At the risk of being argumenative....
Elaine -HFB- Ashton wrote:
>
> Brent Michalski [[EMAIL PROTECTED]] quoth:
> *>
> *>Perl is there already, just many people don't know it. Maybe they
> *>should start being informed. How many "enterprise computing
> *>environments" (*nix) come WITH Perl? A lot.
>
> Yes, but does Perl have an IDE,
Yes.
As tom is fond of pointing out, and it's the UNIX CLI.
If that's not enough, ActiveState has announced two more
graphical IDEs that are more like what you're talking about.
http://www.activestate.com/Corporate/Media_Center/News/Press959117519.html
http://www.activestate.com/Corporate/Media_Center/News/Press959150636.html
And if that's not enough, occasionally when people say
"IDE" they mean "Syntax Coloring". Many vi clones and emacs
(and BBEdit?) support that, so add them to the list.
> corporate support
Yes.
It's called PerlDirect.
http://www.activestate.com/Products/PerlDirect/index.html
> and a slew of mainstream
> corp applications
Last I checked, qmail is written in Perl, and pumps out a
decent amount of email traffic world wide. I can't think
of a more mainstream corporate application than an email server.
http://www.qmail.org/top.html
During the recent lovebug headache, one of our SA's
who had just taken Dan Klein's Perl for SA's class hacked
a portion of qmail to display all of the "I LOVE YOU" messages
queued for delivery, then deleted them all.
Time to resolution: ~15 minutes, for a perl newbie.
The full-time Exchange folk took over 2 hours to get
Exchange happy, and a few more downtimes were required later, IIRC.
And if you prefer Sendmail to qmail, no problem! I think
Sarathy is at USENIX talking about the Perl plugin for sendmail.
http://www.activestate.com/Corporate/Media_Center/News/Press957676267.html
Mail servers are one easy to cite example. Is there a version
of SAP or Oracle written in Perl (or Java)? No. Do we need them? No.
Does their absences speak condemn Perl? No.
Do we need more than this tired touchstone? Most definitely.
If the Web is important to your PHB du jour, I'd point them to
Philip Greenspun's rants about how AOLServer has an embedded scripting
language, and that one factor alone makes dynamic sites so much more
responsive and easier to implement. And then promptly redirect
said PHB's attention to http://www.modperl.org/.
(Avoid if the comparison of AOLServer+TCL to Apache+mod_perl is too
strenuous.)
> to keep the pointy hairs humming in their chairs? No.
My local PHB started humming when he heard there was such a
thing as PerlDirect and ActivePerl for Solaris.
(Using terms "Quality Assured" and "Open Source" in the same
sentence provoked a PHB-in-the-headlights reaction.)
> *>> 2. Access to information will become more fluid as the limitations on
> *>> bandwidth become a thing of the past.
> *>
> *>Great! What does this have to do with Java?
>
> It's a red herring because you have to have content before you can blame
> the bandwidth.....besides, the size of the pipe has nothing to do with its
> fluidity.
Not reading the original article, I'm assuming that "access to information"
paired with "more bandwidth" is supposed to mean that we're going to be
doing more information processing than we used to do 10 years ago. Duh.
It's certainly not a red herring though. I'm constantly munging a ~100GB
dataset, and I'm certainly doing more with it now that it's nfs mounted
against a 100Mbit etherswitch instead of a 10Mbit hub. I'm mining it
more frequently, and I'm mining it more deeply.
Of course, this is a simple NFS mount, so access to network protocols
is a moot point. If I were doing this at a dotcom, I'd want to use
a programming language where I could use SMTP, SNMP, FTP and HTTP
seemlessly.
Like Perl. [*]
> *>Perl is WAP compliant too! All the programmer needs to do is generate
> *>WAP compliant code.
"WAP compliant code" might be running on an embeddable of some sort,
like a Palm IX or Nokia. I can't imagine Perl or VB WAP applications
running on a cellphone, but I'm sure Sun will cook up some kind of
femtoJava or attoJava to prove the point that you can run Java on
your pager.
> *> Hell, Visual Basic could be WAP compliant. The
> *>Perl modules will be developed when there is a real need.
>
> WTF does WAP compliant mean? :)
See http://www.wapforum.org/
> The Perl modules will be written when someone writes them....the need for
> them has been around for a while now.
>From http://www.wapforum.org/what/index.htm
What operating systems are compatible with WAP?
WAP is a communications protocol and application environment.
It can be built on any operating system including PalmOS, EPOC,
Windows CE, FLEXOS, OS/9, JavaOS etc. It provides service
interoperability even between different device families.
Those Perl modules will be WAP compliant, but I can't imagine
they're going to be very useful if they run on Solaris but not
PalmOS.
> *>Wow, I feel like I am in a time warp. Wasn't this being said when Java
> *>first came out? Java has been saying this for years hoping that someday
> *>everyone would buy in to what they are saying. Unfortunately, many
> *>managers have heard it over and over and are starting to believe it. We
> *>need to somehow prove otherwise.
>
> Yes, but it sells because everyone has been wanking and counter wanking
> for years. I wish someone would shout 'show me the money' at one of these
> conferences someday...
Sun demoed Java on Jewelry at JavaOne a few years ago, followed by
Java on a Palm Pilot the next year.
Someone *is* saying 'show me the money', and someone at Sun is giving good demo.
The problem is that on the one hand the marketing machine at Sun
is making unfufilled promises, and the R&D groups at Sun are feeding
the fire with "early demos" of those unfufilled promises, but never
ship product. (Anyone remember HotJava? Anyone remember a non-beta version?)
What does the Perl community do? Ship working code, skipping the demos
and the marketing department in the process.
> *>Large, successful projects to use as reference are helpful. The success
> *>stories are great, but we need more *proof*. Even though Perl is a very
> *>proven language, we are continually having to prove it again and again.
> *>Why?!
Because we tend to be engineers that prefer patches and working code to
marketing. Where would Java be if Kim Polese, Scott McNealy and Bill Joy
weren't on magazine covers at the rate of 12x/year talking about Java?
Probably as well known as Python, Perl or TCL.
How many cover stories has Larry done? Tom? Andreas? Jarkko? A lot fewer.
> I suggested a while back that a slogan of 'use perl; get laid' would work
> smashingly in this sort of market.
Sounds like that would make a good tshirt, but wouldn't do a damn thing
for PHBs who can't find PerlDirect yet insist it doesn't exist. Would
do even less for PHBs who insist Perl is slow, and use C++ instead.
The problem isn't reminding people that Perl exists.
The problem is convincing PHBs that Perl is ready for prime time.
Z.
*: A surprising amount of Java marketing literature reads just as
accurately after s/Java/Perl/g;
Funny, that.
