Can someone remind me why Perl needs to be more popular?
My friend and former colleague John Tsangaris addressed this fairly well, but I'd like to expand on his points a bit.
For me, popularity matters for two reasons:
1. If you like Perl enough that you'd like it to be all or a big part of your day job.
Which leads me to something I've been wondering about lately: How many Boston.pm members can say that their primary income comes from being a Perl programmer? That Perl is front-and-center on their resume. And if you're an independent consultant, that you market yourself as a perl developer/expert.
Perl is a sideline in so many fields that it is quite understandable that language popularity holds minimal interest to some people. In niches like bioinformatics or system administration, Perl will probably continue to be used because it is readily accepted that it does the job better than the alternatives. But a system administrator isn't paid for being a Perl programmer, they're valued for their administration skills that happen to be enhanced by Perl skills.
I came upon Perl by accident back in the late 80's when a system administrator of a system I used at school decided that having a C compiler on the system was just too dangerous. (Back then binary distribution were rare, but anyone with an account on a compatible machine could of course FTP over all the binaries they wanted.) Perl proved to be a great way to bypass that limitation.
Once I learned how useful Perl was, it became my language of choice for personal projects. Professionally I was doing hardware design and software development in other languages, but Perl was always sneaking into projects as a way to automate things.
In the mid-90's when the web hit, I was among the few who was proficient with Perl, and presto, my professional Perl career began. I worked from that point on as a contractor, an independent consultant, employee, and eventually manager, with Perl being the focus.
(The work I did as an independent consultant was more solution oriented, where implementation details, such as the language used, were secondary. The diminishing interest in Perl has some impact on this, but less so than someone seeing work as a Perl contractor.)
But now there seems to be mounting evidence (if you follow job postings) that it is no longer viable to market oneself as a Perl developer. There are still jobs out there, but they're thinning, and a high percentage of the ones that are available are entry level or have amazingly low rates.
Just this week I was talking with a local company seeking a senior Perl developer for a 3 month project, and they were expecting to pay under $20 an hour.
I suspect this year is probably the last time I'll be hired specifically to do Perl coding. (I'm actively working on expanding my solution oriented consulting practice as well as pondering what other languages to invest my time in.)
Though my hope, slight as it is, is that perhaps Perl 6 will address some of the technical/mechanical perceived shortcomings of the language, and over time the language's reputation will improve among the broader IT market. (If we had the "mind share" meeting discussion, I was hoping someone might be able to give some opinions on what Perl 6 advances might address this. If nothing else, a highly successful Parrot implementation that can run Python or C# faster than their native run-times at least provides a platform for sneaking in Perl code.)
I'm also interested in understanding better why the low end of the market is being lost to PHP. I think there are things that can be done around Perl 5 today to address that market better. More on that another day.
2. Even if the job market doesn't influence your use of Perl, you might care about popularity if you'd rather be able to use the latest and greatest applications - wikis, forums, blogs, whatever - and customize them without having to learn PHP, Python, or Java. Sure, there are still lots of Perl choices available, but only half as many as PHP, and the gap is widening. Here are the top 10 languages used for projects hosted at sf.net:
http://sourceforge.net/softwaremap/trove_list.php?form_cat=160
# projects
C++ 14801 C 14311 Java 14070 PHP 10410 Perl 5682 Python 3767 C# 2203 JavaScript 2168 Visual Basic 2037 Delphi/Kylix 1698
(Similar to the language popularity stats at http://www.tiobe.com/tpci.htm that was mentioned on the list recently. Same languages in the top 5, but a different order.)
-Tom
_______________________________________________ Boston-pm mailing list Boston-pm@mail.pm.org http://mail.pm.org/mailman/listinfo/boston-pm
