Commenting for the first time in over ten years...

*EVERYONE* at Google is required to take courses about how to carefully write things with a mind to legal discovery (ie, *anything* you write needs to done with an assumption that it will be used in the most negative possible way, regardless of actual intent on your part). There's no way this person didn't know that.

Technically, you pay to turn on the headlights, on every car.

I'll show myself out.

So how's your lawn doing, this summer? Mine's great, the neighborhood kids all grew up, and taught their own children to respect other people's lawns.

Nah, they thought it was just Voodoo.

I'm scratching my head on that one. It seems like you could have said "for an asparagus eater", for all the difference it makes to your argument.

Oh, I get it. It is a proxy for a lower bound.

Mortality rate of all injected seems like a bogus stat? Seems like you should only concern yourself with final outcomes, when calculating mortality rate.

Only since last century? Pshhaw, I'm here since last millennium. Get off my lawn.

I know you're an AC, but lest some reader thinks your reply is insightful:

* "immuno-compromised" includes people taking immuno-supressants, such as those with Crohn's, Ankylosing Spondylitis, etc.
* some people attending may have a newborn or elderly at home, and they could take a new infection back home with them

ah, it's the ol' Slashdot read-a-roo

Do we say "username checks out" here, or is that only a Reddit thing?

I find it increasingly hard to understand what value people get from trolling on slashdot, now. I mean, the value proposition has always been questionable, but now that /. is well past its heyday, the comment sections are a ghost town. There's not really anyone around to give trolls the attention they crave.

Hold up. You are claiming that an application of De Morgan's Law on that simple SQL query sped up your run-time by 360x?

I don't have much experience with SQL, so take this as just a request for info with merely a tinge of scepticism (rather than full-on scepticism.) But:

(assuming 'a' and 'b' both represent SQL queries)

* Your 'orig' looks like it would often get away with just 1 query (i.e, if 'a' proves to be not-null, then no point querying 'b')
* Your 'new' looks like it would always perform 2 queries

Which leads me to the following observations:

* 'orig' actually looks to be more performant that 'new'
* Even if you just accidentally swapped your telling of 'orig' and 'new' in your comment here, still at best you are looking at a 2x speed up, not the 360x speedup you are claiming.

Thanks for the recommend, I'll look into it.

Systems programmer here, for a semiconductor company. I've used a Chromebook Pixel for 5 years now, at work. I used the browser for our intranet sites (including Outlook web access) and an NaCl-based SSH client to access bigger iron for development (tmux/vim/gcc). Chrome OS worked amazingly well for me as a thin client, where I could work in the office or on the road, and never store company IP on it. Spreadsheets and presentations were done with Google Docs. I did not feel in anyway like I was using a Fisher Price toy or Coleco computer. My only gripe was the mapping of the F1 - F10 keys... which was a problem with the physical device, not the OS.

Now that that model has been EOL'd for new Chrome OS updates, it has been repurposed as a Gallium OS machine. The only improvement that has given me over Chrome OS is X-forwarding.