-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

I was reading L<this|http://use.perl.org/article.pl?sid=07/09/11/0716209>
today, and realized that I'm sick of C<::Tiny> modules.  The reason is
not because I enjoy bloat, but because sticking to standards is important
to me.  L<XML::Tiny> is a great idea, but ... it's not XML.  It doesn't
help anyone to have an XML parser that doesn't parse XML, and it certainly
doesn't help to have Yet Another XML Parser -- there are plenty already!
Why bother writing one that doesn't even work?

The same goes for YAML::Tiny.  Frankly, neither YAML nor YAML::Syck
are hard to install, nor are they all that bloated.  Why bother making
a Tiny version that doesn't even work right?

The linked article is a proposal for JSON::Tiny, and I just don't
think we need it.  There are too many JSON parsers already.  Instead
of writing one that's broken by design, it would be a better use of
time to clean up the existing ones into a single JSON module that has
a nice interface, is well-documented and tested, and works correctly!
I've never heard anyone complain that their JSON parser used too much RAM,
but I have heard that it's a problem to load 5 parsers into one app because
each higher-level dist on CPAN uses a different parser.  Let's not
add another to the mix!

(Come to think of it... anyone want to unify XML::* into the One True
Light XML Parser, and then rewrite the various web service API modules
to use it?  I would chip in some time and funds for that.  I hear 6a 
already does this internaly for Vox.  It's a shame that they had to
waste effort reimplementing stuff that already existed.)
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.6 (GNU/Linux)

iQCVAwUBRueVf9AZeFPdJeQvAQIW1QP/bCEHST5EfAt8xr1K9R7xSpTp+l2K6xeS
hZUnLxZwmcQWuhcmczck7reMey8JI6dcT5oGuCwEC/9K9ILXM0SSJJkBnOhLTVcN
TRhozyiHU6N0rvdmMIhFj08PZsOQmOdXfr+Rx4+4Fp3G8Gkju1Ffxh5pGovrQ5tk
WdT36etMQME=
=Tkqg
-----END PGP SIGNATURE-----