2007.07.16

ArrayML: quick and dirty XML for interop 2007-07-16T06:51:15ZTitled entry permalink

I've just published some code, somewhat cheekily called ArrayML. It's designed to turn PHP associative arrays in to an XML interchange format. Think of it as the SWX of XSLT.

It'll hopefully make writing mashups and screen-scrapers a bit quicker for me, and will mean less time writing PHP and more writing XSLT (that's a good thing, btw).

Comments | TrackBack

Open a bin, take Microsoft pseudo-standards and toss it in. 2007-07-16T14:16:33ZTitled entry permalink

Rick Jelliffe at XML.com has a list of corrections to Microsoft's Office "Open" XML. The biggest correction to Office Open XML is it's existence. It proves a number of things. Firstly, Microsoft Office Open XML proves that a bad data format that gets an XML syntax is still a bad data format. And the Office data formats are terrible. If you judged it like any other data format, you'd see it is terrible.

If Microsoft wanted to be Open, they'd make the Office standards available using crrently existing XML standards - XHTML, DocBook, SVG, XForms, MathML and so on. MSOOXML proves that Microsoft love reinventing standards, since neither MathML nor SVG are used - instead Microsoft have reinvented the wheel for both of these.

But Microsoft don't want to be open, they want to be pseudo-open. Pseudo-open means that you get all the commercial benefits of openness - ie. business and governments who specify that an XML format is used will be able to avoid changing to a non-Microsoft technology - without the actual benefits of open data formats based on technologies like XML. The actual benefits are that there is more competition rather than less.

If you take a look at an XHTML document, you can quite easily figure out how to do something with it. You may need to check the specification, or a normative schema, for the minutiae and particulars. But the actual data is sitting there in a format that human beings can comprehend with ease.

The true test of a data format is when it's cracked open without schema, specifications or reference implementation. If I open it up without syntax highlighting and can last for more than thirty seconds without my head exploding, it's passed the first test.

Microsoft Office Open XML does not pass this test. It's bloated, over-engineered, unsemantic, insanely complicated and there are not enough implementations to make it worth bothering with.

The cool thing about XML and RDF is that you can express what you actually mean. Office XML does not give you that (and it cannot - remember, garbage in, garbage out). HTML can scale from the presentational markup of FONT tags up to the rich semantics of microformats and embedded RDF formats. Plain old XML does this too.

I call for a boycott of Microsoft Office Open XML. If we are to use it, we should write one XSLT 1.0 implementation to turn it in to a sane format, run everything through that and then say "go fuck yourself" to Microsoft. Oh, except that we possibly can't do that because of Microsoft's software patents on the sub-technologies inside MSOOXML.

If they want to be pseudo-open with people's documents, then we'll give them pseudo-interest in response.

I've said it before, and I'll say it again. There are cool people doing interesting things at Microsoft. Why, oh why, aren't they making MSOOXML good? It's because they want to be pseudo-open. Let's not dance around the issue.

Right, rant over.

For more information about MSOOXML, see David Wheeler's article and Free Software Foundation Europe's objections.

Comments | TrackBack

Links from del.icio.us

 

Login with your OpenID:
No. 619
Tom Morris
Currently in: East Sussex, England
Usually in: East Sussex, United Kingdom
AIM: tommorris
YIM: tom.morris

I am a , an , like to code in and noodle about with and the . I also have a BA in philosophy from London, and am in preparation for an MA. My philosophical interests are in Victorian-era German philosophy, Kierkegaard, Robert Nozick, hermeneutics and current approaches to the demarcation problem in the philosophy of science. Musically, I like jazz fusion, soul and P-Funk. My musical nirvana would be a mixture of Beethoven, Miles Davis and George Clinton topped with a side-serving of Erykah, Jill and Angie.

Elsewhere:

  • GPG Key
  • del.icio.us
  • Flickr
  • Twitter
  • digg
  • Jaiku
  • LinkedIn
  • ma.gnolia
  • blip.tv
  • upcoming.org
  • MetaFilter
  • LiveJournal
  • CiteULike
  • Technorati Profile

RSS Feed Subscribe:

RDF

« July 2007 »
SuMoTuWeThFrSa
1234567
891011121314
15161718192021
22232425262728
293031 

View in month context

On this day in: 2006