I have mentioned in another thread that our hospital uses muLisp for
pharmacokinetic modeling. I do not know if there are other lisp programmers
interested in drug dose modeling but if there are the University of Florida
Pharmacy School has a new web site with a discussion of computer simulations
at
http://www.bluemountain.com/cards/box3724e/hft4gnsseypdja.htm
by Dr Guenther Hochhaus. Select the "Discussion" selection box under
"Simulations" in the scrolling window on the left of the screen.
Our muLisp system is still in DOS and the output (graphs, charts, etc) are
still in characters. It would be a wonderful application for a
windows/graphics platform. Any suggestions as to what you would do?
Rusty
Dang it... I posted the wrong URL to the UoF Phamarcy School. The correct
URL
is
http://www.cop.ufl.edu/safezone/pat/pha5127/index.htm
Gee I can not wait to see where the wrong URL points???? OMG
sorry
Rusty
rusty craine wrote in message <············@excalibur.flash.net>...
>I have mentioned in another thread that our hospital uses muLisp for
>pharmacokinetic modeling. I do not know if there are other lisp
programmers
>interested in drug dose modeling but if there are the University of Florida
>Pharmacy School has a new web site with a discussion of computer
simulations
>at
>
>http://www.bluemountain.com/cards/box3724e/hft4gnsseypdja.htm
>
>by Dr Guenther Hochhaus. Select the "Discussion" selection box under
>"Simulations" in the scrolling window on the left of the screen.
>
>Our muLisp system is still in DOS and the output (graphs, charts, etc) are
>still in characters. It would be a wonderful application for a
>windows/graphics platform. Any suggestions as to what you would do?
>
>Rusty
>
>
>
>
On Fri, 23 Oct 1998 17:48:08 -0500, "rusty craine"
<········@flash.net> claimed or asked:
% Dang it... I posted the wrong URL to the UoF Phamarcy School. The correct
% URL
% is
% http://www.cop.ufl.edu/safezone/pat/pha5127/index.htm
%
% Gee I can not wait to see where the wrong URL points???? OMG
Well, they _might_ qualify as medical pictures :-)
Just kidding. I didn't look.
--
David Steuber (ver 1.31.2a)
http://www.david-steuber.com
To reply by e-mail, replace trashcan with david.
"Ignore reality there's nothing you can do about it..."
-- Natalie Imbruglia "Don't you think?"
On Fri, 23 Oct 1998 16:54:35 -0500, "rusty craine"
<········@flash.net> claimed or asked:
% Our muLisp system is still in DOS and the output (graphs, charts, etc) are
% still in characters. It would be a wonderful application for a
% windows/graphics platform. Any suggestions as to what you would do?
I would upgrade to Linux and CMUCL.
Oh, I did. :-) Some of the Garnet demos look pretty cool,
particularly the Motif ones. Garnet is available separately from
cons.org. CMUCL is also available at cons.org.
Linux is everywhere!
--
David Steuber (ver 1.31.2a)
http://www.david-steuber.com
To reply by e-mail, replace trashcan with david.
"Ignore reality there's nothing you can do about it..."
-- Natalie Imbruglia "Don't you think?"
I shall be chastized for posting this "off track" post but...
At work we have always stayed with main stream software vendors and OS's for
the vendor support. QNX instead of Linux RTOS modules. MuLisp vs Xlisp or
Clist, etc. This way we could be an expert in our fields of science or
engineering and apply that directly to the problems with a little help for
*our friends*. With the advent of very high quailty free ware, is the
paradigm changing?. Now the users are the support. Scary for an
application manager!
>>>as written by Dr S. Harms (used with his blessing)
It would be interesting to draw a parallel with a simplified evolution
theory here. Multiple mutations of organisms "trying their luck" out there
competing for the limited successful niches. I guess you could identify at
least two levels of organisms here: the actual software and the companies
(or groups of people) that support them, together as a unified organism,
competing for attention and acceptance by the target organism (the
consumers). It becomes interesting when the consumers become the supportes,
(as in the example of Linux), since the same group of people do the two
functions that previously were separate. Does the focus then turn to on the
performance of the actual software itself or is professional support still
an issue in the commerical world?
>>>
Afraid of the answer
Rusty
David Steuber "The Interloper" wrote in message
<·················@news.newsguy.com>...
>On Fri, 23 Oct 1998 16:54:35 -0500, "rusty craine"
><········@flash.net> claimed or asked:
>
>% Our muLisp system is still in DOS and the output (graphs, charts, etc)
are
>% still in characters. It would be a wonderful application for a
>% windows/graphics platform. Any suggestions as to what you would do?
>
>I would upgrade to Linux and CMUCL.
>
>Oh, I did. :-) Some of the Garnet demos look pretty cool,
>particularly the Motif ones. Garnet is available separately from
>cons.org. CMUCL is also available at cons.org.
>
>Linux is everywhere!
>
>--
>David Steuber (ver 1.31.2a)
>http://www.david-steuber.com
>To reply by e-mail, replace trashcan with david.
>
>"Ignore reality there's nothing you can do about it..."
>-- Natalie Imbruglia "Don't you think?"
On Sun, 25 Oct 1998 11:06:34 -0600, "rusty craine"
<········@flash.net> claimed or asked:
% I shall be chastized for posting this "off track" post but...
% At work we have always stayed with main stream software vendors and OS's for
% the vendor support. QNX instead of Linux RTOS modules. MuLisp vs Xlisp or
% Clist, etc. This way we could be an expert in our fields of science or
% engineering and apply that directly to the problems with a little help for
% *our friends*. With the advent of very high quailty free ware, is the
% paradigm changing?. Now the users are the support. Scary for an
% application manager!
% >>>as written by Dr S. Harms (used with his blessing)
% It would be interesting to draw a parallel with a simplified evolution
% theory here. Multiple mutations of organisms "trying their luck" out there
% competing for the limited successful niches. I guess you could identify at
% least two levels of organisms here: the actual software and the companies
% (or groups of people) that support them, together as a unified organism,
% competing for attention and acceptance by the target organism (the
% consumers). It becomes interesting when the consumers become the supportes,
% (as in the example of Linux), since the same group of people do the two
% functions that previously were separate. Does the focus then turn to on the
% performance of the actual software itself or is professional support still
% an issue in the commerical world?
% >>>
%
% Afraid of the answer
% Rusty
Companies like to pay someone they can go crying to when something
doesn't work. They're kinda funny that way. The joke is on them when
the company they go crying to is Microsoft, and Microsoft says, "we
don't support that configuration."[1]
I don't know the fate of 'free' software. In the case of Linux, the
end users are after something that works. The original end users were
also the developers (starting with just one Finnish college student).
Linux has become something more than a research project. It is now a
useful, industrial strength OS. It is rapidly becoming as easy, if
not easier, to install than the atrocities from Microsoft. There
seems to be a critical mass forming around this OS. Some nice
articles on the matter are published on www.linuxworld.com.
Then there is CMUCL. CMU builds a new lisp as part of some DARPA
research project. Now it is in the public domain. A group of people
who maintain a site at www.cons.org continue to maintain CMUCL. Why
do they do this? I bet it's because they want a good free lisp
around. Public domain is even more free (as in liberty) than GPL
software. You can use CMUCL for both proprietary and open source
development. Since you do in house software, that doesn't really
matter to you. But this might. The people who work on CMUCL have
helped me, through this forum and private e-mail, to get set up and
running with it. The responsiveness and the usefulness of the support
far surpasses anything that Microsoft has ever come up with in my
experience.
You can get Linux support. Red Hat sells it, as do other
distributors. If you are in Europe, SuSE is right there. Of course,
with e-mail, geography is a moot point.
Now I don't want to give the impression that I am an expert or
anything. Quite the reverse. While I am not the least common
denominator end user of OS and development software, I do like things
to plug right in and work. The Mac would suit me fine, but I do also
like to tinker under the hood, even if that means failing emissions.
The bottom line, for you and other commercial users, is what is the
risk that support will be un-available or that the product will become
un-maintained? In the case of Linux, I would hazard a speculation
that the risk is near zero. In the case of CMUCL, I can't say. It
seems like a greater risk, but I get the impression that the
maintainers will keep at it for a while longer. Even if they don't,
the source is there. That just means becoming your own maintainer
(the same can apply to Linux). I should point out that the risk is
probably higher when going with a company like Microsoft. They don't
care about you. IE 3.0 has a serious memory leak bug for posted
forms. Microsoft's answer is for you to upgrade to 4.0. Netscape is
really no better in that respect. They don't support the older
versions. So based on my admittedly narrow world view, I would say
that you are actually reducing your risk when you go with certain free
software packages. If you pick good quality software, you have a
reasonable chance that other people are doing the same thing.
My conjecture is that Linux and CMUCL qualify as quality software. I
would say that I have had much better support for CMUCL than I have
had for Linux. Possibly that is because the user community is
smaller. Possibly because I've been a squeaky wheel. Lisp in general
is not a mainstream language. People look at the (((()))))) and get
intimidated. Lisp hasn't achieved credibility yet. I can't say that
it will ever be more than a niche language, but it isn't going to die.
How many languages can return a function from a function call? Neat
stuff like that doesn't just disappear.
You may of course decide to upgrade to Windows 98 where only
commercial Lisps will work for you. It is a more expensive solution.
It may be more risky too, but I have no way to prove it. It's your
call. I've made mine.
--
David Steuber (ver 1.31.2a)
http://www.david-steuber.com
To reply by e-mail, replace trashcan with david.
"Ignore reality there's nothing you can do about it..."
-- Natalie Imbruglia "Don't you think?"
In article <·················@news.newsguy.com>,
········@david-steuber.com wrote:
> matter to you. But this might. The people who work on CMUCL have
> helped me, through this forum and private e-mail, to get set up and
> running with it. The responsiveness and the usefulness of the support
> far surpasses anything that Microsoft has ever come up with in my
> experience.
Sorry, I don't buy it. If you want something special from
the maintainers - you'll end up implementing it on your
own. Great - but I don't want to be an expert on
threads, FFI, low-level software debugging, TCP/IP
interfacing etc. I can get software where it already works.
> You can get Linux support. Red Hat sells it, as do other
> distributors. If you are in Europe, SuSE is right there. Of course,
> with e-mail, geography is a moot point.
What is that different to buying support from Microsoft?
> The Mac would suit me fine, but I do also
> like to tinker under the hood, even if that means failing emissions.
I don't understand that. Are you under the impression that
you can't do low-level or system-level programming on a Mac?
> That just means becoming your own maintainer
> (the same can apply to Linux).
Good luck maintaining a multi-thousand lines software
doing really difficult stuff. Try to debug the type inferencer.
> I should point out that the risk is
> probably higher when going with a company like Microsoft. They don't
> care about you. IE 3.0 has a serious memory leak bug for posted
> forms. Microsoft's answer is for you to upgrade to 4.0.
Great. They have fixed the bug. You get a new version.
That's what I'd expect. What would you want more than that?
*I*'d be happy.
> really no better in that respect. They don't support the older
> versions.
Why should they? I don't care about old buggy software.
Does anybody support or maintain Emacs 17.23?
> My conjecture is that Linux and CMUCL qualify as quality software.
I seem to have a different view on "quality software".
You mean a system that is misdesign from the start,
a copy of what is already there is quality software?
Try to upgrade you Linux system. Will your software
still run? How many hours would you need hunting the
right shared libraries? Correcting you scripts?
Upgrading to new versions of your software?
Which installer do you use? Which one can you use?
> You may of course decide to upgrade to Windows 98 where only
> commercial Lisps will work for you. It is a more expensive solution.
How do you measure "expensive"? What does your time cost?
What does a failed project cost? What will hiring some
hacker to get it fixed cost? What does a day late delivery
cost?
Can you always choose your platform? We currently are
contemplating a project where the customer wants
a GUI-based software for Windows NT. Who am I to
tell an aircraft company what platform they should
run the software on? With luck they want it in Lisp.
We have been using so the called "free" and "quality software".
At the end you will need to do C programming to fix
basic stuff - you even can try to get the GC really
working, or even reading from input. Some of these
can really be **show**stoppers**.
Careful, I'm not saying that Linux or CMU CL is generally
bad (well, o.k., Unix in general is evil). But you need
to look around and see that there are some more
constraints than just that it has to be "free", "cheap" or
"open". Whatever these buzzwords mean.
Still, ACL, LispWorks, MCL and Genera (those I have seen)
are in many respects light years ahead of CMU CL.
I have great respect for all these people who are not
whining, but writing and maintaining useful
software (like CMU CL, CLisp, SIOD, CL-HTTP, ...). And I have
the same respect for people working in companies
providing their excellent tools. Whining about the
current situation will not produce one line of useful
code. Same for the people who apologize all the
time. If you are apologizing that you want a certain
project to be Lisp-based (because it's more fun or whatever),
than you don't have the spirit to get it done. I mean,
who needed a Lisp-based web server? Today it is running
in the White House, maintained there by really
ordinary (well, almost ;-) ) support people, delivering
documents to the ordinary citizen (I even once saw a TV report
where people told how happy they are with this service - it does
improve democracy by providing documents unfiltered
by mass media). O.k., enough preaching. ;-)
--
http://www.lavielle.com/~joswig
In article <·······················@pbg3.lavielle.com>,
······@lavielle.com says...
> Sorry, I don't buy it. If you want something special from
> the maintainers - you'll end up implementing it on your
> own. Great - but I don't want to be an expert on
> threads, FFI, low-level software debugging, TCP/IP
> interfacing etc. I can get software where it already works.
It's one of those choice issues. Software that will please the largest
number of people might not please an individual. It's great to have a
choice, but not so great if you're forced to make that choice, where
"choice" means access to the source code and the right to modify it.
"Shut up, be happy. The conveniences you demanded are now mandatory."
-- Jello Biafra
This quote expresses an alternate view, which is that conveniences
exclude choices. I have doubts about this. However, I don't believe that
choices always exclude conveniences.
IMHO, it depends rather more on the assumptions made by the people who
create the software - and the people who use it. It takes two to tango.
> > That just means becoming your own maintainer
> > (the same can apply to Linux).
>
> Good luck maintaining a multi-thousand lines software
> doing really difficult stuff. Try to debug the type inferencer.
Agreed. Access to the source code and the right to modify it and use the
results is not (always) enough. For example, even if I have the source
for a device driver, I may prefer not to debug it myself. Even if I were
paid to do it.
Likewise, I understand the principles of type inferencing, but life is
too short to debug, say, CMU CL's type inferencer. There are others who
are much better qualified and _motivated_ to do that.
This is why consultants exist. ;)
> Try to upgrade you Linux system. Will your software
> still run? How many hours would you need hunting the
> right shared libraries? Correcting you scripts?
> Upgrading to new versions of your software?
> Which installer do you use? Which one can you use?
This sounds dangerously like an anti-Linux argument, even if you are
asking some good questions. Similar questions can be asked about other
operating systems, too.
Even upgrading DOS has caused nightmares.
> How do you measure "expensive"? What does your time cost?
> What does a failed project cost? What will hiring some
> hacker to get it fixed cost? What does a day late delivery
> cost?
These are also good questions.
> Can you always choose your platform? We currently are
> contemplating a project where the customer wants
> a GUI-based software for Windows NT. Who am I to
> tell an aircraft company what platform they should
> run the software on? With luck they want it in Lisp.
With luck, yes. It'll take a lot more than a hard sell. Productivity is
meaningless in a world of ideology. Big customers, like aircraft
companies or anything remotely corporate, may have their own internal
ideology, plus who knows what kind of political infighting.
Where does a 300 pound gorilla sit?
> We have been using so the called "free" and "quality software".
> At the end you will need to do C programming to fix
> basic stuff - you even can try to get the GC really
> working, or even reading from input. Some of these
> can really be **show**stoppers**.
They're certainly critical. At best, they can impact the performance of
your code. If anyone cares about performance, then this can hurt you.
> Careful, I'm not saying that Linux or CMU CL is generally
> bad (well, o.k., Unix in general is evil). But you need
> to look around and see that there are some more
> constraints than just that it has to be "free", "cheap" or
> "open". Whatever these buzzwords mean.
> Still, ACL, LispWorks, MCL and Genera (those I have seen)
> are in many respects light years ahead of CMU CL.
I've used ACL, LW and CMU CL, and my #1 choice would be ACL, followed by
LW. The latter wins on price, which can matter if your budget is small.
> I have great respect for all these people who are not
> whining, but writing and maintaining useful
> software (like CMU CL, CLisp, SIOD, CL-HTTP, ...). And I have
Likewise.
> the same respect for people working in companies
> providing their excellent tools. Whining about the
> current situation will not produce one line of useful
> code. Same for the people who apologize all the
> time. If you are apologizing that you want a certain
> project to be Lisp-based (because it's more fun or whatever),
> than you don't have the spirit to get it done. I mean,
For me, Lisp is a simple choice. I'm more productive in Lisp than in
anything else. That's not an apology. However, pointing it out has got me
a few blank looks. So does the same point about Perl. Go figure.
My answer to any question about my choice of tools is simple: "When you
pay for my time, you can tell me what to use." _That_ might be an
apology, but see my sig.file for an alternate phrasing.
> who needed a Lisp-based web server
> in the White House, maintained there by really
> ordinary (well, almost ;-) ) support people, delivering
> documents to the ordinary citizen (I even once saw a TV report
> where people told how happy they are with this service - it does
> improve democracy by providing documents unfiltered
> by mass media). O.k., enough preaching. ;-)
Anyway, you're preaching to the converted. ;) I've yet to see a TV report
that mentions Apache, never mind CL-HTTP, but I live in hope.
--
Remove insect from address to email me | You can never browse enough
will write code that writes code that writes code for food
In article <··························@news.demon.co.uk>,
···@wildcard.butterfly.demon.co.uk (Martin Rodgers) wrote:
> Likewise, I understand the principles of type inferencing, but life is
> too short to debug, say, CMU CL's type inferencer. There are others who
> are much better qualified and _motivated_ to do that.
>
> This is why consultants exist. ;)
Or vendors.
> > Try to upgrade you Linux system. Will your software
> > still run? How many hours would you need hunting the
> > right shared libraries? Correcting you scripts?
> > Upgrading to new versions of your software?
> > Which installer do you use? Which one can you use?
>
> This sounds dangerously like an anti-Linux argument, even if you are
> asking some good questions. Similar questions can be asked about other
> operating systems, too.
Sure, but some people seem to think that Linux is different.
It is not. It is worse in some respects. I have updated
my Mac yesterday from 8.1 to 8.5 in two hours. Everything
still works. Only faster.
> > tell an aircraft company what platform they should
> > run the software on? With luck they want it in Lisp.
>
> With luck, yes. It'll take a lot more than a hard sell. Productivity is
> meaningless in a world of ideology. Big customers, like aircraft
> companies or anything remotely corporate, may have their own internal
> ideology, plus who knows what kind of political infighting.
Luckily there are current version is also in Lisp - also a bit
exotic.
> They're certainly critical. At best, they can impact the performance of
> your code. If anyone cares about performance, then this can hurt you.
Sure, especially if you have to serve hundreds or thousands
users a day.
> For me, Lisp is a simple choice. I'm more productive in Lisp than in
> anything else.
So, what have been your experiences lately? We had some discussions
earlier this year. Anything you are allowed to tell?
--
http://www.lavielle.com/~joswig
In article <·······················@pbg3.lavielle.com>,
······@lavielle.com says...
> > This is why consultants exist. ;)
>
> Or vendors.
I didn't mention them coz I'm a consultant. ;)
> Sure, but some people seem to think that Linux is different.
> It is not. It is worse in some respects. I have updated
> my Mac yesterday from 8.1 to 8.5 in two hours. Everything
> still works. Only faster.
Everything is "worse in some respects". Hence the popularity of the
phrase, "Your Mileage May Vary", or YMMV for short.
I've been hassled and flamed by Linux Advocates telling me I can't manage
without Linux, or how the fate of humanity depends on me renouncing NT in
favour of Linux. So I remind people that I'm a consultant, and I use the
same OS as the people who pay me. Pay me to use Linux and I will.
FWIW, I have two machines here. This one runs NT, and the other Linux.
If anyone wants me to use a Mac, they can pay me like anyone else.
Speed isn't relevant. Only the money is. Everything else is a luxury.
> Sure, especially if you have to serve hundreds or thousands
> users a day.
Or if your code has to run on very old hardware.
> So, what have been your experiences lately? We had some discussions
> earlier this year. Anything you are allowed to tell?
The situation is grim. I think my sig.file puts it rather neatly.
--
Remove insect from address to email me | You can never browse enough
will write code that writes code that writes code for food
···@wildcard.butterfly.demon.co.uk (Martin Rodgers) writes:
> I've been hassled and flamed by Linux Advocates telling me I can't manage
> without Linux, or how the fate of humanity depends on me renouncing NT in
> favour of Linux.
Absolute Truths Are Always A Bad Thing!
And You Cannot Disagree With That!
Stig Hemmer,
Jack of a Few Trades.
From: Klaus Schilling
Subject: Re: Web Site if you are Interested
Date:
Message-ID: <87hfwq2e7b.fsf@ivm.de>
······@lavielle.com (Rainer Joswig) writes:
> In article <·················@news.newsguy.com>,
> ········@david-steuber.com wrote:
>
> > You can get Linux support. Red Hat sells it, as do other
> > distributors. If you are in Europe, SuSE is right there. Of course,
> > with e-mail, geography is a moot point.
>
> What is that different to buying support from Microsoft?
The difference is that with microcrap, one is forced to buy support from Gate$,
with Linux, there's free competition.
>
> > The Mac would suit me fine, but I do also
> > like to tinker under the hood, even if that means failing emissions.
>
> I don't understand that. Are you under the impression that
> you can't do low-level or system-level programming on a Mac?
>
The only thing one can do with an Apple is verifying gravitation.
Klaus Schilling
In article <··············@ivm.de>, ···············@home.ivm.de says...
> The difference is that with microcrap, one is forced to buy support from Gate$,
> with Linux, there's free competition.
Ah, ideology. You wouldn't be a Linux Advocate, would you?
> The only thing one can do with an Apple is verifying gravitation.
A most provocative statement. Almost a troll.
--
Remove insect from address to email me | You can never browse enough
will write code that writes code that writes code for food
From: Christopher Browne
Subject: Re: Web Site if you are Interested
Date:
Message-ID: <7166p6$s7p$5@blue.hex.net>
On Tue, 27 Oct 1998 19:44:08 -0000, Martin Rodgers
<···@wildcard.butterfly.demon.co.uk> wrote:
>In article <··············@ivm.de>, ···············@home.ivm.de says...
>> The difference is that with microcrap, one is forced to buy support
>> from Gate$, with Linux, there's free competition.
>
>Ah, ideology. You wouldn't be a Linux Advocate, would you?
There are "advo-kids," and there are advocates, and there are Advocates.
Klaus is not so much of a Linux Advocate as he is a GPL advocate. I
expect that if Hurd were released in other than "alpha" form, he'd not
be nearly so much of an apparent advocate of Linux.
>> The only thing one can do with an Apple is verifying gravitation.
>
>A most provocative statement. Almost a troll.
"Almost" is not the right word, and you evidently haven't seen Klaus
getting "most provocative." This isn't even close...
--
"I don't know what OS we will be running in 40 years, but it will probably
be called UNIX...." -- Chris Mikkelson <········@maroon.tc.umn.edu>
········@ntlug.org- <http//www.hex.net/~cbbrowne/lsf.html>
In article <············@blue.hex.net>, ········@news.hex.net says...
> There are "advo-kids," and there are advocates, and there are Advocates.
> Klaus is not so much of a Linux Advocate as he is a GPL advocate. I
> expect that if Hurd were released in other than "alpha" form, he'd not
> be nearly so much of an apparent advocate of Linux.
It's still ideology, and this isn't even an OS newsgroup.
>> A most provocative statement. Almost a troll.
> "Almost" is not the right word, and you evidently haven't seen Klaus
> getting "most provocative." This isn't even close...
Compared to what I've read in other newsgroups, it's pretty tame. OTOH,
I'm often amazed at how sensitive people can be. Almost anything can be a
troll, these days. I hope it's a just a millenial thing...
--
Remove insect from address to email me | You can never browse enough
will write code that writes code that writes code for food
······@lavielle.com (Rainer Joswig) writes:
> In article <·················@news.newsguy.com>,
> ········@david-steuber.com wrote:
>
> > matter to you. But this might. The people who work on CMUCL have
> > helped me, through this forum and private e-mail, to get set up and
> > running with it. The responsiveness and the usefulness of the support
> > far surpasses anything that Microsoft has ever come up with in my
> > experience.
>
> Sorry, I don't buy it. If you want something special from
> the maintainers - you'll end up implementing it on your
> own. Great - but I don't want to be an expert on
> threads, FFI, low-level software debugging, TCP/IP
> interfacing etc. I can get software where it already works.
>
> > You can get Linux support. Red Hat sells it, as do other
> > distributors. If you are in Europe, SuSE is right there. Of course,
> > with e-mail, geography is a moot point.
>
> What is that different to buying support from Microsoft?
the difference is this. there are two kinds of software support ---
hand-holding and programming. in the first case, microsoft or redhat
guide you step by step through an install or configuration. in the
second, you report a problem/issue/bug and you can get a fixed program
that does what you want and not what you don't. even have them add a
feature to help to out even if the program isn't broken.
commercial software often does have both kinds of support. if you
want true programming support, you have to pay a lot of money.
synopsis, opnet, and to some extent matlab give good support of both
kinds.
microsoft offers hand-holding as far as i can see. although i
suppose if you give them enough money, even microsoft will provide
programming. if you're not a fortune 500 company, however, you
probably don't have that kind of money.
with linux, the vendors are hungrier (they don't have 50 billion in
paper assets). the vendors are many -- don't like redhat, go to suse.
don't like them, hire the recent college grad or kid down the street.
you can get both hand-holding and programming support. a certain
extent of the programming support comes to you for free as the
upgrades don't cost you anything! since everyone, even you, has
access to the source, there's room for actual competition here to
insure that you get a good deal.
the difference is clear. with microsoft, you get a large bureaucratic
organization with monopoly of control over their software. with
linux, you have some semblance of a free market.
microsoft may have some good (or necessary) applications. linux is
weak in office suites with universally accepted file formats (they
aren't MS-Office). but as for support, linux is a good choice and has
potential to really win in the long run.
--
johan kullstam
In article <·············@res.raytheon.com>, Johan Kullstam
<·······@idt.net> wrote:
> microsoft offers hand-holding as far as i can see. although i
> suppose if you give them enough money, even microsoft will provide
> programming. if you're not a fortune 500 company, however, you
> probably don't have that kind of money.
There are a lot of vendors providing programming support even
in the Windows area.
> the difference is clear. with microsoft, you get a large bureaucratic
> organization with monopoly of control over their software. with
> linux, you have some semblance of a free market.
I don't like Microsoft and I find Windows really weak (Windows NT is a
joke as a server system, IMHO). I was using Oracle under
Windows NT. Pathetic. Currently I'm reading to the
installation of Oracle under Solaris. Pathetic, too.
> microsoft may have some good (or necessary) applications. linux is
> weak in office suites with universally accepted file formats (they
> aren't MS-Office).
There are a lot of productivity applications not available
on Linux. Freehand, Xpress, Illustrator, etc. are
standards in the graphics industry. You'll find nothing
remotely that usable under Linux.
> but as for support, linux is a good choice and has
> potential to really win in the long run.
I don't think so. Linux inherits all the weaknesses of the typical
Unix. User friendliness basically is foreign to Unix.
Desktops under Unix are still trying to copy 15 year old
paradigms. Where is progress?
I can only hope that Unix does not infect the minds of the
people, otherwise the current dark ages will last even longer.
I'm really tired of this stuff. If there
would be a really good Lisp operating system available for
popular platforms, I'd switch for a lot of tasks. I want
for example to get rid of the usual mail programs
like SENDMAIL. I also would like to get of PERL nightmares
like Majordomo. Who replaces the crontab by a sane piece
of software? inetd.conf? One must have a real lack of
fantasy to not imagine that this stuff could be beautifully
replaced by a Lisp OS with objects and persistence (
Unfortunately I don't have the time to write the new
computing infrastructure in my spare time - things can be frustrating).
No more crappy adhoc scripting or configuration languages.
Not a multitude of "interesting" formats of configuration
files. An understandable and debuggable OS with clear
abstractions would improve execution safety and security.
Strict end user focus with an easy UI. Too, bad - the
last attempt on this - the Newton OS - failed. It
had dynamic objects with persistence and a great UI.
It even didn't have a "save" command. Everything was
stored/compressed/indexed immediately - without
the need to save.
My advice: look around, read some books (start with an old
book by Erik Sandeval where the Xerox Lisp stuff
is being described in one chapter), get a wider view.
At some point you will see that Linux is neither the first nor
the last OS.
--
http://www.lavielle.com/~joswig
Rainer Joswig <······@lavielle.com> wrote:
>My advice: look around, read some books (start with an old
>book by Erik Sandeval where the Xerox Lisp stuff
>is being described in one chapter), get a wider view.
>At some point you will see that Linux is neither the first nor
>the last OS.
You probably mean Erik Sandewall, as in:
Interactive programming environments / editors, David R. Barstow, Howard
E. Shrobe, Erik Sandewall. New York : McGraw-Hill, c1984. xii, 609 p. :
ill. ; 24 cm.
LC CALL NUMBER: QA76.6 .I525 1984
ISBN: 0070038856 : $34.95
It's a depressing read. 14 years later and we're still trapped in
inferior environments.
John
In article <············@news.NERO.NET>, ·······@bronze.CS.ORST.EDU (John
Atwood) wrote:
> Rainer Joswig <······@lavielle.com> wrote:
> >My advice: look around, read some books (start with an old
> >book by Erik Sandeval where the Xerox Lisp stuff
> >is being described in one chapter), get a wider view.
> >At some point you will see that Linux is neither the first nor
> >the last OS.
>
> You probably mean Erik Sandewall, as in:
>
> Interactive programming environments / editors, David R. Barstow, Howard
> E. Shrobe, Erik Sandewall. New York : McGraw-Hill, c1984. xii, 609 p. :
> ill. ; 24 cm.
> LC CALL NUMBER: QA76.6 .I525 1984
> ISBN: 0070038856 : $34.95
Right. Great book. Thanks for the data.
> It's a depressing read. 14 years later and we're still trapped in
> inferior environments.
At the University years ago we heard about things like Xerox Lisp machines,
KEE, SmallTalk - still we were using character terminals.
Later I and a friend discovered that a research department
had some of these machines and . . . a Symbolics 3600 called "Hans"
(Martin, is Hans still alive?).
--
http://www.lavielle.com/~joswig
······@lavielle.com (Rainer Joswig) writes:
> In article <·············@res.raytheon.com>, Johan Kullstam
> <·······@idt.net> wrote:
>
> > microsoft offers hand-holding as far as i can see. although i
> > suppose if you give them enough money, even microsoft will provide
> > programming. if you're not a fortune 500 company, however, you
> > probably don't have that kind of money.
>
> There are a lot of vendors providing programming support even
> in the Windows area.
yes, but how many can fix bugs in ms-word or windows? can you
influence new features?
> > the difference is clear. with microsoft, you get a large bureaucratic
> > organization with monopoly of control over their software. with
> > linux, you have some semblance of a free market.
>
> I don't like Microsoft and I find Windows really weak (Windows NT is a
> joke as a server system, IMHO). I was using Oracle under
> Windows NT. Pathetic. Currently I'm reading to the
> installation of Oracle under Solaris. Pathetic, too.
nod.
> > microsoft may have some good (or necessary) applications. linux is
> > weak in office suites with universally accepted file formats (they
> > aren't MS-Office).
>
> There are a lot of productivity applications not available
> on Linux. Freehand, Xpress, Illustrator, etc. are
> standards in the graphics industry. You'll find nothing
> remotely that usable under Linux.
yes, but, for example, the US DoD requires all docs in `MS-Word 6.0 for
PC Windows' format. this is an impediment to choice.
> > but as for support, linux is a good choice and has
> > potential to really win in the long run.
well, i was trying to contrast free vs proprietary software and say
that real proprietary software programming support can be had but it
generally costs a lot of money.
> I don't think so. Linux inherits all the weaknesses of the typical
> Unix. User friendliness basically is foreign to Unix.
> Desktops under Unix are still trying to copy 15 year old
> paradigms. Where is progress?
nod. i have read and agreed with the unix-hater's handbook.
curiously, most every flaw of unix is either worse in a microsoft
operating system or that MS/OS is too feeble to support the required
functionality in order to exhibit the flaw.
> I can only hope that Unix does not infect the minds of the
> people, otherwise the current dark ages will last even longer.
> I'm really tired of this stuff. If there
> would be a really good Lisp operating system available for
> popular platforms, I'd switch for a lot of tasks. I want
> for example to get rid of the usual mail programs
> like SENDMAIL. I also would like to get of PERL nightmares
> like Majordomo. Who replaces the crontab by a sane piece
> of software? inetd.conf? One must have a real lack of
> fantasy to not imagine that this stuff could be beautifully
> replaced by a Lisp OS with objects and persistence
all these programs, sendmail, perl, the /etc configs could be done
better in a lisp. the configs are usually some form of associated
list in a weird format. C is a rotten language for general
programming work. using lisp for the config files and programming
makes a lot of sense.
> (Unfortunately I don't have the time to write the new
> computing infrastructure in my spare time - things can be
> frustrating).
unfortunately, we seem to be stuck with the `low bidders' in software
forever.
> No more crappy adhoc scripting or configuration languages.
> Not a multitude of "interesting" formats of configuration
> files.
agreed.
> An understandable and debuggable OS with clear
> abstractions would improve execution safety and security.
> Strict end user focus with an easy UI. Too, bad - the
> last attempt on this - the Newton OS - failed. It
> had dynamic objects with persistence and a great UI.
> It even didn't have a "save" command. Everything was
> stored/compressed/indexed immediately - without
> the need to save.
>
> My advice: look around, read some books (start with an old
> book by Erik Sandeval where the Xerox Lisp stuff
> is being described in one chapter), get a wider view.
> At some point you will see that Linux is neither the first nor
> the last OS.
unix sucks. perhaps parts of it could be improved but that is
impossible since it would `break old scripts' which seems to be
anathema to the unix community. thus unix is doomed to zero progress.
however, i need a real alternative in operating systems, not some pipe
dream. the choice presented to me is basically windows(9[58]|NT) or
some kind of unix. i pick linux as the lesser of the two evils. do i
have any other option?
--
Johan Kullstam [·······@idt.net] Don't Fear the Penguin!
In article <··············@sophia.axel.nom>, Johan Kullstam
<·······@idt.net> wrote:
> > There are a lot of vendors providing programming support even
> > in the Windows area.
>
> yes, but how many can fix bugs in ms-word or windows?
Depends on the bugs, I guess. A lot stuff is closed.
> can you influence new features?
Depends.
> yes, but, for example, the US DoD requires all docs in `MS-Word 6.0 for
> PC Windows' format. this is an impediment to choice.
So, let your software generate this format. Sigh.
> however, i need a real alternative in operating systems, not some pipe
> dream. the choice presented to me is basically windows(9[58]|NT) or
> some kind of unix. i pick linux as the lesser of the two evils. do i
> have any other option?
As a client machine I prefer a Mac. With MacOS X the
OS will be improved, too.
--
http://www.lavielle.com/~joswig
Rainer Joswig wrote:
[someone else said:]
>> yes, but, for example, the US DoD requires all docs in `MS-Word 6.0 for
>> PC Windows' format. this is an impediment to choice.
>
> So, let your software generate this format. Sigh.
This format, which is (last I heard, anyway) undocumented
by the people who created it. Which is likely to be replaced
at any moment by another, incompatible format with substantially
the same name (though one can plausibly hope that the DoD will
wait a while before switching), without anyone outside Microsoft
knowing what the new format looks like.
It's not so simple.
--
Gareth McCaughan Dept. of Pure Mathematics & Mathematical Statistics,
·····@dpmms.cam.ac.uk Cambridge University, England.
Johan Kullstam <·······@idt.net> writes:
> yes, but, for example, the US DoD requires all docs in `MS-Word 6.0 for
> PC Windows' format. this is an impediment to choice.
Reading things like this is really frightening if you think about it:
The MS Word binary format, starting with version 6.0, with its
macro viruses, is probably _the_ most severe computer-related security
threat in todays society (hey! what if Xerox TEdit had become the
standard? Would we have had Interlisp viruses today?).
--
Espen Vestre
Espen Vestre <··@nextel.no> writes:
<
< Johan Kullstam <·······@idt.net> writes:
<
< > yes, but, for example, the US DoD requires all docs in `MS-Word 6.0 for
< > PC Windows' format. this is an impediment to choice.
<
< Reading things like this is really frightening if you think about it:
< The MS Word binary format, starting with version 6.0, with its
< macro viruses, is probably _the_ most severe computer-related security
< threat in todays society (hey! what if Xerox TEdit had become the
< standard? Would we have had Interlisp viruses today?).
There is no Word binary format! The ascii text is stored in the center
of the file; strings --all somefile.doc, will work every time! The
300% size increase of the document has almost no relevance to
information storage and representation. They are not concerned about
storing a document in a format which is well designed and safe because
it doesn't further promote the illusion of planned obsolescence. How
can software stop working unless it wasn't meant, and therefor didn't,
work to begin with? Is the "media" so powerful that that consumers
have come to embrace products that are _not_ expected to live up to
their promises?!?! Maybe not living up to their promise is their
promise (by promise I mean brand). I dunno, too tired to think...Sigh.
Steve Gonedes <········@worldnet.att.net> writes:
> There is no Word binary format!
When I used the term 'binary format' in this context, I was using
referring to the standard file format as opposed to the somewhat
less evil RTF format.
> 300% size increase
I've even seen a case of 1100% increase - when importing a Word for mac 5.0
doc. into a newer PC version!
Anyway, this is getting way off topic (my excuse for my previous
posting of course was my little joke about TEdit ;-))...
--
Espen Vestre
On 28 Oct 1998 07:35:59 -0500, Johan Kullstam <·······@idt.net>
claimed or asked:
% yes, but, for example, the US DoD requires all docs in `MS-Word 6.0 for
% PC Windows' format. this is an impediment to choice.
I wonder how that affects the DOJ case against Microsoft. I would
expect TEX or LATEX to be a better format for the simple reason the
specification is a published standard. There is also HTML if
typesetting isn't required.
I was also under the impression that the U.S. Govt was required to use
ANSI standard software and formats.
I guess Bill wins.
--
David Steuber (ver 1.31.2a)
http://www.david-steuber.com
To reply by e-mail, replace trashcan with david.
"Ignore reality there's nothing you can do about it..."
-- Natalie Imbruglia "Don't you think?"
Johan Kullstam wrote:
> yes, but, for example, the US DoD requires all docs in `MS-Word 6.0 for
> PC Windows' format. this is an impediment to choice.
This is really funny: When you say "save" in Word, it does not store
the text you are working on, but a bunch of memory, including
data from other programs called before word etc. (just work a
while in Win, than in Word, save, and look at the *.doc with a
hex editor).
IMHO this risk alone should be a sufficient reason to outlaw the use
of Word in sensitive areas.
Ralf
On Tue, 27 Oct 1998 12:36:51 +0100, ······@lavielle.com (Rainer
Joswig) claimed or asked:
% In article <·················@news.newsguy.com>,
% ········@david-steuber.com wrote:
%
% > matter to you. But this might. The people who work on CMUCL have
% > helped me, through this forum and private e-mail, to get set up and
% > running with it. The responsiveness and the usefulness of the support
% > far surpasses anything that Microsoft has ever come up with in my
% > experience.
%
% Sorry, I don't buy it. If you want something special from
% the maintainers - you'll end up implementing it on your
% own. Great - but I don't want to be an expert on
% threads, FFI, low-level software debugging, TCP/IP
% interfacing etc. I can get software where it already works.
Perhaps you've had much better support from Microsoft than I have.
If you don't like the free or nearly free software, go ahead and buy
the commercial stuff. With decent free software around, the
commercial products have to be that much better to justify their cost
in money.
I'm just talking from my experience. Perhaps it has been tainted by
dealing with Microsoft.
--
David Steuber (ver 1.31.2a)
http://www.david-steuber.com
To reply by e-mail, replace trashcan with david.
"Ignore reality there's nothing you can do about it..."
-- Natalie Imbruglia "Don't you think?"