|Perl: the Markov chain saw|
Re: Can a new name or version number change the way Perl is perceived?by BrowserUk (Pope)
|on Feb 15, 2013 at 20:06 UTC||Need Help??|
Re-branding doesn't work in the long term.
Years ago I rode motorbikes. I, like almost all my biker friends, insured myself (not my bike!) using what was known as a "Norwich Union Rider Policy". This insurance policy from the oldest and most well respected Insurance company (in the world), insured the rider to ride any bike up to the capacity you specified. You changed your bike, no problem as long as it was under the insured capacity. You borrowed your mate's bike; same thing.
It was reasonably priced; simple to administer; and this old, staid, even boring company did a steady and profitable trade amongst this group of customers that other, more 'hip'n'trendy' companies looked down their collective noses at.
Some time in the early 2000's as the dotcom bubble was about to burst, and there was this flurry of internet-based virtual Insurance companies sprang up -- usually with ridiculous names and worse, but prominent advertising; this old, staid, boring, but trusted, fair and commercially successful company decided that it needed a makeover. They became "Aviva". And with it, they threw away all the associations with the "oldest", "most trusted", "fairest", "dependable" that people read into the "Norwich Union" name.
As Aviva; the went for catchy adverts featuring a well-known comedian; and ... nothing. Many years on they are still having to advertise their wares on a daily basis; and they are seen as just another commodity insurance company offering new customers the same £XXX savings that every other insurance company is offering -- until you sign up at which point your premiums will inexorably rise year on year until you decide to switch. (And in the process, the innovative, unique products they used to offer, bit the dust.)
Re-brand and you may attract a little new attention in the short term, but unless you have something new to offer, the truth will out. And in the process, you may well alienate and loose your core customers.
Perl(5) still has a lot of advantages over it 'competitors'. It just needs to make greater play of those; instead of continually navel gazing over the dark, distended corners. It also needs to innovate rather than titivate. 5.8 brought threading; 5.10 brought defined-OR, say, state, named-captures, (given/when but that's probably best not mentioned). What has come since? Does anyone outside p5p know without going off and looking them up?
There are so many areas of Perl that could be improved; but that require more than one random individual beavering away alone to tackle. (From my own areas of interest: a) threads::shared could be so much more flexible, faster and less crufty; but I've looked and looked and I don't know how to begin. b) function call overhead is just too high; c) arrays and hashes are incredibly flexible; but far too memory hungry for single-type data -- which is common case.)
And CPAN. Some call it Perl's greatest asset, but for every good module there are 100 dogs. And for every great module, there are a 1000 that could be substantially improved with collective effort; but that requires a mechanism for collaboration. A "Selected CPAN" would be a great asset. A subset of modules that are collectively chosen; and collaboratively maintained and improved; and renamed into a Std::* hierarchy. A place where people could go first for generically applicable modules in the knowledge that they are maintained in an ongoing, best-of-bred manner by the collective wit; and not likely to fall into disrepair when their author gets bored, moves on, or dies.
As for Perl6 ... I'm too scared to comment.
With the rise and rise of 'Social' network sites: 'Computers are making people easier to use everyday'
Examine what is said, not who speaks -- Silence betokens consent -- Love the truth but pardon error.
"Science is about questioning the status quo. Questioning authority".
In the absence of evidence, opinion is indistinguishable from prejudice.