawing. And for geometry and lighting, as they move
into hardware.
Jonathan
>>>>> "Jonathan" == Jonathan <········@meanwhile.freeserve.co.uk> writes:
Jonathan> When I first came to Lisp I was expecting to find Tk and
Jonathan> maybe Wx available.
GCL used to (still does) have an interface to Tk. I think it ran by
starting up a Tk app and sending commands via a pipe or socket. I
think someone ported this to CMUCL, perhaps CLISP.
Don't know if this satisfies your expectations or not.
Ray
Raymond Toy <···@rtp.ericsson.se> writes:
> >>>>> "Jonathan" == Jonathan <········@meanwhile.freeserve.co.uk> writes:
>
> Jonathan> When I first came to Lisp I was expecting to find Tk and
> Jonathan> maybe Wx available.
>
> GCL used to (still does) have an interface to Tk. I think it ran by
> starting up a Tk app and sending commands via a pipe or socket. I
> think someone ported this to CMUCL, perhaps CLISP.
>
> Don't know if this satisfies your expectations or not.
You may want to check out John Harper's librep[1], which has a GTK
interface. It's rather small and meant to be used as an extention
language. He's done some really cool stuff with it, however, like Sawmill[2]
Footnotes:
[1] <URL:http://www.dcs.warwick.ac.uk/~john/sw/librep.html>
[2] <URL:http://www.dcs.warwick.ac.uk/~john/sw/sawmill/index.html>
--
# Erik Arneson ····@starseed.com Webring Software Engineer #
# Yahoo! Inc. PGP ID: 2048/84413E19 (541) 482-3000x114 #
# "There's such a fine line between stupid and clever." Spinal Tap #
Raymond Toy <···@rtp.ericsson.se> wrote in message
···················@rtp.ericsson.se...
> >>>>> "Jonathan" == Jonathan <········@meanwhile.freeserve.co.uk> writes:
>
> Jonathan> When I first came to Lisp I was expecting to find Tk and
> Jonathan> maybe Wx available.
>
> GCL used to (still does) have an interface to Tk. I think it ran by
> starting up a Tk app and sending commands via a pipe or socket. I
> think someone ported this to CMUCL, perhaps CLISP.
>
> Don't know if this satisfies your expectations or not.
>
> Ray
>
It's nice to know, but I for one need Windows. What I'm asking here is
whether people think that it might be best to have a cross OS and cross
language standard that Lisp was part of.
Jonathan
From: Clint Hyde
Subject: Re: CLIM, does anyone really use this beast??
Date:
Message-ID: <3810D0CE.E60DAEC7@bbn.com>
in late to this discussion...I've been building systems with Lisp/CLIM since
1992, and currently have systems
in the field. just demo'd yesterday/today to folks at the Pentagon.
my current system runs on both Unix and Win NT (clim is not available for Win
98). It has some very interesting stuff
in it, incl drag/drop in several places, lots of mouse-sensitivity (probably far
more than typical Windoze users are
used to).
if you want to see a presentation on it, or read the tutorial:
http://phaedrus.gteinetva.bbnplanet.net/
so yes, I use CLIM. it is WAY powerful--far more so than other UI toolkits I've
used (plain X/Motif, Java, TCL/TK).
It is also simpler, in that your objects don't have to inherit anything in order
to become mouse-sensitive.
In fact, I am doing a pseudo-port of behavior/concepts to Java these days. (so
far I have mouse-sensitive objects on screen,
a mouse-doc window, commands/tables, presentations, pres-translators.) the one
gaping hole I don't have filled is
"format-graph-from root", but I expect to be writing that this fall.
-- clint
as I recall, Nick Ing-Simmons (?) created a lisp interface to libtk.so, so that
one could build a TK UI. I haven't tried
this out. I sure with CLIM had a text-widget of equal power to TK's text widget.
>
> It's nice to know, but I for one need Windows. What I'm asking here is
> whether people think that it might be best to have a cross OS and cross
> language standard that Lisp was part of.
>
> Jonathan
of course. good standards are always good things to have.
From: Arthur Lemmens
Subject: Re: CLIM, does anyone really use this beast??
Date:
Message-ID: <38116578.C3A7551B@simplex.nl>
Clint Hyde wrote:
> (clim is not available for Win 98)
The Professional Edition of Lispworks for Windows includes CLIM.
The CLIM demos I've been playing with run fine on Windows 98.
Arthur Lemmens