Interesting article of the "Church of Objectology" type.
I actually liked this version of the story a lot better than Cringley's.
And I liked Rohit's original uncut version better than Cringley's too.
Needless to say, I don't quite agree, but I'm sympathetic to the points
being made. Java seen as merely job security for software developers,
indeed. Hey! You wouldn't want to put all of us sw developer
types on the dole, now would you? :-)
Incidentally, just to prove this whole interconnectedness thing really
works, I found it amusing that from the url above, you can follow a link
to
"InactiveX: What you won't find in Hell's Kitchen."
http://www.webreview.com/97/05/23/feature/index2.html
and from there to
"So, what exactly is Hell's Kitchen?"
http://www.webreview.com/97/05/23/feature/hellontheside.html
and from there to
http://www.hellskitchen.com ,
where it's possible to find a FoRKer
who quacks a lot, and who has a link to
http://www.hellskitchen.com/friends.htm
with several other prominent FoRKers. E.g. I can follow
from there to
http://www.cs.caltech.edu/~adam/adam.html
and quickly on to
http://www.cs.caltech.edu/~adam/LOCAL/people.html
and hence to
http://www.interlog.com/~resnick/forr/forr1.html
which isn't exactly me, but is as close as Adam comes to linking
to me. So, I counted 7 links from original Webreview article to me -
pretty close to 6 degrees of separation, right?
All together now, .. "it's a
small world after all, it's a ..." Whap!
Ron.
Here's a snip from the article:
----
...I think it's obvious that Java's
employment on the web is far more specialized and
limited than would be expected from the fuss made
over it.
Basically, these sites are using Java to solve
bandwidth problems. Heavy interaction in the
Web's client/server model does not work well over
a network. Otherwise, these sites could just create
a GIF on the fly at the server, send it to the client as
either an image map or an image within a form, and
get the same results, at least for the examples
shown above.
I'm fascinated by Java, and see it as the first good
chance in years against Microsoft's operating
system monopoly. Some recent tests I did on Java
performance blew me away: with Just-In-Time
(JIT) compilers, Java code achieves performance
comparable to an optimized C binary.
But Java's supporters claim the language
represents a completely new "paradigm" (see, for
example, the pretentious quotation from Taiichi
Sakaiya's The Knowledge-Value Revolution in
Sun's Java white paper). The Web --
HTTP+HTML+JavaScript+CGI -- certainly
represents a major shift in software development,
but Java at best merely complements these
revolutionary protocols.
So what's the matter with a bit of hype, or even a
gross amount of hype? Marketing Java, not as a
useful tool in some circumstances, but as a veritable
"paradigm shift" in software development, creates
serious misconceptions.
For example, I recently received mail from a
long-time OS/2 conference, saying they had
switched their entire agenda to Java:
"After producing five conferences
devoted to OS/2 programming, we
have concluded that we can no
longer focus exclusively on OS/2. The
majority of our customers have told
us that their companies are
reluctantly moving away from OS/2;
we feel that we must, with even
greater reluctance, do the same...a
dramatic new force is at work in the
marketplace, one which makes the
operating system and hardware
platform largely irrelevant. That force
is Java."
Now, if they had said that there is a dramatic new
force blah blah, and that this force is
HTTP+HTML+JavaScript+CGI, I would
understand, and agree. These simple web
standards are the real revolution in software
development. Plain Web standards are providing
the ability to create new types of applications, such
as Amazon.com. As noted above, check out the
IBM chess home page for an example of the
effects that can be achieved with HTML; yes, as
we've seen that site does use Java, but as we've
also seen it's hardly indispensible.
HTTP, HTML, JavaScript (and yes, CGI) are
undermining the old Microsoft-dominated way of
developing applications in C or C++ for Windows,
using complicated and poorly-documented APIs.
These Web standards make it radically easier to
create applications (which then forces you to focus
more on content).
I'm afraid that this is one reason why there's such
fuss being made over Java and ActiveX. HTML,
HTTP, JavaScript, and CGI are too simple for
professional programmers. Even if everything they
need to do can be done with these simple standards,
they're still too simple. Developers often want
"power" without being sure what they're going to
do with it. Thus, professional programmers often
"get" Java or ActiveX in a way that they don't
really "get" the power of HTML.
and ActiveX look almost like the
counter-revolution: An attempt to reintroduce
complexity into the development of applications
and, not coincidentally, provide job security for
software developers, and for people such as myself
who write about software. I mean, how many
books can you do about HTML? But Java: there's
an endless topic. Even if 99% of the dynamic
content on the Web is produced using simple
protocols like CGI and JavaScript, the 1% that's
handled by Java and ActiveX seems a lot more
familiar to many of us with a vested interest in the
old way of developing software.