> [apologies to Scott for accidently sending this to him off list, too.
> Stupid email program....]
[I think the problem is in Sun's mailing list server.]
Summary of your message: Product software has no value. Distribution, tech
support and consulting have value, but writing software is worthless.
Implication: No sane business will create software for JSP.
Maybe you're right. But I want you to think hard about the secondary
implications of JSP with no commercial support.
Compare the market for JSP and ASP. Not the product, the market. ASP has say
100,000 developers? JSP has 1,000? Maybe? ASP has support from many
development/editing tools. It has many web sites with tutorials and examples
and huge libraries of code. It has several books and tutorials. ASP has many
consultants. It is available from ISPs. ISPs won't touch JSP, nor servlets,
because there is still no solid deployment for them commercial or
non-commercial.
If you're a consultant, how valuable is your knowledge of JSP? Zero? Do you
even bother putting it on your resume? ASP consultants brag about it on their
resume. ASP is valuable to consultants. JSP is close to worthless.
If you're working in a larger company, how hard is it to convince management
to use JSP?
Why is JSP in severe trouble? In part because of spec problems. Mostly,
though, because no sane business will spend any effort on JSP. That means no
commercial-quality implementations (and Apache JServ is not commercial
quality.) That means no technical support beyond this list. That means
neither books, nor tutorials, nor consultancies.
Let me give you two real, not hypothetical examples.
The open source servlet runner, Apache JServ, is slow, it's very difficult to
install, it's still in beta and it still missing the servlet 2.1 API. And
servlets are much more mature than JSP. We've got a prototype which cuts the
latency from 35ms to 11ms. (That's including the fixed 6ms Apache
overhead.) I think we can make installation easy even for novices. Will you
ever see it? No. Why? Because it's not worth our time to develop it. So
you can wait and hope someone will volunteer to improve JServ. It may
happen.
Our primary product, Resin, quickly approaching beta, brings JavaScript to
JSP. Even those of you who are Java hackers should see the value of making
JSP available to non-Java programmers. The more people who use JSP, the more
valuable your JSP experience is, and the easier it is to convince management
or customers to let you use JSP.
We're 90% convinced to drop JSP and adopt ASP. It's an easier sell. People
who develop with ASP can then deploy on whatever server they want with our
product and since it's in Java, can use Java components instead of COM if they
want.
Why abandon JSP? Mostly because the market is small, the customers are cheap,
and it's not likely that it'll really take off, especially if it continues to
discourages non-experts and non-Java scripts. Also the 1.0 spec is months
late, incomplete, and all indications are that it is unfriendly to non-Java
languages.
I don't expect you to care about one tiny company's decision to abandon
JSP support. Just decide for yourself if it's an exception or a symptom why
JSP isn't receiving support.
Scott Ferguson
Caucho Technology
p.s. The statement "what about making money from distributing computers,
..., providing tech support, from consulting?" is the most asinine statement
I've heard in months. You can also make money professional wrestling,
flipping hamburgers, or pimping. If you value software development less than
tech support, don't be surprised if no one jumps to write software for you.
> Well, what about making money from distributing computers with the
> software installed, from providing tech support, from consulting? I
> actually really like the idea filling needs for companies using 'free'
> software. That way they get products that have good reason to stay
> around for a long time, growing and developing, and if they have the
> need they can find others to help them out. I worked on one assignment
> for a big consulting company, working for a huge client, and we had a
> need for daily document conversion. There was a company that had an
> ideal product, just needed a little massaging in the code, a change in
> perspective. We called them up, dropped the names, and told them we'd
> pay them whatever they wanted. Anything. And they wouldn't do it. I
> like the idea of GNU software being free, not necessarily from cost,
> but from the fear that it's secrets will die with the company.
>
> And as far as the viability of the business model, I'd like nothing
> more than to have a consultancy that works primarily with freeware.
> Anyone else feel the same?
>
> Josh
===========================================================================
To unsubscribe, send email to [EMAIL PROTECTED] and include in the body
of the message "signoff JSP-INTEREST". For general help, send email to
[EMAIL PROTECTED] and include in the body of the message "help".
