"Michael Mathews" schreef:
> [attribution repaired] Ruud:
>> [attribution repaired] Michael:
(Michael previously sent me an independent off-list reply; we're back on
the list now)
>>> As I gradually learn how Parrot works, I see that perhaps the idea
>>> of decompiling byte-code into language ___ is only a pipe-dream.
>>> But the point still remains--using the fact that one *could* mix
>>> languages X, Y, and P into your company's source tree is a very
>>> weak argument for Parrot/Perl6. I would say it is a non-argument.
>> Not really. Think about a Cobol-to-Parrot translator. You could for
>> example use Perl (glue) to add GUI stuff to old Cobol programs.
>> Just see it as a way to solve real problems. You don't have to use it
> I'll try to be more clear. The original question was seeking opinions
> on what the big gains were for companies to switch to Parrot/Perl6
> (someday). My point was that saying it would allow a mixture of
> languages to be used in an application is, in my real-world
> experience, not something companies are currently seeking, plus you
> don't need to upgrade to do it, so it isn't really a selling point
> worth bringing up to your local IT Manager type person.
Then don't use this point to sell it to that manager.
>>> it also means she now has to keep a
>>> Java/PHP/Perl programmer around and happy whenever one of three
>>> different languages might throw a buggie.
>> No, it is not limited by that.
> Huh? Let me give an actual example. A major broadcasting company I was
> contracted with needed to change part of their gigantic code base that
> dealt with a data source of live sports scores (which were
> automatically displayed on air). The code was all Perl :-) except for
> one chunk in Python--so guess where the problem was. This had to be
> fixed FAST and no one in the office knew Python well enough to do it,
> including me (the guy who wrote the Python was long gone to work at
> Google). In the end it was decided to rewrite that chunk in Perl. I
> can tell you, there definitely was cursing in the office that day, and
> I doubt anyone there would see it as a plus to have the ability to mix
> languages more easily. I just wouldn't put it that way if I were
> trying to sell Perl6 to a business manager.
I already knew from the start that you had a bad experience in this
Of course you don't want your car built partly with metric and partly
with English measured parts.
<http://en.wikipedia.org/wiki/Metrication> But that is not what this is
about. The feature is not designed to crash a spacecraft, but of course
somebody will still use it for that.
> In my experience Perl has an (undeserved) bad reputation in regards to
> large, long-term projects because it allows "too much" flexibility.
That will not change.
> Really that means management has to do work to set and enforce
> standards, but those managers aren't going to be impressed by hearing
> Perl 6 makes it easy to mix lots of different languages together.
> Nevermind how cool/useful I personally think that is.
I like it that Perl5 and C go so well together. I dislike Java, so I
don't care for it.
Let's just make Perl6/Parrot the ultimate Cobol "shell" as well.
"Gewoon is een tijger."