Initially, I found it strange that some companies rejected the idea of mandatory, but non-exclusive, support for Ogg/Theora in the standard because of the possibility that unknown patents may apply.

The formats they suggested instead are definitely covered by known patents and are, like every format in the world, still exposed to the possibility of other unknown patents...

But this blog post makes a good point I didn't think of before: A lot of companies are already exposed to the risk of unknown patents for the codecs they support already; continuing to use those codecs doesn't add any new risk but adding support for new codecs does. For them it's both implementation effort and additional risk for no reward, given the similar quality of the codecs in question. So maybe it's not as crazy as it first seems.

(At the same time, it isn't crazy for the people who care about free standards, and who haven't invested in the known-patented codecs, to push for Ogg/Theora.)

Until recently it seemed like adoption of web standards was at the whim of the browser makers (and MS in particular, due to their market share, though AFAIK they don't have much involvement in the codec dispute). I wonder if a huge site like Wikipedia, which doesn't depend on advertising hits, can change the balance somewhat. I imagine if Wikipedia say "we're only supporting Ogg/Theora" (or whatever) then it will push the browser makers to support the format, whether they like it or not, to avoid losing customers who want to view the content.

Until now the formats sites use have been based on what works in the popular browsers (and mainly IE), but having the equation driven from the other end (within reason) seems like a more healthy situation. (I mean, cripes, we're still suffering from IE's late support for PNG alpha channels due to all the people still using IE6.)

OTOH, maybe Wikipedia will put the videos up in a way which suits all the browsers. (I hope not as if they don't push it I can't think who else will and it'd be nice if there was some single, agreed format and embedding. I don't personally care which format it is.)

That's a good point. Even more so with Vista and especially Win7 (if what I read about new codecs is still true) providing more codecs out-of-the-box.

I think the H.26x codecs being proposed are only in Win7 (if at all), so the browser/media-plugin vendors will probably still need to provide their own implementations, at least for people on older versions of Windows.

I'm assuming that indemnity regarding in-box code doesn't extend to 3rd party code that implements the same algorithms, but I might be wrong.

Sigh, software patents seem to cause so many more problems than they solve.

I do call into question the competency of people who claim to be battling browser differences whilst doing their work.

I think it's mostly an issue of experience.

If you've done web development for a while you already know which techniques tend to work and which tend to run into compatibility problems across browsers. People who only dabble in web stuff (like me) have to keep trying different things (and searching the web for solutions) until they find a method which works when faced with each problem. Having found something that works they now have a bit more experience and know how to solve that particular problem.

Experience is part of competence, of course, but I don't see a problem with inexperienced people complaining about something being more difficult than it should be. There are a lot of perfectly reasonable things you can do with HTML + CSS which don't work in some browsers for obscure reasons. You've just got to trip over all those holes in the road to know that they're there.

It's a bit like C++, which is more my bread & butter than web stuff. To be a good C++ programmer takes years, even if you're already a good programmer in other languages. There are so many caveats in the language which you don't know are there (or don't realise the importance of) until you trip over them. Most of those caveats happened by accident rather than design, and newer languages got the benefit of hindsight and were able to avoid many of them.

When people new to C++ complain about aspects of it I usually (not always!) agree with them and wish it wasn't so, and try to help them understand what it was they just tripped over, but I still really like the language. I don't think I'd call them incompetent, though. Maybe, technically, they are incompetent at C++, but I'd call them inexperienced instead.