« Seeking Venue for Teaching Modern Perl | Main | New Catalyst to Plack Port Ready for Your Review »

06/27/2011

Comments

Feed You can follow this conversation by subscribing to the comment feed for this post.

Mithaldu

I once applied for a company here in germany, Idealo, that does price comparisons. They are owned by the country's biggest media conglomerate. Their stake in Perl is that it drives the data retrieval processes.

When i sat there to talk with them they told me these things:

They were awaiting Perl 6 because it would be "much faster than Perl 5". They are frustrated that Perl 6 is so long in the coming. As such their management decided to migrate to Java because Perl 5 is clearly going nowhere.

This was less than a year ago.

Chip  Salzenberg

If mockery were not a powerful weapon, a lot of political battles would have gone differently.

The incredibly roundabout path to 6 as well as the 5/6 split open us to mockery, and are thus inherently problems, if only of image.

oylenshpeegul

Counterexample: I once had a manager-type ask me, "what's this about Perl 6 not being backwards compatible with Perl 5?"

john

Dear oylenshpeegul,

Can you give us a bit more context? For example was this recently or a few years ago? I'm only asking this because I did hear questions like this in the late 1990's and early 21st century but not at all least 4-5 years. I'm trying to figure out how big a problem this really is.

Dear Chip,

I can understand that, I do notice the occasional riff on slashdot by a troll. I am just not sure if there's a solution that is not worse than the truth cost of the problem.

Dear Mithaldu,

Did that Java rewrite ever happen do you know? My experience with hearing "we are rewriting this Perl application in Java", whatever the excuse is said there's a subtext that is the driving force. Often has to do with the Java dev team wanting to expand its realm of power within a company, for example.

Dear Everyone,

I agree that having too many easy excuses to diss Perl is not so good. Just I think if it wasn't about Perl5/6 confusion it would be something else. Back when I started programming in Perl, like in the mid 1990's, I was working at a company that was using heavy PL/SQL and the main developers of those projects were looking at Java a lot. I inherited a pile of Perl apps that nobody wanted because 'there was no future in Perl', as I was told (nearly 27 years ago!) That was long before Perl 6 was even dreamed of. So I think that there are other factors driving this Perl negativity amongst the out community Peer groups. And I think that that negativity is driven more by solvable problems (things I mentioned about, like deployment standards, co-development standards, etc., awesome documentation and communication) than by anything about Perl5/6. Although there is a bit of truth to the idea that this confusion hurts us, we are much more hurt by other things, and if Perl6 did not exist, or if we renamed the projects, etc., unless we solve the true underlying problems all the negativity and objections will just recenter on something else.

Just my opinion, and I have nothing except my gut to say this. Thanks for all the feedback!

john

EDIT for above comment, it was 17 not 27 years ago (Perl is not quite that old yet!)

Max

The real problem with perl is that it is not in the news as Java, Python, Ruby or PHP. It is not a problem of 5, 6 or more.

A basic manager (like many) will act functions of resources. Take a look at any job site, many many many people "knows" (note double-quotes) java and PHP. So manager wants to have the choice, so he chooses java or PHP, so the schools learn java and PHP to students... Cyclic dependency.

What they don't realize, is that it is as well hard to hire a good perl dev than a good java one. They have the choice over the number. But quantity has nothing to do with quality.

PHP allows to do web development easily, like perl for CGI 15 years ago. But PHP code without a framework, is as bad as perl code in CGI yesterday. They replaced HTML in code, by code in HTML. The same bad solution, but it works for beginners and they are happy. They can evolve to a framework later... or not. Between, PHP won on the web.

Ruby is a fully object language. Cool. Rails made it popular. Nothing can be done in Ruby that can't be done in perl. We have many frameworks now to do web dev, perl has autobox, perl is faster, but Ruby won: it was a new, clean language without the "X is write-only language" sentence hook to it because nothing has been written before in Ruby, so it can't be bad... :)

Java is very specific, because Sun then Oracle imposed it via its wonderful marketing, with the help of dollar$. For companies, it is comforting: Sun/Oracle is a big company, so java should be a good language, well supported. Do you remind why java appears on earth? Sun promises us a new chip. The java VM had to be temporary, waiting for the chip... Sun never released the chip, but java is now fixed with VM-JITized to overcome the absurdity of a VM emulating a "real" chip. Java won.

For Python, I don't know really why it is preferred over Perl. I think that when someone prefers Python over Perl it is because of the reputation of Perl, not for an advantage to Python. But perhaps I am wrong. Since, Google adopted it. It is a great help when a big company adopt a language.

The problem with perl, is that you don't have any linux distribution with perl NOT installed by default, or with the first packages. Perl is here and work fine. But it keeps this reputation of scripting-language.

That is wrong. Perl is not only a scripting-language.

With perl we can make big-app, long-running-daemon, without any problem, while consuming less memory than equivalent in java or python without sacrificing speed.

We have to talk about that in conferences, Perl programmers are not marketing-men, but we have to change that. Perl Iron Man Challenge is a big step, but it will not be enough... It will not be easy.

On the other side, we have to make choices in the community about the future of perl 5. Integrate a decent object system in core for example. Many ways to do one thing is good, but not really for beginners...

yarin reuven

hello
thank you very much for your great post.

i have learned a lot about it.

wish you all the best

yarin reuven

iSolution

Pearl X solves anything for me, I don`t see the need in any "newer pearl`s", unless something like hologram usage be added to newer ones, I`m fine with what I got.

The comments to this entry are closed.