Quoting Milt Epstein ([EMAIL PROTECTED]):
> A big problem with your analogy of WebMacro to Perl (or Apache) is
> that Perl is a standalone language/technology, and something like
> WebMacro is more like a plug-in/add-on to an existing
> language/technology (Java/servlets). Perl does not sit on top of
> something else that may be a moving target, and does not have
> "competition" from another complete Perl-type system. A better
> analogy would be to a particular Perl library/module. It has happened
> that multiple Perl libraries/modules have been written with similar
> functionality, and some fall by the wayside as one becomes more
> standard and/or gets more integrated into the Perl core
> (e.g. cgi-lib.pl vs. CGI.pm).
There's some truth here. Neither WebMacro nor JSP nor any other add-on
will ever be as widely used as the technologies they require (namely
Java and Apache, etc.) Also Perl and Apache do have lots of direct
competitors that solve exactly the same task. Python is a direct and
strong competitor to Perl; and there are lots of competitors to Apache.
At any rate Jason's point was that opensource technologies succeed
without being standards, and in this respect Apache and PERL are good
examples. The market for a tool like WebMacro is no doubt smaller
than the market for PERL and Apache, but the principle is the same.
The primary argument for standards in software systems is that it gives
the customer some assurance they won't be hung out to dry. If one vendors
product is too buggy, or not well enough supported, or the vendor goes
bankrupt, then the customer has the option to switch to another vendor.
With proprietary software you NEED that assurance, since there isn't anything
else that you can do about it.
With opensource software you have other choices. If the software is buggy,
you can fix it. Or you can find someone else who will fix it. Or someone
else will just fix it on their own. If the development team building the
software abandons it, someone else can freely take over and pick up right
where they left off--and if it was a useful technology, someone always
will. If the program isn't well enough documented, then someone
can read the source code and document it, or you can set up a
mailing list and exchange ideas, etc.
Support for JSP in tools has less to do with JSP being a standard, I
think, and more to do with the amount of money Sun is spending to
promote it. That's a little bit more cynical than your take, but it
still results in the same thing--there will no doubt be tool support
for JSP. WebMacro can get there too, but via momentum based on merit,
rather than on marketing and promotion.
But, in the end, a successful opensource project winds up being better
supported than a corresponding proprietary one--even if the proprietary
one is a standard. Many people find this hard to believe, but experience
bears it out--people using opensource software tend to support one
another in the beginning; and later on companies find that, since
they have the same rights to the opensource software as the original
author, they can earn a good living by supporting it.
Now I'm clearly biased in this debate as the primary developer of
WebMacro. However, I will hold this argument up to the light of day in
any domain you like--in all cases, standardization is less important for
opensource software, and support winds up being better, and cheaper.
> I haven't really used JSP or WebMacro (or any other "template"
> system), but it strikes me that being standard is a big win,
> especially if that standard is moving towards incorporating the
> functionality and advantages that the other systems have (and that
> seems to be the case).
Technologies integrating the functionality of other technologies always
have a bit of an advantage, but that has nothing to do with standards.
That's called a feature war. Open source software can win some kinds of
feature wars, and tends to ignore others. For example, Apache has been
very successful in the feature war in the webserver market. On the
other hand, Mozilla may find it tougher to win the feature war against
IE. Time will tell.
In order to win a feature war against JSP, WebMacro will have to attract
a significant number of developers. But, I am very close to releasing
the core functionality as 1.0, and when I do that I plan to open my CVS
archives up to the world. I'mve modulaarized WebMacro similar to the way
that Apache has been modularized just so that people can write useful bits
and contribute them to the core.
So, I am confident that WebMacro can win a feature war against JSP, because
I am confident that I can amass a large body of programming expertise
for free, just like all the other opensource projects :-)
Our biggest competitor from JSP will wind up being gnuJSP and it will
quickly abandon the JSP standard in favour of sensible design and
implementation as people add to it what they need, rather than what
some committee says they need. But this doesn't worry me either because
opensource technologies don't really compete with one another--since either
project can freely integreate technologies from the other, they tend to
strenghten one another (ie: Linux vs. FreeBSD has been good for both).
Justin
___________________________________________________________________________
To unsubscribe, send email to [EMAIL PROTECTED] and include in the body
of the message "signoff SERVLET-INTEREST".
Archives: http://archives.java.sun.com/archives/servlet-interest.html
Resources: http://java.sun.com/products/servlet/external-resources.html
LISTSERV Help: http://www.lsoft.com/manuals/user/user.html
