Not surprising -- gasoline has a higher energy content than ethanol -- 34 vs 24 MJ/L.

So you really are getting less energy when you buy a gallon of E10 or E15 vs. a gallon of pure gasoline.

I've currently got VoIP through Google Voice. I paid $20 one time to port my number over and no other fees. Service has been perfect, using an OBI box to handle it.

Alas, google is discontinuing this any day now. But it was a great deal.

$7 in 1997 seems too cheap.

Did you get metered service back then? I remember that being an option where you only got X calls for the month and that reduced the price by like 75%.

Of course, even then you had to not pay extra for caller ID or touch tone (really? touch tone was extra?!?!?!) to get it that cheap. I suspect that there is some change in service between these two figures.

Sure, but I'm not sure that I'd call their use of Gamespy back in the day as a mistake.

Instead, it was a business decision with benefits (they don't need to roll their own and it already enjoyed a large customer base) and risks (the service might go away or be changed.)

Had they they benefits of your 20/20 hindsight, they might have made a different decision, but given the information that they had at the time ... it sounds like a good decision. And really, even if they'd known that the service was going away in 2014, they might have still made the same decision -- Gamespy served a need, and for a long while it served that need better than anything else available.

Nice combination of loaded question and strawman.

EA is incompetent because they didn't write Gamespy's servers for them in such a way to make them easy to migrate? Nevermind that the 1) EA didn't write them at all, and 2) the servers don't even need migrating -- the problem is that Gamespy is turning them off, presumably because they're not making them money any more.

Maybe EA made a mistake back when in actually using Gamespy ... but at the time, Gamespy was quite popular, and if this was a mistake, it was a mistake made by many companies.

These games use Gamespy's servers to find other users who want to play, then once found the computers involved talk amongst themselves. EA isn't running servers for these games that I know of, but even if they did ... Gamespy is still being used to find other users.

With Gamespy disappearing, EA can't just "migrate its servers to another host" to fix the problem. They have to update all the games and release patches, mucking with code that they haven't touched in many years. All this for a game that barely sells any copies at all any more.

In this case, I think EA is making a wise business decision. So no, in this case, EA is not incompetent.

It doesn't take a "megacorp" to need to go to all that trouble.

Even a relatively small company is likely to go through a similar process when their entire company depends on this system and it's complicated enough to span multiple computers. They probably won't have a dedicated team for it (and note that anything that requires a dedicated team is not pocket change, even for a megacorp) but it still requires a lot of resources -- it's usually way more than just rsyncing some stuff around, though if the OS is *nix, there's likely to be some use of rsync in there somewhere.

You really don't have a valid point, as the point you're trying to make doesn't even apply to the situation we're discussing at all, because EA doesn't own Gamespy's servers. What EA would need to do here is pay programmers to pull their old game source code out of mothballs and update them to support something other than Gamespy -- and this is likely not a trivial matter at all, and needs to be repeated for each game. Games that aren't making EA more than a tiny bit of money any more.

And don't forget that the hardware itself may be compromised.

... and this has already happened, with new stuff coming to light all the time.

And so far, most people don't really seem to care. Not enough to do anything about it, anyways.

1) who knows how far NSA has its fingers into everything. If they've hacked the voting machines ... perhaps they *can* control who we vote for.

2) it doesn't have to be the NSA. They may have the most resources and the most support from our government, but China could do similar things. And the part about getting back doors into open source software doesn't require a government agency at all.

The most recent poster child of vulnerabilities that nobody noticed was of course Heartbleed, but who knows how many other problems either 1) have been detected but not reported to anybody, or 2) were deliberately added but made to look benign? And it's always possible that the vulnerabilities aren't where you think they are -- for example, the idea of hacking the C compiler to detect when it's compiling /bin/login and adding a back door if it is is decades old, and it's only one of oodles of possible scenarios.

You can't trust open source either.

Devices like these often have "binary blobs" that aren't open source and could contain backdoors (one of the reasons RMS has been rallying against them, but probably not the primary reason), but even more fundamentally than that, it would be naive to assume that the NSA can't hire programmers to contribute to these projects and that they can't be good enough at what they do to make a backdoors that would pass a code review without being detected.

That said, at least with open source you have the chance to find such things, so there is that. But either way ... I think we're screwed.

Again, by your definition, in the real world ... almost nothing is programmed correctly.

And I imagine that Gamespy is far more than a single server. The server side is probably at least a rack of servers, with databases and who knows what else. And it's owned by a totally different company than EA, a company that wants to shut it down (probably because it doesn't make them any money) so it's not just a matter of "migrating a server".

Companies often spend weeks planning migrations of their services, and often the migration itself takes dozens of people weeks to complete. They often test their migrations on totally separate hardware just to make sure they understand all of the issues that might come up and make sure they can overcome them.

And even with all that planning and testing and redundancy ... they often still screw something up.

Blame it on being programmed poorly if you want ... but it's reality.

Longer than "rsync --archive --verbose /var/www/html/. new-host:/var/www/html/." takes to type.

But again, that's the extreme simple case. That'll serve you well for somebody's 1993 web site, though their "contact us!" form may require a little more work (though I do realize that this form doesn't fit into the "just static files" restriction I mentioned.)

But even back in 1993 that was simpler than most "real" services. Scott Adams gave a nice example of how people viewed complexity back in 1994 (and it's still accurate.). You can argue that anything that is complicated is not properly programmed ... and that's fine, but then again ... by that definition, the vast majority of stuff must not be properly programmed.

I don't know how complicated Gamespy's services are, I don't know how it's built. But I seriously doubt it can be replicated with a simple rsync to the new server. (Unless you rsync *everything*, and the new server has similar hardware to the old server and will sit at the same address in the same datacenter.) And of course EA doesn't even own Gamespy so they can't rsync it to begin with.

Yeah, good luck finding *anything* that's that simple.

Even moving the simplest possible website (just static files, nothing dynamic) to a new host is more work than that. (You could move the content itself with rsync or tar (though not with the command lines you gave), but the new server needs to be configured, the web server still needs to be set up, etc.)

If your definition of "programmed correctly" is that migration to a new host is as simple as you think it is, let me give you a hint ... by that definition, almost nothing of any value is programmed correctly. And modern systems, with clustered setups with failover across multiple nodes, multiple databases, connections to billing systems and the like are several orders of magnitude more complicated than you seem to think they should be.

In any event, this is moot. It's Gamespy that's shutting down, not some server that EA runs that's currently sitting under somebody's desk. In order to fix this, EA would need to dig the source for their old games out of storage, make sure they can still build it (for a game that hasn't been touched in a decade by them, this is real concern), pay a programmer to replace the bits that Gamespy uses to use something else, build it, run it through some minimal testing and release it. All this for a game that may not have made EA any money in years, and it needs to be repeated for a large number of older games.

It's a business decision. To update every game ever made by them would cost a bunch, so EA is wisely deciding to only support the more recent games or the games with sufficient demand. We could argue that they're not using the ideal criteria in deciding what should be updated, but ultimately they do have to draw the line somewhere.

My guess is that Gamespy has had very little development done in a long time and mostly just sits in a room of servers somewhere mostly running on autopilot -- costing money in hosting and power costs. I'm not sure how it is about making money -- do game publishers pay to use it? Advertising? In any event, if it's costing money but not making money, they probably told the developers if they didn't pay up they'd shut it down, and the developers didn't pay up sufficiently, so ... shut it down.

Preemptive correction: My 130,000 terawatt figure was a rough "back of the napkin" calculation. A more accurate figure would be 173,000 terawatts. The "250 watts/m^2" figure I gave was a rough estimate for what actually reaches the surface (1000 watts for the Sun directly overhead, divided by 4 for the ratio of a sphere's surface area vs its cross-section), but of course energy that's absorbed by the air also should be counted.

Either way, the heat we generate is so small that doesn't even really register. But by releasing CO2 (and some other gasses) into the atmosphere we can cause it to be better at trapping the heat from the Sun, and so that *does* warm up the Earth. Note that the greenhouse effect is not a bad thing for life on Earth -- without it, the average temperature of the Earth's surface would be something around freezing -- but we change how effective it is at our peril.

We may also be able to affect things by causing more or less of the Sun's energy to be reflected back into space by changing the overall albedo of the Earth. Our effect on this so far is small as far as I know, but it may grow more in the future.

True, but the heat that we generate is miniscule compared to the energy that the Sun throws at the Earth. The Sun throws about an average of 250 watts of energy at every square meter of the Earth -- that adds up fast.

The entire human race uses about 15 terawatts as of 2008. But the Sun throws about 130,000 terawatts at the Earth -- what we generate doesn't even compare, about 1% of 1%.

Now, these mirrors in space I mentioned could be used to cool the Earth too -- don't shine the reflected light on the Earth, but instead use them to shade part of the Earth. I think there are some international laws against such things right now, but such things could be a possible stop-gap solution to the problem of global warming. I don't know how practical the idea is -- it's probably more science fiction for now and has plenty of problems, but it's not totally unfeasable.