[Lispweb] Mod_lisp 2.0 released

Andrew K. Wolven awolven at redfernlane.org
Mon Jun 18 02:14:59 CDT 2001

Sunil Mishra wrote:

> Andrew,
> There are many, many reasons for which mod_lisp is a good idea.
> I don't know if you have ever worked in an environment where you
> have a support staff operating all the back end machinery for a
> web site. They don't know lisp. Period.

I don't work for [computer] people who don't know lisp.  Period.  I would
much rather trim their hedges.

> It is not reasonable to
> expect them to learn. The lisp way of doing things is alien to
> them, they will not understand it, and they are much more likely
> to make mistakes. To find someone that *is* able to understand
> lisp is going to be far more difficult and far more expensive.

It is not my job, role, mission, duty, pleasure, or hobby to support bad
programming style.  I would suggest for them electroshock treatment so that
they could start from scratch with the right learning materials.  For the
time being I will hire their teenager.

> And besides, you would probably want them doing something other
> than maintaining your web server.
> And web servers do need maintenance. They need someone to go
> through the logs, figure out problems, implement needed
> functionality, for example an approach to security. You never
> want your core application developers to be doing that work.

If they can run Apache, they can run CL-HTTP or Allegroserve.  If they
cannot figure out CL-HTTP on Lispworks on sparc then they should not be an
administrator.  Between me and my non-lisp using administrator I have set up
the server to run as a daemon it dumps logs into a file and/or database and
he uses perl/analog to work with the data.  He can set up directories for
static content.  Your non core developers should be able to figure this out
too.  (He has actually been spending much more time debugging DNS on linux
than anything else.)

> They can do it, but it would be a waste of their time. Like
> asking an electrical engineer to repair a broken socket.

I am an electrical engineer and I make it a point to know [personally] how
to repair a broken socket.

> I love lisp. I think its the best language I have worked with.
> But putting down other languages, even if they don't quite
> measure up, does not help lisp's cause one bit. If anything, it
> alienates other engineers and support people with whom you have
> to work. And that kind of antagonism has no place in an
> engineering organization that needs to deliver a product.

Are you talking about NASA or K-Mart?  Which pansey was it who could not say
to his boss in 1983 that you cannot launch solid rocket boosters on a cold

> We had been working on a distributed client/server application
> at my last job. The front end was java, and the back end was
> lisp. It was java because there has been a lot more work done
> with web development with java, and the problems are a little
> better understood in that context. It would be easier to hire
> people to build that front end. And the front end would have to
> be handed over to clients. Doing this work in lisp would be a
> non-starter.

> The back end was in lisp, of course. Well, part of it was in
> lisp. We also had a transactional queue, a persistence manager
> (also in java) talking to a database. While lisp is great for
> solving hard problems, doing routine things is sometimes far
> more difficult. Because the routine things generally don't need
> a solution through programming.
> IMHO, in the present day you cannot expect lisp alone to form a
> complete product. Period. Lisp is better, sure, but going
> forward lisp absolutely *must* learn to play with what is.

> What mod_lisp gives is very important. It allows lisp developers
> to continue programming in lisp. And it allows the company to
> hire drones to handle the apache server, or whatever server they
> wish to use. It is easier for managers to justify using lisp.
> And last but not least, no lisp implementation on unix is
> currently multi-threaded.

Lisp based schedulers don't count?

> So building a truly scalable lisp-only
> solution is more or less out of the question. Throwing the http
> server into a separate process (apache) will ultimately also
> result in a service that scales better.
> So before you put down someone else's work, ask, listen, and try
> to understand. And if you don't, say so politely.

The last half of your comments I agree with mostly except for the last
paragraph.  It is important to me that you note that I am not slamming
Marc's work.  I am slamming the fact that he insults the lisp community by
perpetrating the idea that marketability through buzzwords implies value.  I
think that it is sleazy that he would go around telling customers that they
should go with apache vs. a native lisp webserver beacause it saves time.
This is not service, it is a scam.  This bullshit [techno babble of those
whold reach into the comode at a resturaunt to retrieve a penny] has invaded
the innards of our military industrial complex to the point that it is
measurably more physically dangerous for people to protect the very building
in which you and your apache guys do business from.  These are friends of
mine, Sunil, as you and I should be, but they deserve justice.  Tell me my
friend, do you believe in non-violence?  Have you ever been drafted?  I
haven't, but I sure as hell know what it feels like.

This civilization needs ball stompers in the right places otherwise corrupt
methodology will cause our human ecosystem to disintegrate under our own
If computers are not [one of] the most important technologies for the
maintenance of our species than Bill Gates would not be so rich.  Do you
drive on the freeway, do you use an elevator?  Ever been in a jumbo jet?  Do
you take any medications for serious conditions?


> Sunil
> On Sunday, June 17, 2001, at 09:04 PM, Andrew K. Wolven wrote:
> > So the point that you are trying to make is that your employer
> > had you dinking around
> > with inferior java technology in order to support a mostly
> > inferior web server in order
> > to support an inferior operating system in order to support yet
> > another mostly inferior
> > programmer who calls button-pushing a 'science'.  (All of this
> > having nothing to do with
> > lisp or web except to justify a system which could have just
> > used fastcgi or jserve in
> > the first place if you must sell buzztalk.)
> >
> > Consider this common abbreviation of one of my employers:
> > AIMD,  Aviation Intermediate
> > Maintenance Depot.  The only Tomcat technology AIMD bothers to
> > support is F-14 Tomcat.
> > In this environment, you may find yourself facing both a court
> > martial and the wrath of
> > a widow if money makes you complacent and you cannot do your homework.
> >
> > Of course, that example is way extreme, let us consider a less
> > mission critical
> > application environment:  an Online store.  Do you want to
> > rewrite the entire system
> > from scratch to make the jump from 100 customers a month to
> > 100,000?  It's up to you, it
> > is your time, your money and your carpal tunnel syndrome.
> >
> > AKW
> >
> _______________________________________________
> Lispweb mailing list
> Lispweb at red-bean.com
> http://www.red-bean.com/mailman/listinfo/lispweb

More information about the lispweb mailing list