Various Topics on Lisp

..k..

Kenny Tilton wrote:
>
>
> mikel wrote:

>> He's right that there is money to be made in it; I made a little of it
>> myself.
>>
>> I worked for a company that built and spun out other companies. One of
>> our successes was a business that built a knowledge-based product that
>> automated the process of installing and configuring servers on a
>> network. I wrote the knowledge-representation bits in Lisp and then
>> ported them to Java (because that was the required target). It was
>> certainly an easier problem to solve in Lisp than in Java, and the
>> code got bulkier and more restrictive as it was ported.
>
>
> And that was my point: more Lisp, not less.

Sure.

You're right: it was far easier to write the app in Lisp.

He's right: ease of writing the app was much the smaller part of the
problem.

..ni.. ..rl..

Chris Capel <ch.ris@iba.nktech.net> wrote in message news:<10kri8h62t6a3fa@corp.supernews.com>...
> Viaweb doesn't count. I'm probably not nearly as smart as Paul Graham.
> Orbitz doesn't count. It's just a backend. It deals with entirely different
> issues that something like Cliki does. And Cliki doesn't have (as far as I
> know) lots and lots of traffic.

I don't know what your definition of "lots and lots" is, but here's a
couple of numbers for Araneida apps that I do know. Back in February
2004 (the latest time I have a report for), CLiki was serving around
20Mb a day, fairly evenly spread out over those 24 hours. Another
Lisp site on the same machine was doing 70Mb, mostly to UK customers
(most traffic served at UK lunchtime and evenings). That's average,
and the uptime on those apps is ... well, indefinite. (Sometimes I
restart CLiki when I've deleted a bunch of pages because it's the
easiest way to get them out of the indexes.) The other day we had a
sudden rush and were serving about 10-20 requests a second at 10% cpu
utilisation.

What does this say about Lisp in high load situations? Well, even
***uming that you consider 10-20 requests a second as "high", still
not very much -

- every page in the site (aside from graphics) is programmatically
generated, but
- there's a cache/proxy in front of the Araneida, and we go to
reasonable lengths to ensure that all our pages have sensible
Last-Modified and Expires headers. So, programmatically generated but
infrequently updated pages are cached "for free". Whatever your
"scripting" language, a page which doesn't use it at all is probably
going to be served faster than one that does.
- and the database interface is still the bottleneck. (Actually,
literally so: there's a bug in the installed version of SBCL that
means the stream connected to the database is unbuffered and we do a
syscall per character)

This is all on a single Athlon 2600 with half a gig of ram: perhaps
not a machine you can find in a cornflake packet yet, but hardly
difficult or expensive to obtain either.

> PHP does handle those kind of loads (though it does need things like
> precompilers and such) and does it regularly. (According to Troutgirl at
> Friendster, it does it better than java.) Lisp is sort of a question mark

That's not what Troutgirl said. I point you especially at
http://troutgirl.com/blog/index.php?...ll_column.html :
"He makes exactly the right point: the programming language itself was
not really the issue". Oddly enough, this is the same point as most
people in this thread have made.

> I think this is a special case of the more general "Microsoft anxiety" that
> particularly affects developers, like me, that have grown up using
> Microsoft software--OS, dev. tools, etc. The perception is, with MS, /or/
> any other widely used platform, that even if something doesn't work, it's
> either /your/ fault, or a hundred people have already had the problem and
> have posted workarounds on the web, so you're never going to be screwed (at
> least by the software itself).

I wish this were true, but my experience has been more along the lines
of "three people have already had the problem and nobody could help
them with it, unless that post in (some language that neither I or
Babelfish speaks) contained a solution". Maybe I have weird problems.

..in.. ..sw..

Henry Lenzi <void@void.org> wrote in message news:<87wtyogudz.fsf@samadhi.i-did-not-set--mail-host-address--so-shoot-me>...
> Hi Chris --
>
> Do some Google searching on this group.
> IIRC, Howard Dean's web site ran on CL-HTTP.
>
> Cheers.
> Henry L.

No, some internal application for the campaign
was written with CL-HTTP.

Before that, the publications server (not the default website!)
for the Whitehouse (Clinton/Gore) ran on Open Genera with CL-HTTP.

Plus there were some other applications in that area like "Open Meeting"
(Vice President AL Gore's Open Meeting on the National Performance Review.)
were written with CL-HTTP.

Orbitz seemed to use some HTML generation from CL-HTTP.
(http://www.itasoftware.com/acknowledgements/)

www.booble.com (adult search engine, be careful) seems to
use CL-HTTP's template engine. (See the HTML source of the homepage).

Etc.

..m ..adsh..

Kenny Tilton wrote:
> And my point is that it sounds like Mikel already solved the problem
> which you believed would be the same no matter what language you
used.
> By using Lisp to write an application to solve what you believed was
a
> sys admin task.

Well, as he said:

You're [Kenny] right: it was far easier to write the app in Lisp.

He's [tfb] right: ease of writing the app was much the smaller
part of the problem.

That's the point I'm trying to make in a nutshell.

Or, to put it another way: the problems I'm describing are technical
to only a small degree. A technical fix can only fix that small
degree. It is a cl***ic programmer mistake to ***ume that *most* of
the problems of large systems are programming-related: they aren't.
--tim

..t.. ..ib..

joswig@corporate-world.lisp.de (Rainer Joswig) writes:

> Henry Lenzi <void@void.org> wrote in message news:<87wtyogudz.fsf@samadhi.i-did-not-set--mail-host-address--so-shoot-me>...
>> Hi Chris --
>>
>> Do some Google searching on this group.
>> IIRC, Howard Dean's web site ran on CL-HTTP.
>>
>> Cheers.
>> Henry L.
>
> No, some internal application for the campaign
> was written with CL-HTTP.
>
> Before that, the publications server (not the default website!)
> for the Whitehouse (Clinton/Gore) ran on Open Genera with CL-HTTP.

Speaking of which, is it just me or are many of the links at

<http://www.ai.mit.edu/projects/iiip/doc/cl-http/home-page.html>

broken? Such as all the links to ftp.ai.mit.edu and wilson.ai.mit.edu.

-Peter

--
Peter Seibel peter@javamonkey.com

Lisp is the red pill. -- John Fraser, comp.lang.lisp

..a.. ....

Peter Seibel <peter@javamonkey.com> wrote:

> Speaking of which, is it just me or are many of the links at
>
> <http://www.ai.mit.edu/projects/iiip/doc/cl-http/home-page.html>
>
> broken? Such as all the links to ftp.ai.mit.edu and wilson.ai.mit.edu.

Nearly no link worked, and the few external links which worked looks like
they are served by Apache (see my posting in this thread:
http://groups.google.de/groups?selm=....netcologne.de )

Perhaps the server is as mature as described on the page, but the
broken links doesn't look good.

--
Frank Buß, fb@frank-buss.de
http://www.frank-buss.de, http://www.it4-systems.de

..sc.. ..stan..

Frank Buss wrote:
> Peter Seibel <peter@javamonkey.com> wrote:
>
>
>>Speaking of which, is it just me or are many of the links at
>>
>> <http://www.ai.mit.edu/projects/iiip/doc/cl-http/home-page.html>
>>
>>broken? Such as all the links to ftp.ai.mit.edu and wilson.ai.mit.edu.
>
> Nearly no link worked, and the few external links which worked looks like
> they are served by Apache (see my posting in this thread:
> http://groups.google.de/groups?selm=....netcologne.de )
>
> Perhaps the server is as mature as described on the page, but the
> broken links doesn't look good.

The MIT AI Lab doesn't exist anymore but has merged with another lab.
See http://www.ai.mit.edu/ - maybe this explains why some of their pages
don't work correctly anymore.


Pascal

--
Tyler: "How's that working out for you?"
Jack: "Great."
Tyler: "Keep it up, then."