oiuytgyuio

Yes, but that's no different from encrypting a file locally using truecrypt/pgp etc and then uploading that.

The problem is that Dropbox can't offer seamless encryption while still having lazy file sharing.

After various UK government departments lost just about every USB key some new rules were introduced about how they had to be encrypted.

Naturally this was hidden in a procedures manual several 1000 pages long.

But because the procedure dealt with encryption of classified data - the procedures manual saying how it had to be protected was of course secret! And couldn't be issued to the workers.

Depends who you are. If you're MSFT, IBM, etc then you don't even send a no thanks letter, you have nobody who isn't a certified HR person talk to a candidate and you don't allow any off the cuff remarks. The lawyer target cross section is too high.

If you are a small startup in a high tech hub trying to attract people, and especially if you are in a market where the programers you reject might be future customers then you're risk is different.

Companies often forget that the people they reject don't disappear. It used to be a big thing in merchant banks in the last boom to really grind and humiliate candidates - you aren't smart enough to work here etc. What they forgot is that those people went to work for companies that might need a merchant bank and that the victim is suddenly the customer.

Of course there is a flip side, if you interview 100 programmers in a small high tech town and piss off 99 of them by not even sending a 'no thanks' letter.

And each of them tell all their friends that you are an a-hole, then when every potential hire in Cambridge/ NYC/ Stanford etc has heard that XYZ software are a-holes you are going to have one hell of a job hiring.

Diligently following the prescribed rules and making careful measurements and calculations = Engineering.

Deciding how to measure something so you can determine what it's doing without knowing the internal rules = experimental science.

You don't have any company lawyers do you?

Most companies in the land of the lawyer don't even send rejection letters, just in case the candidate decides that it contains some grounds to sue you over.

It doesn't even have to be good grounds - they just need a sleazy enough lawyer. The more info you give them the more ammo they have.

Yes it sucks to simply not hear from a company, and it leads to negative recomendations about them to friends /colleagues /potential other recruits. You can still be friendly in the letter but I would be careful about telling someone why they aren't hired.

If two files uploaded by different users used the same key (ie the server stores the key) then you have no security anyway.

If you can detect that two files encrypted with different secure keys are the same then you are similarly screwed.

If the flake is 10m long it weighs many tons, adding 75kg of climber in the same gravity direction is negligible. Putting a torque on the rock along the breaking line is a different story.

(probably more like 50kg for skinny zero % body fat climber types)

The dropbox business model doesn't really work for encryption. They hash files and avoid storing multiple copies of the same file, this is great for our business - we can 'instantly' upload large files if another user on our system already has an identical copy of that file - but it means that files can't be securely encrypted.

Basically trust Dropbox about as much as you trust email, if it matters encrypt it locally

I think programming is becoming a science.

Computer programming used to be applied maths. You followed the correct lemmas in the language and if you didn't make any errors you had a correct program.

Now there are so many layers of indirection between you and the world (huge complex APIS, graphics drivers, optimizing compilers, hyperthreading, cache, etc) that the actual behaviour of a program - especially in high performance, real time world - is no longer possible to calculate from the published specs.

It becomes experimental science, you adjust various parameters and measure the result. Certainly to get performance but sometimes just to get APIs to work together.