Much more interesting than writing a CV is, well, almost
everything. But there’s a whole class of activities that you can, if
you wave your hands in a certain way, justifiably consider part of
writing your CV, even though you’re not actually writing your
CV. You could call this writing your meta-CV, if you were some kind of
Obvious example: deciding how you want the CV to look. Classic
meta-CV writing, lots of scope for experimenting. Mmm,
But you know me. I need to take it just that little bit
Two things stood out straight away: I wanted the CV under version
control, and I didn’t want to write it in Word (I’d rather have a PDF
version). The first was easy: I’ve got a Subversion setup for version
control, which I use for my web site. New repository for CV and other
non-web site creations: five minutes.
PDF? Well, OpenOffice.org
exports to PDF natively in version 1.1. But its standard file format
is zipped XML files, which annoyed me. How would I do diffs? For
version control zipped XML is as bad as any other binary
format. Explode, commit, implode? Too complicated.
Bah, I thought, and naturally decided to roll my own XML. Easy
version control. Easy editing (xemacs, of course!). And when writing
I would concentrate on the content, and not constantly fiddle with
layout. I already had a good idea of the structure I wanted to use
for the content, so the schema would be pretty easy.
How to get from XML to PDF? Through the strategic application of
more acronyms. Take one XML document. Whisk with XSLT to produce XSL-FO. Bake using Apache FOP. And hey poncho:
PDF. And while I’m at it, I could use a similar XSLT transform to give
me HTML sprinkled with CSS, no baking required.
And that’s where I am today: an XML CV, built to HTML and PDF
(with layouts more or less identical). All the content is tagged so I
can build targeted CVs by keyword – so much easier than
maintaining multiple documents in parallel.
Now bow down before my meta-CV mastery!
(No you can’t see it… yet.)