OOXML Headache

Yesterday I took some time to read some opinions of various people taking part in the OOXML debate. Today, just even thinking about the debate gives me a headache, so I'll briefly summarize my observations.
  • As with most debates of this nature, it seems to be a battle of halfwits. I didn't get the impression that anyone had actually taken the time to read the actual standard. This, however, didn't seem to stop people from participating with gusto. On both sides.
  • Apparently the thing is 6000 pages. Well then, do you have any idea how much information 6000 pages is. Try to visualize a 1000 page book describing a programming language, an API or a protocol. Now imagine six of those books and that you are responsible for implementing everything described in them with a team of programmers. Even just carrying six books a thousand pages each is going to be physically difficult. What common standards can you mention that have seen widespread adoption and a multitude of independent implementations, all perfectly interoperable, that represent this amount of information? Not many.
  • Those in favor of OOXML seem oddly persistent in their insistence that this is the right thing to do. The whole thing seems a bit rabid, not very informed and a bit too...coreographed. It has the feel of a campaign: make enough noise, ignore criticisms and try to make the thing stick by having thousands of PR people throw it against the wall again and again. Even when confronted with the facts that they are paid by Microsoft to blow the OOXML horn, some proponents come across as borderline dishonest. Hey, if Microsoft pays you money to push OOXML you are representing Microsoft and you should stop waffling whenever the question comes up.
  • ODF keeps popping up all the time. This isn't a debate about ODF. ODF having faults doesn't make OOXML better (or worse). Why not keep the debate focused on OOXML?
  • What's with the licensing? It is unclear to me under what conditions and terms you can implement the spec and share or market the implementations (or the spec for that matter). There is a lot of handwaving going on here and it isn't particularly confidence-inspiring. There has to be exactly zero uncertainties in this area for OOXML to even be something one would consider supporting.

No comments:

Post a Comment