Re: [tlug] Do you whitelist or blacklist utf-8?

[Raymond Wan <rwan.kyoto@example.com>]

On Sun, Feb 27, 2011 at 9:33 AM, Darren Cook <darren@example.com> wrote:
> P.S. I hate language wars, even when I'm joining in them. But it matters
> out there in the Real World: for many web projects there will be a PHP
> Quote, and a Java Quote, and typically the features and prices and
> schedule will be comparable. In truth the security, reliability and
> speed of the underlying languages are also comparable (meaning
> equivalent enough that it won't matter for the success of the project),
> and what the decision maker should be doing is comparing the likelihood
> of each development team being able to do what they say they can.
> It is just very annoying when the inferior team is chosen because of
> something the decision maker heard in a bar from a language advocate. :-)


Another IMHO important is the person's familiarity with the language.
A while back, I read about this study on speed of programming
languages where a set of programmers coded the same algorithm in
multiple languages [I think it was a professor who got a bunch of his
students/guinea pigs :-)].  Of course, compiled ones are faster than
interpreted ones.  However, there was a lot of variation between
people ... if I remember correctly, it was enough to not make the
difference statistically significant.  i.e., if you gave me a
functional language like ML, I'd probably try to code like a C
programmer.  I wouldn't be surprised if security and reliability are
equally difficult to measure.

Of course, information is all out there on the Web, but you still need
some experience to know what questions to ask or what to search for.

And managers like the one in Dilbert :-) would also take into account
how much they want to pay their staff, when to get the product out,
and whether having a 99% secure system is "acceptable".

Ray