Are you suggesting that if a measurable amount of CO2 was coming from natural sources that it would somehow be harder to sequester that carbon than if it was man-made CO2? Or that cutting our emissions in the face of a natural uptick wouldn't effectively offset it?

Most of the man made arguments are out there to satisfy the nutty 5 percent who are denying that it's a problem. Stop letting them set the agenda and stop responding to them.

What does it matter if it's man's fault? If it turned out that this was all part of a natural cycle that was going to kill us all, then we would have just as much reason to do something about it as if it was our fault.

The real question has always been how much do we spend in order to prevent the damage that is coming. This report seems to be saying "Please turn off everything". So, in order to prevent a large portion of humanity dying, we should stop using the technology that is currently keeping many of them alive. Studies like this are yet another diversion from a real practical discussion of how we make the best of the situation we're in.

I prefer that the county use a proper retention system. This was one of our case studies we used when selling such a system.

In order of best protection of taxpayers to worst:

-Do retention right. The email should have been easily accessible and delivered within a few hours.
-Don't do retention at all. The dirtbag walks, but we save millions of dollars (the $190K was from a single incident). Not ideal, but probably worth it.
-Retain data in a way that is very hard to retrieve, or hard to give a compelling answer as to why you can't get it. This avoids this expensive conversation: "Give me all the email between the Mayor and dirtbag X. I'm not sure if I can, we have this garbage bag full of backup tapes, it might be in there. OK, get started looking through them.

Yeah, they screwed up the implementation. But at least they provide a consistent way to pass a list-type argument to a database command. Once the implementation is hardened, everyone who uses it will be a little more secure.

Not only is it ironic, it's a good thing. If people use this module, then they are a single patch away from fixing every occurrence of this bug. If they don't use this module, then they have to find all of their code that is similarly flawed and fix each instance individually. It's not like doing a database lookup with an IN is a rare thing; roll-your-own implementations are likely to be broken too.

Microsoft SQL Server has both an XML data type and a table-valued parameter that can be used to pass an arbitrarily long list of values in a single parameter. Does MySQL not have an equivalent, or maybe it does and PHP doesn't support them?

I used to work at a company with 50,000 employees and a 2 billion dollar IT budget. We were not an IT company. Scaling that up would be 80 billion dollars for 2 million employees. The US federal government has a bit less than 5 million employees, so the number seems to be not only plausible, but quite low.

The retention policy itself is the accountability; I'm only advocating rigorously adhering to the policy. If it needs to be set longer, that's up to the people who set the policy. The important takeaway of my anecdote is that if the IT staff do a half-asses job of keeping documents beyond the required date, it can cost them a lot. Do it right or don't do it at all.

The county government in my example wasted a tremendous amount of other people's money by implementing a poor documentation retention plan.

The goal of an effective document retention policy is to identify documents that can be destroyed and destroy them as soon as it is permissible to do so. Old documents are a court case with a broad discovery order away from becoming a big cost. It's very cheap to say "the retention policy says these documents are only kept five years and we physically destroy them shortly after this date".

I know of a county government in New York that kept their backups tapes from their mail server as a method of retention. There was some political trouble with a mayor (who used the county's email system) and a contractor - suspicion of giving no-bid contracts or something like that. A request came to the county's doorstep for all of the email correspondence between the two for the four years the mayor was in office. The county had to buy a spare server and restore each monthly tape to it and manually pick out the email messages. It cost them $190,000. It would have been better for them to either have an effective archiving plan, or to have deleted them. Keeping stuff "just in case" is a horrible idea.

Of course, if these documents are being singled out for aggressive purging and other documents are not, then there may be some funny business going on.

I'm arguing because the statement that we both agree on is completely meaningless from a practical perspective. I stated it up front to show that I wasn't insane enough to disagree with the fact that unlimited is a very strong word and no plan could ever be "purely unlimited". Then I went on to show that AT&T's plan not only doesn't meet the academic definition of unlimited, but doesn't meet any practical definition of "unlimited cellular data plan".

You chose to point out that I agree with the academic statement while completely ignoring the fact that I said that it doesn't matter whether it does or not. If the only way you can think of the word "unlimited" is from an unrestricted academic sense, then you are speaking a different version of English than the rest of us. That's why this FTC position, or almost everyone else's view, doesn't make sense to you.

Nothing is unlimited from every angle, but that doesn't make unlimited a meaningless word. If AOL cut my 56k unlimited access back to 5k after 10 hours, I would ask for credit for anything I paid above a 10 hour plan. The lifetime supply of Doritos was likely advertised as a fixed number of bags per month, so that isn't similar. And if someone puts a tent over my property, they've taken away my unlimited access to sunlight.

Unlimited means that I get fair access to available bandwidth and I never get charged extra no matter how much I use.

AT&T's implementation of unlimited means that if I use more total data in a billing period than they think a typical customer should use, then my service is degraded more than other customers service is degraded. In many cases, they are throttling unlimited users at times when there is surplus bandwidth - that's just vindictive.

Tell me how unlimited has any meaning at all if it's interpreted that way and the company fiddles with your assigned bandwidth based on your cumulative monthly usage. That's the misleading part.

I used to have AT&T unlimited, but changed to a limited plan when I realized that my unlimited plan didn't behave in a way consistent with any definition of "unlimited" and behaved nearly identical to a hypothetical "3GB per month" plan.

Red light violation ticket costs are way out of proportion with the potential damage done. For example: I go through about 40 traffic lights as part of my daily commute. If I sneak through only one of them every day, then I could potentially owe about $40,000 in fines each year. I'm certain the safety aspect of a few extra cars going through the end of a red doesn't constitute enough of a safety issue to warrant fines at that level.

People in cars have good intuition of what is right. If a rule is being ignored, then it's probably a bad rule. For example, if a right-turn lane forms on the shoulder of an intersection every day at rush hour, it means another lane is needed, not that an officer should be stationed there to beat the people into submission. If people habitually speed on a section of road, then the speed limit is probably set artificially low. If people in a congested city go at the end of a red, then it's probably best. What they're signaling is that the dead time between the two directions going is too long and their using that time productively.

Also, I assure you that a few extra cars getting through a red light doesn't promote gridlock at the next one. There's really no conceivable scenario where the next light wouldn't have the same capacity as the previous one, or that if there is a lack of capacity then a backup would form and people would eventually not be able to sneak through the red light. In the case of mismatched capacity, the backup is inevitable, so it wouldn't be caused by red light behavior. The state of traffic engineering is pretty dismal.

Your opinion that red light cameras would help with traffic flow is just a gut feeling, not data. Every time cameras have been added to a traffic problem, the result has been more problems, not less. I strongly suspect that adding a red light camera to an intersection would not allow more people to go through per hour. Even if it helped, a traffic circle would help more, so why bother with a solution that costs good people money and creates huge conflicts of interest when a simpler and better solution is available?

Policing should not be automated. Probably not ever, definitely not until there are new traffic rules written for the "surveillance age".