XML size (was: no subject)

[Rob Browning]

linas@linas.org (Linas Vepstas) writes:

> Except that this statement is completely incorrect.  It wasn't brittle,
> it had no endian or architecture issues, it was easy to extend (a *lot*
> easier than the current xml format) and it didn't need to be a dead
> end.

FWIW, I'm fairly certain there were endian/architecture issues.  I say
that because just before the switch to XML I spent a little while
trying to fix them on the alpha.  However I certainly won't argue that
XML is the "great solution" so many people seem to think it is.  Heck,
I'd probably rather see us using scheme forms for the data.  Then you
get your lexical bits for free with a simple "(read port)".

> I, for one, would support work to put a good binary file format back
> into gnucash.

I'd rather see that effort put into investigating the use of
sqlite/postgres as the only backend.  Since we can now easily run a
dedicated postgres in a local sandbox if we want to, that makes things
a lot easier.

-- 
Rob Browning
rlb @defaultvalue.org, @linuxdevel.com, and @debian.org
Previously @cs.utexas.edu
GPG=1C58 8B2C FB5E 3F64 EA5C  64AE 78FE E5FE F0CB A0AD