On Perl Names And Numbers
This was originally posted in response to this discussion on the p5-porters mailing list and similar discussions on blogs.perl.org. I am also reposting it here.
I started cooking a big editorial essay in my head, but I'm just going to try and boil it down…
So long as a fixed number ("5" in this case) is *prominently* attached to Perl, it will be seen as stuck. Specifically, it is stuck behind that thing called "Perl6".
You may understand the distinction between "Perl" and "Perl5" and "perl-5.16.2", but the rest of the world (including many developers) does not.
Others (notably Apple, Java, Mozilla) understand how to manage perceptions through names and numbers. We should learn from them.
Continuing the affiliation between Perl5 and Perl6 is harmful to both camps. Negative or false perceptions about either needlessly end up affecting both.
The Perl5 community must pursue its own best interests and cannot be held hostage by the name and number of another language.
Establishing a way forward through both minor and major (i.e. non-backcompat) releases to Perl5 will help to energize and accelerate development.
The Perl6 community will not take offense at anything the Perl5 community does. We all understand that we have common roots, but we are now on separate paths.
I will not argue whether Perl is thriving or dying, but I will assert that the language and community does not receive as much "buzz" as it deserves.
Rebranding, renaming, and/or renumbering are useful tools for generating "buzz." But these alone are not sufficient -- a compelling message is also required.
Initial reactions to these kinds of changes are always mixed. But those feelings will fade and the benefits will come over time.
Yes, this is mostly about marketing and image, but that doesn't mean it isn't worthwhile. Perceptions do matter, regardless of whether they are rational.
Not everyone in the community cares about marketing (nor should they), but we must not impede or discourage those who do. We are all on the same team.
If we want things to *be* different, than we must *do* things differently. Let us not be afraid to try.
Hacker, speaker, author, dad.
I agree with almost every point you make. I think some people are getting lost in the idea that we need a new version number in an of itself to attract new blood and hype to Perl. But that's not it at all.
The problem is not ever breaking backwards compatability. Yes it's great that Perl has a history of being extraordinarily compatible with it's past mistakes, but you can't be saddled with that baggage forever.
If we can't have a new major version of a Perl 5-style language then we can't ever break back compat in a meaningful way. And if we can't do that, then I don't see how we can progress past just minor fixes.
Hi
Glad to read some new opinions.
As for the version #, can those proposing a change for marketing reasons, please explain what happens if there is no resultant buzz. Just saying "Oops" is not enough.
My position is that any such change has to mean something special with regard to the structure/syntax of the language, and with its deployment and usability (meaning it ships with a small but significant change in the list of included modules).
I don't think anyone is saying we should change the version just for marketing reasons, at least I haven't seen anyone say that.
A change of version number shall reflect an advance, not solely be a marketing instrument.
We have quite some chances to do that now.
The different works on creating a usable MOP would be a real reason to do that. Modern perl has it's own hype in the perl camp and the syntax seems quite settled (think Moose or Moo). Having finally a single sane implementation in core would be a very good reason for a major version (even if it's named 5.20, I don't really care) and with some buzz around it.
Even having a roadmap would be considered a reason IMHO ;-)