Tell me, who cares. I don't as there are better platforms to develop software
with
and have been in the hands of fossies for quite some time now.
Steven J. Vaughan-Nichols asks whether Java's move to GPL is too late? and my answer is a firm ``Yes, Sir.'' I don't care what took
them so long to make that step.
They f***ed all those people for nearly a decade that tried
to cooperate with them to get Java running on different platforms than
Windows.
If you followed the history of Blackdown Java 2 you know how stingy Sun has always been to answer technical questions
that these guys asked again and again without getting a sufficient
answer. Blackdown tried hard to cooperate with Sun to bring Java to Linux
but their stack was never a vital alternative to using the original
distribution downloadable by Sun for free (as in free beer). And programmers
that first of all want to get their
own problems solved don't care as
you simply risk to get fired for using non-standard tools if things go belly up due to incompatibilities between different Java engines. For a long long
time there was a pressure to get a free Java implementation. Just count
the lines of code other people where willing to write just to get around using
Sun's Java. Good code, that will probably never flow back into an open sourced
Java.
Yes, Java and all its APIs still runs best on Windows, Microsoft not being
a particular friend of Sun and vice versa as you know. Nevertheless,
Sun developed Java for Windows first as this is the biggest market.
They always had to catch up with the underlying operating system and its moving
APIs trying to lock out competitors. That surely was not cheap.
One of Java's key selling point has always been platform independence: ``write once -- run everywhere''. So far the theory but things are always different in real life.
Trolltech had a nice twist on that meme rephrasing it ``with write once --
compile everywhere''. The differences between
both approaches to platform independence are compelling.
Java plain fails to run on multiple platforms for technical reasons or just
because Sun did not deliver the equivalent software stack for other platforms than Windows; Java is slow as its byte code is still to be interpreted by a virtual engine even though you need a separate compile step.
Qt, Trolltech's key product, on the other hand, is simply yet another library you link your programs against the result of which is a highly optimized executable. And you can do so on whatever platform
you decide to support. Plus the code is more efficient.
I could go on comparing Java's APIs with Qt for quite some time where
Qt wins hands down, e.g. library design. But let's stop here.
In some respects Java on servers has to be judged differently than in
browsers or on the desktop. Let me make it short: it is OK on servers but
sucks everywhere else. Applets are dead. Swing sucks. It is too complicated. It looks bad.
It is ugly, out of the box.
It does not integrate with the rest of your desktop. If it does then it
only integrates with a Windows desktop. Fonts are a nightmare. Themability
and skinnability of its widgets is voodoo -- at least to me.
Next point: who is interested not only going to looking at the masses of source code but also understand and fix it? The anonymous Open Source masses?
This project is much too large now to. It was locked away for too long. Things are fine if a community is given a chance to mature and grow as the software does it is grouped around. This is obviously not the case
for Java. Just have a look at Open Office or Firefox. These projects have
big big problems to get more coders involved, people that not only work around
symptoms but understand and fix the underlying issues besides envision sensible
enhancements. How many coders work on Firefox? Who is currently investigating
buggy printing? How
long does that bug persist? Is this issue being a key feature for browsers really
5 years, 11 months, 3 weeks, 6 days, 14 hours old with a long thread of comments circling
around it. To be fair, we all have such bugs buried in you bug tracker too.
But hey printing is so essential ... gosh.
One comment from a guy called Wells dating back to 2006-10-13 reads like this:
While I wish I had the skills necessary to fix a bug like this I don't,
but I have hung around now a year and a half to see what would happen
and I find this bug an excellent case study in the Cathedral and Bazaar
model of software development. It seems to defy traditional claims that
the bazaar model is a superior development model when it comes to fixing
top100 rated bugs. The wider community has not hunkered down and come
up with a fix to this critical printing problem. Now the bug fix is slated
for Firefox 3, if we should be so lucky. Keep up the good work!
Not quite encouraging, eh? Don't forget that this browser has its own knotty history and this particular bug may be a good indicator of what the situation
currently is for the Firefox project. As I trust in history I expect things to be even worse for Java. The community already moved on to different
techniques. Will it come back to Java? Let's place a bet on ``no''.
Tags: community, firefox, java, linux, microsoft, qt, trolltech