It seems that people are often coming and complaining that ACL is too
pricey. So why aren't people complaining about LispWorks?
( Partial answer: because they want ACL more than they want LispWorks.
Why? )
--------------------------------------------------
Thaddeus L. Olczyk, PhD
Think twice, code once.
Thaddeus L Olczyk <······@interaccess.com> writes:
> It seems that people are often coming and complaining that ACL is too
> pricey. So why aren't people complaining about LispWorks?
> ( Partial answer: because they want ACL more than they want LispWorks.
> Why? )
This question is worded weirdly, but the obvious answer is:
Because they think LispWorks is priced as they expect.
The wording of your question also suggests that you think the cost of ACL
and LispWorks are approximately the same. If so, you are quite wrong.
The pricing and licensing restrictions of these two vendors are like
apples and oranges, and as much different from each other as they are
different from any other Lisp offering.
On 09 May 2003 15:31:23 -0400, Kent M Pitman <······@world.std.com>
wrote:
>Thaddeus L Olczyk <······@interaccess.com> writes:
>
>> It seems that people are often coming and complaining that ACL is too
>> pricey. So why aren't people complaining about LispWorks?
>> ( Partial answer: because they want ACL more than they want LispWorks.
>> Why? )
>
>This question is worded weirdly, but the obvious answer is:
>Because they think LispWorks is priced as they expect.
>
>The wording of your question also suggests that you think the cost of ACL
>and LispWorks are approximately the same. If so, you are quite wrong.
>The pricing and licensing restrictions of these two vendors are like
>apples and oranges, and as much different from each other as they are
>different from any other Lisp offering.
I didn't say they were. I see these arguments that ACL should be lower
priced ( or even free ) and I wonder why ACL draws attention and not
LispWorks.
--------------------------------------------------
Thaddeus L. Olczyk, PhD
Think twice, code once.
"Appologize for the e-mail it was an accident.
----- Original Message -----
From: "Michael Israel" <·································@hotmail.com>
To: <······@interaccess.com>
Sent: Friday, May 09, 2003 4:58 PM
Subject: Re: Why aren't people complaining that LispWorks is too pricey?
>
> ----- Original Message -----
> From: "Thaddeus L Olczyk" <······@interaccess.com>
> Newsgroups: comp.lang.lisp
> Sent: Friday, May 09, 2003 4:45 PM
> Subject: Re: Why aren't people complaining that LispWorks is too pricey?
>
>
> > On 09 May 2003 15:31:23 -0400, Kent M Pitman <······@world.std.com>
> > wrote:
> >
> > >Thaddeus L Olczyk <······@interaccess.com> writes:
> > >
> > . I see these arguments that ACL should be lower
> > priced ( or even free ) and I wonder why ACL draws attention and not
> > LispWorks.
>
> Allegro CL cost $3,000 student price w/o CLIM support.
> + Runtime fees to distrubite apps.
>
> LispWorks cost $900 commercail $525 student /w CLIM support
> + Licence 2 Distrubute Programs w/o Runtime.
>
> I was just informing Franz that the price was rather high.
> I know they can set there price as high as they'd like 2
> but if someone needs ANSI CL + CLIM they'd prob. choose
> LispWorks.
>
> We're just trying to tell Franz to sep. Support & Product costs &
> they might get new users.
>
> If they don't want to, it's their call.
>
> I also pointed out that if an older version of any Commercial Lisp was put
> in the public domain that had three things: 1.) Ran on Linux, Windows, and
> Mac/OS 2.) Delivered executibles 3.) Conformed to the ANSI specs. -- it
> might bring more users into Lisp kind of like Linux brought more users
into
> C/C++ and the Web brough more users into Java.
>
> I'm not saying that any one should, has too, or is wrong by not doing
> so--only that if thay did it would prob. help the Lisp community in the
long
> run because Lisp would be popular.
>
> However, I realize that if Lisp became too popular and M$ made VisualLisp,
> all hope would prob. be lost--because it would be a
> crappy version. ;) LOL
>
> I am in no way implying that any money would be made--the new users could
go
> to another Lisp--and I am not just talking about commercial Lisps is a
> freeware Lisp did the job it would pull more people into the language.
(Few
> people would by <$500 to learn a lang. Most intro. prog. books include a
> free compiler--to aid the student, and most of those free compilers and
make
> exe files :) )
>
>
In the last exciting episode, Thaddeus L Olczyk <······@interaccess.com> wrote:
> On 09 May 2003 15:31:23 -0400, Kent M Pitman <······@world.std.com>
> wrote:
>
>>Thaddeus L Olczyk <······@interaccess.com> writes:
>>
>>> It seems that people are often coming and complaining that ACL is too
>>> pricey. So why aren't people complaining about LispWorks?
>>> ( Partial answer: because they want ACL more than they want LispWorks.
>>> Why? )
>>
>>This question is worded weirdly, but the obvious answer is:
>>Because they think LispWorks is priced as they expect.
>>
>>The wording of your question also suggests that you think the cost of ACL
>>and LispWorks are approximately the same. If so, you are quite wrong.
>>The pricing and licensing restrictions of these two vendors are like
>>apples and oranges, and as much different from each other as they are
>>different from any other Lisp offering.
> I didn't say they were. I see these arguments that ACL should be
> lower priced ( or even free ) and I wonder why ACL draws attention
> and not LispWorks.
The answer is simple: Nobody is surprised by LispWorks pricing; its
pricing conforms with peoples' expectations.
There is no reason for anyone to get excited; no grounds for
argumentative debate; no major complaints of "If it were *vastly*
cheaper, I might consider it."
ACL draws attention because the licensing pricing schemes are much
more interesting and worthy of debate.
--
output = reverse(··········@" "enworbbc")
http://cbbrowne.com/info/lisp.html
Rules of the Evil Overlord #221. "My force-field generators will be
located inside the shield they generate."
<http://www.eviloverlord.com/>
Thaddeus L Olczyk <······@interaccess.com> wrote in message news:<··································@4ax.com>...
> It seems that people are often coming and complaining that ACL is too
> pricey. So why aren't people complaining about LispWorks?
I think when people spend time working with the free version
they are looking ahead to getting the professional version in
the near future. When using the free version of Lispworks you
can relax about the price and spend your time learning it and
doing useful things with it. But those who use the free version
of ACL might spend way too much of their time worrying asking
themselves if they are ever going to be able to afford the
professional version or get approval for it on their budget.
That worry might hold them back, causing them to learn it more
slowly and to accomplish less with it. And in fact those
complaints you see might be mostly just a symptom of that
worry.
But the good news is that Lispworks is an incredibly powerful
and useful software development environment, even in its free
version. Whatever minor advantages ACL might or might not have
over it are probably not worth a minute's worry unless you are
already aware of a specific advantage you need for a specific
reason.
Thaddeus L Olczyk wrote:
> It seems that people are often coming and complaining that ACL is too
> pricey. So why aren't people complaining about LispWorks?
> ( Partial answer: because they want ACL more than they want LispWorks.
> Why? )
One obvious possibility would be that ACL is more expensive
than LispWorks. I have seen no official ACL prices, so I can't
comment with great confidence, but the numbers I've heard
bandied around for ACL are considerably higher than those
I've seen for LispWorks.
--
Gareth McCaughan ················@pobox.com
.sig under construc
From: Espen Vestre
Subject: Re: Why aren't people complaining that LispWorks is too pricey?
Date:
Message-ID: <kwbry8e0cw.fsf@merced.netfonds.no>
Gareth McCaughan <················@pobox.com> writes:
> One obvious possibility would be that ACL is more expensive
> than LispWorks. I have seen no official ACL prices, so I can't
> comment with great confidence, but the numbers I've heard
> bandied around for ACL are considerably higher than those
> I've seen for LispWorks.
The price for the development environment is at least approximately
within the same order of magnitude. But what makes ACL expensive
are the runtime licenses. LispWorks runtime licenses are either
'a bargain' (unix) or don't exists at all (linux + windows).
--
(espen)
From: Marc Battyani
Subject: Re: Why aren't people complaining that LispWorks is too pricey?
Date:
Message-ID: <b9h4md$q1d@library2.airnews.net>
"Thaddeus L Olczyk" <······@interaccess.com> wrote
> It seems that people are often coming and complaining that ACL is too
> pricey. So why aren't people complaining about LispWorks?
Because we are happy with it. And it's probably the same for ACL users.
The people who are whining about this are just trolling. They can use a lot
of free or commercial Lisp implementations if they really want to do
something. A wrong analogy with cars have been given but a better one is that
it's the same for cars: You have cheap ones and more expensive ones and
whiners should go to try and buy a luxury car for the price of an entry level
Korean car. And explain to the luxury car maker that if they lower the price
of their cars they would make more money. Not everybody needs of even wants a
such a car.
> ( Partial answer: because they want ACL more than they want LispWorks.
Wrong. They only want to cry, so they choose the higest price target. because
they would be even more ridiculous to cry for a lower price.
BTW When I don't write Lisp, I design electronics and I can tell you that ACL
prices are a joke compared to those of electronics design automation tools.
Just my Mentor Expedition PCB license has a list price of more than $30K. But
for some strange reason people don't cry about this in electronics groups.
It's purely a Lisp related syndrome.
Marc
From: anonymous by necessity
Subject: Re: Why aren't people complaining that LispWorks is too pricey?
Date:
Message-ID: <gYKdnXlVbMSMyCGjXTWcrg@surfnetusa.com>
> of their cars they would make more money. Not everybody needs of even
wants a
> such a car.
>
Feh.
Here's a car/IDE/Language analogy for you:
Visual Studio.NET is a Mercedes Benz IDE with a Yugo language
ACL/LispWorks are Yugo IDEs with a Mercedes Benz language
For a Common Lisp IDE, ACL might be ok. But compare it to Visual Studio, or
JBuilder, and I'm sorry but you're going to find that ACL is completely,
utterly, patently, comically, _tragically_, and in ALL other ways
overpriced. Same for Lispworks, except I might be inclined to leave out the
words "comical" and "tragically".
I love Common Lisp as much as the next guy. But I wont even sneeze in the
general direction of ACL, and that includes the "free" version. Honestly,
how is it supposed to help me? I can't develop a stand-alone application
with it, and the GUI tools are 1989 technology at best. IMO Franz is
completely out to lunch; they as a company live in dreamland. XAnalys is a
bit better, maybe I could say they are merely snoozing, as opposed to being
all the way full bore into REM sleep dreamland. At least XAnalys isn't
bone-headed enough to charge runtime royalties. OTOH they can't even seem to
figure out how to put a "recently used files" option on their IDE menu...
> BTW When I don't write Lisp, I design electronics and I can tell you that
ACL
> prices are a joke compared to those of electronics design automation
tools.
> Just my Mentor Expedition PCB license has a list price of more than $30K.
But
> for some strange reason people don't cry about this in electronics groups.
> It's purely a Lisp related syndrome.
>
Oh, I dunno- maybe, like in the CL world, there isn't really any serious
competition so people just accept such looney-tunes pricing schemes as
"normal". I don't know anything about the electronics design industry. If
you look around, are there other tools/design environements that do 10x
more, are easier to use, give you more options, etc., but cost less (much
less, much MUCH less) than half the $30K license, and that don't charge you
runtime distribution royalties? Look around the software language IDE
business for even two seconds, and you'll find there ARE such products.
Anyway...you CL guys, Franz especially, you can just keep snoooozing away,
sawing logs, selling to people who either don't know any better or are just
_trapped_ in a competitionless niche. It's not doing much to promote Common
Lisp to the unwashed masses, though.
Good luck to you all...
From: Friedrich Dominicus
Subject: Re: Why aren't people complaining that LispWorks is too pricey?
Date:
Message-ID: <87r8779twv.fsf@fbigm.here>
"anonymous by necessity" <······@nomail.org> writes:
> > of their cars they would make more money. Not everybody needs of even
> wants a
> > such a car.
> >
>
> Feh.
> Here's a car/IDE/Language analogy for you:
>
> Visual Studio.NET is a Mercedes Benz IDE with a Yugo language
> ACL/LispWorks are Yugo IDEs with a Mercedes Benz language
>
> For a Common Lisp IDE, ACL might be ok. But compare it to Visual Studio, or
> JBuilder, and I'm sorry but you're going to find that ACL is completely,
> utterly, patently, comically, _tragically_, and in ALL other ways
> overpriced. Same for Lispworks, except I might be inclined to leave out the
> words "comical" and "tragically".
You are a troll therefor no name.
>
> I love Common Lisp as much as the next guy.
Hardly, if you would love it you won't write such a crap.
From: Tim Bradshaw
Subject: Re: Why aren't people complaining that LispWorks is too pricey?
Date:
Message-ID: <ey3of2b113o.fsf@cley.com>
* anonymous by necessity wrote:
> Oh, I dunno- maybe, like in the CL world, there isn't really any serious
> competition so people just accept such looney-tunes pricing schemes as
> "normal". I don't know anything about the electronics design industry. If
> you look around, are there other tools/design environements that do 10x
> more, are easier to use, give you more options, etc., but cost less (much
> less, much MUCH less) than half the $30K license, and that don't charge you
> runtime distribution royalties? Look around the software language IDE
> business for even two seconds, and you'll find there ARE such products.
There are such products, of course, because the people who write these
products either sell them in enormous numbers to a captive market
(Windows), or lose money on them to get people to use the other stuff
they make, where they make money[1] In both cases enormous effort is
expended on the products to make them swish. Lisp vendors do not have
this luxury.
I think someone should write a paper on `economics for programmers'
and put it on the web somewhere, because there is *so* much *truly*
clueless stuff like this. Though I dread to think that maybe
programmers aren't smart enough to understand how this kind of thing
works.
--tim
Footnotes:
[1] or, in the free software model, huge numbers of people
voluntarily impoverish themselves for the greater good of the
cause.
"Marc Battyani" <·············@fractalconcept.com> writes:
> BTW When I don't write Lisp, I design electronics and I can tell you that ACL
> prices are a joke compared to those of electronics design automation tools.
> Just my Mentor Expedition PCB license has a list price of more than $30K. But
> for some strange reason people don't cry about this in electronics groups.
> It's purely a Lisp related syndrome.
not just Lisp. this syndrome is related to software development tools
in general.
I think it has similar roots to the popular demand for free stuff on
the Internet. like, practically nobody is willing to pay for reading
www.nytimes.com. like, people would rather see online publications
go bankrupt and then whine about it than pay. they'd rather wade
through tons of obnoxious ad popups than pay for the content.
the root cause is, I think, that those people got addicted to the
"market" where valuable things are given to them for free in order to
promote something else entirely. as the time goes, most of those
people are much more likely to learn to mentally "tune out" the ads
than they are to start thinking that perhaps /paying/ for things they
value might turn out more beneficial to them.
it's the same with software tools, ever since the leading
(market-wise) tool vendors started basically giving away the tools in
order to "win developer mindshare". the currently predominant
software tool marketing model is more optimized for people who are
shopping for platforms to standartize on than it is for people who
want to write software. of course there are much more people of the
latter type than the former, but that's OK: most of them are "smart"
enough to just mentally "tune out" the promotions and enjoy the low
prices. so there we go.
--
I knew you weren't really interested.
According to Marc Battyani <·············@fractalconcept.com>:
> It's purely a Lisp related syndrome.
Periodically people also whine in c.l.smalltalk about the pricing and
licensing of Cincom's main Smalltalk product. Sometimes they are the same
people who whine here about ACL.
--
Ng Pheng Siong <····@netmemetic.com>
http://firewall.rulemaker.net -+- Manage Your Firewall Rulebase Changes
http://www.post1.com/home/ngps -+- Open Source Python Crypto & SSL
From: Richard C J Putman
Subject: Re: Why aren't people complaining that LispWorks is too pricey?
Date:
Message-ID: <3cjld3ea.fsf@physics.org>
"Marc Battyani" <·············@fractalconcept.com> writes:
> "Thaddeus L Olczyk" <······@interaccess.com> wrote
> > It seems that people are often coming and complaining that ACL is too
> > pricey. So why aren't people complaining about LispWorks?
>
> Because we are happy with it. And it's probably the same for ACL users.
>
> The people who are whining about this are just trolling...
I've lurked in this ng for more years than I care to count and there
have been many threads complaining about ACL pricing policy. I've
never gotten involved before because Franz's marketing tactics are
frankly nobody's business but theirs. However, we've just had two
employees from Franz attempting to justifying that cost and so I think
I'm entitled to dispute their assertions without fear of being accused
of whining or trolling.
Firstly, Steven Haflich posted an example that told us the compiler
cost included, in one instance, payment for staff getting shipped off
to Japan to fix a bug. This seems astonishing if true. I worked for
several years at a company producing Fortran compilers, in some
respects, a similar niche market to Lisp. None of our customers
expected the product to be bug free (except a few Germans for some
reason) and that problems occurring with, for instance, a new piece
of hardware were either fixed at the customers expense or they waited
for the problem to run through the development cycle.
Secondly, Duane justified the cost of ACL to hobbyists on the basis
that many hobbies are expensive. I cannot believe many hobbyists buy
software costing many times the price of their hardware.
Maybe Franz could issue a license for ACL which included database
support etc. but with installation support only and allowed a single
user to create executables to run internally to that company. I'd be
interested to know if Franz would think they would loose money from
customers switching to this license. Or whether their existing
customers would be unaffected. I guess a hobbyist could be considered
as a company with one badly paid employee :)
Of course I want Franz to be a profitable company, its in all our
interests that it remains so. I would never suggest it reduce its
cost to the level of the most hardup user. However, if it is true
that the support is bundled with the cost of the compiler I'd suggest
that it pushes it out of the reach of many software project budgets.
If that is by design, so that they only attract the top end of the
market, I think I would take a very dim view of Franz as a company.
> ... They can use a lot
> of free or commercial Lisp implementations if they really want to do
> something. A wrong analogy with cars have been given but a better one is that
> it's the same for cars: You have cheap ones and more expensive ones and
> whiners should go to try and buy a luxury car for the price of an entry level
> Korean car. And explain to the luxury car maker that if they lower the price
> of their cars they would make more money. Not everybody needs of even wants a
> such a car.
>
I evaluated all the Lisp compilers I could find a few years ago. ACL
was the best on several counts. I don't view ACL as anything to do
with luxury. It is simply a more effective tool.
Thanks to Peter Seibel for some suggestions about how to explain difficulties
to Franz, but I have exchanged emails with their sales several times
before. The last time this happened, even trying to get some sales
estimates out of our company, at the bequest of Franz in order to work
out a quote, was enough to kill the project dead. In any case, does
it not strike you odd that I should appeal to Franz as being some sort
of special case for consideration? I'm merely trying to introduce ACL
into a company who doesn't consider Lisp for software projects.
> > ( Partial answer: because they want ACL more than they want LispWorks.
> Wrong. They only want to cry, so they choose the higest price target. because
> they would be even more ridiculous to cry for a lower price.
>
> BTW When I don't write Lisp, I design electronics and I can tell you that ACL
> prices are a joke compared to those of electronics design automation tools.
> Just my Mentor Expedition PCB license has a list price of more than $30K. But
> for some strange reason people don't cry about this in electronics groups.
> It's purely a Lisp related syndrome.
>
OK so you decide that what you really want to do in life it design
PCB's with this great but expensive tool. It would not be unrealistic
to go out and try and find a job at an electronic firm where they use
this product. Suppose you decide ACL is the tool you want to work
with; what chances are there that you can hope to find a Lisp job?
A much more likely option would be to introduce it on a project you're
already working on or use it as a hobby until you can use it for
consultancy etc. What I'm arguing is that Franz's pricing policy
provides little chance of these two routes working. I'd hesitate
recommending ACL to any new lispers as, in my experience, the chances
of them getting use the full product are too slim.
IMHO Lispworks is a very good product and fairly priced and I'll
probably be purchasing it shortly. Its just a shame I don't appear to
have any other choice.
Richard
From: Matthew Danish
Subject: Re: Why aren't people complaining that LispWorks is too pricey?
Date:
Message-ID: <20030511101947.I22493@mapcar.org>
On Sun, May 11, 2003 at 02:22:21PM +0100, Richard C J Putman wrote:
> I evaluated all the Lisp compilers I could find a few years ago. ACL
> was the best on several counts. I don't view ACL as anything to do
> with luxury. It is simply a more effective tool.
I don't know what your criteria are, of course, but a lot has changed in the
last few years so it might be worthwhile re-evaluating the various Lisp
compilers.
--
; Matthew Danish <·······@andrew.cmu.edu>
; OpenPGP public key: C24B6010 on keyring.debian.org
; Signed or encrypted mail welcome.
; "There is no dark side of the moon really; matter of fact, it's all dark."
From: Marc Battyani
Subject: Re: Why aren't people complaining that LispWorks is too pricey?
Date:
Message-ID: <b9m997$800@library2.airnews.net>
"Richard C J Putman" <··········@physics.org> wrote
> "Marc Battyani" <·············@fractalconcept.com> writes:
[About ACL price]
> > The people who are whining about this are just trolling...
>
> I've lurked in this ng for more years than I care to count and there
> have been many threads complaining about ACL pricing policy. I've
> never gotten involved before because Franz's marketing tactics are
> frankly nobody's business but theirs. However, we've just had two
> employees from Franz attempting to justifying that cost and so I think
> I'm entitled to dispute their assertions without fear of being accused
> of whining or trolling.
OK, this post does not look like a troll. I won't comment on your suggestions
for Franz as I don't have any marketting data on the Lisp market. But
choosing a profitable business model is incredibly hard to do. So doing it
without data...
[...]
> I evaluated all the Lisp compilers I could find a few years ago. ACL
> was the best on several counts. I don't view ACL as anything to do
> with luxury. It is simply a more effective tool.
I think you should update your tests. Of course it depends on what weight you
give to different points.
Personally I never use any proprietary extension if I don't have it's sources
or if it's not a widely available one (like sockets for example). So I only
look at the Common Lisp implementation (compiler performance, supported on
the OS I use, portability, tools (debugger, profiler, inspector,...), etc.)
For instance in LW, though Common SQL is a cool thing, I don't use it. I use
CL-SQL. I also use ASDF instead of lw:defsystem, etc., etc. This does not
means that I don't like the Lispworks versions or that they are bad, just
that I prefer not to depend on them if I can. It's not only Lisp related, I
was doing the same when I was working in C++.
What has changed in the recent years is that they are more and more Lisp
libraries available. Just look at www.cliki.net or at the impressive list of
Lisp packages maintained by Kevin Rosenberg if you want to be convinced of
this.
It's here: http://b9.com/debian.html
You can also look at this benchmark:
http://b9.com/archives/000037.html
So tell us what is missing ?
[...]
> > BTW When I don't write Lisp, I design electronics and I can tell you that
ACL
> > prices are a joke compared to those of electronics design automation
tools.
> > Just my Mentor Expedition PCB license has a list price of more than $30K.
But
> > for some strange reason people don't cry about this in electronics
groups.
> > It's purely a Lisp related syndrome.
> OK so you decide that what you really want to do in life it design
> PCB's with this great but expensive tool. It would not be unrealistic
> to go out and try and find a job at an electronic firm where they use
> this product. Suppose you decide ACL is the tool you want to work
> with; what chances are there that you can hope to find a Lisp job?
Well, there was no job where I could design analog and digital electronics
(an ultra low noise differential photodiode amplifier with spread-spectrum
synchronous demodulation for instance), and program in Lisp (real time
industrial applications and web applications) and in VHDL (for the digital
stuff), and spend 20 to 30% of my time to learn new things, and earn money...
This is why I created my own company. ;-)
So if I decide that I need ACL, I would call Franz to negotiate the price and
if we could arrive at some mutually satisfying agreement, I would buy it. I
would not go to c.l.l to explain why they should lower their price.
[...]
> IMHO Lispworks is a very good product and fairly priced and I'll
> probably be purchasing it shortly. Its just a shame I don't appear to
> have any other choice.
You won't regret it, Lispworks is really nice.
At the ILC2002, I've met lots of nice people from Franz and Xanalys and none
of them seemed particularly dumb. They have just chosen different business
models, so you have a choice. And don't forget Corman Lisp too.
Marc
From: Duane Rettig
Subject: Re: Why aren't people complaining that LispWorks is too pricey?
Date:
Message-ID: <43cjkef0d.fsf@beta.franz.com>
Richard C J Putman <··········@physics.org> writes:
> Firstly, Steven Haflich posted an example that told us the compiler
> cost included, in one instance, payment for staff getting shipped off
> to Japan to fix a bug. This seems astonishing if true.
Steve's point was certainly not that we went to Japan to fix a bug.
The emphasis I read in his post was that the goal was the success of
the customer. That we fixed a bug as a means to that goal is
certainly true. "For lack of a nail ..."
> I worked for
> several years at a company producing Fortran compilers, in some
> respects, a similar niche market to Lisp. None of our customers
> expected the product to be bug free (except a few Germans for some
> reason) and that problems occurring with, for instance, a new piece
> of hardware were either fixed at the customers expense or they waited
> for the problem to run through the development cycle.
Well, Fortran and Common Lisp tend to be different in the way they are
patched, so perhaps expectations might be different. Again, though,
we were not talking about bugs, per se, but about relationships and
what it takes to truly help the customer to be successful.
> Secondly, Duane justified the cost of ACL to hobbyists on the basis
> that many hobbies are expensive. I cannot believe many hobbyists buy
> software costing many times the price of their hardware.
OK, you are at least the second poster who has misunderstood this. I
admit that since I am talking to a group of programmers, that my
statement was likely to be misunderstood - that it is likely that
some of you have no hobbies outside of software. In fact, if I had
thought about how likely it is that some people here might either
have a career not in software and are software hobbyists, or who in
fact are career programmers and take on software also on the side,
I would have qualified my statement a little more carefully.
I was not, however, talking at all about software hobbyists. Instead,
I was talking about hobbies like snow and water skiing, photography,
woodworking, collections of various kinds, model train sets, music, etc.
All of these are activities that one may take on in varying degrees,
with various commitments of money. And unless the activity is being
taken on as a career, where the goal is to make enough money for
sustenance, a hobby tends to be an end unto itself.
When it comes to software, it's a lot harder to describe what a
hobbyist really is. I personally consider software hobbyists to
be rare, and most cases, what one might describe as a hobbyist is
more of a "dabbler" to me. On the activity side, a dabbler would
be one who knows how to waterski, but who owns neither a ski nor
a boat, and who gets his skiing in by having a friend who owns a
boat and ski (i.e. a hobbyist) to take him out skiing.
--
Duane Rettig ·····@franz.com Franz Inc. http://www.franz.com/
555 12th St., Suite 1450 http://www.555citycenter.com/
Oakland, Ca. 94607 Phone: (510) 452-2000; Fax: (510) 452-0182
Duane Rettig <·····@franz.com> wrote in message news:<·············@beta.franz.com>...
> When it comes to software, it's a lot harder to describe what a
> hobbyist really is. I personally consider software hobbyists to
> be rare, and most cases, what one might describe as a hobbyist is
> more of a "dabbler" to me. On the activity side, a dabbler would
There seems to be a lot of agreement that Common Lisp
needs a bigger market. If the hobbyist submarket and
the dabbler submarket are not considered worth the time
and effort it would take to serve them, it might be
time to rethink such assumptions. Those hobbyists and
dabblers obviously have lives outside of Lisp, and in
those lives they may influence a large number of other
people to become interested in Lisp.
It seems clear from large numbers of postings to
comp.lang.lisp over the years that there is a significant
market for being able to build binary executables while
learning Lisp. I don't know how much money such people
would be willing to pay, but clearly they would be willing
to pay a lot more than they are paying now, which in most
cases is nothing, and in many cases causes them to choose
other programming languages.
The desire to build binary executables seems clear to me.
They want to distribute software to people who don't have
the skills or inclination to install any Lisp software on
their computers. Maybe they even want to make money from
such binary executables, but in most cases thair work is
far from commercial, because they're just learning and they
have no real expectation of earning significant money while
they learn. Maybe they want to practice starting their own
software company, by coming out with software products and
setting up websites to sell those software products. But
it's still not commercial if they're just learning and not
really earning enough to make any difference. Most
"companies" are actually hobbies. If the Lisp vendors would
recognize that fact, they could serve such hobbyist "companies"
and maybe vastly increase the marketshare and mindshare of Lisp.
If Lisp can make a significant fraction of such "companies"
succeed and become real companies, that's exactly what it needs
to become popular. Paul Graham multiplied by thousands, so
the success can no longer be denied.
Why did Paul Graham use Clisp? The commercial Lisp vendors
should think very seriously of ways of capturing that market
and making it grow.
"Eric Smith" <········@yahoo.com> ha scritto nel messaggio ································@posting.google.com...
> Why did Paul Graham use Clisp?
Because at the time it had the best performance for PG's application.
Not a question of price, I understand.
P.
From: Duane Rettig
Subject: Re: Why aren't people complaining that LispWorks is too pricey?
Date:
Message-ID: <43cjj99gy.fsf@beta.franz.com>
"Pierpaolo BERNARDI" <··················@hotmail.com> writes:
> "Eric Smith" <········@yahoo.com> ha scritto nel messaggio
································@posting.google.com...
>
> > Why did Paul Graham use Clisp?
>
> Because at the time it had the best performance for PG's application.
>
> Not a question of price, I understand.
Yes, this was true. In 1987 Paul came to us for porting advice and
optimization questions, in order to port his application from Clisp
(which he had been using at the time) to Allegro CL, so that he could
deploy his application on the fastest lisp possible. To his and our
surprise, it ran faster on Clisp, due to our usage of the Gray streams
implementation. We tried doing some optimizations for him, and did
succeed in speeding his app slightly, but it was still much slower
than on Clisp (as I recall, some of the suggestions I gave him for
speeding up on Allegro, based on the profiler outputs he was giving me,
were also speeding up his app on clisp :-). He eventually decided to
stay with clisp, which had a very simple implementation of ANSI streams
which did not include gray streams (I think it was also mostly also
implemented in C).
It was this experience, and the knowledge that we weren't going to be
able to compete in the relatively new web and server market with such
a slow streams implementation, that got us started working on a design
that eventually became simple-streams.
--
Duane Rettig ·····@franz.com Franz Inc. http://www.franz.com/
555 12th St., Suite 1450 http://www.555citycenter.com/
Oakland, Ca. 94607 Phone: (510) 452-2000; Fax: (510) 452-0182
On Mon, 12 May 2003 16:37:55 GMT, "Pierpaolo BERNARDI"
<··················@hotmail.com> wrote:
>
>"Eric Smith" <········@yahoo.com> ha scritto nel messaggio ································@posting.google.com...
>
>> Why did Paul Graham use Clisp?
>
>Because at the time it had the best performance for PG's application.
>
>Not a question of price, I understand.
>
>P.
Paul Graham used CMUCL not CLisp.
I let it go at first, figure someone else would mention it.
I guess others did so too.
--------------------------------------------------
Thaddeus L. Olczyk, PhD
Think twice, code once.
From: Sam Steingold
Subject: Re: Why aren't people complaining that LispWorks is too pricey?
Date:
Message-ID: <m3of28yq4f.fsf@loiso.podval.org>
> * In message <··································@4ax.com>
> * On the subject of "Re: Why aren't people complaining that LispWorks is too pricey?"
> * Sent on Mon, 12 May 2003 18:04:02 GMT
> * Honorable Thaddeus L Olczyk <······@interaccess.com> writes:
>
> Paul Graham used CMUCL not CLisp.
Paul Graham used CLISP, not CMUCL, for ViaWeb.
the original poster was correct.
--
Sam Steingold (http://www.podval.org/~sds) running RedHat9 GNU/Linux
<http://www.camera.org> <http://www.iris.org.il> <http://www.memri.org/>
<http://www.mideasttruth.com/> <http://www.palestine-central.com/links.html>
The only intuitive interface is the nipple. The rest has to be learned.
From: Tim Bradshaw
Subject: Re: Why aren't people complaining that LispWorks is too pricey?
Date:
Message-ID: <ey3u1c07zwi.fsf@cley.com>
* Thaddeus L Olczyk wrote:
> Paul Graham used CMUCL not CLisp.
No, he didn't, he used CLISP. I was in the room when he said this.
--tim
On Mon, 12 May 2003 18:04:02 GMT, Thaddeus L Olczyk
<······@interaccess.com> wrote:
[about Yahoo! Store]
> Paul Graham used CMUCL not CLisp.
Paul Graham posted this message to the `clisp-list' mailing list on Oct 16,
1998:
I just sent Bruno some email thanking him and his collaborators
for their work on Clisp, and mentioning that we had used it to
implement Yahoo! Store (http://store.yahoo.com). Bruno suggested
I send mail to clisp-list about this use of Clisp.
Yahoo! Store is one of the most complex server-based apps so far,
and we have found Lisp to be the perfect language for writing this
type of program. Y! Store is also a good example of a successful
mainstream Lisp application. It is by far the most popular offering
in the extremely competitive e-commerce field. We currently have
over 2300 live users, about twice as many as the #2 product. We've
used Clisp since the beginning, and have found it to be a great CL
implementation.
--pg
Paolo
--
Paolo Amoroso <·······@mclink.it>
From: Daniel Barlow
Subject: Re: Why aren't people complaining that LispWorks is too pricey?
Date:
Message-ID: <87ptmo2lga.fsf@noetbook.telent.net>
········@yahoo.com (Eric Smith) writes:
> It seems clear from large numbers of postings to
> comp.lang.lisp over the years that there is a significant
> market for being able to build binary executables while
> learning Lisp. I don't know how much money such people
There is nothing (except perhaps the ~20Mb installed size, which is
roughly what a JRE weighs in at anyway) stopping people from
distributing CMUCL or SBCL along with their application. No runtime
licence, no GPL "taint". In SBCL there's even a command line option
to disable the debugger, and a contrib module that makes it easy to
generate an "executable" that starts up the lisp environment and runs
the user's program: all you need is an installed SBCL and the single
file containing the application.
-dan
--
http://www.cliki.net/ - Link farm for free CL-on-Unix resources
"Daniel Barlow" <···@telent.net> wrote in message
···················@noetbook.telent.net...
> ········@yahoo.com (Eric Smith) writes:
>
> > It seems clear from large numbers of postings to
> > comp.lang.lisp over the years that there is a significant
> > market for being able to build binary executables while
> > learning Lisp. I don't know how much money such people
>
> There is nothing (except perhaps the ~20Mb installed size, which is
> roughly what a JRE weighs in at anyway) stopping people from
> distributing CMUCL or SBCL along with their application. No runtime
> licence, no GPL "taint". In SBCL there's even a command line option
> to disable the debugger, and a contrib module that makes it easy to
> generate an "executable" that starts up the lisp environment and runs
> the user's program: all you need is an installed SBCL and the single
> file containing the application.
>
>
> -dan
>
> --
>
> http://www.cliki.net/ - Link farm for free CL-on-Unix resources
The people we want to convice to try Lisp, are the C++/Java programmers
looking for a better language to use. Most of these
people will not have a Lisp running and there system.
Some would prob. not like it if our app installed a CL imp. on
there computer.
People already running CLisp, CMUCL, or SBCL prob. already
know about Lisp--we want to hook people on a cool app. before
telling them that it is written in Lisp, and could not have been
written in another language as easily.
Or, we want it to have a Lisp-style scripting language--so when the people
extend our app. they are learning Lisp. (This is why EMACS
attracted so many people to Lisp in the past.)
A Lisp-Scripting lang. for the Web, or for Computer Games will attract the
young crowd who have not yet choosen a lang. and are looking at all options.
Most end users don't what to DL at 20 Meg file when other programs that do
the same things are 5 Megs. -- if Lisp people don't realize this, no wonder
people choose C++/Java over Lisp. When the Lisp people wake up--only then
can we attract more People.
Lisp is like the Matrix, people need to realize it exists--before they'll
want to use it. :)
From the length of these Lisp price/ value of a small exe file post and the
amount of excuses that Lisp people come up with for not wanting small exe
files--it's clear to me that the Lisp community has not yet woken up.
People wake up; We need more people to use Lisp, or it might end up a dead
Language.
The More Speakers a Lang. Has the Less Likely It will Become a Dead
Toung. ;)
From: Tim Bradshaw
Subject: Re: Why aren't people complaining that LispWorks is too pricey?
Date:
Message-ID: <ey33cjk9i6a.fsf@cley.com>
* Franz Kafka wrote:
> Some would prob. not like it if our app installed a CL imp. on
> there computer.
How would they *know* that this thing that says `libcmucl.so.1" is a
slightly wrapped Lisp system?
> Most end users don't what to DL at 20 Meg file when other programs
> that do the same things are 5 Megs. -- if Lisp people don't realize
> this, no wonder people choose C++/Java over Lisp. When the Lisp
> people wake up--only then can we attract more People.
Every time I visit the BBC news site using Phoenix it asks me if I
want to download the JRE plugin, which is ... 20MB.
No. The real problem with Lisp is that almost everyone who claims to
want to use it would rather stand around whining on CLL than writing
applications, because standing around whining avoids having to fact
the terrible fact that they have no good ideas at all.
Put your money where you mouth is for once: quit complaining, and go
out there and WRITE SOME APPLICATIONS.
--tim
From: David Steuber
Subject: Re: Why aren't people complaining that LispWorks is too pricey?
Date:
Message-ID: <878ytbre9i.fsf@verizon.net>
Tim Bradshaw <···@cley.com> writes:
> Put your money where you mouth is for once: quit complaining, and go
> out there and WRITE SOME APPLICATIONS.
But that would be hard.
--
(describe 'describe)
From: Yarden Katz
Subject: Re: Why aren't people complaining that LispWorks is too pricey?
Date:
Message-ID: <86smr9kswl.fsf@underlevel.net>
David Steuber <·············@verizon.net> writes:
> Tim Bradshaw <···@cley.com> writes:
>
>> Put your money where you mouth is for once: quit complaining, and go
>> out there and WRITE SOME APPLICATIONS.
>
> But that would be hard.
I think that starting an active online Lisp magazine (kind of like XML.com
only for Lisp) where people publish very brief and focused articles
with code snippets would inspire more people to write applications. I
considered actually setting up a server/domain name for such magazine
and host it, but for this to really work I'd need a couple of
volunteers who are more knowledgeable than I am to review submitted
articles and write the initial ones.
--
Yarden Katz <····@underlevel.net> | Mind the gap
From: Tim Bradshaw
Subject: Re: Why aren't people complaining that LispWorks is too pricey?
Date:
Message-ID: <ey3y911cbez.fsf@cley.com>
* Yarden Katz wrote:
> I think that starting an active online Lisp magazine (kind of like
> XML.com only for Lisp) where people publish very brief and focused
> articles with code snippets would inspire more people to write
> applications.
While it might, I think the most important thing is just to go out
there and write things. There seems to be an endless stream of people
in CLL who sit around and complain at great length about how they
can't use Lisp because it has no interface to C++, it's case
insensitive, it's too expensive, Erik Naggum is too frightening and so
on. Yet here I am at 3AM trying to debug some annoying CORBA problem
in a Lisp application I use from tens to thousands of times a
day. Tomorrow I expect to need to spend some time writing some more
support code for our documentation preparation system which is also in
Lisp and which produces all but one of our web pages (OK `all but one'
is, only about 20 user-visible ones). I'm also going to ship a
version of an application, in Lisp, to a client.
Somehow, none of these terrible things that should have stopped me
doing any of this actually did. Why not? Well I think because these
terrible problems don't actually exist other than as excuses for
people to sit around and whine rather than actually doing something.
So I'd like to suggest that people quit complaining and try and *do
something*: no one is going to get things done but you.
--tim
In article <···············@cley.com>, Tim Bradshaw <···@cley.com>
wrote:
> So I'd like to suggest that people quit complaining and try and *do
> something*: no one is going to get things done but you.
>
Somehow this reminds of a scene in Life of Brian. ;)
Pascal
P.S.: I think you're right.
From: Yarden Katz
Subject: Re: Why aren't people complaining that LispWorks is too pricey?
Date:
Message-ID: <867k8kkz5t.fsf@underlevel.net>
Tim Bradshaw <···@cley.com> writes:
> * Yarden Katz wrote:
>> I think that starting an active online Lisp magazine (kind of like
>> XML.com only for Lisp) where people publish very brief and focused
>> articles with code snippets would inspire more people to write
>> applications.
>
> While it might, I think the most important thing is just to go out
> there and write things. There seems to be an endless stream of people
> in CLL who sit around and complain at great length about how they
> can't use Lisp because it has no interface to C++, it's case
> insensitive, it's too expensive, Erik Naggum is too frightening and so
> on. Yet here I am at 3AM trying to debug some annoying CORBA problem
> in a Lisp application I use from tens to thousands of times a
> day. Tomorrow I expect to need to spend some time writing some more
> support code for our documentation preparation system which is also in
> Lisp and which produces all but one of our web pages (OK `all but one'
> is, only about 20 user-visible ones). I'm also going to ship a
> version of an application, in Lisp, to a client.
While I agree with your analysis in this case, and do think that people need
to go out there and write applications, the fact that you're writing a
a highly practical application in Lisp for your company's/client's use
has absolutely no impact on how many people will use Lisp.
Many Lisp companies out there that spend big bucks on LispWorks or
Allegro support have access to (or have written) many "practical"[1]
Lisp libraries and tools that the complaining individual users on c.l.l. are
looking for. However, if the tools are not available freely with
open source for John Lisper Doe to use, then it cannot possibly be
considered as motivation for individual users to write applications in
Lisp. If I tell someone about an amazing application my company wrote
to edit movie clips but that someone cannot easily find a single
Lisp library that is able to manipulate, say, MPEG files, it
certainly will not motivate him/her to first write an MPEG library in
Lisp only to then apply Lisp's power, unless of course that someone
is set on using Lisp's power in a specific multimedia application
which will motivate him to write the library first. This
obviously means that the user has been heavily exposed to and
influenced by Lisp, and that is often not the case for those who complain
about Lisp's lack of practical libraries/tools on public forums.
Please do not interpret this post as an argument for why all libraries
should be free, or open source, or both: it's not. I just think it's
not realistic that an individual user who can find practical libraries
X, Y, Z prepackaged (and free) with language A will want to, by their
argument, "reinvent the wheel" in Lisp. Regardless of whether this
argument is justified or not, claiming that X, Y, Z has been
accomplished in Lisp but is not available can only add frustration.
I think that companies like yours that write commercial Lisp applications
could help the situation by repackaging only the generic elements of
their code (e.g. code to handle NNTP) in the form of a library and
releasing it open-source. That being said, this should definitely not
be an expectation or a demand, and if you made some money writing an
XML library in Lisp that's better than the open-source alternatives
and do not want to hand it over, more power to you.
Best wishes,
--
Yarden Katz <····@underlevel.net> | Mind the gap
[1] multimedia, net, encryption, graphics etc. libraries
From: Tim Bradshaw
Subject: Re: Why aren't people complaining that LispWorks is too pricey?
Date:
Message-ID: <ey3brxw1acu.fsf@cley.com>
* Yarden Katz wrote:
> While I agree with your analysis in this case, and do think that
> people need to go out there and write applications, the fact that
> you're writing a a highly practical application in Lisp for your
> company's/client's use has absolutely no impact on how many people
> will use Lisp.
No. But if everyone got off their backside and started writing
practical applications rather than standing around complaining, *that*
would. Which was my point.
--tim
From: Jeff Massung
Subject: Re: Why aren't people complaining that LispWorks is too pricey?
Date:
Message-ID: <vcnefgb59alheb@corp.supernews.com>
Tim Bradshaw <···@cley.com> wrote in ····················@cley.com:
> * Yarden Katz wrote:
>> While I agree with your analysis in this case, and do think that
>> people need to go out there and write applications, the fact that
>> you're writing a a highly practical application in Lisp for your
>> company's/client's use has absolutely no impact on how many people
>> will use Lisp.
>
> No. But if everyone got off their backside and started writing
> practical applications rather than standing around complaining, *that*
> would. Which was my point.
>
> --tim
>
True, but most programmers today suck. I don't think it is a problem with
not wanting to duplicate efforts or "re-invent the wheel". Most
programmers today, if there doesn't exist: AutoSyntaxHighlightMyEditField
() function in a library somewhere, they just give up or assume that it
can't be done (as an example). This is sad...
Jeff
--
;=========================================================
(defun reply (type)
(cond
((emailp type) ···········@NOSPAM.mfire.com")
((webp type) "http://www.simforth.com")))
;=========================================================
Tim Bradshaw wrote:
> [...] if everyone got off their backside and started writing
> practical applications rather than standing around complaining, *that*
> would. Which was my point.
Your point is valid, of course. But the people complaining about the state
of CL sometimes have a point, too (occasionally not a very original,
though).
If I try to summarize a lot of CL critics heard over the last year or so
(that's when I started reading this group) it boils down to one point: CL
is hard to get accepted because of the diversity of its implementations and
the small size of its user community. The diversity makes it hard to
support libraries which can be used with different CL implementations and
it makes it hard to write realistic code which runs on a number of
different platforms. The fact that people effectively are not writing
Ansi-CL, but write Ansi-CL with libs running under Allegro, LW, Clisp only,
lessens the impact each developer can have (by writing libs) and it makes
distribution of applications more complicated. This is a systematic
problem which the typical one-implementation-only language doesn't have and
which won't go away if just more people start coding now.
Ways out? Things like UFFI (extended by an introductory tutorial) and
everything which reduces the _unneccessary_ diversity between
implementations (i.e., an agreement on using one specific defsystem, an
agreement on how libs should be installed, how (that!) they should be
documented). An example on how such agreement can be found: In the C++
world there is the boost effort (www.boost.org) where developers were not
satisfied with their standard libs and decided to develop peer-reviewed,
portable, documented libs. This effort brought to light some really cool
stuff ("cool" meaning as-cool-as-c++-can-get, of course) at the price that
the developers invested quite a lot of time dicussing questions of design
and implementation with each other on their mailing list. (Thus boost is
more than just another repository where you can dump your code for others
to use.)
But then, I could just shut up and start coding now. I'll do. ;-)
Matthias
From: Kenny Tilton
Subject: Re: Why aren't people complaining that LispWorks is too pricey?
Date:
Message-ID: <3ECCC659.9040909@nyc.rr.com>
Matthias wrote:
> Tim Bradshaw wrote:
>
>>[...] if everyone got off their backside and started writing
>>practical applications rather than standing around complaining, *that*
>>would. Which was my point.
>
>
> Your point is valid, of course. But the people complaining about the state
> of CL sometimes have a point, too (occasionally not a very original,
> though).
>
> If I try to summarize a lot of CL critics heard over the last year or so
> (that's when I started reading this group) it boils down to one point: CL
> is hard to get accepted because of the diversity of its implementations and
> the small size of its user community. The diversity makes it hard to
> support libraries which can be used with different CL implementations and...
Folks say lots of things when asked to justify their prejudice. You can
prove this with a thought experiment: imagine all those issues you
mentioned (uffi, defsystem, diff gui layers (implicit in the vendor
library thang)) have been addressed. Will your critics now adopt Lisp?
No, they will find new reasons, or just say they are happy with C++.
Also ask yourself how pristine are Python or perl or Ruby, esp. in the
early days when folks gravitated to them. Look at the "do list" on
Python. The damn thing is a research project, yet has quite a following.
Besides, how many real developers other than tools developers worry
about cross-platform issues? Most have adopted a platform and can just
pick a Lisp and have a ball.
What is needed is really just More Time. People already want to use
Lisp, they just have not yet worked it out. But some early adopter types
in the statlang camp /have/ found their way to Lisp, and the rest will
follow. Then the valid issues your critics are using to make themselves
feel better about living in Hell will be addressed.
--
kenny tilton
clinisys, inc
http://www.tilton-technology.com/
---------------------------------------------------------------
"Everything is a cell." -- Alan Kay
On Thu, 22 May 2003 12:50:03 GMT, Kenny Tilton <·······@nyc.rr.com> wrote:
> Besides, how many real developers other than tools developers worry
> about cross-platform issues? Most have adopted a platform and can just
> pick a Lisp and have a ball.
My experience is surely not typical, but of my three employers, two
have considered cross-platformness essential, because that way you can
sell to more customers. Since this is server side, this really
matters: my current employer has a majority of customers on Unix, but
there are a couple who insist on Windows.
But then, had we used Lisp, Allegro or LispWorks would have worked
quite well, I expect, so this is hardly much of an anti-Lisp
argument. But cross-platformness is important, especially if you can
get it nearly for free (which is what you get with Java on the server
side, and probably with ACL or LispWorks, as well, though there I have
no experience). At least for a certain class of applications.
From: Tim Bradshaw
Subject: Re: Why aren't people complaining that LispWorks is too pricey?
Date:
Message-ID: <ey3llwzuplj.fsf@cley.com>
* Matthias wrote:
> But then, I could just shut up and start coding now. I'll do. ;-)
Yes, indeed, I think that the answer to the issues you mentioned
(which are valid) is the same answer I propose - stop complaining and
do something, and preferably something which isn't recreating stuff
that already exists.
--tim (which is not to imply you are complaining!)
Matthias wrote:
> If I try to summarize a lot of CL critics heard over the last year or so
> (that's when I started reading this group) it boils down to one point: CL
> is hard to get accepted because of the diversity of its implementations and
> the small size of its user community. The diversity makes it hard to
> support libraries which can be used with different CL implementations and
> it makes it hard to write realistic code which runs on a number of
> different platforms. The fact that people effectively are not writing
> Ansi-CL, but write Ansi-CL with libs running under Allegro, LW, Clisp only,
> lessens the impact each developer can have (by writing libs) and it makes
> distribution of applications more complicated. This is a systematic
> problem which the typical one-implementation-only language doesn't have and
> which won't go away if just more people start coding now.
I see several issues here.
- The situation with Common Lisp can't possibly be that bad. Common Lisp
is just one dialect of a family of Lisp dialects. It's pretty easy to
find one-implementation-only dialects of Lisp. (for example, OpenLisp -
http://www.eligis.com/, STELLA -
http://www.isi.edu/isd/LOOM/Stella/index.html, lush -
http://lush.sourceforge.net/) The fact that most people here stick to
either Common Lisp or Scheme indicates that these two dialects have some
real advantages.
- Languages that are based on standards rely on different _competing_
implementations. Otherwise it wouldn't make sense to have a standard. If
you think that a particular implementation provides clear benefits over
another one, then just go with that implementation. Don't worry first
about making your application portable. If it's successful and there is
some demand for portability, you can add that later on. I think that
when you worry about portability in the first step you are in fact
setting the wrong priorities.
- How much are you willing to spend on portable libraries? I don't
understand why people think that commercial Lisp implementations are too
expensive. Think about the big advantages Common Lisp gives you. Think
about how much time you would need to invest to get the same expressive
power in other languages. Divide the price of a Common lisp
implementation by that time and you get an idea how "expensive" these
implementations really are.
In the last few months I have been able to prototype language extensions
and ideas on my own that would have taken a few diploma theses when
implemented in Java and would still not be as good. Common Lisp means
such an amazing advantage over the "competition" that I am really happy
to invest a few bucks to make use of it.
Pascal
--
Pascal Costanza University of Bonn
···············@web.de Institute of Computer Science III
http://www.pascalcostanza.de R�merstr. 164, D-53117 Bonn (Germany)
From: Lars Brinkhoff
Subject: Re: Why aren't people complaining that LispWorks is too pricey?
Date:
Message-ID: <85u1bnqfry.fsf@junk.nocrew.org>
Pascal Costanza <········@web.de> writes:
> It's pretty easy to find one-implementation-only dialects of
> Lisp. (for example, OpenLisp - http://www.eligis.com/
Is there really only one implementation of ISLISP? Is the list of
implementations at islisp.info known to be exhaustive? A Google
search turns up TISL, but it's not clear it's a full implementation.
Lars Brinkhoff wrote:
> Is there really only one implementation of ISLISP? Is the list of
> implementations at islisp.info known to be exhaustive? A Google
> search turns up TISL, but it's not clear it's a full implementation.
It's my impression that OpenLisp is the only serious implementation of
ISLISP.
Pascal
--
Pascal Costanza University of Bonn
···············@web.de Institute of Computer Science III
http://www.pascalcostanza.de R�merstr. 164, D-53117 Bonn (Germany)
Pascal Costanza <········@web.de> writes:
> Lars Brinkhoff wrote:
>
> > Is there really only one implementation of ISLISP? Is the list of
> > implementations at islisp.info known to be exhaustive? A Google
> > search turns up TISL, but it's not clear it's a full implementation.
>
> It's my impression that OpenLisp is the only serious implementation of
> ISLISP.
No, I don't think so. I haven't finished contacting all the people who
I think have ISLISP implementations. I'm pretty sure there are quite
a lot of implementations (if you count different platforms) and
somewhat fewer but not just one sources of such implementations, maybe
2-5? (That is, some people maintain more than one implementation.)
Check back in a few weeks.
From: Eduardo Muñoz
Subject: Re: Why aren't people complaining that LispWorks is too pricey?
Date:
Message-ID: <ullx09q4u.fsf@terra.es>
* Tim Bradshaw <···@cley.com>
| [...] Yet here I am at 3AM trying to debug some annoying CORBA problem
| in a Lisp application I use from tens to thousands of times a
| day. Tomorrow I expect to need to spend some time writing some more
| support code for our documentation preparation system which is also in
| Lisp and which produces all but one of our web pages [...]
Are you hiring?
I don't have demonstrable experience, but could cheat if
needed ;)
--
Eduardo Mu�oz | (prog () 10 (print "Hello world!")
http://213.97.131.125/ | 20 (go 10))
From: Tim Bradshaw
Subject: Re: Why aren't people complaining that LispWorks is too pricey?
Date:
Message-ID: <ey3znlfzxbd.fsf@cley.com>
* Eduardo Mu�oz wrote:
> Are you hiring?
No, sorry.
But this wasn't the point. I didn't want to make some big thing of
how *I'm* doing clever stuff (which I'm not, mostly). What I'm doing
doesn't matter a damn. The point was meant to be that if you want
lisp to succeed, don't stand around complaining why you can't use it,
*use it* and write cool applications. The barriers turn out to be in
your own mind (both the invented ones that come out as `I can't do
this because Lisp doesn't do x' and the real ones which come out as `I
can't do this because I'm not smart enough').
--tim
From: Edi Weitz
Subject: Re: Why aren't people complaining that LispWorks is too pricey?
Date:
Message-ID: <87el2tt90a.fsf@bird.agharta.de>
Yarden Katz <····@underlevel.net> writes:
> I think that starting an active online Lisp magazine (kind of like
> XML.com only for Lisp) where people publish very brief and focused
> articles with code snippets would inspire more people to write
> applications. I considered actually setting up a server/domain name
> for such magazine and host it, but for this to really work I'd need
> a couple of volunteers who are more knowledgeable than I am to
> review submitted articles and write the initial ones.
You're not the first one to have this idea. See
<http://www.lisp-p.org/>
and search Google Groups for "yadda lisp".
You might also want to add something to the CL cookbook at
<http://cl-cookbook.sf.net/>.
Good luck.
Edi.
"Franz Kafka" <Symbolics _ XL1201 _ Sebek _ Budo _ Kafka @ hotmail . com> writes:
> The people we want to convice to try Lisp, are the C++/Java programmers
> looking for a better language to use. Most of these
> people will not have a Lisp running and there system.
>
> Some would prob. not like it if our app installed a CL imp. on
> there computer.
As others have said before, Java has the same "problem", as has .NET,
not to mention all the "scripting" languages that need to have an
interpreter around.
More anecdotal evidence: I guess most FreeBSD users have a Ruby
interpreter installed, even if few may use or even know the
language. The reason is that there is a collection of useful Ruby
programs that help with package management. I've never heard anyone
complain about this (and we are talking about users of an operating
system whose developers rewrote significant amounts of code in C and
sh recently just to get rid of perl in the base system).
I guess it's a better idea to focus on writing useful applications
rather than to bother too much whether the user starts them by
executing a native executable or a one-line shell script.
In fact, I have never heard any users of Lisp applications complain
either, only people saying they would really like to use Lisp, but
can't because of the percieved lack of ways to build executables. I
would have expected the kind of snake oil like SBCLs
pseudo-executables, or Linux' binfmt_misc or whatever, to satisfy
these people, it seems to have worked for Python with py2exe. One
possible interpretation is that Lispniks in general are smarter than
Pythoneers and realize that this doesn't really make anything better,
but that doesn't explain why they don't just become enlightened and
learn to love the way things are in the Lisp world ;-)
> Most end users don't what to DL at 20 Meg file when other programs that do
> the same things are 5 Megs.
Well, then write programs that do things others don't ;-)
Regards
Henrik
* Franz Kafka <>:
> Some would prob. not like it if our app installed a CL imp. on there
> computer.
Why? Why wouldn't they like it? How many of them wouldn't "like it"? Is your
application so unattractive that they need a reason like this to not use it?
> Or, we want it to have a Lisp-style scripting language--so when the people
> extend our app. they are learning Lisp. (This is why EMACS attracted so
> many people to Lisp in the past.)
So use something like ECL or, on Windows, Corman Common Lisp. I've not
personally played with this aspect of it but as I understand it you can ship
a single DLL to get access to Corman Common Lisp from any other language
capable of dealing with COM. Would your users for your application complain
about you installing a CL implementation that way?
Why?
How would they know?
If they do know doesn't that provide you with the knowledge you wish they'd
gain:
> Lisp is like the Matrix, people need to realize it exists--before they'll
> want to use it. :)
This doesn't make sense at all. First you don't want "people" to know that
it exists, then you do. Really, what do you *really* want and what are you
willing to do to get it?
> From the length of these Lisp price/ value of a small exe file post and
> the amount of excuses that Lisp people come up with for not wanting small
> exe files--it's clear to me that the Lisp community has not yet woken up.
Contrast this with the list of excuses that people come up with for not
using Lisp. What you dismiss as "excuses" are generally reasons why people
haven't had the problems you're coming up with. In other words people who
*have* done stuff with Lisp are saying why these issues haven't been a
problem. On the other hand you've got people coming up with problems,
apparently to save themselves the hassle of actually getting down and
writing something.
Really, just get on and do it.
--
Dave Pearson
http://www.davep.org/lisp/
From: Thien-Thi Nguyen
Subject: Re: Why aren't people complaining that LispWorks is too pricey?
Date:
Message-ID: <7gk7cuvmbk.fsf@gnufans.net>
Dave Pearson <··········@davep.org> writes:
> Really, just get on and do it.
i call "usenet buckyball" on Kafka! (rigid self-supporting
argumentation surrounding what is, in essense, a void. ;-)
ok, back to lurking...
thi
From: Raymond Wiker
Subject: Re: Why aren't people complaining that LispWorks is too pricey?
Date:
Message-ID: <86ptmmwxfc.fsf@raw.grenland.fast.no>
Thien-Thi Nguyen <···@glug.org> writes:
> Dave Pearson <··········@davep.org> writes:
>
> > Really, just get on and do it.
>
> i call "usenet buckyball" on Kafka! (rigid self-supporting
> argumentation surrounding what is, in essense, a void. ;-)
>
> ok, back to lurking...
*I* think Kafka == ilias.
--
Raymond Wiker Mail: ·············@fast.no
Senior Software Engineer Web: http://www.fast.no/
Fast Search & Transfer ASA Phone: +47 23 01 11 60
P.O. Box 1677 Vika Fax: +47 35 54 87 99
NO-0120 Oslo, NORWAY Mob: +47 48 01 11 60
Try FAST Search: http://alltheweb.com/
"Dave Pearson" <··········@davep.org> wrote in message
······························@hagbard.davep.org...
>
> > Lisp is like the Matrix, people need to realize it exists--before
they'll
> > want to use it. :)
>
> This doesn't make sense at all. First you don't want "people" to know that
> it exists, then you do. Really, what do you *really* want and what are you
> willing to do to get it?
>
People need to get hooked onto Lisp, by using it as a Scripting
Language--after that more people will know how to use Lisp.
If people have used Lisp as a Scripting Language it will be easier to
convince them to use it as programming language.
However, it needs to be an app. that many people want to use. AutoCad's
AutoLisp was a good example--but AutoCad cost
too much. Emacs ELISP is an other good idea.
We need more apps. like this--people'll want to extend they apps. and will
be willing to try a scripting lang. As a bonus they'll learn how to
use Lisp.
* Franz Kafka <>:
> "Dave Pearson" <··········@davep.org> wrote in message
> ······························@hagbard.davep.org...
>
> > This doesn't make sense at all. First you don't want "people" to know
> > that it exists, then you do. Really, what do you *really* want and what
> > are you willing to do to get it?
>
> People need to get hooked onto Lisp, by using it as a Scripting
> Language--after that more people will know how to use Lisp.
Excellent idea. What's this killer application you're writing that will do
this? How is it better than previous applications that have done this?
> We need more apps. like this--people'll want to extend they apps. and will
> be willing to try a scripting lang. As a bonus they'll learn how to use
> Lisp.
So, what is your killer application? When can we see it?
--
Dave Pearson: | lbdb.el - LBDB interface.
http://www.davep.org/ | sawfish.el - Sawfish mode.
Emacs: | uptimes.el - Record emacs uptimes.
http://www.davep.org/emacs/ | quickurl.el - Recall lists of URLs.
From: Damond Walker
Subject: Re: Why aren't people complaining that LispWorks is too pricey?
Date:
Message-ID: <BAE706B7.14EE7%damosan@comcast.net>
On 5/13/03 12:36, in article
···················@news02.roc.ny.frontiernet.net, "Franz Kafka" <Symbolics
_ XL1201 _ Sebek _ Budo _ Kafka @ hotmail . com> wrote:
>
> People need to get hooked onto Lisp, by using it as a Scripting
> Language--after that more people will know how to use Lisp.
>
The downside to this is that, after some amount of time, people will think
of Lisp as *only* a scripting language. Even then what percentage of any
given user base will spend the time required to learn how to use it?
What's *really* needed is a mechanism that will show the unwashed masses
that Lisp is a good tool when faced with the hard problems. :)
Damond
Damond Walker <·······@comcast.net> writes:
> On 5/13/03 12:36, in article
> ···················@news02.roc.ny.frontiernet.net, "Franz Kafka" <Symbolics
> _ XL1201 _ Sebek _ Budo _ Kafka @ hotmail . com> wrote:
>
> >
> > People need to get hooked onto Lisp, by using it as a Scripting
> > Language--after that more people will know how to use Lisp.
> >
>
> The downside to this is that, after some amount of time, people will think
> of Lisp as *only* a scripting language. Even then what percentage of any
> given user base will spend the time required to learn how to use it?
>
> What's *really* needed is a mechanism that will show the unwashed masses
> that Lisp is a good tool when faced with the hard problems. :)
If free software, or at least open source was more current, they could
see that bad software or easy problem software was developed in C++
and that good software or hard problem software was developed in Lisp. ;-)
--
__Pascal_Bourguignon__ http://www.informatimago.com/
----------------------------------------------------------------------
Do not adjust your mind, there is a fault in reality.
On Tue, 13 May 2003 20:31:51 -0400, Damond Walker
<·······@comcast.net> wrote:
>On 5/13/03 12:36, in article
>···················@news02.roc.ny.frontiernet.net, "Franz Kafka" <Symbolics
>_ XL1201 _ Sebek _ Budo _ Kafka @ hotmail . com> wrote:
>
>>
>> People need to get hooked onto Lisp, by using it as a Scripting
>> Language--after that more people will know how to use Lisp.
>>
>
>The downside to this is that, after some amount of time, people will think
>of Lisp as *only* a scripting language. Even then what percentage of any
>given user base will spend the time required to learn how to use it?
>
>What's *really* needed is a mechanism that will show the unwashed masses
>that Lisp is a good tool when faced with the hard problems. :)
Said by someone who doesn't know. Scripting languages can
do quite complex things. There are scripts out there that do some
pretty sophistcated things.
--------------------------------------------------
Thaddeus L. Olczyk, PhD
Think twice, code once.
From: Damond Walker
Subject: Re: Why aren't people complaining that LispWorks is too pricey?
Date:
Message-ID: <BAE84404.15124%damosan@comcast.net>
On 5/14/03 13:26, in article ··································@4ax.com,
"Thaddeus L Olczyk" <······@interaccess.com> wrote:
>
> Said by someone who doesn't know. Scripting languages can
> do quite complex things. There are scripts out there that do some
> pretty sophistcated things.
Oh, on the contrary, I know exactly what scripting is capable of. But
your response doesn't really address my original message in the least.
While you and I and a host of others may know what's possible there are a
whole crew of those who do not.
What I said in my original post was that if Lisp becomes widespread as a
scripting languge then folks who come to Lisp via this path might assume
that scripting is all that Lisp is good for. Which isn't the case is it?
Damond
From: Tim Bradshaw
Subject: Re: Why aren't people complaining that LispWorks is too pricey?
Date:
Message-ID: <ey3u1bwy2t2.fsf@cley.com>
* Damond Walker wrote:
> What I said in my original post was that if Lisp becomes widespread
> as a scripting languge then folks who come to Lisp via this path
> might assume that scripting is all that Lisp is good for. Which
> isn't the case is it?
Yes, it's all it's good for. Because there isn't, actually, any
difference between scripting and programming.
--tim
After a long battle with technology,Tim Bradshaw <···@cley.com>, an earthling, wrote:
> * Damond Walker wrote:
>> What I said in my original post was that if Lisp becomes widespread
>> as a scripting languge then folks who come to Lisp via this path
>> might assume that scripting is all that Lisp is good for. Which
>> isn't the case is it?
>
> Yes, it's all it's good for. Because there isn't, actually, any
> difference between scripting and programming.
There was, back in the day when "scripting" meant JCL or something not
much better.
But when people actually deploy substantial programs where the only
code they wrote was in Python/Perl/Ruby, you're certainly right that
there's no meaningful difference between using a "scripting" language
and a "real" one (whatever /that/ distinction is...).
--
select 'aa454' || ·@' || 'freenet.carleton.ca';
http://www.ntlug.org/~cbbrowne/spreadsheets.html
As of next Thursday, ITS will be flushed in favor of TOPS-10.
Please update your programs.
Centuries ago, Nostradamus foresaw when Thaddeus L Olczyk <······@interaccess.com> would write:
> On Tue, 13 May 2003 20:31:51 -0400, Damond Walker
> <·······@comcast.net> wrote:
>>The downside to this is that, after some amount of time, people will
>>think of Lisp as *only* a scripting language. Even then what
>>percentage of any given user base will spend the time required to
>>learn how to use it?
>>What's *really* needed is a mechanism that will show the unwashed
>>masses that Lisp is a good tool when faced with the hard problems.
>>:)
> Said by someone who doesn't know. Scripting languages can do quite
> complex things. There are scripts out there that do some pretty
> sophistcated things.
At one time, "scripting languages" were merely 'glue' used to paste to
control the 'real' programs. (Think "JCL"; the idea being to attach
the file descriptors referenced in the program to the files you
_actually_ want to use...)
Bourne shell and REXX extended this by providing the ability to do a
bit of "structured programming."
Perl took it further, linking in hordes of Unix libraries.
What happens *NOW* is that there is commonly nothing about the
"scripting" languages that forcibly associates them with the notion of
"scripting."
-> They are more than likely compiled. Certainly true for Python, and
plans for Perl involve compiling to a sorta-Riscy VM assembly code.
So you may NOT be running a text "script."
-> Since they probably have more control structures than C has, they
are more than likely more usable as "programming languages" than
many of the traditional "real programming languages."
If you write an application in Perl/Python/Ruby/... that isn't merely
a wrapper for compiled apps written in C/C++/..., then you're not
using it as a "scripting language" nor in a mode that could reasonably
considered "scripted."
--
If this was helpful, <http://svcs.affero.net/rm.php?r=cbbrowne> rate me
http://cbbrowne.com/info/scripting.html
The easiest way to find something lost around the house is to buy a
replacement.
From: Björn Lindberg
Subject: Re: Why aren't people complaining that LispWorks is too pricey?
Date:
Message-ID: <hcswugt1ym8.fsf@tjatte.nada.kth.se>
"Franz Kafka" <Symbolics _ XL1201 _ Sebek _ Budo _ Kafka @ hotmail . com> writes:
> "Dave Pearson" <··········@davep.org> wrote in message
> ······························@hagbard.davep.org...
> >
> > > Lisp is like the Matrix, people need to realize it exists--before
> they'll
> > > want to use it. :)
> >
> > This doesn't make sense at all. First you don't want "people" to know that
> > it exists, then you do. Really, what do you *really* want and what are you
> > willing to do to get it?
> >
>
> People need to get hooked onto Lisp, by using it as a Scripting
> Language--after that more people will know how to use Lisp.
If you are referring to "scripting language" in the same way languages
like Python, Perl and Ruby are scripting languages, Clisp supports
such use, at least under Unix. With Clisp you can write shebang
scripts (a text file beginning with #!/usr/bin/env clisp for example),
similarly to eg Python scripts.
Bj�rn
Daniel Barlow <···@telent.net> wrote in message news:<··············@noetbook.telent.net>...
> ········@yahoo.com (Eric Smith) writes:
>
> > It seems clear from large numbers of postings to
> > comp.lang.lisp over the years that there is a significant
> > market for being able to build binary executables while
> > learning Lisp. I don't know how much money such people
>
> There is nothing (except perhaps the ~20Mb installed size, which is
> roughly what a JRE weighs in at anyway) stopping people from
> distributing CMUCL or SBCL along with their application.
Nothing except a desire to be successful. Java is arguably the most
popular language in the world right now, and it hasn't made a dent in
C/C++ for client-side apps because of the huge boat anchor of a
runtime a Java app has to drag around with it (and the fact that its
philosophy requires that it use that boat anchor for *everything*,
even for things that the local OS does better).
C apps require a runtime, too, but that runtime is either called the
"OS" or is installed with the OS, so C apps are small and look as good
as the standard apps on that OS.
Languages that let you use the features of the underlying OS for most
of their work and only require small runtimes statically linked to the
app itself are the languages that have been successful for apps that
are sold to the open market (not in-house). That includes C, C++,
Pascal, Delphi, Objective-C, etc.
Languages that require the installation of a separate runtime have
only been successful on machines the developers can personally manage
(servers or in-house clients), despite any advantages of the languages
themselves.
From: Daniel Barlow
Subject: Re: Why aren't people complaining that LispWorks is too pricey?
Date:
Message-ID: <8765oezgik.fsf@noetbook.telent.net>
········@hotmail.com (Tuang) writes:
>> There is nothing (except perhaps the ~20Mb installed size, which is
>> roughly what a JRE weighs in at anyway) stopping people from
>> distributing CMUCL or SBCL along with their application.
>
> Nothing except a desire to be successful. Java is arguably the most
> popular language in the world right now, and it hasn't made a dent in
> C/C++ for client-side apps because of the huge boat anchor of a
> runtime a Java app has to drag around with it (and the fact that its
I said "nothing except the installed size", so you say "the installed
size is a problem". Can we establish exactly how much of a problem
we're talking about? 40Gb of disk practically comes free with
breakfast cereals these days.
:; du -s /usr/lib/mozilla/
38184 /usr/lib/mozilla
> Languages that require the installation of a separate runtime have
> only been successful on machines the developers can personally manage
> (servers or in-house clients), despite any advantages of the languages
> themselves.
Sure. That's why all Windows apps are delivered as single EXE files.
Clearly InstallShield must have been something I dreamt one night
after eating too much cheese.
If you want to talk about poor integration with the host OS as a
reason that people are not delivering applications in Lisp, fine, go
ahead (I'd even agree to a certain extent: integration with the host
OS is suboptimal in free unix lisps), but at least start by admitting
that you're moving the goalposts. We _were_ talkin about generating
"standalone executables"
-dan
--
http://www.cliki.net/ - Link farm for free CL-on-Unix resources
From: Espen Vestre
Subject: Re: Why aren't people complaining that LispWorks is too pricey?
Date:
Message-ID: <kw4r3yvxp3.fsf@merced.netfonds.no>
Daniel Barlow <···@telent.net> writes:
> > Languages that require the installation of a separate runtime have
> > only been successful on machines the developers can personally manage
> > (servers or in-house clients), despite any advantages of the languages
> > themselves.
>
> Sure. That's why all Windows apps are delivered as single EXE files.
> Clearly InstallShield must have been something I dreamt one night
> after eating too much cheese.
InstallShield is something you dreamt of last night after eating
the wrong kind of mushrooms. And your dream was true.
It's frustrating that you have to build those stupid installers even
when you actually _do_ provide single EXE files (as I in fact do for
my Common Lisp application, languages do NOT "require the installation
of separate runtimes", that's up to the implementations) , because the
average Windows users has absolute no sensible idea of what a
'program' is and hasn't got the faintest idea that you can actually
put things into the startmenu all by yourself.
--
(espen)
Espen Vestre <·····@*do-not-spam-me*.vestre.net> writes:
> Daniel Barlow <···@telent.net> writes:
>
> > > Languages that require the installation of a separate runtime have
> > > only been successful on machines the developers can personally manage
> > > (servers or in-house clients), despite any advantages of the languages
> > > themselves.
> >
> > Sure. That's why all Windows apps are delivered as single EXE files.
> > Clearly InstallShield must have been something I dreamt one night
> > after eating too much cheese.
>
> InstallShield is something you dreamt of last night after eating
> the wrong kind of mushrooms. And your dream was true.
Is this a "Lathe of Heaven" moment?
marc
From: Espen Vestre
Subject: Re: Why aren't people complaining that LispWorks is too pricey?
Date:
Message-ID: <kwof25smk0.fsf@merced.netfonds.no>
Greg Menke <··········@toadmail.com> writes:
> So theres nothing stopping a program from being distributed as a .exe
> file, but all the menu/icon/registration crud will then have to be
> done by hand- and doing that is at least as messy as you could imagine
> it being.
...adding a program to the start menu is a simple right-click operation
in XP (don't remember if is in the older versions). It is simple,
but still Joe Average User will get totally confused, since they
_expect_ installations to be complicated.
--
(espen)
From: Jan Rychter
Subject: Re: Why aren't people complaining that LispWorks is too pricey?
Date:
Message-ID: <m2y9174cwe.fsf@tnuctip.rychter.com>
>>>>> "Espen" == Espen Vestre <·····@*do-not-spam-me*.vestre.net>:
Espen> Greg Menke <··········@toadmail.com> writes:
>> So theres nothing stopping a program from being distributed as a
>> .exe file, but all the menu/icon/registration crud will then have to
>> be done by hand- and doing that is at least as messy as you could
>> imagine it being.
Espen> ...adding a program to the start menu is a simple right-click
Espen> operation in XP (don't remember if is in the older versions). It
Espen> is simple, but still Joe Average User will get totally confused,
Espen> since they _expect_ installations to be complicated.
I don't understand. I'm not a "Joe Average Windows User", but I keep
dreaming of the day when all new functionality (software) will be a
single click (or command) to install. And another one to
deinstall. Properly. Including all the start menus, associations, or
whatever else.
No, I don't want to think about where I should put the executable. No, I
don't want to add anything manually. No, I don't want to "configure"
paths. A lot of software is designed with total disrespect for user's
time and requires a lot of things to be done manually.
Also, I would expect to have a standardized interface for installation
of new software. Installshield is close to that.
So, I don't really understand the point of this discussion. IMHO
installers do have their place if they are simple, standardized, and
work well for both installation and deinstallation. Whatever the OS at
hand (I don't use Windows).
--J.
Stop dreaming and switch to Mac OS X. ;)
Pascal
Jan Rychter wrote:
> I don't understand. I'm not a "Joe Average Windows User", but I keep
> dreaming of the day when all new functionality (software) will be a
> single click (or command) to install. And another one to
> deinstall. Properly. Including all the start menus, associations, or
> whatever else.
>
> No, I don't want to think about where I should put the executable. No, I
> don't want to add anything manually. No, I don't want to "configure"
> paths. A lot of software is designed with total disrespect for user's
> time and requires a lot of things to be done manually.
>
> Also, I would expect to have a standardized interface for installation
> of new software. Installshield is close to that.
>
> So, I don't really understand the point of this discussion. IMHO
> installers do have their place if they are simple, standardized, and
> work well for both installation and deinstallation. Whatever the OS at
> hand (I don't use Windows).
--
Pascal Costanza University of Bonn
···············@web.de Institute of Computer Science III
http://www.pascalcostanza.de R�merstr. 164, D-53117 Bonn (Germany)
From: Espen Vestre
Subject: Re: Why aren't people complaining that LispWorks is too pricey?
Date:
Message-ID: <kwfznfkzrx.fsf@merced.netfonds.no>
Pascal Costanza <········@web.de> writes:
> Stop dreaming and switch to Mac OS X. ;)
Huh? Some OS X developers actually require me to _drag_ the application
to the application folder (or whereever I want to put it), and
they even expect me to put the icons into the dock all by myself!
--
(espen)
Espen Vestre wrote:
> Pascal Costanza <········@web.de> writes:
>
>>Stop dreaming and switch to Mac OS X. ;)
>
> Huh? Some OS X developers actually require me to _drag_ the application
> to the application folder (or whereever I want to put it), and
> they even expect me to put the icons into the dock all by myself!
So do you have any problems with that? Is there anything in that model
that's hard to understand?
BTW, I think the "whereever I want to put it" part is the most important
part. "Whereever I want to put it" might be the trash can. This means
that "installing" and "deinstalling" are essentially the same operation.
No more "do you really want to delete this DLL?"...
Pascal
--
Pascal Costanza University of Bonn
···············@web.de Institute of Computer Science III
http://www.pascalcostanza.de R�merstr. 164, D-53117 Bonn (Germany)
From: Espen Vestre
Subject: Re: Why aren't people complaining that LispWorks is too pricey?
Date:
Message-ID: <kwbry3kyys.fsf@merced.netfonds.no>
Pascal Costanza <········@web.de> writes:
> So do you have any problems with that? Is there anything in that model
> that's hard to understand?
No, not at all, quite in contrary - I love it.
I was trying to be ironic, since my original point was that my
applications created with LW are just as easy to install under
Windows. I can put them wherever I want and run them from there, since
they don't come with their own dll-s and don't depend on residing in
specific directories.
--
(espen)
Espen Vestre wrote:
> Pascal Costanza <········@web.de> writes:
>
>>So do you have any problems with that? Is there anything in that model
>>that's hard to understand?
>
> No, not at all, quite in contrary - I love it.
>
> I was trying to be ironic, since my original point was that my
> applications created with LW are just as easy to install under
> Windows. I can put them wherever I want and run them from there, since
> they don't come with their own dll-s and don't depend on residing in
> specific directories.
Oh, yes, that's nice. ;)
Every once in a while even Windows applications get it right. ;)
Pascal
--
Pascal Costanza University of Bonn
···············@web.de Institute of Computer Science III
http://www.pascalcostanza.de R�merstr. 164, D-53117 Bonn (Germany)
Espen Vestre <·····@*do-not-spam-me*.vestre.net> writes:
> Daniel Barlow <···@telent.net> writes:
>
> > > Languages that require the installation of a separate runtime have
> > > only been successful on machines the developers can personally manage
> > > (servers or in-house clients), despite any advantages of the languages
> > > themselves.
> >
> > Sure. That's why all Windows apps are delivered as single EXE files.
> > Clearly InstallShield must have been something I dreamt one night
> > after eating too much cheese.
>
> InstallShield is something you dreamt of last night after eating
> the wrong kind of mushrooms. And your dream was true.
>
> It's frustrating that you have to build those stupid installers even
> when you actually _do_ provide single EXE files (as I in fact do for
> my Common Lisp application, languages do NOT "require the installation
> of separate runtimes", that's up to the implementations) , because the
> average Windows users has absolute no sensible idea of what a
> 'program' is and hasn't got the faintest idea that you can actually
> put things into the startmenu all by yourself.
I like it on my side of the looking glass, where I can have users drop
a tarball in /usr/local/packages, or a .app in /Applications.
--
/|_ .-----------------------.
,' .\ / | No to Imperialist war |
,--' _,' | Wage class war! |
/ / `-----------------------'
( -. |
| ) |
(`-. '--.)
`. )----'
········@hotmail.com (Tuang) wrote in message news:<····························@posting.google.com>...
> Languages that require the installation of a separate runtime have
> only been successful on machines the developers can personally manage
> (servers or in-house clients), despite any advantages of the languages
> themselves.
There are roughly 15 million Visual Basic programmers out there,
successfully marketing end-user applications even though it does need
a runtime.
Boethius wrote:
> ········@hotmail.com (Tuang) wrote in message
> news:<····························@posting.google.com>...
>
>> Languages that require the installation of a separate runtime have
>> only been successful on machines the developers can personally manage
>> (servers or in-house clients), despite any advantages of the languages
>> themselves.
>
> There are roughly 15 million Visual Basic programmers out there,
> successfully marketing end-user applications even though it does need
> a runtime.
Scripting languages are accepted (have to install Python, Perl, Ruby, ...)
as is Visual Basic in the Win-World. But for CL-as-we-know-it you are
asking the user not for installing one but several Lisps: Depending on if
the developers choose CLisp, CMUCL, ECL, name-your-favourite-free-Lisp and
did not write 100% pure ansi-only CL. Also it might be required that a user
downloads and installs separate packages (sockets,
your-favourite-defsystem, ...) for one of these Lisps.
CL is a full-blown programming language with multiple, different
implementations (as opposed to classical scripting languages or VB). This
is good as these implementations make different design desicions each and
you may find one which fits especially well to your needs. This is bad
because if you want to distribute applications to a wide audience each
potential user needs to retrieve-install-extend-update a number of
different implementations. As many Lisp implementations and libraries are
relatively difficult to install (sometimes you have to know Lisp or
some-developers-favourite-defsystem) compared to popular scripting
languages (which usally come in an rpm or self-installing exe), there is a
good chance that many users won't be able to do it. Unfortunately, simply
shipping your own Lisp-environment with your application is no option:
Imagine Lisp gets popular and the users have 100 Lisp programs, each
shipping its own implementation...
In short: If you want to distribute an application to a wide audience who
doesn't yet have many technical skills the free CLs are a bad choice.
Exception 1: Of course, you can always write a program which the user
percieves as extremely useful and without competition. Then installing a
CL is obviously worth the effort. (Maxima is an example for a program
which has no serious competition in the free software world and might be
percieved as very useful.)
Exception 2: For some OS platforms there are free CLs reported which are
said to install out-of-the box (haven't checked myself).
Matthias
In article <············@trumpet.uni-mannheim.de>,
Matthias <····@yourself.pl> wrote:
> Scripting languages are accepted (have to install Python, Perl, Ruby, ...)
> as is Visual Basic in the Win-World. But for CL-as-we-know-it you are
> asking the user not for installing one but several Lisps: Depending on if
> the developers choose CLisp, CMUCL, ECL, name-your-favourite-free-Lisp and
> did not write 100% pure ansi-only CL. Also it might be required that a user
> downloads and installs separate packages (sockets,
> your-favourite-defsystem, ...) for one of these Lisps.
In the Java world, usually each application comes with its own JVM
because of differences between the several JDKs. So again, this doesn't
seem to prevent people from using Java.
Just choose the right Common Lisp implementation that allows you to
easily deploy your application.
(Several years ago I have worked for a web company, and we have used
SchemaText in one of our projects. At that time I wasn't aware of Common
Lisp, but noticed that that application was developed in Allegro Common
Lisp. I don't recall any problems whatsoever wrt installation, it just
behaved like any other Win app.)
Pascal
From: Raymond Wiker
Subject: Re: Why aren't people complaining that LispWorks is too pricey?
Date:
Message-ID: <868yt8wr9r.fsf@raw.grenland.fast.no>
Pascal Costanza <········@web.de> writes:
> In article <············@trumpet.uni-mannheim.de>,
> Matthias <····@yourself.pl> wrote:
>
> > Scripting languages are accepted (have to install Python, Perl, Ruby, ...)
> > as is Visual Basic in the Win-World. But for CL-as-we-know-it you are
> > asking the user not for installing one but several Lisps: Depending on if
> > the developers choose CLisp, CMUCL, ECL, name-your-favourite-free-Lisp and
> > did not write 100% pure ansi-only CL. Also it might be required that a user
> > downloads and installs separate packages (sockets,
> > your-favourite-defsystem, ...) for one of these Lisps.
>
> In the Java world, usually each application comes with its own JVM
> because of differences between the several JDKs. So again, this doesn't
> seem to prevent people from using Java.
Right. On my Linux development machine, I have installed
Lotus/Domino, and Oracle 9. Both of these come with their own Java
environements. In addition, I have installed a separate Java version
for general use.
I have *no* machines with more than two Common Lisp
implementations.
--
Raymond Wiker Mail: ·············@fast.no
Senior Software Engineer Web: http://www.fast.no/
Fast Search & Transfer ASA Phone: +47 23 01 11 60
P.O. Box 1677 Vika Fax: +47 35 54 87 99
NO-0120 Oslo, NORWAY Mob: +47 48 01 11 60
Try FAST Search: http://alltheweb.com/
From: Raymond Wiker
Subject: Re: Why aren't people complaining that LispWorks is too pricey?
Date:
Message-ID: <864r3wwpf7.fsf@raw.grenland.fast.no>
Raymond Wiker <·············@fast.no> writes:
> Right. On my Linux development machine, I have installed
> Lotus/Domino, and Oracle 9. Both of these come with their own Java
> environements. In addition, I have installed a separate Java version
> for general use.
The same machine also has two versions of Python installed:
1.5 (or something) for system use, and 2.2 for actually doing stuff
with.
> I have *no* machines with more than two Common Lisp
> implementations.
--
Raymond Wiker Mail: ·············@fast.no
Senior Software Engineer Web: http://www.fast.no/
Fast Search & Transfer ASA Phone: +47 23 01 11 60
P.O. Box 1677 Vika Fax: +47 35 54 87 99
NO-0120 Oslo, NORWAY Mob: +47 48 01 11 60
Try FAST Search: http://alltheweb.com/
········@hotmail.com (Tuang) wrote in message news:<····························@posting.google.com>...
...
> Languages that require the installation of a separate runtime have
> only been successful on machines the developers can personally manage
> (servers or in-house clients), despite any advantages of the languages
> themselves.
I'm not sure how "Languages" can require that. Maybe "implementations".
Here is a cool example what is possible with a FREE implementation
of Common Lisp.
If you look for example at OpenMCL for Mac OS X. There is an application
called Alpaca (http://alpaca.sourceforge.net/) written in OpenMCL.
Alpaca is a specialized editor (Lisp-Programmable Author's
Creative Assistant). It is in early stages of its development, but very
useful as an example for people who want to see how to develop
GUI-based (using the Cocoa-Libs) applications for Mac OS X with OpenMCL.
Installation is as follows:
- download a disk image
- double click on the disk image to mount it
- copy the application to any place you wish
- double click the application -> runs
I'm pretty sure ANY enduser can do that, even though the application
is written in Common Lisp and includes Common Lisp.
Alpaca contains a version of OpenMCL *with* the Common Lisp native
code compiler. The Alpaca installation disk image size is
just 1.5 Mbytes.
This is surely an example where ANY enduser can easily install
an application (in this case a specialized editor) written
in Common Lisp.
Rainer Joswig
In article <····························@posting.google.com>,
······@corporate-world.lisp.de (Rainer Joswig) wrote:
> Here is a cool example what is possible with a FREE implementation
> of Common Lisp.
>
> If you look for example at OpenMCL for Mac OS X. There is an application
> called Alpaca (http://alpaca.sourceforge.net/) written in OpenMCL.
> Alpaca is a specialized editor (Lisp-Programmable Author's
> Creative Assistant). It is in early stages of its development, but very
> useful as an example for people who want to see how to develop
> GUI-based (using the Cocoa-Libs) applications for Mac OS X with OpenMCL.
>
> Installation is as follows:
>
> - download a disk image
> - double click on the disk image to mount it
> - copy the application to any place you wish
> - double click the application -> runs
>
> I'm pretty sure ANY enduser can do that, even though the application
> is written in Common Lisp and includes Common Lisp.
This level of simplicity is largely due to the underlying operating
system.
Other things you can usually do on Mac OS X:
- Move applications ot other places at any time.
- Yes, any time! An experiment I have made once: I have started
Microsoft Word, and while it was running moved the Word application out
of the application folder to the desktop. Word just continued to work.
Not that you want to do this regularly, but such things give you faith
in the reliability of the operating system.
[This is not meant as a counter argument - to the contrary, the ease of
installation doesn't have _anything_ to do with the
language/implementation thereof it is implemented in.]
Pascal
Pascal Costanza <········@web.de> wrote in message news:<······························@news.netcologne.de>...
> In article <····························@posting.google.com>,
> ······@corporate-world.lisp.de (Rainer Joswig) wrote:
>
> > Here is a cool example what is possible with a FREE implementation
> > of Common Lisp.
> >
> > If you look for example at OpenMCL for Mac OS X. There is an application
> > called Alpaca (http://alpaca.sourceforge.net/) written in OpenMCL.
> > Alpaca is a specialized editor (Lisp-Programmable Author's
> > Creative Assistant). It is in early stages of its development, but very
> > useful as an example for people who want to see how to develop
> > GUI-based (using the Cocoa-Libs) applications for Mac OS X with OpenMCL.
> >
> > Installation is as follows:
> >
> > - download a disk image
> > - double click on the disk image to mount it
> > - copy the application to any place you wish
> > - double click the application -> runs
> >
> > I'm pretty sure ANY enduser can do that, even though the application
> > is written in Common Lisp and includes Common Lisp.
>
> This level of simplicity is largely due to the underlying operating
> system.
The level of simplicity is largely due to the fact that OpenMCL
does enable the use of all the OS mechanisms that makes it possible.
The GUI stuff even can be designed using Apple's "Interface Builder"
application - which btw. was once was a Lisp application on the Mac
(SOS Interface from Jean-Marie Hullot in 1984).
...
From: Tim Bradshaw
Subject: Re: Why aren't people complaining that LispWorks is too pricey?
Date:
Message-ID: <ey31xz45wd1.fsf@cley.com>
* Duane Rettig wrote:
> I was not, however, talking at all about software hobbyists.
> Instead, I was talking about hobbies like snow and water skiing,
> photography, woodworking, collections of various kinds, model train
> sets, music, etc. All of these are activities that one may take on
> in varying degrees, with various commitments of money. And unless
> the activity is being taken on as a career, where the goal is to
> make enough money for sustenance, a hobby tends to be an end unto
> itself.
This is worth reinforcing. Maybe, as you say, lots of programmers
have no hobbies (I suspect this), but I have some - too many in fact -
and if I programmed as a hobby it would likely not be that
expensive...
Sailing dinghies (which I don't do any more) - I don't know how much a
competitive new boat would be of the class I used to race, but I've
found second-hand but clearly serious boats for 5000-8000 pounds.
Less serious - older - boats will come down to ~ 1000 I suspect, below
that and they'd be too wet and soft I think. Cost to race for a
season might easily be 1000 pounds.
Photography - well, a decent medium format camera is probably
1000-2000, say as much again on darkroom stuff. And that's if you're
not obsessed about having a google lenses. Want to scan things? A
medium format scanner is between 1800 and 3000.
Music - I don't know how much I've spent on guitars over the years,
but my lovely ES175 is probably 2-3k, add on amplification &c &c. And
what about CDs & records - I know it's now fashionable to steal music
rather than paying for it, but I must have spent a few tens of
thousands of pounds on records & CDs over the years.
--tim
From: Greg Menke
Subject: Re: Why aren't people complaining that LispWorks is too pricey?
Date:
Message-ID: <m3ptmo8hnc.fsf@europa.pienet>
Tim Bradshaw <···@cley.com> writes:
> * Duane Rettig wrote:
>
> > I was not, however, talking at all about software hobbyists.
> > Instead, I was talking about hobbies like snow and water skiing,
> > photography, woodworking, collections of various kinds, model train
> > sets, music, etc. All of these are activities that one may take on
> > in varying degrees, with various commitments of money. And unless
> > the activity is being taken on as a career, where the goal is to
> > make enough money for sustenance, a hobby tends to be an end unto
> > itself.
>
> This is worth reinforcing. Maybe, as you say, lots of programmers
> have no hobbies (I suspect this), but I have some - too many in fact -
> and if I programmed as a hobby it would likely not be that
> expensive...
>
> Sailing dinghies (which I don't do any more) - I don't know how much a
> competitive new boat would be of the class I used to race, but I've
> found second-hand but clearly serious boats for 5000-8000 pounds.
> Less serious - older - boats will come down to ~ 1000 I suspect, below
> that and they'd be too wet and soft I think. Cost to race for a
> season might easily be 1000 pounds.
>
> Photography - well, a decent medium format camera is probably
> 1000-2000, say as much again on darkroom stuff. And that's if you're
> not obsessed about having a google lenses. Want to scan things? A
> medium format scanner is between 1800 and 3000.
>
> Music - I don't know how much I've spent on guitars over the years,
> but my lovely ES175 is probably 2-3k, add on amplification &c &c. And
> what about CDs & records - I know it's now fashionable to steal music
> rather than paying for it, but I must have spent a few tens of
> thousands of pounds on records & CDs over the years.
I agreee as well. If the op wanted to set up a machine shop to fiddle
around working with metal in a non-trivial manner, it will require at
least about $3,000 in tooling and hardware- even if the big stuff is
purchased used. Probably more like $5,000 if you want some of it to
be nice as opposed to sufficient. By "non-trivial", I mean actually
making things that work and have good fit & finish, as opposed to
screwing together chunks of roughly cut metal & filing off corners
until the clearances are adequate.
OTOH, its easy to pick up inexpensive lathes/mill drills in the $500
to $750 range. Unfortunately, these end up only being able to handle
fairly small cuts in aluminum, lacking power and precision to handle
steel or considerably large parts.
If all you need is a workbench, hacksaw and vise then you don't need
to spend so much, but there is simply no way (short of being some kind
of genius metalworker) that you'll be able to do what having good
machinery will allow you to.
Gregm
Tim Bradshaw <···@cley.com> writes:
> This is worth reinforcing. Maybe, as you say, lots of programmers
> have no hobbies (I suspect this), but I have some - too many in fact -
> and if I programmed as a hobby it would likely not be that
> expensive...
>
> Sailing dinghies (which I don't do any more) - I don't know how much a
> competitive new boat would be of the class I used to race, but I've
> found second-hand but clearly serious boats for 5000-8000 pounds.
> Less serious - older - boats will come down to ~ 1000 I suspect, below
> that and they'd be too wet and soft I think. Cost to race for a
> season might easily be 1000 pounds.
I don't think this analogy reflects the situation with software very
well, unless I missed something and you can get perfectly fine,
"serious" - but maybe not exactly stylish, and perhaps lacking a
maintenance contract option - dinghies for free somewhere.
From: Tim Bradshaw
Subject: Re: Why aren't people complaining that LispWorks is too pricey?
Date:
Message-ID: <ey3r874ib3f.fsf@cley.com>
* Henrik Motakef wrote:
> I don't think this analogy reflects the situation with software very
> well, unless I missed something and you can get perfectly fine,
> "serious" - but maybe not exactly stylish, and perhaps lacking a
> maintenance contract option - dinghies for free somewhere.
Well, I gave away mine, and I'm reasonably sure you could get a
perfectly acceptable boat for a few hundred, or less if you are lucky,
and quite possibly free. It just wouldn't be that competitive at
national/international level.
Similarly if you want software which is competitive with the best, you
pay for it.
--tim
What is quite evident is that there is a sizable variation in the
"priciness" of hobbies.
- Some are cheap. If you noodle around with a guitar, it may
suffice to pay next to nothing for a used one.
- If you want a /nice/ electric guitar, the price for hardware may get
as high as a few thousand dollars. And if you want to get sheet
music, that can get pricey fast...
- Some recreational pastimes require considerable capital costs, and
thus get remarkably costly. Playing golf in Japan is the
pathological example, but in any case, playing golf commonly
involves fairly hefty "greens" fees, while those that ski find that
it is not cheap to get to a ski hill that is being well groomed.
- Those that collect and restore antique cars or that race cars or
boats can, if they must, spend relatively little, or, if they wish,
spend stunning amounts.
Nobody should find it the slightest bit remarkable if those that are
"computer hobbyists" can both be:
a) Stingy and
b) Gluttons for spending.
--
If this was helpful, <http://svcs.affero.net/rm.php?r=cbbrowne> rate me
http://cbbrowne.com/info/sgml.html
--Kill Running Inferiors--
Christopher Browne <········@acm.org> wrote:
> What is quite evident is that there is a sizable variation in the
> "priciness" of hobbies.
> - Some are cheap. If you noodle around with a guitar, it may
> suffice to pay next to nothing for a used one.
> - If you want a /nice/ electric guitar, the price for hardware may get
> as high as a few thousand dollars.
It may, but if you know where to look for used axes, you can easily play
serious music on an instrument that costs a few hundred dollars. That's
true of classical (my specialty), as well as steel string acoustic and
electric. I have a fairly nice electric guitar (G&L series II) that
cost me around $500 new 20+ years ago, and the current used price would
be around $300. Okay, when you throw in the amp and a few effects boxes
the whole thing probably cost around a grand (US), and would sell for
maybe $500 now. But, while I couldn't fill a concert hall with my amp,
I can certainly play (and have when I was in practice on the evil skinny
fingerboard) coffeehouses and other small hall sessions.
So I guess you could say it's a good analogy. It's possible to play
really good music without paying a lot of money. But once you get
there, you will probably be drooling over the more expensive stuff, even
if you decide it's not worth the money for you to get it.
The biggest expense, if you want to get good quickly and not pick up bad
habits, is lessons.
> - Those that collect and restore antique cars or that race cars or
> boats can, if they must, spend relatively little, or, if they wish,
> spend stunning amounts.
Like when my uncle built a house for himself that had a 5-bay garage?
Michael
Quoth Tim Bradshaw <···@cley.com>:
> * Christopher Browne wrote:
>
>> - Some are cheap. If you noodle around with a guitar, it may
>> suffice to pay next to nothing for a used one.
>
> This is way off topic by now, but I think this isn't really true.
> I've come to the conclusion that if you want to not have a guitar
> which will be significantly hard to play (really because of having a
> too-high action and/or crap intonation), you have to spend a
> reasonable amount (not thousands, but a few hundred).
>
> Just my opinion, of course.
I don't think it is off topic; what is clearly true is that there are
big variations in opinion as to what can be suitable for "hobbyist"
work.
Different peoples' "minimum requirements" will differ. I agree that
there will be good reason to want something better than is likely to
be found for $25 at a garage sale; that still does not prevent someone
thrifty from searching garage sales for "hidden gems," the magical
cases where someone can find a cast-off guitar that is better than you
could have possibly expected.
Fountain pen collectors (mentioned elsewhere in the thread, I
believe?) have occasionally found pens worth hundreds of dollars and
paid but pennies for them; finding an unexpectedly good guitar would
not be _totally_ shocking.
Your minimum standards for a "hobbyist's Lisp" may vary from mine, in
much the same fashion.
--
(reverse (concatenate 'string ·············@" "enworbbc"))
http://cbbrowne.com/info/lisp.html
"I'm not sure it is of as much general concern as, say, coke-machines."
-- Marvin Minsky (out of context), on the subject of death.
Tim Bradshaw <···@cley.com> writes:
>> I don't think this analogy reflects the situation with software very
>> well, unless I missed something and you can get perfectly fine,
>> "serious" - but maybe not exactly stylish, and perhaps lacking a
>> maintenance contract option - dinghies for free somewhere.
>
> Well, I gave away mine, and I'm reasonably sure you could get a
> perfectly acceptable boat for a few hundred, or less if you are lucky,
> and quite possibly free. It just wouldn't be that competitive at
> national/international level.
>
> Similarly if you want software which is competitive with the best, you
> pay for it.
If the last sentence means "... you have to pay for it", not "... it's
OK for you to pay whatever it takes", I don't think that's
neccessarily true.
Sure, ACL and Lispworks have some really nice features that, say,
CMUCL and SBCL lack, like an IDE, Allegro's Java interface or the
LispWorks ORB, but not everyone needs that. Then again, there are
areas where the free ones might actually be better. From what I
understand, perfomance would be such a thing.
The free Lisps are definitly not just toys for "dabblers". They were
competetive enough for ViaWeb and Orbitz, after all. Just looking at
the price and licensing conditions, infering that the worse they are,
the better the product must be, might not tell you everything you
should consider.
(Not to mention other kinds of software. Linux, JBoss, Apache, SAP DB
and Mozilla all seem to be quite competetive in one way or the other.)
Regards
Henrik
From: Tim Bradshaw
Subject: Re: Why aren't people complaining that LispWorks is too pricey?
Date:
Message-ID: <ey3llxcz13a.fsf@cley.com>
* Henrik Motakef wrote:
> The free Lisps are definitly not just toys for "dabblers". They were
> competetive enough for ViaWeb and Orbitz, after all. Just looking at
> the price and licensing conditions, infering that the worse they are,
> the better the product must be, might not tell you everything you
> should consider.
Sure, and that's just like lots of things. To switch to photography
as an example - you can buy a perfectly reasonable east european MF
camera, or an old Japanese one for really not very much & similarly
for darkroom equipment, lenses &c (some of the east european lenses
are very good indeed, and the lens is pretty much the only thing in a
a camera that actually *counts* (other than the incompetence of the
photographer, in my case). And you can use these things to make
pictures as good as anyone's. *But* if you want something that will
be really reliable, will not have rattly lenses or other
idiosyncracies, will have a reasonably state of the art metering
system, have lots of accessories, be light, be supported, have spares
availability &c, why then you buy a new Japanese one, for a lot more
money. And almost all professionals and many amateurs will do that.
--tim
From: Edi Weitz
Subject: Re: Why aren't people complaining that LispWorks is too pricey?
Date:
Message-ID: <87vfwgyorc.fsf@bird.agharta.de>
Tim Bradshaw <···@cley.com> writes:
> To switch to photography as an example - you can buy a perfectly
> reasonable east european MF camera, or an old Japanese one for
> really not very much & similarly for darkroom equipment, lenses &c
> (some of the east european lenses are very good indeed, and the lens
> is pretty much the only thing in a a camera that actually *counts*
> (other than the incompetence of the photographer, in my case). And
> you can use these things to make pictures as good as anyone's.
> *But* if you want something that will be really reliable, will not
> have rattly lenses or other idiosyncracies, will have a reasonably
> state of the art metering system, have lots of accessories, be
> light, be supported, have spares availability &c, why then you buy a
> new Japanese one, for a lot more money.
No, you buy a Leica... :)
Edi.
From: Tim Bradshaw
Subject: Re: Why aren't people complaining that LispWorks is too pricey?
Date:
Message-ID: <ey3llxb977q.fsf@cley.com>
* Edi Weitz wrote:
> No, you buy a Leica... :)
I think that's for the people with altogether too much money.
--tim
From: Edi Weitz
Subject: Re: Why aren't people complaining that LispWorks is too pricey?
Date:
Message-ID: <878ytbu50q.fsf@bird.agharta.de>
Tim Bradshaw <···@cley.com> writes:
> * Edi Weitz wrote:
>
> > No, you buy a Leica... :)
>
> I think that's for the people with altogether too much money.
Nah, I buy mine used and that isn't more expensive than a shiny new
top-notch Nikon SLR. But it'll most likely last longer and you have
guaranteed spare-part availability (not that I ever needed one) for
the next 30 years (although they'll happily repair cameras that are
twice as old).
OK, enough OT rant... :)
Edi.
From: Nils Goesche
Subject: Re: Why aren't people complaining that LispWorks is too pricey?
Date:
Message-ID: <87ptmn7s3j.fsf@darkstar.cartan>
Tim Bradshaw <···@cley.com> writes:
> * Duane Rettig wrote:
>
> > I was not, however, talking at all about software hobbyists.
> > Instead, I was talking about hobbies like snow and water skiing,
> > photography, woodworking, collections of various kinds, model train
> > sets, music, etc. All of these are activities that one may take on
> > in varying degrees, with various commitments of money. And unless
> > the activity is being taken on as a career, where the goal is to
> > make enough money for sustenance, a hobby tends to be an end unto
> > itself.
>
> This is worth reinforcing. Maybe, as you say, lots of programmers
> have no hobbies (I suspect this), but I have some - too many in fact -
> and if I programmed as a hobby it would likely not be that
> expensive...
>
> Sailing dinghies (which I don't do any more) - I don't know how much a
> competitive new boat would be of the class I used to race, but I've
> found second-hand but clearly serious boats for 5000-8000 pounds.
> Less serious - older - boats will come down to ~ 1000 I suspect, below
> that and they'd be too wet and soft I think. Cost to race for a
> season might easily be 1000 pounds.
>
> Photography - well, a decent medium format camera is probably
> 1000-2000, say as much again on darkroom stuff. And that's if you're
> not obsessed about having a google lenses. Want to scan things? A
> medium format scanner is between 1800 and 3000.
>
> Music - I don't know how much I've spent on guitars over the years,
> but my lovely ES175 is probably 2-3k, add on amplification &c &c. And
> what about CDs & records - I know it's now fashionable to steal music
> rather than paying for it, but I must have spent a few tens of
> thousands of pounds on records & CDs over the years.
Yeah. I don't know what these people's problem is -- I thought
programmers are supposed to earn quite a lot of money? Even the
numbers you mentioned are a fucking joke -- marry a woman who
looks good in jewelry and haute couture, start collecting DVDs,
develop a taste for haute cuisine, and the cost for a commercial
Lisp system seems like a mere drop in the ocean that doesn't make
any difference whatsoever. Or simply make her next diamond one
carat lighter and feed her a bottle of Veuve Clicquot before
handing it over and she won't notice any difference, anyway, thus
getting you a decent Lisp system for free, basically.
Just joking (really!),
--
Nils G�sche
Ask not for whom the <CONTROL-G> tolls.
PGP key ID #xD26EF2A0
Duane Rettig <·····@franz.com> writes:
> Richard C J Putman <··········@physics.org> writes:
[SNIP]
> > Secondly, Duane justified the cost of ACL to hobbyists on the basis
> > that many hobbies are expensive. I cannot believe many hobbyists buy
> > software costing many times the price of their hardware.
>
> OK, you are at least the second poster who has misunderstood this. I
> admit that since I am talking to a group of programmers, that my
> statement was likely to be misunderstood - that it is likely that
> some of you have no hobbies outside of software. In fact, if I had
> thought about how likely it is that some people here might either
> have a career not in software and are software hobbyists, or who in
> fact are career programmers and take on software also on the side,
> I would have qualified my statement a little more carefully.
>
> I was not, however, talking at all about software hobbyists. Instead,
> I was talking about hobbies like snow and water skiing, photography,
> woodworking, collections of various kinds, model train sets, music, etc.
> All of these are activities that one may take on in varying degrees,
> with various commitments of money. And unless the activity is being
> taken on as a career, where the goal is to make enough money for
> sustenance, a hobby tends to be an end unto itself.
Some of my hobbies require cash investments from time to time. At
least one of the two I can think of off-hand does (alas) have
not-cheap pieces of carefully manufactured steel as "consumables".
As a comparison, yesterday I spent around �150 on:
* one electrical epee (�65, also a "consumable")
* one pair of fencing breeches (�45)
* a pair of long cotton socks
* two body wires (one sabre/foil and one epee).
Also note taht my primary weapon is sabre, so this is at least �65 for
a side-line (I aim to be at least half-arsed in all three modern
fencing weapons).
With another hobby, I recently splurged �200 on a 70-300 mm zoom lens
(hopefully not a consumable) and quite a lot on film and
developing/printing of film.
> When it comes to software, it's a lot harder to describe what a
> hobbyist really is. I personally consider software hobbyists to
> be rare, and most cases, what one might describe as a hobbyist is
> more of a "dabbler" to me. On the activity side, a dabbler would
> be one who knows how to waterski, but who owns neither a ski nor
> a boat, and who gets his skiing in by having a friend who owns a
> boat and ski (i.e. a hobbyist) to take him out skiing.
*shrug* I consider programming as a hobby. Something I do partly
because I think it's fun and partly because there are programs in my
head that *scream* to be implemented. Typical itch-scratching I
think. I have worked as a programmer and developer as my day job and I
really don't want to do it full-time, it requires discipline I don't
have.
See it as the programming equivalent of an amateur author. If people
like what I do, it's a good ego-boost. If people would give me money
for something I wrote for fun, fine. If people make money from
something I made for fun, fine (but I wouldn't mind having a small
share of it).
//Ingvar
--
Self-referencing
Five, seven, five syllables
This haiku contains
The great thing about LispWorks Professional is that you don't have
to pay any runtime royalties (unlike ACL).
I have a (unfortunately low sales volume) product written
in Common Lisp and I appreciate the fact that I can use LispWorks
to make relatively small, compact, and robust native executables
and not have to worry about separate royalties. BTW, I would use
LispWorks for many more projects if it had great web services
support (SOAP, WSDL, perhaps UDDI).
-Mark
Thaddeus L Olczyk <······@interaccess.com> wrote in message news:<··································@4ax.com>...
> It seems that people are often coming and complaining that ACL is too
> pricey. So why aren't people complaining about LispWorks?
> ( Partial answer: because they want ACL more than they want LispWorks.
> Why? )
> --------------------------------------------------
> Thaddeus L. Olczyk, PhD
> Think twice, code once.
What is the environment you would want to use Lispworks in for such
projects? CL-HTTP, Allegroserve, Apache/mod_lisp (not an exhaustive
list)? What would you use for database access? What about the database
itself? What other components would be involved and what specifics can
you mention for them?
It would be very interesting if some people wanted to configure similar
systems and see what holes are left, then maybe work towards filling them.
Do you have a web reference for the Lispworks product you market?
Mark Watson wrote:
...
>
BTW, I would use
> LispWorks for many more projects if it had great web services
> support (SOAP, WSDL, perhaps UDDI).