The problem is that the tools we use make it easier to do the wrong thing than to do the right thing. Take SQL injection, for example:

When I started using SQL (late 80s / early 90s), parameterized SQL was pretty much the only way to go. I did actually figure out how to execute an arbitrary string of SQL and capture the resulting columns, whose names and data types couldn't be known until runtime. But it was a royal pain, and I doubt most people would have been that persistent - if at all possible, they'd have stuck with the easy way, which involved using SQL parameters.

Fast-forward through a few decades of "advancement" in tools, and now it's trivial to execute a string of arbitrary SQL (just begging for SQL injection), and SQL parameters are this strange thing that people may have heard of, but they're hard to use and they seem like more trouble than they're worth. So far too many people just do it the easy (i.e. INSECURE) way!

I'd like to hear more about these ways of proving that the machines weren't hacked.

Oh, and by the way, none of that high-falutin' "crypto-whatever" nonsense that only a PHD can understand. Those are the same propeller-heads that can't seem to go a week without some major website getting hacked.

If it's going to convince the common man (and perhaps especially the uncommonly-stupid man), it's got to be very simple and straightforward. If your plan is too complex for a simpleton to understand it in 10 minutes, then it's ALREADY failed at the task of maintaining confidence in the system - no matter how technically correct it might be.

See, this is why electronic vote counting is such an abysmal idea. It's not just that the vote totals can theoretically be hacked (though that's bad enough), it's that there's simply NO WAY to prove the totals WEREN'T hacked. If a group of people decides that the election was hacked, there's no real evidence one way or the other. This undermines faith in the system REGARDLESS of whether the election was or was not hacked!

So we're putting the foundations of our system of consensual government at risk just to save 1 sheet of paper every 4 years? Look, I'm all for saving the environment, but is this really the best way to do it? Maybe the newspapers can agree to sell ONE LESS PAGE of advertising on ONE DAY out of every 1,461 days instead? Or maybe we can all agree to buy one less book in our lifetimes? Or maybe we just agree that this is one situation that really IS worth "killing trees" for!

But however we justify it to ourselves, can we PLEASE go back to paper ballots?

Where are my moderator points when I need them???

So... every time I pull the phone out of my pocket, it's going to activate some feature? No thanks. P.S. Frist!

I'm afraid they're not very good at math: 4,000 people at $17K - $24K per year for 3 years is anywhere from $204M to $288M. So it's going to cost roughly 5X the $50M they claim. And is there any chance at all that they're going to be able to stop the program 3 years from now???

I just like the name "Offline Streaming". Streaming is, by definition, online.

As opposed to coal and other fossil fuels, where we have a very effective way of dealing with the waste products: just let them go up the smokestack!

P.S. You do know that coal mining releases more radiation into the air, and kills more people, than nuclear power - right?

Where are my humor mod points when I need them?

The "supermoon" may look slightly larger than the moon normally looks when it's close to the earth (i.e. every two weeks), but not significantly. In particular, it pales in comparison to the psychological effect of viewing the moon low in the sky (when it looks larger, even though it's actually further away) vs high in the sky (when it looks smaller, even though it's actually closer).

But there's still no reason it should be in the news: regular people wouldn't even notice if it wasn't pointed out to them, and astronomers know it's just a cyclical thing that doesn't mean anything. Imagine if, every spring, the news was full of breathless reports of the upcoming "supersun" which would be much brighter than the "winter sun" and which would stay in the sky for hours longer than the winter sun. That's pretty much the level of amazement we're talking about.

...and in other news, Generalissimo Francisco Franco is still dead and water is still wet.

I agree - it seems like every other month or so there's "gee whiz" story about a "supermoon" or "supermars" or some "celestial alignment" or something. But at best it's probably no more than a fraction of a percent bigger than normal, and it doesn't actually mean anything. If they didn't make a big deal out of it, no one would notice or care.

It actually could mean something: if there used to be an older 2-gallon-per-flush valve, then the 1-gallon valves save 1 gallon - and the 1 pint valves saves 1.87 gallons (compared to the 2-gallon valves), which is 87% more than the 1-gallon valves did. But I seriously doubt that's what they mean, and even if they did, do they actually expect people to do that math while they're peeing?

There's one I see regularly that baffles (and disappoints) me: on the top of the flush valve for public urinals (sorry, I'm a compulsive reader) it says "This 1-pint-per-flush valve saves 87% more than standard 1-gallon valves". What the hell does "saves 87% more" mean? Uses 87% less, fine - but saves 87% more??? WTF???

No, it was a totally different problem.

Y2K was about an optimization made early in the history of software development, when every bit and byte was precious, and it was expected that the software would be replaced long before it became a problem. Well, not all of it got replaced before then - but everyone knew the problem was there, and exactly when it would bite us, so a lot of people worked hard patching system so that there were no major problems. And before you sneer at the short-sightedness of early developers, let me ask you this: how many of YOUR programs are Y10K compatible? Or Y2037 compatible? Or Y65536 compatible?

This is about security flaws (some due to criminally-negligent designs, some due to inevitable software bugs made even by skilled developers) that are NOT known about in advance, and that CANNOT be patched when they suddenly become a problem.

The definition of a networked system is "one you can't use because some computer you never heard of is down".