Back when working at Sun, I did in fact see Gosling drinking tea in the cafeteria.

See extended data figure 1 (page 14) for comparisons with other polyols, including Xylitol.

The whooshing sound being played by a SoundBlaster AWE32!

I passed by the Creative Labs office in Milpitas, CA last week. Their company sign, unfortunately, is in a terrible state of disrepair, but thought about how great their sound cards were back in the 90's.

ChatGPT vs Dr. Sbaiso? Bring it on.

Hey, floppies were a great start! I came home from computer camp in 1994 with parts of a Slackware Linux distribution on 40 floppies. I was excited... my parents not so much. The compiler package fit on a whole 10 1.44MB disks and you could load up a usable system on top of a FAT16 filesystem (with metadata for long file names/unix permissions) if you didn't want to take the plunge and re-partition your hard disk.

There's no relation, but I'd happily do anything the song describes

8=D

Bok bok bok! Just barely!

The analogy is if you suspect someone of stealing your wallet, you are allowed to break into their house, search through it to find and take back your wallet, destroy a few things here and there to prevent them from pickpocketing in the future, and then call in the police to arrest the guy.

Oh, but if you made a mistake and destroyed some random person's stuff, well, you were still acting within the law.

Isn't associating non-Google app and web site activity with Google information sort of necessary in order for Google Now and Google Home to work properly?

House of Cards distribution rights are not solely owned by Netflix. Wikipedia indicates distribution is jointly owned by Netflix and Sony. Compare to Daredevil or Luke Cage, which I think was launched simultaneously everywhere, where Netflix is listed as the only distributor.

if I'd had a daughter when I stopped really giving a shit about this place, she'd have eaten cum at least once by now.

probably a son too.

but it's nice to hear from/about people other than Vlad. cheers.

Case law could be considered LD_PRELOAD.

What happened to all the /. posts about how there is an excess of qualified U.S. candidates and companies asking to raise the H1-B cap are just trying to pay people less?

Anyway, OP's problem is one I think is very common when you're actually looking for someone really good. Even if crypto or security is not the primary job, a senior architect/developer/designer will be able to do a much better job knowing about crypto and security for the same reasons such a person would do a much better job knowing about multi-threading or cache behaviors. Knowledge and skill in those areas will ensure the design and code starts out in a better state than otherwise. In today's increasingly security-conscious world even the most basic of applications and devices need team and project leads to consider security as a fundamental aspect of development.

A lot of answers to this post are basically stating security considerations are not important to the job or the questions are too specific. I disagree with that. (Although I do think it would be OK for people to make a few mistakes around details in an interview as long as they demonstrated proper understanding.)

Maybe a candidate does know how to set up a web site to use HTTPS instead of HTTP. Does that same candidate know why certain cipher suites should not be used? And that really only secures the public network communication. What ensures user passwords are not easily accessed while in use and not just while at rest? How do you protect sensitive keys, symmetric or private, like the one used to encrypt user data?

If you're putting together something super simple and turnkey like a personal blog then maybe you can get by just following examples you read online. But if you're actually developing a new application or device then your solutions will need to be customized to your needs and capabilities. And that's not something you can copy/paste out of a Google search.

I had tried using GnuTLS for a while in one of my builds (with libcurl, I think), but found it didn't always work right while OpenSSL did. I'm not sure if that is because I had to do something different with GnuTLS, but it just wasn't happy as a drop-in replacement.

Anyway, I don't think "trust should be earned" works. If you visit a banking or shopping web site, in what way are they supposed to earn your trust before you do business with that web site? I can't think of a particularly good way (scalable, understandable, and convenient) other than the "I trust X and X trusts Y so I can trust Y" approach we are using today.