Ya, they have me on HTML5 now, I am feeling better, no longer have the iphone fits I used to have and my jobs got better.

There - fixed that for you.

Check http://www.scribd.com/documents/30964170/Scribd-in-HTML5, it has a blue box to the right, with the title "Reading just got better", where you can switch to HTML5 mode
(I'd really say "HTML mode" since it works in IE too.... the whole HTML5 vs Flash argument for Scribd is just flamebait/publicity stunt).

Furthermore, I find their "major reason" that HTML5 supports all the major points of the site's previous functionality to be a blatant lie. To give one example - ok, HTML5 supports webfonts... but how exactly are you going to license the fonts from Adobe (or any other font foundry that doesn't give away the font for free)?

Don't get me wrong: the ability to select, search (*) and so on is great, and could be a very good reason per se to switch. But I don't think that the solution is to flame things up.... just go the Google way, they added HTML5 video on youtube (where possible(!) ) and didn't make so much fuss about scrapping a plugin that enabled them to have a business in the first place.

I'm pretty sure that this is going to backfire for scribd in the future, as they have set some not-so-realistic expectations with their messaging, in the hopes of getting lots of publicity. This whole HTML5 craze reminds me of the similar period when XML was fashionable and thought (by some) that it will replace SQL databases, and would become the universal-good-for-all-storage-format. Guess what, Oracle is still around :)

(*) Search doesn't really work in my experience... check http://www.scribd.com/documents/30964170/Scribd-in-HTML5. If you select text in a box you can then search (& find stuff in that box), but not in all boxes; for instance, try searching "me three".

You have to use it... it's actually very useful - at least for a n00b like me :) (but I'm sure it saves time for a "pro", too; I only played with it for a few minutes on a colleague's computer, at work).
Sure, it won't work directly on all situations, but you quickly grasp what it takes for it to work. For example, in the case of the penguins, I can bet that I could've removed them all in less than a minute (and remember that I have no artistic bone in my body). Just start right-to-left, make an approximate freeform selection for each penguin and press the "del" key; if not perfect the first time, re-select the artifacts and "delete" them again.
I haven't worked with GIMP, but from what I saw in an online demo, it's slightly more cumbersome to use. You can get similar results with it, too... but hey, if you're willing to put in the time, I'm told you can get similar results with the "clone brush" :) (so yes, there's probably nothing you can do with PS that you can't do with GIMP, it's just that you'll probably do it quicker with PS).

You see, the issue is that Adobe's reader ALREADY HAS this protection. It always did! Try reading the "researcher's" (notice the quotes) so-called attack, use a version of Adobe Reader however old, and see how it works - guess what, you get a warning telling you that the PDF is trying to execute code and you should only allow it in case you trust it.

Read the report people, this is a non-issue where Adobe's name was only mentioned because it is fashionable to bash Adobe for whatever "security" issues (saying Foxit had a security issue - because it did! - would not have been news; but put Adobe too in the press release - now you have something that people will read! ).