You're attacking my use of the "obvious" word. Yes, the obviousness needs to be applied at the moment of the invention -- I agree wholeheartedly. I also agree that talking about obviousness of the already patented invention today is pointless: the information is published and everybody can claim that they had the same idea. We're on the same page here.
Now, the differences. The idea of storing markup and data in separate containers, at least one of which is XML-based is not something anyone with knowledge of XML can not think of. Let me re-read the patent in question. Yep, I got it right. While I can't turn off my hindsight, I can't also think that CSS doesn't fit into this, as well as many other technologies. BTW, i4i, in case you're reading, I didn't give you this idea and you can't use it, at least because CSS is at least 2 years older than issuing date of patent 5787449. Which reminds me of the prior art arguments, but I digress.
All that is neither here-nor there. You say it's not obvious -- I say it is. What really matters is the goal of the exercise. Patents were created when inventions were _hard_. Coming up with the idea of pumping air out of the glass tube with wolfram wire in it required lots of costly tests, trials and errors. Do you know how many stores stocked wolfram back then? Patents were created to _encourage_ innovation on the assumption that not too many people can afford innovating in the same field at the same time. Fast forward to today. Coming up with software patents today requires a whiteboard, a Word document and a phone to call a lawyer. I know, I have a few of those (equally embarrassingly obvious, I must add). Does it protect me and my invention _today_? Nope, since I can't afford to sue somebody even if I can prove beyond a reasonable doubt that the patent has been infringed on. Big companies plow over individual investors with their army of lawyers. The opposite, however, is not true (the in-app purchasing patent story is the point in case).
Another aspect of patent situation today is the shier vastness of the field. Who can afford doing patent searches for literally every line of code before it is written? So we all are forced to live on the minefield. This serves as an argument against innovation by individuals. And that brings results which are directly opposite to the original goals of the patent system. Which is not a good thing.