I think moving the scribd link to the right of the source domain might help. Something like this:

http://i.imgur.com/PAElgoK.png

http://i.imgur.com/GPa0DS8.png (Standard browser zoom level)

That makes it a bit more clear that it's a different source for the same document.

I agree that that would be way better.

I'll bite - what's wrong with scribd ? (besides the fact you might "stumble" upon documents you're not supposed to read)

JavaScript, holds PDFs hostage, adds zero value (my browser can display PDFs just fine, thank you very much), operates a business based on copyright violation.

The last time I've checked, the worst thing was that it demands money once the document is old enough(!) Then, it required from you to have an account just to download the document. Finally, it didn't work without Javascript. For me, instead of adding value, it reduces it, so I decided to avoid it to save my time.

I know Scribd is a Y Combinator's company.

Not just Javascript but Javascript from about a dozen different hosts last time I tried it and quite a few of them seemed to be required. I avoid it too.

> I'll bite - what's wrong with scribd ?

For a platform that is supposed to enable the user to read documents online, it fails by not enabling the user to, you know, read documents online. Oh, sure, sometimes I might get lucky and be able to read the first 4 of 37 slides without the application stopping to respond or crashing my tab.

It's still very slow. They have improved in this regard, but the original performance was so bad that saying that is not a compliment. There's a pdf viewer built in to firefox now, which in my opinion is superior. With noscript, the text is not formatted correctly. With noscript off there's a banner across the screen, "download full view" etc. This banner is rendered correctly in chromium but not my version of iceweasel, where it is about an inch down from the address bar and obscures the text. Pages after the first one don't appear until some time after scrolling down. Finally scribd want me to sign in to download a pdf. I'm not sure what you mean by "stumble across documents I'm not supposed to read".

It used to have a really obnoxious and entirely useless UI around the actual document. But recently links to scribd point to plain PDF so I made my peace with it.

An alternative is Google Viewer: https://docs.google.com/viewer?url=https://noggin.intel.com/...

Isn't this done automatically with all submitted PDFs? At least I seem to recall that there is some integration between HN and Scribd (YC '06 incidentally, which might explain it), but there's nothing in the FAQ or anything.