Quoting the Free On Line Dictionary:
"trend - a general direction in which something tends to move;"
Define the "something" and the area of movement and then you will be
talking business.
Is there a trend of developers moving to Ruby and Python? Sure. Is there
a trend of developers that contribute to CPAN, as CPAN is expanding?
Sure there is. And it can be easily quantified with numbers.
So, the question is which trend should one pay attention to characterize
a language as trendy, or buzzy, or whatever. Depends who you talk to. As
other folks pointed out, go to an OS developer and ask whether C and C++
is trendy (and then you might elaborate whether you have seen many
usable OSes written at their very core in another language). Shall I
remind of other OS debates (monolithic kernel versus micro kernels) to
have fun, and remind what was (is?) trendy and usable?
Then I paused to read Steve Yegge's rant. I am not sure I would pay
attention to something that someone writes for fun between glasses of
wine. When I drink wine I become less hostile (obviously wine has
different effects on different folks, or Steve was confusing wine with
Red Bull or other equivalent ) and I have better (in my view) things to
do. I would agree that designing a language is very hard. I would also
agree that Perl could have better OO semantics. I don't agree with his
view of references and the building of nested data structures. Views
that Perl is a human centric invention at its core are also wrong, don't
particularly care to discuss it here and now. I agree that depending on
the task you have ahead of you, you should pick the right tool.
Going back to the idea of being trendy...If I am a sysadmin and I need
to push an update ro 1000 machines and check dependencies my own way or
block an SSH probe looking at 1000 logs, I still do not see why I should
use Python or Ruby. I am not saying you can't do that with Python or
Ruby, but I can't see the point. Especially with CPAN's wealth and the
fact that my time is valuable to re-invent things.
If I am a bioinformatician, what's the relative amount of development
traffic and modules written in the BioPerl community relative to that of
BioPython and BioJava? Why Perl is still one of the best choices to bang
prototype systems to crunch biological sequences or even base entire
high profile projects in the area? Examples, well, have a look at
project ENSEMBL http://www.ensembl.org/index.html and other Perlisms of
the bioinformatics area?
If I had to do number crunching (core functionality) on a commodity
cluster, none of our "trendy" languages would be able to help, as C/C++
and some specific libraries (PVM/MPI) and other Domain Specific
Languages excel there and shape the trend.
The things that work under the cover are not always trendy. Trends vary
in different fields.
(I use Java for GUIs and client side, Perl for core text processing and
system automation, C++/C for HPC number crunching on a daily to daily
basis. I would not necessarily interchange roles).
GM
Shlomi Fish wrote:
Hi all!
There is an interesting discussion about Perl here:
http://discuss.joelonsoftware.com/default.asp?joel.3.546349.20
Quoting it:
<<<<<<<<<
For some reasons there are lots of buzz about Ruby, Javascript, Python but not
too much about Perl. Why is that? What makes Perl less trendy than those
languages?
As some people note, Perl is not as new as these languages are, and so has
become less trendy and more "well-established". There are other comments too.
Regards,
Shlomi Fish
---------------------------------------------------------------------
Shlomi Fish [EMAIL PROTECTED]
Homepage: http://www.shlomifish.org/
If it's not in my E-mail it doesn't happen. And if my E-mail is saying
one thing, and everything else says something else - E-mail will conquer.
-- An Israeli Linuxer
