[Chicago-talk] Ovious question

Steven Lembark lembark at wrkhors.com
Sat May 23 22:34:12 PDT 2020


> No I'm not using Raku. I'm bitter about the language to be honest. I
> feel that it's extended gestation contributed to the stagnation of
> Perl5 and it's fall from grace. At $work, we have decided to do no
> new development in perl and actively migrate away from it. We were
> having problems finding candidates that knew perl well enough. The
> recent college grads we've interviewed didn't know perl at all.
> 
> So, no I'm not doing anything in Raku.

It's more like the Perl community shot outselves in the collective
foot marketing the language. 

Recall that no university ever taught Perl, so no college grads ever 
knew it. It was alwyas hard finding good Perl programmers, even in
the 1990s. The language is about getting things done, not fulfilling
the demands of a theoretical framework. At that point nobody has 
ever taught it, people picked it up on the job hacking *NIX servers
and websites.

Result: The people who did know Perl were largely SysAdmins and such, 
pople like MJD, Damian, and Larry are the minority in this community. 
None of us spent much time countering the various nitpicks, we just 
got our varous jobs done.

The world learned to hack dynamic code on Perl. We were *all* still
learning how to program as a profession in the 80's and 90's. As a
result, some really *shitty* code was writtin in Perl and a lot of 
people looked at it as said "Hey, shitty code gets written in Perl!"

Enter people who didn't like Perl becuase it wasn't theoretically
clean in whatever way they measure languages. They all did a much
better job of marketing than the Perl community. If people didn't
use Python they screamed bloody murder while the Perl community 
was saying "Fine, enjoy yourself, do what works." When Sun/Oracle
pushed Java as they way to make monkeys write Perfectly Kleen Kode
we did the same.

Net result; nobody really pushing Perl in the marketplace it got used 
less and less, which led to fewer people learning it, etc. 

Please note that none of this had anything whatever to do with Perl
itself, it was a massive lack of marketing by the Perl commuity. We
are by far our own worst enemies.

Calling the next language Perl6 was another massive marketing idiocy.
If our only purpose for developing Perl is ourselves then fine. But
in that case we don't really care if anyone uses it, or has any
interest in learning it, or if we can find a pool of people who know 
it. Otherwise, showing that we had a new language that looks more like
what people from other programming backgrounds expect these days then
picking a new name was helpful. By not changing the name until
after it was released we gave lots of people who don't think Perl is a 
"real" language the opportunity to pre-brand it as "more of that",
which we know it isn't.

Maybe with a new name people will begin teaching the language.

If you look carefully at the changes since around Perl 5.8 they
largely came from developments in Raku. They were largely not 
going to happen any other way because, frankly, many people who
work on Perl have no interest in changing it -- hell it's hard to
even get security fixes into it because so many people have "their
way" and fixing an obvioius bug would break it. 

One result of that was stagnation in the language. When everyone 
else started expecting coding to be a matter of plugging ever-
smaller pieces into frameworks we all wrote code that still looked
like C. 

Net result: Perl doesn't do things the way people understand how to
do them these days. That means people don't stop to learn it.

mod_perl is a perfect example: Designed by a bunch of C hackers,
we ripped the face off of Apache and dealt directly with all the
mess underneath. Writing anythig significant in mod_perl is hell.

Compare it to Plack: Much more hackable, flexiable, maintainable...
it's "modern". 

The entire "modern Perl" effort is an effort to drag many of the 
Perl community out of the 1990's midset on programming. Catch is
that the langauge can do what so many people cannot: Adapt.

The one suggestion I'd give you is you care at all about the Perl
language and community that comes with it, learn and contribute 
to Raku before the entire effort we put into it, defining the 
nature of dynamic programming in the first place, becomes a dead
end.

-- 
Steven Lembark
Workhorse Computing
lembark at wrkhors.com
+1 888 359 3508


More information about the Chicago-talk mailing list