Hello,
I am writing in the line of "Is there a futyre for LISP". As a
biologist, I came onto LISP through the imperative pathway (C, Ada,
Eiffel...), and all things consider LISP syntax is quite logical,
simple and consistant (what is the big deal between 1+1; (+ 1 1) or
(add 1 1)?). I found that once you get use to parsing LISP it is
actually easier than going through C code for example. As a biologist,
I would have prefered learned about LISP, Prolog and SQL rather than
Pascal or C. At least the end result would have been practical!
However, (pertinent ) comments from gurus like Paul Graham are
worrisome, and bring up the problem of the old guard (the people that
grew up with LISP, improved it and made it what it is today) versus the
(smaller?) "new guard" that has now access to free LISP system and can
explore its power. But, what would happen as the old guard retires and
less and less people would use LISP? In othe words, is there enough
people (with extensive computing knowledge) using LISP to keep it at
its current level?
As a pratical example, we can take the development of Object Oriented
Databases, written in java in many instances. This is, in my opinion, a
perfect case where LISP use would be more suited. This is also a reason
to review LISP standard. Why is there no reference to distributive and
parallel processing in LISP standard (this is left to the implemetors
but breaks the standard and portability of the code. Moreover, as one
of the language of AI, I would thing that at least parallel processing
should be important for AI - our brain does that all the time!?!).
So to conclude, would it be a good time to gather with the smart people
that made LISP what it is today, check on its user base (is it stable,
growing... What is the nature of LISP users - professional,
dilettante...) and improve on LISP (as Paul Graham said about Common
LISP, compared to Scheme, its code is uglier but its is for real life
application. Which means that there is room for improvement). It could
even be a reason to create a specific discussion group?!?
Hope I did not bore you to death, just concern that a beautiful
language goes to waste because C# and jave are "coooool".
Alex Meyer.
Alex wrote:
> Why is there no reference to distributive and
> parallel processing in LISP standard (this is left to the implemetors
> but breaks the standard and portability of the code. Moreover, as one
> of the language of AI, I would thing that at least parallel processing
> should be important for AI - our brain does that all the time!?!).
If you're curious about the historical reasons, "The Evolution of Lisp"
is a good source. Lisp once ran on parallel machines which shipped with
~64,000 processors. As well as constrained machines. So this is not in
the Common Lisp spec, which was more of a lowest common denominator.
(Nowadays, we often hear that we must return to these relatively
advanced parallel architectures, due to Moore's Law problems.)
http://www.dreamsongs.com/Essays.html
The spec push had special political forces which led to
standardization. I vaguely remember reading that the spec was a MacLisp
response to Interlisp threatening to walk away with all the Darpa
funding. Comments from people in the usenet archives make it seem like
the process was antagonistic, and biased in favor of politically
powerful groups like Symbolics. Therefore, I'd wonder if the solution
to many of the popularity problems might lie elsewhere than "official"
standardization.
For example, a cl-gardeners project could be built to have a central
website around something like LispBox, with people who smoketest
package installations, maybe have nice little matrices showing what
platforms something currently installs successfully on. In addition,
the controller of the lisp.org domain recently offered it to anyone
wanting to create a serious lisp entrypoint.
> As a pratical example, we can take the development of Object Oriented
> Databases, written in java in many instances. This is, in my opinion, a
> perfect case where LISP use would be more suited. This is also a reason
> to review LISP standard.
If you mean persistent OO databases, there's Elephant, the upcoming
Rucksack, etc.
Tayssir
I think the biggest barriers to Lisp adoption is probably the steep
learning curve and just the fact that setting things up take enormous
patience for a newbie. I have spent the last 2 months learning just the
basics of setting up the environment, learning emacs, working through 2
Lisp books, etc. Only now am I up to the point of starting to install
third party libraries to explore stuff on my own. With ruby on rails I
got there on the third day. 2 months vs 2 days is a big difference in
learning curve which means a lot less people will stick around long
enough to become productive with the language. That having been said,
I'm still not fully productive with lisp because I'm not familiar with
the libraries and I still have to look a lot of things up.
I should also mention that setting Lisp up on a windows machine is
harder than on an UNIX machine and like it or not most people still use
Windows. I bought a Mac mini just to run Lisp without having to put up
with all the problems created by windows.
Most of my friends are engineering majors, a supposedly technically
inclined group. Even among this group, I had a hard time convincing them
to use Latex to type up reports and papers. (and Latex is a lot easier
to learn and setup than Lisp) Their argument was that Latex is too much
trouble to learn and setup (For windows, Latex is just 2 downloads and
some tutorials away from productivity). Compare this to lisp, where you
have to setup emacs, learn emacs, setup some lisp, setup slime to hook
it up, then learn lisp, and learn a bunch of tips and tricks on
configuration files and customizations and 3rd party library
installations before you become productive.
So for lisp to become popular setting it up would have to be as easy as
making instant noodles. Lispbox is a move in the right direction but I
couldn't figure out how to edit the lisp and emacs configuration files
to customize things so I ended up doing a complete installation by hand.
And it doesn't help that there are those in the Lisp community who think
that this steep learning curve is useful for weeding out those who are
not deserving of Lisp.
-Jeff
Alex wrote:
> Hello,
>
> I am writing in the line of "Is there a futyre for LISP". As a
> biologist, I came onto LISP through the imperative pathway (C, Ada,
> Eiffel...), and all things consider LISP syntax is quite logical,
> simple and consistant (what is the big deal between 1+1; (+ 1 1) or
> (add 1 1)?). I found that once you get use to parsing LISP it is
> actually easier than going through C code for example. As a biologist,
> I would have prefered learned about LISP, Prolog and SQL rather than
> Pascal or C. At least the end result would have been practical!
> However, (pertinent ) comments from gurus like Paul Graham are
> worrisome, and bring up the problem of the old guard (the people that
> grew up with LISP, improved it and made it what it is today) versus the
> (smaller?) "new guard" that has now access to free LISP system and can
> explore its power. But, what would happen as the old guard retires and
> less and less people would use LISP? In othe words, is there enough
> people (with extensive computing knowledge) using LISP to keep it at
> its current level?
> As a pratical example, we can take the development of Object Oriented
> Databases, written in java in many instances. This is, in my opinion, a
> perfect case where LISP use would be more suited. This is also a reason
> to review LISP standard. Why is there no reference to distributive and
> parallel processing in LISP standard (this is left to the implemetors
> but breaks the standard and portability of the code. Moreover, as one
> of the language of AI, I would thing that at least parallel processing
> should be important for AI - our brain does that all the time!?!).
> So to conclude, would it be a good time to gather with the smart people
> that made LISP what it is today, check on its user base (is it stable,
> growing... What is the nature of LISP users - professional,
> dilettante...) and improve on LISP (as Paul Graham said about Common
> LISP, compared to Scheme, its code is uglier but its is for real life
> application. Which means that there is room for improvement). It could
> even be a reason to create a specific discussion group?!?
> Hope I did not bore you to death, just concern that a beautiful
> language goes to waste because C# and jave are "coooool".
>
> Alex Meyer.
>
Amen, Jeff. It us more than a month since I started to try to learn
lisp and I haven't even figured out how to set up the environment yet,
much less how to actually work in the language! I'm very close to
giving up and learning Ruby.
From: Ken Tilton
Subject: Re: LISP and Object Oriented Databases
Date:
Message-ID: <QxU1g.32$Mx6.3@fe09.lga>
Jacob Lyles wrote:
> Amen, Jeff. It us more than a month since I started to try to learn
> lisp and I haven't even figured out how to set up the environment yet,
> much less how to actually work in the language! I'm very close to
> giving up and learning Ruby.
>
When the installer window appears, just click on all the defaults. If
you have trouble with that, go to the Mouse control panel before running
the installer and set it to automatically position the mouse cursor over
the default button. A cat could help you with the button. I admit there
is one tricky bit. You have to move the mouse manually to the radio
button that agrees to the license conditions, and then manually bring it
back to the default button. You should have pen and paper ready to make
a note of which is the default button before moving the mouse away from
it...
oh, god. Lisp /is/ hard! Let's go shopping!!
:)
ken
--
Cells: http://common-lisp.net/project/cells/
"Have you ever been in a relationship?"
Attorney for Mary Winkler, confessed killer of her
minister husband, when asked if the couple had
marital problems.
Ah yes, mock the newbies. Great way to get more people using Lisp.
Jacob, I started with Lispbox from http://www.gigamonkeys.com/book/lispbox/
Just download and run, no install, setup, configuration or anything.
That got me through 2 Lisp books and the basics of the language. I only
tried to setup my own environment after hitting the limitations of the
Lispbox setup, that's when the real pain started.
Ruby is actually pretty close to Lisp in functionality for most
purposes. The only reason why I'm sticking with Lisp is because Ruby is
a relatively new language and most of the AI books I read have sample
code in Lisp (or Scheme). So if you're not doing AI work, Ruby is
probably just as good.
Ken Tilton wrote:
>
>
> Jacob Lyles wrote:
>
>> Amen, Jeff. It us more than a month since I started to try to learn
>> lisp and I haven't even figured out how to set up the environment yet,
>> much less how to actually work in the language! I'm very close to
>> giving up and learning Ruby.
>>
>
> When the installer window appears, just click on all the defaults. If
> you have trouble with that, go to the Mouse control panel before running
> the installer and set it to automatically position the mouse cursor over
> the default button. A cat could help you with the button. I admit there
> is one tricky bit. You have to move the mouse manually to the radio
> button that agrees to the license conditions, and then manually bring it
> back to the default button. You should have pen and paper ready to make
> a note of which is the default button before moving the mouse away from
> it...
>
> oh, god. Lisp /is/ hard! Let's go shopping!!
>
> :)
>
> ken
>
From: Bill Atkins
Subject: Re: LISP and Object Oriented Databases
Date:
Message-ID: <87fyk79zyu.fsf@rpi.edu>
Jeffery Zhang <····@cornell.edu> writes:
> Ruby is actually pretty close to Lisp in functionality for most
> purposes.
> So if you're not doing AI work, Ruby is probably just as good.
I disagree with both of these statements. I'll assume you're not
talking about this from a Turing perspective. Ruby lacks the
following important (to me) bits of functionality:
- macros (of course)
- CLOS (e.g. before/after/around methods)
- support for interactive development (this is key)
- s-expressions for code, lack of syntax
- multiple values, dynamic scoping
- support for low-level optimizations
- a compiler, often a native compiler (Ruby code is interpreted right
off the parse tree, I believe)
- the package system
- read macros
- more
I'm not sure of precisely what you mean by functionality, but for me
these features set Lisp and Ruby (and every other language) a couple
of light-years apart (I've never done any AI work outside of classes).
--
"...and when, another time, I discovered that he considered not
unworthy of reflection in one of those mirrors of absolute truth which
were his writings a remark similar to one which I had had occasion to
make about our friend M. Legrandin, ...then it was suddenly revealed
to me that my own humble existence and the realms of the true were
less widely separated than I had supposed, that at certain points they
actually collided, and in my newfound confidence and joy, I had wept
upon his printed page as in the arms of a long-lost father."
Jeffery Zhang wrote:
> Ah yes, mock the newbies. Great way to get more people using Lisp.
You can whine, O Great Friend of Noobs, or you can explain to the world
how in hell you were not able to install either AllegroCl or Lispworks
on win32 to get started with Lisp. That really is a much more
interesting question., which you seem to be avoiding.
I think you just want to spread FUD so you can keep Lisp to your self.
Understandable. :)
kt
In article <··············@fe09.lga>, Ken Tilton <·········@gmail.com>
wrote:
> Jacob Lyles wrote:
> > Amen, Jeff. It us more than a month since I started to try to learn
> > lisp and I haven't even figured out how to set up the environment yet,
> > much less how to actually work in the language! I'm very close to
> > giving up and learning Ruby.
> >
>
> When the installer window appears, just click on all the defaults. If
> you have trouble with that, go to the Mouse control panel before running
> the installer and set it to automatically position the mouse cursor over
> the default button. A cat could help you with the button. I admit there
> is one tricky bit. You have to move the mouse manually to the radio
> button that agrees to the license conditions, and then manually bring it
> back to the default button. You should have pen and paper ready to make
> a note of which is the default button before moving the mouse away from
> it...
>
> oh, god. Lisp /is/ hard! Let's go shopping!!
>
> :)
>
> ken
he, that was funny! :-)
--
http://lispm.dyndns.org/
Ken Tilton wrote:
>
>
> Jacob Lyles wrote:
>
>> Amen, Jeff. It us more than a month since I started to try to learn
>> lisp and I haven't even figured out how to set up the environment yet,
>> much less how to actually work in the language! I'm very close to
>> giving up and learning Ruby.
>>
>
> When the installer window appears, just click on all the defaults. If
> you have trouble with that, go to the Mouse control panel before running
> the installer and set it to automatically position the mouse cursor over
> the default button. A cat could help you with the button. I admit there
> is one tricky bit. You have to move the mouse manually to the radio
> button that agrees to the license conditions, and then manually bring it
> back to the default button. You should have pen and paper ready to make
> a note of which is the default button before moving the mouse away from
> it...
>
> oh, god. Lisp /is/ hard! Let's go shopping!!
Good lord, but the smug asshole^H^H^H^H^H^H^HLispnik quotient is going
up in here.
From: Ken Tilton
Subject: Re: LISP and Object Oriented Databases
Date:
Message-ID: <Xke2g.548$LG5.81@fe10.lga>
Larry Elmore wrote:
> Ken Tilton wrote:
>
>>
>>Jacob Lyles wrote:
>>
>>
>>>Amen, Jeff. It us more than a month since I started to try to learn
>>>lisp and I haven't even figured out how to set up the environment yet,
>>>much less how to actually work in the language! I'm very close to
>>>giving up and learning Ruby.
>>>
>>
>>When the installer window appears, just click on all the defaults. If
>>you have trouble with that, go to the Mouse control panel before running
>>the installer and set it to automatically position the mouse cursor over
>>the default button. A cat could help you with the button. I admit there
>>is one tricky bit. You have to move the mouse manually to the radio
>>button that agrees to the license conditions, and then manually bring it
>>back to the default button. You should have pen and paper ready to make
>>a note of which is the default button before moving the mouse away from
>>it...
>>
>>oh, god. Lisp /is/ hard! Let's go shopping!!
>
>
> Good lord, but the smug asshole^H^H^H^H^H^H^HLispnik quotient is going
> up in here.
It gets worse. And we prefer "smug lisp weenies" or "savages of
comp.lang.lisp".
kennj
--
Cells: http://common-lisp.net/project/cells/
"Have you ever been in a relationship?"
Attorney for Mary Winkler, confessed killer of her
minister husband, when asked if the couple had
marital problems.
Jeffery Zhang wrote:
> I think the biggest barriers to Lisp adoption is probably the steep
> learning curve and just the fact that setting things up take enormous
> patience for a newbie.
Wrong. Well, OK, ASDF is a pain, but you make it sound like just
installing and using a Lisp is a pain. Wrong.
> I have spent the last 2 months learning just the
> basics of setting up the environment, learning emacs, working through 2
> Lisp books, etc. Only now am I up to the point of starting to install
> third party libraries to explore stuff on my own. With ruby on rails...
OK. That is one library that is tres chique just now, because the
application is tres au courrant. Lisp has been kicking ass for almost
fifty years, and apparently no Lispnik is chasing fads (tho Lisp Web
application stacks seem to be sprouting everywhere).
> I
> got there on the third day. 2 months vs 2 days is a big difference in
> learning curve which means a lot less people will stick around long
> enough to become productive with the language. That having been said,
> I'm still not fully productive with lisp because I'm not familiar with
> the libraries and I still have to look a lot of things up.
That will change after ten years of heads-down use. :) What you are
complaining about is that CL offers so much. Out of the box, as people
like to say.
>
> I should also mention that setting Lisp up on a windows machine is
> harder than on an UNIX machine and like it or not most people still use
> Windows. I bought a Mac mini just to run Lisp without having to put up
> with all the problems created by windows.
Are you quite simply out of your mind? Not one but two free commercial
Lisps available for newbies, complete with IDEs, GUIs, sockets, threads,
FFIs? And InstallWizard installers? Are you out of your mind?
:)
I am more sympathetic on the installation of libraries, tho. Never fun.
But I think we need to differentiate between noobs wanting to learn Lisp
and noobs trying to build a stock exchange system in their first week.
Sounds like on day one you want Lisp, on day two you want a Web
application up, on day three... and you found one heavily-used system
RoR that gave you that and now you want everything on a silver platter.
ken
--
Cells: http://common-lisp.net/project/cells/
"Have you ever been in a relationship?"
Attorney for Mary Winkler, confessed killer of her
minister husband, when asked if the couple had
marital problems.
Where do you get off saying "WRONG" to someone describing their pain?
Did you stay home the day the talked about the importance of listening
to customer experience? It certainly isn't just him. I walked this
path myself on two platforms and I know from my own experience that
much he says is quite true. Of course I knew Emacs going in so that
helped. But ASDF had problems itself on some OS / Lisp combinations.
Many of the ASDF packages worked fine in SBCL but not so with OpenMCL
or even Lispworks.
Installing and using Lisp is a pain if the tools to build up your
working environment and take advantage of others work are a pain.
Lisp web stacks are sprouting everywhere? Yes, but how many are
actually mature? Do any have the coverage and useability of Ruby on
Rails? No. I am no fan of RoR as I see how much it glosses over and
the persistence stuff is a joke. But still.
Lisp does not "offer so much out of the box". It offers a lot after
you climb in the box long enough and wallow around until thoroughly
coated in lispiness and show the patience of a saint.
Are you out of your mind? Threads are a pain point on many free lisps.
IDEs don't exist for most or even for very expensive commercial lisps
on many platforms.
You are a prime example of what is wrong in this community that drives
people, especially newbies away. I hope you enjoy thinking you are all
uber because you are on the inside of your little box.
Jeffery Zhang wrote:
> I think the biggest barriers to Lisp adoption is probably the steep
> learning curve and just the fact that setting things up take enormous
> patience for a newbie. I have spent the last 2 months learning just the
> basics of setting up the environment, learning emacs, working through 2
> Lisp books, etc. Only now am I up to the point of starting to install
> third party libraries to explore stuff on my own. With ruby on rails I
> got there on the third day. 2 months vs 2 days is a big difference in
> learning curve which means a lot less people will stick around long
> enough to become productive with the language. That having been said,
> I'm still not fully productive with lisp because I'm not familiar with
> the libraries and I still have to look a lot of things up.
>
> I should also mention that setting Lisp up on a windows machine is
> harder than on an UNIX machine and like it or not most people still use
> Windows. I bought a Mac mini just to run Lisp without having to put up
> with all the problems created by windows.
>
> Most of my friends are engineering majors, a supposedly technically
> inclined group. Even among this group, I had a hard time convincing them
> to use Latex to type up reports and papers. (and Latex is a lot easier
> to learn and setup than Lisp) Their argument was that Latex is too much
> trouble to learn and setup (For windows, Latex is just 2 downloads and
> some tutorials away from productivity). Compare this to lisp, where you
> have to setup emacs, learn emacs, setup some lisp, setup slime to hook
> it up, then learn lisp, and learn a bunch of tips and tricks on
> configuration files and customizations and 3rd party library
> installations before you become productive.
>
> So for lisp to become popular setting it up would have to be as easy as
> making instant noodles. Lispbox is a move in the right direction but I
> couldn't figure out how to edit the lisp and emacs configuration files
> to customize things so I ended up doing a complete installation by hand.
You can sidestep all these issues by using one of the free editions of
the commercial Common Lisp implementations.
Pascal
--
3rd European Lisp Workshop
July 3-4 - Nantes, France - co-located with ECOOP 2006
http://lisp-ecoop06.bknr.net/
Hi, Jeff:
I have been learning lisp since a year ago. I still don't know how to
setup some libs, but Clisp is worth learning and is easy to learn with
so many good references.
It is an incemental process, just like that lisp is incemental.
I didn't even know emacs before I started clisp, and now I can do my
some routines in clisp rather than in c/c++.
P.B. has a good plan:
http://groups.google.de/group/comp.lang.lisp/browse_thread/thread/ce484c85dc5379cc/13fe9a5d1bbf0a18?q=Pascal+Bourguignon&rnum=176#13fe9a5d1bbf0a18
keyboard
Jeffery Zhang wrote:
> I think the biggest barriers to Lisp adoption is probably the steep
> learning curve and just the fact that setting things up take enormous
> patience for a newbie. I have spent the last 2 months learning just the
> basics of setting up the environment, learning emacs, working through 2
> Lisp books, etc. Only now am I up to the point of starting to install
> third party libraries to explore stuff on my own.
Just curious: If you wanted to learn Common Lisp why didn't you start
with e.g. the Allegro Trial Edition?
--
Jens Axel S�gaard
<http://www.scheme.dk/blog/>
In article <············@ruby.cit.cornell.edu>,
Jeffery Zhang <····@cornell.edu> wrote:
> I think the biggest barriers to Lisp adoption is probably the steep
> learning curve and just the fact that setting things up take enormous
> patience for a newbie. I have spent the last 2 months learning just the
> basics of setting up the environment, learning emacs, working through 2
> Lisp books, etc. Only now am I up to the point of starting to install
> third party libraries to explore stuff on my own. With ruby on rails I
> got there on the third day. 2 months vs 2 days is a big difference in
> learning curve which means a lot less people will stick around long
> enough to become productive with the language. That having been said,
> I'm still not fully productive with lisp because I'm not familiar with
> the libraries and I still have to look a lot of things up.
>
> I should also mention that setting Lisp up on a windows machine is
> harder than on an UNIX machine and like it or not most people still use
> Windows. I bought a Mac mini just to run Lisp without having to put up
> with all the problems created by windows.
>
> Most of my friends are engineering majors, a supposedly technically
> inclined group. Even among this group, I had a hard time convincing them
> to use Latex to type up reports and papers. (and Latex is a lot easier
> to learn and setup than Lisp) Their argument was that Latex is too much
> trouble to learn and setup (For windows, Latex is just 2 downloads and
> some tutorials away from productivity). Compare this to lisp, where you
> have to setup emacs, learn emacs, setup some lisp, setup slime to hook
> it up, then learn lisp, and learn a bunch of tips and tricks on
> configuration files and customizations and 3rd party library
> installations before you become productive.
>
> So for lisp to become popular setting it up would have to be as easy as
> making instant noodles. Lispbox is a move in the right direction but I
> couldn't figure out how to edit the lisp and emacs configuration files
> to customize things so I ended up doing a complete installation by hand.
>
> And it doesn't help that there are those in the Lisp community who think
> that this steep learning curve is useful for weeding out those who are
> not deserving of Lisp.
Hmm, I've put a CD of LispWorks into my Mac's disk drive.
Than I double clicked the installer, a few okays
and then I had an installion of LispWorks in the
programs folder. Another double click and
it running. With editor, listener, compiler, tons of
other tools, sockets, database interfaces, ...
--
http://lispm.dyndns.org/
If you got working database interfaces from LW then you paid $3000 for
the privilege. Folks seeking to see what the excitement is about for
themselves aren't about to do that. Nor are they going to pay $1000
for the professional edition. They will likely shy away from the feer
edition of LW or Franz if they are serious programmers put off by the
constraints on these. So they will likely hit SBCL, etc. first. If
we want the community to grow then the free lisps must reasonably
straight forward to get up and competent on. And before anyone makes
any pathetic shopping cracks, I am a hacker who earned my spurs before
a lot of you got past Basic. I have been all over the software
landscape for 26 years now. Yet I found the Lisp path much more
painful than it should be. The pain is real. Ignore it at your peril
and the peril of Lisp.
From: Bill Atkins
Subject: Re: LISP and Object Oriented Databases
Date:
Message-ID: <87acac6q71.fsf@rpi.edu>
"samantha" <········@gmail.com> writes:
> If you got working database interfaces from LW then you paid $3000 for
> the privilege. Folks seeking to see what the excitement is about for
> themselves aren't about to do that. Nor are they going to pay $1000
> for the professional edition. They will likely shy away from the feer
> edition of LW or Franz if they are serious programmers put off by the
> constraints on these. So they will likely hit SBCL, etc. first. If
> we want the community to grow then the free lisps must reasonably
> straight forward to get up and competent on. And before anyone makes
> any pathetic shopping cracks, I am a hacker who earned my spurs before
> a lot of you got past Basic. I have been all over the software
> landscape for 26 years now. Yet I found the Lisp path much more
> painful than it should be. The pain is real. Ignore it at your peril
> and the peril of Lisp.
>
CLSQL?
If you really want to see The Future of Lisp (and even in a
Biocomputing context!), try BioBike (www.biobike.org) -- biocomputing
in Lisp direct through your browser, no muss no fuss, nothing to
download, fully persistent threads, data, sessions. (Well, mostly
persistent, until the server dies.) Be sure to check out the demo
server and the Live Tutorials.
> I am writing in the line of "Is there a futyre for LISP". As a
> biologist, I came onto LISP through the imperative pathway (C, Ada,
> Eiffel...), and all things consider LISP syntax is quite logical,
> simple and consistant (what is the big deal between 1+1; (+ 1 1) or
> (add 1 1)?). I found that once you get use to parsing LISP it is
> actually easier than going through C code for example. As a biologist,
> I would have prefered learned about LISP, Prolog and SQL rather than
> Pascal or C. At least the end result would have been practical!
> However, (pertinent ) comments from gurus like Paul Graham are
> worrisome, and bring up the problem of the old guard (the people that
> grew up with LISP, improved it and made it what it is today) versus the
> (smaller?) "new guard" that has now access to free LISP system and can
> explore its power. But, what would happen as the old guard retires and
> less and less people would use LISP? In othe words, is there enough
> people (with extensive computing knowledge) using LISP to keep it at
> its current level?
> As a pratical example, we can take the development of Object Oriented
> Databases, written in java in many instances. This is, in my opinion, a
> perfect case where LISP use would be more suited. This is also a reason
> to review LISP standard. Why is there no reference to distributive and
> parallel processing in LISP standard (this is left to the implemetors
> but breaks the standard and portability of the code. Moreover, as one
> of the language of AI, I would thing that at least parallel processing
> should be important for AI - our brain does that all the time!?!).
> So to conclude, would it be a good time to gather with the smart people
> that made LISP what it is today, check on its user base (is it stable,
> growing... What is the nature of LISP users - professional,
> dilettante...) and improve on LISP (as Paul Graham said about Common
> LISP, compared to Scheme, its code is uglier but its is for real life
> application. Which means that there is room for improvement). It could
> even be a reason to create a specific discussion group?!?
> Hope I did not bore you to death, just concern that a beautiful
> language goes to waste because C# and jave are "coooool".
>
> Alex Meyer.
@Ken:
Please understand that your comments come across as hostile. It's
almost like you're saying, "You're just plain stupid." This is the way
it seems to me and I think this is the way it seems to newbies.
Instead of:
>Are you quite simply out of your mind? Not one but two free commercial
>Lisps available for newbies, complete with IDEs, GUIs, sockets, threads,
>FFIs? And InstallWizard installers? Are you out of your mind?
Try:
>There are two free commercial Lisps available for newbies, complete with
>IDEs, GUIs, sockets, threads, and FFI's. They are:_____ and ____.
In most cultures, to say that someone is out of their mind means that
that person is crazy, or has gone mad. In most cultures, madness is
denigrated and considered a "lesser state". Thus, asking if the person
is out of his mind can come across as asking if he is crazy and in a
lesser state, which can be considered insulting.
I suppose you've met too many idiot noobs who've rejected your advice
even if they were framed that way, but in my personal opinion it's no
reason to act as if every random newbie is a complete idiot. They
might simply be misinformed. Please, try to understand: not everyone
else is an idiot. Some are smart, they're just ignorant of details
they've never encountered before. We're here to teach them if they're
willing to learn. We're not here to rob them of their willingness to
learn.
From: Ken Tilton
Subject: Re: LISP and Object Oriented Databases
Date:
Message-ID: <tB72g.9$LG5.0@fe10.lga>
Alan Manuel K. Gloria wrote:
> @Ken:
> Please understand that your comments come across as hostile. It's
> almost like you're saying, "You're just plain stupid." This is the way
> it seems to me and I think this is the way it seems to newbies.
> Instead of:
>
>>Are you quite simply out of your mind? Not one but two free commercial
>>Lisps available for newbies, complete with IDEs, GUIs, sockets, threads,
>>FFIs? And InstallWizard installers? Are you out of your mind?
>
> Try:
>
>>There are two free commercial Lisps available for newbies, complete with
>>IDEs, GUIs, sockets, threads, and FFI's. They are:_____ and ____.
>
>
> In most cultures, to say that someone is out of their mind means that
> that person is crazy, or has gone mad. In most cultures, madness is
> denigrated and considered a "lesser state". Thus, asking if the person
> is out of his mind can come across as asking if he is crazy and in a
> lesser state, which can be considered insulting.
>
> I suppose you've met too many idiot noobs who've rejected your advice
> even if they were framed that way, but in my personal opinion it's no
> reason to act as if every random newbie is a complete idiot. They
> might simply be misinformed. Please, try to understand: not everyone
> else is an idiot. Some are smart, they're just ignorant of details
> they've never encountered before. We're here to teach them if they're
> willing to learn. We're not here to rob them of their willingness to
> learn.
>
<sob> What a beautiful speech! </sob>
I notice no one has yet explained how getting a Lisp running on Windows
is "even harder". I guess that is why the only thing you dweebs want to
do is prattle on about your hurt feelings. All I can say, it's a good
thing you do not hang out on Usenet, because over there it is even...
doh! This /is/ Usenet!
OK, look, I will make it easy for you. On all your behalf, I will
apologize for saying it is hard to get Lisp running on Windows. Our
mistake was in trying to use Emacs and Slime when there were perfectly
good truly integrated Lisp development environments to be had on a trial
basis, including the Cadillac (Lexus?) of environments, AllegroCL.
See, that was not so bad, was it? It's a lot easier than getting me to
be nice.
:)
ken
--
Cells: http://common-lisp.net/project/cells/
"Have you ever been in a relationship?"
Attorney for Mary Winkler, confessed killer of her
minister husband, when asked if the couple had
marital problems.
>I notice no one has yet explained how getting a Lisp running on Windows
>is "even harder".
Getting asdf and asdf-install running on CLISP and WinXP is very hard,
for one - you need Cygwin to get asdf-install running. If you decide to
just use asdf (without asdf-install), you have to manually install the
dependencies, and you have to constantly update your
asdf:*central-registry* variable each time you install a new package.
Granted, the two Windows installations you didn't mention (which I
assume are LispWorks and Allegro CL trial versions) may be installable
in two mouse moves and a series of clicks, but the problem is: "TRIAL
VERSIONS". They have some limitations; LispWorks will run for five
hours, then force a quit, for example. They have heap limitations.
And lastly, they say "TRIAL" somewhere.
Noobies want to try stuff out, true, but they want to stick around with
it also. If they feel they can't afford the full version - $99 for
Allegro CL (for students at that), $660 per license for Lispworks -
they won't even try it out, because if they like it, they'll just be
pissed that it's so good but they can't afford it.
Noobs are noobs, but the best noobs want to stop being noobs someday.
They'll try their best to stop being noobs.
> I guess that is why the only thing you dweebs want to
>do is prattle on about your hurt feelings. All I can say, it's a good
>thing you do not hang out on Usenet, because over there it is even...
>doh! This /is/ Usenet!
And this is reason enough to be rude?
>OK, look, I will make it easy for you. On all your behalf, I will
>apologize for saying it is hard to get Lisp running on Windows.
When did you say this? I thought your point was how easy it was?
> Our
>mistake was in trying to use Emacs and Slime when there were perfectly
>good truly integrated Lisp development environments to be had on a trial
>basis, including the Cadillac (Lexus?) of environments, AllegroCL.
Why is this a mistake? Why does it seem that a majority of expert CL
users seem geared to be using Emacs and Slime, when, as you point out,
there are truly integrated dev environments to be had on a trial basis?
Look, ken: We would like to reach your level of competency in CL. It
*appears* that most people work in CL using Emacs and Slime, because it
seems that Emacs and Slime is discussed here more often than AllegroCL.
Do you use Emacs and Slime? If so, why? If not, why not? If you,
someone who is competent enough to create Cells, are using or not using
Emacs/Slime, we newbies would be curious. Could your use/non-use Emacs
have something to do with your competency?
>See, that was not so bad, was it? It's a lot easier than getting me to
>be nice.
I'm pretty sure that being nice would be a lot easier for you. For one
thing, you could type less. You don't need to type "Are you out of
your mind?" any more, for example.
From: Bill Atkins
Subject: Re: LISP and Object Oriented Databases
Date:
Message-ID: <878xpy964z.fsf@rpi.edu>
"Alan Manuel K. Gloria" <········@gmail.com> writes:
>>I notice no one has yet explained how getting a Lisp running on Windows
>>is "even harder".
> Getting asdf and asdf-install running on CLISP and WinXP is very hard,
> for one - you need Cygwin to get asdf-install running. If you decide to
> just use asdf (without asdf-install), you have to manually install the
> dependencies, and you have to constantly update your
> asdf:*central-registry* variable each time you install a new package.
>
> Granted, the two Windows installations you didn't mention (which I
> assume are LispWorks and Allegro CL trial versions) may be installable
> in two mouse moves and a series of clicks, but the problem is: "TRIAL
> VERSIONS". They have some limitations; LispWorks will run for five
> hours, then force a quit, for example. They have heap limitations.
> And lastly, they say "TRIAL" somewhere.
>
> Noobies want to try stuff out, true, but they want to stick around with
> it also. If they feel they can't afford the full version - $99 for
> Allegro CL (for students at that), $660 per license for Lispworks -
> they won't even try it out, because if they like it, they'll just be
> pissed that it's so good but they can't afford it.
>
> Noobs are noobs, but the best noobs want to stop being noobs someday.
> They'll try their best to stop being noobs.
>
>> I guess that is why the only thing you dweebs want to
>>do is prattle on about your hurt feelings. All I can say, it's a good
>>thing you do not hang out on Usenet, because over there it is even...
>>doh! This /is/ Usenet!
> And this is reason enough to be rude?
>
>>OK, look, I will make it easy for you. On all your behalf, I will
>>apologize for saying it is hard to get Lisp running on Windows.
> When did you say this? I thought your point was how easy it was?
>
>> Our
>>mistake was in trying to use Emacs and Slime when there were perfectly
>>good truly integrated Lisp development environments to be had on a trial
>>basis, including the Cadillac (Lexus?) of environments, AllegroCL.
> Why is this a mistake? Why does it seem that a majority of expert CL
> users seem geared to be using Emacs and Slime, when, as you point out,
> there are truly integrated dev environments to be had on a trial basis?
>
> Look, ken: We would like to reach your level of competency in CL. It
> *appears* that most people work in CL using Emacs and Slime, because it
> seems that Emacs and Slime is discussed here more often than AllegroCL.
> Do you use Emacs and Slime? If so, why? If not, why not? If you,
> someone who is competent enough to create Cells, are using or not using
> Emacs/Slime, we newbies would be curious. Could your use/non-use Emacs
> have something to do with your competency?
>
>>See, that was not so bad, was it? It's a lot easier than getting me to
>>be nice.
> I'm pretty sure that being nice would be a lot easier for you. For one
> thing, you could type less. You don't need to type "Are you out of
> your mind?" any more, for example.
>
Deep, cleansing breaths, man.
Alan Manuel K. Gloria wrote:
> Granted, the two Windows installations you didn't mention (which I
> assume are LispWorks and Allegro CL trial versions) may be installable
> in two mouse moves and a series of clicks, but the problem is: "TRIAL
> VERSIONS". They have some limitations; LispWorks will run for five
> hours, then force a quit, for example. They have heap limitations.
> And lastly, they say "TRIAL" somewhere.
>
> Noobies want to try stuff out, true, but they want to stick around with
> it also. If they feel they can't afford the full version - $99 for
> Allegro CL (for students at that), $660 per license for Lispworks -
> they won't even try it out, because if they like it, they'll just be
> pissed that it's so good but they can't afford it.
>
> Noobs are noobs, but the best noobs want to stop being noobs someday.
> They'll try their best to stop being noobs.
By the time they have learned enough Lisp to be really productive, they
will also be able to figure how to configure their preferred
environment. Until then, the trial versions are absolutely sufficient,
and the danger of being locked in is very low. Their fears are unjustified.
Pascal
--
3rd European Lisp Workshop
July 3-4 - Nantes, France - co-located with ECOOP 2006
http://lisp-ecoop06.bknr.net/
@Pascal:
>Until then, the trial versions are absolutely sufficient,
>and the danger of being locked in is very low. Their fears are unjustified.
Hmm. It was still my fear, anyway. And from my point of view, it's
easier to learn how to configure my preferred environment if I am using
that environment - using the trial versions seems to be installing yet
another program you're going to end up using for a month or so, then
ignore completely because you've learned how to hack CLISP/SBCL/CMUCL.
In the end, you still have to learn how to make CLISP load ASDF and
ASDF-INSTALL run on WinXP, figure out which libraries work in CLISP,
figure out how to make libraries which work on CLISP+Unix work on
CLISP+WinXP, etc. The trial versions seem more like diversions than
helps in these.
Just out of curiousity, how many of you used the trial versions and
then switched to a free version eventually, because of limitations on
the trial versions?
@Alex:
> However, (pertinent ) comments from gurus like Paul Graham are
>worrisome, and bring up the problem of the old guard (the people that
>grew up with LISP, improved it and made it what it is today) versus the
>(smaller?) "new guard" that has now access to free LISP system and can
>explore its power.
Hmm. Interesting problem... have you made any observations on this
problem so far?
Alan Manuel K. Gloria wrote:
> @Pascal:
>> Until then, the trial versions are absolutely sufficient,
>> and the danger of being locked in is very low. Their fears are unjustified.
> Hmm. It was still my fear, anyway. And from my point of view, it's
> easier to learn how to configure my preferred environment if I am using
> that environment - using the trial versions seems to be installing yet
> another program you're going to end up using for a month or so, then
> ignore completely because you've learned how to hack CLISP/SBCL/CMUCL.
>
> In the end, you still have to learn how to make CLISP load ASDF and
> ASDF-INSTALL run on WinXP, figure out which libraries work in CLISP,
> figure out how to make libraries which work on CLISP+Unix work on
> CLISP+WinXP, etc. The trial versions seem more like diversions than
> helps in these.
>
> Just out of curiousity, how many of you used the trial versions and
> then switched to a free version eventually, because of limitations on
> the trial versions?
I have started to learn Common Lisp with the student edition of
Macintosh Common Lisp 4.3, which costed around 90 Euro at that time. I
have then upgraded to Macintosh Common Lisp 5.0, which costed around 500
Euro. To me, my time is more valuable than my money, and there are
hobbies that are much more expensive than that.
By now, I have found an employer who has payed for a LispWorks
Professional Edition, and that's my main development environment for the
stuff I am working on. For porting my stuff to other Common Lisp
implementations, I just use a terminal window with the basic command
line, and every now and then I use SLIME in Aquamacs. But generally I
prefer LispWorks because I find it a lot more convenient. (I like the
fact that I can always fall back to using the mouse when I don't have a
clue what key binding I could use to achieve a certain task.) If my
employer hadn't payed for LispWorks Professional, I would probably use
the trial edition on a regular basis.
Pascal
--
3rd European Lisp Workshop
July 3-4 - Nantes, France - co-located with ECOOP 2006
http://lisp-ecoop06.bknr.net/
@Pascal:
You're lucky! Hmm. I've heard some nice things about LispWorks. I'll
try it, then, although I doubt I can use the free version for anything
as ambitious as a webserver (a webserver that has to be rebooted every
five hours is not a good webserver).
To help put things in perspective: I earn PHP12,068 (Philippine Pesos)
a month, plus a professional pay of about PHP1,000 per month. Then
there are taxes, so the base PHP12,068 goes down to something like
PHP10,500. Now, the conversion is 1EU = PHP63.7077. 500EU is about
PHP31,000 - more than two month's pay for me (and almost three). Like
you, to me, time is more valuable than my money, and a few months' work
and saving to get non-free software is too much time. Hence my
inherent fear of non-free software.
Yes, there are people who get paid that low. It's called "living in a
corrupt ****** banana republic under a corrupt ******* government and
wishing you were, like, born in the USA, or at least, Europe." Thank
God there's an Internet. Still, maybe I'll look at the trial edition
of LispWorks, if you think highly about it enough. Maybe I can even
find a pirated full version here for PHP100 (about 1.6EU), LOL! (living
in a banana republic has its compensations, if you're unscrupulous
enough.)
The only non-free software installed on my computer is Windows XP
(which came with the PC my dad bought me for my birthday) and some
games (which I begged off a friend who was tired of them).
Alan Manuel K. Gloria wrote:
> @Pascal:
> You're lucky! Hmm. I've heard some nice things about LispWorks. I'll
> try it, then, although I doubt I can use the free version for anything
> as ambitious as a webserver (a webserver that has to be rebooted every
> five hours is not a good webserver).
Hm, maybe I am missing something here. From what I can tell from the
rest of your posting, you are a hobbyist doing this at home. Do you
seriously consider running a webserver at home? (This is not a
rhetorical question.)
> To help put things in perspective: I earn PHP12,068 (Philippine Pesos)
> a month, plus a professional pay of about PHP1,000 per month. Then
> there are taxes, so the base PHP12,068 goes down to something like
> PHP10,500. Now, the conversion is 1EU = PHP63.7077. 500EU is about
> PHP31,000 - more than two month's pay for me (and almost three). Like
> you, to me, time is more valuable than my money, and a few months' work
> and saving to get non-free software is too much time. Hence my
> inherent fear of non-free software.
Sorry for my eurocentricity.
Again, what's important is that for learning Common Lisp, the trial
editions are fine, and the danger that you are tied to the respective
implementation is quite low. Apart from that, yes, you will probably
have to invest more time then.
Pascal
--
3rd European Lisp Workshop
July 3-4 - Nantes, France - co-located with ECOOP 2006
http://lisp-ecoop06.bknr.net/
Pascal Costanza <··@p-cos.net> writes:
>
> Hm, maybe I am missing something here. From what I can tell from the
> rest of your posting, you are a hobbyist doing this at home. Do you
> seriously consider running a webserver at home? (This is not a
> rhetorical question.)
I do--http://latakia.dyndns.org/ is my home machine; it hosts my blog,
my home page, a beer tasting notes app (which is probably down atm) and
some miscellaneous files for other folks.
Why _wouldn't_ one run a small webserver at home?
--
Robert Uhl <http://public.xdi.org/=ruhl>
People who are willing to rely on the government to keep them safe are
pretty much standing on Darwin's mat, pounding on the door, screaming
`Take me, take me!' --Carl Jacobs
Robert Uhl <·········@NOSPAMgmail.com> wrote:
+---------------
| Pascal Costanza <··@p-cos.net> writes:
| > Do you seriously consider running a webserver at home?
| > (This is not a rhetorical question.)
|
| I do--http://latakia.dyndns.org/ is my home machine; it hosts my blog,
| my home page, a beer tasting notes app (which is probably down atm) and
| some miscellaneous files for other folks.
+---------------
I do the same, except with static IPs & DNS: <http://rpw3.org> is
my web server [with CMUCL-based apps behind Apache], mail server,
DNS primary, and main shell home [e.g., where I read news from
(though often remotely, via SSH, as at this moment)]. It & its
DSL modem [and the telephone answering machine & another small
BSD system] share a small UPS good for ~2 hours -- enough that
they've never lost power since I put in the UPS. [Before that,
though, they were crashing several times a month (or more!) due
to frequent "tiny" power outages of a few seconds to a minute.]
+---------------
| Why _wouldn't_ one run a small webserver at home?
+---------------
Beats me. As you demonstrate, given DynDNS one doesn't even need
a static IP anymore.
-Rob
-----
Rob Warnock <····@rpw3.org>
627 26th Avenue <URL:http://rpw3.org/>
San Mateo, CA 94403 (650)572-2607
Robert Uhl <·········@NOSPAMgmail.com> writes:
> Why _wouldn't_ one run a small webserver at home?
Well, for me there's a very simple reason: I prefer to turn off our
computers when we go on holiday - less risk of fire. I also disconnect
the adsl line from the network equipment - to avoid firestorm-induced
damage (there has been a lot of that around here).
--
(espen)
Espen Vestre <·····@vestre.net> writes:
> Robert Uhl <·········@NOSPAMgmail.com> writes:
>
>> Why _wouldn't_ one run a small webserver at home?
>
> Well, for me there's a very simple reason: I prefer to turn off our
> computers when we go on holiday - less risk of fire. I also disconnect
> the adsl line from the network equipment - to avoid firestorm-induced
> damage (there has been a lot of that around here).
Haven't you heard of this guy Benjamin Franklin? He invented a couple
of nice devices...
That said, I've got my server at home too with 99.9% uptime since 7
years (and never shut down my computers much before either), and the
only fire hazard I've witnessed involved candles, not computers.
--
__Pascal Bourguignon__ http://www.informatimago.com/
I need a new toy.
Tail of black dog keeps good time.
Pounce! Good dog! Good dog!
Pascal Bourguignon <···@informatimago.com> writes:
> Haven't you heard of this guy Benjamin Franklin? He invented a couple
> of nice devices...
Sure, but a lightning rod won't help my network equipment when lightning
strikes the telephone lines... I've already destroyed one modem and one
isdn router a few years ago, so I always unplug the telephone line when
going on summer holiday.
--
(espen)
Espen Vestre <·····@vestre.net> writes:
> Pascal Bourguignon <···@informatimago.com> writes:
>
>> Haven't you heard of this guy Benjamin Franklin? He invented a couple
>> of nice devices...
>
> Sure, but a lightning rod won't help my network equipment when lightning
> strikes the telephone lines... I've already destroyed one modem and one
> isdn router a few years ago, so I always unplug the telephone line when
> going on summer holiday.
Ah well, in more civilized countries, phone (and electric) lines are
underground... We avoid lightning, but not rats :-)
--
__Pascal Bourguignon__ http://www.informatimago.com/
PLEASE NOTE: Some quantum physics theories suggest that when the
consumer is not directly observing this product, it may cease to
exist or will exist only in a vague and undetermined state.
+ Pascal Bourguignon <···@informatimago.com>:
| Ah well, in more civilized countries, phone (and electric) lines are
| underground...
So they are in Norway, in urban and suburban neighbourhoods developed
since the sixties or so. But with older neighbourhoods it's a
different story.
--
* Harald Hanche-Olsen <URL:http://www.math.ntnu.no/~hanche/>
- It is undesirable to believe a proposition
when there is no ground whatsoever for supposing it is true.
-- Bertrand Russell
Harald Hanche-Olsen <······@math.ntnu.no> writes:
> + Pascal Bourguignon <···@informatimago.com>:
>
> | Ah well, in more civilized countries, phone (and electric) lines are
> | underground...
>
> So they are in Norway, in urban and suburban neighbourhoods developed
> since the sixties or so. But with older neighbourhoods it's a
> different story.
>
I hope you mean "developed" or "industrialised" rather than civilised.
>
--
tcross (at) rapttech dot com dot au
Pascal Bourguignon <···@informatimago.com> writes:
> Ah well, in more civilized countries, phone (and electric) lines are
> underground... We avoid lightning, but not rats :-)
Well, the lines along the main roads is underground where I live, too,
but it's a very hilly and rocky place with not that many inhabitants
and many of old houses, so I assume they won't get all of it
underground for the next half century or so...
(Cicilized? Hah! I used to live in Germany close to the french border,
and the easiest way to recognize when you actually crossed the border
was the telephone and power lines which were underground in Germany,
but not in France. But probably you don't count Lorraine to the
civilized part of the world ;-))
--
(espen)
Espen Vestre <·····@vestre.net> writes:
> Pascal Bourguignon <···@informatimago.com> writes:
>
>> Ah well, in more civilized countries, phone (and electric) lines are
>> underground... We avoid lightning, but not rats :-)
>
> Well, the lines along the main roads is underground where I live, too,
> but it's a very hilly and rocky place with not that many inhabitants
> and many of old houses, so I assume they won't get all of it
> underground for the next half century or so...
>
> (Cicilized? Hah! I used to live in Germany close to the french border,
> and the easiest way to recognize when you actually crossed the border
> was the telephone and power lines which were underground in Germany,
> but not in France. But probably you don't count Lorraine to the
> civilized part of the world ;-))
Probably less civilized than German, I concede.
All right, so the question is city vs. country.
--
__Pascal Bourguignon__ http://www.informatimago.com/
This is a signature virus. Add me to your signature and help me to live.
Ken Tilton <·········@gmail.com> writes:
> I notice no one has yet explained how getting a Lisp running on
> Windows is "even harder".
I could explain, or at least give you my theory:
First, "being harder" is relative, to the population.
If you take 100 random linux users and 100 random windows users,
then the linux group will find it easier to fetch and install a free
lisp implementation than the windows group.
Just because amongst the linux group, you'll have much more
programmers, hackers and otherwise people used to download, compile
and install free software, while in the windows group you'll have much
more plain users who don't know how a computer works.
Then, there's the fact that linux is a free software system: there are
sources everywhere, the compilers are (almost) always installed
automatically by the distributions, so even a non-programmer can be
directed to open a terminal window (he KNOWS what it is!, the icon is
there in KDE, GNOME or whatever), and type:
wget http://...
tar jxf ...
cd ...
./configure && make && make install
...
and that's it.
On the other hand, windows is a closed system: people don't know what
a source is, they don't know what a terminal is, they can hardly find
the so called "MS-DOS" window, (I've even heard it disappeared from
the Start menu in the latest versions of MS-Windows), and even if you
find it, what can you do from a MS-DOS prompt on MS-Windows?
Is there a wget?
Is there a tar or unzip?
Is there a compiler?
Ok, so first you'll have to download the source tarball from Internet
Explorer. You'll have the icon right there in the window, but what
directory is that?
Then you'll have to download an utility to untar or unzip it. That'll
be a binary, otherwise, you go back to the start case. Check it has
no virus/worm.
Then you'll have to download a compiler. Happilly, I hear Microsoft
distributes his MSVC for free.
Then you'll be able to unarchive the sources, and to compile it.
Actually, *I* wouldn't bother. I'd point at cygwin, download
http://www.cygwin.com/setup.exe execute it, install a sizeable set of
tools, then open the cygwin icon, and you're back to the linux case.
But then, people would complain that the compiled programs don't have
the MS-Windows GUI and don't integrate well with MS-Windows, blah blah
blah.
--
__Pascal Bourguignon__ http://www.informatimago.com/
ATTENTION: Despite any other listing of product contents found
herein, the consumer is advised that, in actuality, this product
consists of 99.9999999999% empty space.
On Fri, 21 Apr 2006 19:46:56 +0200, Pascal Bourguignon <···@informatimago.com> wrote:
>
> Ken Tilton <·········@gmail.com> writes:
> > I notice no one has yet explained how getting a Lisp running on
> > Windows is "even harder".
>
> I could explain, or at least give you my theory:
I don't want to get into an OS flamewar, especially not on
this newsgroup, but I could not let this sub-thread pass
without commenting (and with all due respect to Pascal)...
> First, "being harder" is relative, to the population.
>
> If you take 100 random linux users and 100 random windows users,
> then the linux group will find it easier to fetch and install a free
> lisp implementation than the windows group.
>
> Just because amongst the linux group, you'll have much more
> programmers, hackers and otherwise people used to download, compile
> and install free software, while in the windows group you'll have much
> more plain users who don't know how a computer works.
Your sampling method is somewhat flawed -- Windows users
who are interested in experimenting with Lisp are likely
to be developers (professional or otherwise). As a rule,
Windows developers are able to download and configure lots
and lots of tools to customize their environments. There is
a huge variety of choice, and also there are some well-known
standard choices.
I wrote "somewhat flawed" because in general, I agree
that the amongst the overall population of Windows users,
relatively few are developers. My point is that you
were using the wrong subset to back up your theory.
> Then, there's the fact that linux is a free software system: there are
> sources everywhere, the compilers are (almost) always installed
> automatically by the distributions, so even a non-programmer can be
> directed to open a terminal window (he KNOWS what it is!, the icon is
> there in KDE, GNOME or whatever), and type:
>
> wget http://...
> tar jxf ...
> cd ...
> ./configure && make && make install
> ...
>
> and that's it.
There are tons and tons of free utilities for Windows, and there is
no evidence[1] that it is any harder to download and install them than
it is on Linux. Again, assuming the sub-population of Windows users
that would be interested in Lisp or programming at all.
True, the vast majority of such tools are not "source available."
Installation is a matter of download the setup.exe (or setup.msi)
and double-click. Sometimes it's just a matter of unzipping.
[1] -- I'll concede that installation can be more of a pain because
of the bad habits of many Windows developers. Savvy Windows
users understand they should not be running with Administrator
rights by default; unfortunately this often complicates installation
a bit because many developers are sloppy/lazy and don't support folks
running in a LUA. Whereas sudo[2] on Unix/Linux/OSX is a piece of cake.
[2] -- Yes, I know about "runas" in Windows. Unfortunately, there
are situations where the install program still borks things up
so badly that the end result needs to be fixed manually.
> On the other hand, windows is a closed system: people don't know what
> a source is, they don't know what a terminal is, they can hardly find
> the so called "MS-DOS" window, (I've even heard it disappeared from
> the Start menu in the latest versions of MS-Windows), and even if you
> find it, what can you do from a MS-DOS prompt on MS-Windows?
> Is there a wget?
> Is there a tar or unzip?
> Is there a compiler?
Yes to all of the above. Just because they are not command-line
apps by default does not lessen their value. WinZip, for instance,
is able to handle zip, tar, and tar.gz files quite easily, and
integrates into the Explorer Shell.
It's usually easier to find interactive utilities than non-interactive
utilities, so I think your mention of wget is a little unfair. But
there may be an existing equivalent that I haven't found yet. I just
use FTP Commander and I'm happy.
There are several free compilers for Windows, not just MSVC.
Borland and Watcom are both free as well.
> Ok, so first you'll have to download the source tarball from Internet
> Explorer. You'll have the icon right there in the window, but what
> directory is that?
>
> Then you'll have to download an utility to untar or unzip it. That'll
> be a binary, otherwise, you go back to the start case. Check it has
> no virus/worm.
>
> Then you'll have to download a compiler. Happilly, I hear Microsoft
> distributes his MSVC for free.
>
> Then you'll be able to unarchive the sources, and to compile it.
>
> Actually, *I* wouldn't bother. I'd point at cygwin, download
> http://www.cygwin.com/setup.exe execute it, install a sizeable set of
> tools, then open the cygwin icon, and you're back to the linux case.
While I have cygwin and MSYS and GnuWindow, as I mentioned in a thread
about asdf-install, I think requiring any of those to use any app is a
burden that Windows developers should not place on their users. Sometimes
it's the only (or best) choice, certainly! But requiring any of those
compatibility environments is usually not the *only* choice a developer
nor the best.
> But then, people would complain that the compiled programs don't have
> the MS-Windows GUI and don't integrate well with MS-Windows, blah blah
> blah.
What you characterize as a "complaint", I call a "preference".
They are equally valid.
--
Jack Unrue
On Fri, 21 Apr 2006 19:06:37 GMT, Jack Unrue <·······@example.tld> wrote:
>
> While I have cygwin and MSYS and GnuWindow, as I mentioned in a thread
^^^^^^^^^
I meant GnuWin32.
--
Jack Unrue
Jack Unrue <·······@example.tld> writes:
> While I have cygwin and MSYS and GnuWindow, as I mentioned in a thread
> about asdf-install, I think requiring any of those to use any app is a
> burden that Windows developers should not place on their users. Sometimes
> it's the only (or best) choice, certainly! But requiring any of those
> compatibility environments is usually not the *only* choice a developer
> nor the best.
Ah but this is another problem: most free software developers ARE NOT
MS-Windows developers (AFAIK). That explains why most free software
is unix (linux, *bsd) software, not MS-Windows, and that more free
software are ported from unix to MS-Windows than the other way.
Windows users are a burden on free software developers ;-)
Can we find MS-Windows developers willing to port (or develop) free
software for MS-Windows users?
>> But then, people would complain that the compiled programs don't have
>> the MS-Windows GUI and don't integrate well with MS-Windows, blah blah
>> blah.
>
> What you characterize as a "complaint", I call a "preference".
> They are equally valid.
Of course. And MS-Windows developers should volunteer to write these
OS specific adaptations. Unix developers have already made the effort
of writing portable software, and compiling it on MS-Windows.
--
__Pascal Bourguignon__ http://www.informatimago.com/
HANDLE WITH EXTREME CARE: This product contains minute electrically
charged particles moving at velocities in excess of five hundred
million miles per hour.
On Fri, 21 Apr 2006 22:09:51 +0200, Pascal Bourguignon <···@informatimago.com> wrote:
>
> Jack Unrue <·······@example.tld> writes:
> > While I have cygwin and MSYS and GnuWindow, as I mentioned in a thread
> > about asdf-install, I think requiring any of those to use any app is a
> > burden that Windows developers should not place on their users. Sometimes
> > it's the only (or best) choice, certainly! But requiring any of those
> > compatibility environments is usually not the *only* choice a developer
> > nor the best.
>
> Ah but this is another problem: most free software developers ARE NOT
> MS-Windows developers (AFAIK). That explains why most free software
> is unix (linux, *bsd) software, not MS-Windows, and that more free
> software are ported from unix to MS-Windows than the other way.
That might have been true 10-15 years ago, but in this day and age
(with Java, Python, Perl, and Ruby all well supported on Windows
in addition to surprisingly long-lived C/C++) I'm not so sure.
See below.
> Windows users are a burden on free software developers ;-)
Well, that's debatable :-)
What I meant, though, was 'burden' in the sense that if I make
an application available, my userbase shouldn't have to worry
about Cygwin just because I didn't feel like adapting my tools
to Windows. They don't care about my issues as a developer.
I can extend this to users (who are developers themselves)
of my library code.
> Can we find MS-Windows developers willing to port (or develop) free
> software for MS-Windows users?
Well, I'm one. I definitely would like to see more Windows-based
Lisp OSS developers, and hope that as time goes on the number
increases.
If you're going to respond that by "free" you specifically
mean GPL licensed, well stop reading because you will disagree
with the rest of this post a priori.
I think you have a bias towards Unix-based (or maybe even Gnu)
development tools and traditions that are clouding your perspective.
In the Java community alone, there are a great many free software
developers whose primary platform is Windows. Examples that I know
from actual, personal experience include people involved with:
Eclipse
Spring
Apache (I know a few people in various projects, but there are
a *huge* number of Apache projects and I won't bother
naming them all)
I would feel confident that there are a non-trivial number of
primarily-Windows-based folks working on:
JBoss
JDOM / DOM4J / name any other XML-oriented library or tool
Boost
Qt
wxWidgets
SharpDevelop (a free IDE for C# and VB.NET developers)
I could go on for a whole day writing down OSS projects that
count Windows developers amongst their numbers.
For these people, Windows support is an *assumption*.
> >> But then, people would complain that the compiled programs don't have
> >> the MS-Windows GUI and don't integrate well with MS-Windows, blah blah
> >> blah.
> >
> > What you characterize as a "complaint", I call a "preference".
> > They are equally valid.
>
> Of course. And MS-Windows developers should volunteer to write these
> OS specific adaptations. Unix developers have already made the effort
> of writing portable software, and compiling it on MS-Windows.
Although my primary platform is Windows, I am well-versed in Unix,
I have two Linux boxes running on my LAN at all times, and in fact
I developed on a great many different platforms in the past. I would
expect (and I'm sure most people would as well) any well-rounded
developer to have serious cross-platform experience. And by that,
I mean knowing more than how to type "./configure; make; make install" :-)
The bottomline is that if you are making an assumption about
Windows developers having no recourse other than the generosity
of Unix developers to get access to FOSS, this is a historical
debt to which we all (current Unix and Windows and Mac developers)
owe those who came before, but it hardly represents the actual
situation today.
--
Jack Unrue
As an aside, OODBMS were mostly written in C until very recently.
There were a few notable exceptions. I was there in the middle of the
fray about sixteen years ago. I believe an OODBMS was done in Lisp in
part by MCC around that time. But most DB engines used C at their core
because that is about as close to the metal as you can comfortably get
and, especially in those days, you wanted as much perfromance as
possible at the levels close to spinning disks.
samantha wrote:
> As an aside, OODBMS were mostly written in C until very recently.
> There were a few notable exceptions. I was there in the middle of the
> fray about sixteen years ago. I believe an OODBMS was done in Lisp in
> part by MCC around that time. But most DB engines used C at their core
> because that is about as close to the metal as you can comfortably get
> and, especially in those days, you wanted as much perfromance as
> possible at the levels close to spinning disks.
Not many databases have been written in Lisp, but some:
Graphael's Gbase (the first commercial OODBMS, 1984)
Statice (1988) from Symbolics, one of the first commercial OODBMS (some
the developers later went to develop ObjectStore), later Portable
Statice
Zeitgeist (from TI)
Orion, research system, then commercialized as Itasca
AllegroCache from Franz (new)
Similar:
Persistent CLOS
Plob
Postgres was originally partly written in Lisp, too.