> [CDC] isn't usually so sweeping in its statements, but with a holiday coming...

Seriously? If (like me) you were wondering whether that clanger came from the CDC itself or the vapid press (Wired in this case), it's the latter.

I'm pretty sure that the CIA and friends would pay all of the money (*all of it*) to have a box that could crack public key encryption. How feasible is this? Is it on the horizon, or one of those things (like practical fusion) that always will be?

That can't be the reason. On the margin, raising wages ought to make some of that work unprofitable, leading to the workers in question being let go.

Not taking a position on the study, but if raising the minimum wage doesn't lead to workers being let go, there must be some other explanation.

You meant that as a joke, but this is actually a good solution to the problem of bullshit self-valuations: Pass a law allowing anyone to buy a property for (say) 3x the owner's valuation of it, with no recourse by the owner.

Problem solved.

A more interesting question is how they managed to go from 59.6% white techies in 2014 to 50.7% white techies in 2018. That's an 8.9% drop in four years. Presumably they're not preferentially firing white techies, nor are they preferentially quitting. That leaves hiring, which for some reason has fallen off a cliff, versus non-whites. Is there any innocent explanation for this?

On top of that, female techies have gone from 16.6% to 21.4% over the same four years. Since many of these new hires are undoubtedly white, that means that white male techies have fallen even more than the above stats would indicate. We need more data, but I'm guessing 15 or 20% over four years.

This seems almost beyond belief, and yet no one else seems to be discussing it. Am I all wet here? Or maybe Google screwed up their numbers?

As a data point, I did an extensive job search this last year at 53. Although I'd heard that market value decreases with age, it was a bit surprising to me how sharp the drop-off was. I do, of course, hope that your experience is not like mine. But prudence suggests that you assume that your market value at 55 is similar to what your market value at 35 was, and will decrease from there. Plan accordingly and godspeed.

No. This problem is so difficult that even our best minds are unable to attack it...

I'm sorry, but you're too late. All of the programming jobs are already gone. :-)

It's stunning that someone could write an article like this and not be aware of Christopher Alexander's work on the subject. Highly recommend his book A Pattern Language.

Security clearances aren't about being "fair". They're simply about an estimate of the likelihood that the subject will turn out to be a bad actor, able to be compromised, etc. Fairness isn't a part of that.

Is that like making the Kessel Run in 12 parsecs?

It's okay. We won't think any worse of you than we do of Apple's marketing department...

These were noble experiments, but none are useful for practical programming, AFAIK.

I feel a bit wistful asking this, but is it really worse? Deadmau5 isn't necessarily better or worse than the NY Philharmonic, just different. And while computer-generated scores might be mediocre now, soon enough they will probably exceed what a human can do.

T-Mobile was great until I made the mistake of moving. Coverage unusable at new place, even though it was in an urban area. After many months of run-around from T-Mobile, finally gave up and ate the contract penalty, which was about $800. Never again.