You are correct, I don't run a 501c3. But in another life I worked on the calculation of tax deductibility for (among other entities) several 501c9's so I have a pretty good idea about how complex tax rules for nfp's can be, and how someone whose business is something other than filling out IRS paperwork can fail to have the information necessary to fill out a 990. The "Oh crap, can you help us recreate the last 3 year's books?" scenario, while rare, does occasionally happen.

I'll drop the tinfoil hat when the government stops spying on me (oops, I mean collecting my metadata) and IRS employees enforce the rules evenhandedly to all types of organizations. 2 years ago I would have thought this position was crazy too, but given all of the stuff that has come out recently, I think that the a priori assumption should be that the government is abusing its power and it should have to show that it is not. For example, if the IRS came out with a statement that said "We looked at a sample of 10,000 registered 501c3's, and found that 53 of them had not filed a return in the last 2 years. We revoked the 501c3 status on 51 of them, and are looking into assertions by 2 of them that they did file and we [the IRS] must have lost them." I'd look upon this action a lot more favorably.

The problem is that the rules are phenomenally complex. It's easy to say that they should have just followed the rules, but IRS rules are a serious PITA to satisfy. It is quite likely that no matter what Xorg had done, the IRS could have found some error in their compliance that would enable them to revoke 501c3 status.

So the real issue is that by making it so hard to comply with the rules, regulations, and laws, it raises the question of whether the government is using "selective enforcement" to punish people, organizations, and views that they don't like. Did this happen because of a general review of nonprofits, in which case this was a simple case of good enforcement, or are "hackers" being targeted by the government (for lots of reasons, e.g. resistance to NSA monitoring), and any one of a number of technical violations would have led to the IRS' actions? In that way it is similar to the Aaron Schwartz case, and is something that should be noted, if not actively resisted.

The worst I know of is the fact that the prefix 'in-' means "not" or "the opposite of" and "flammable" means easily set on fire. So therefore, "inflammable" must mean "not easily set on fire." As in, "Don't worry about accidentally burning the house down, I coated the walls of the fireplace with inflammable material"

However, the "literally" thing is probably the most annoying perversion of the language.

...but if they authorize the re-write, they could be held responsible for getting it right (especially if they don't use HIPAA-compliant encryption TM). Kind of the epilogue to an "I'll be gone, you'll be gone" strategy. As you point out, the main position of the typical middle manager is CYA, whether warranted or not.

This is the biggest problem with highly regulated businesses like insurance. Lots of business rules in that industry are ultimately derived from legal requirements, and the people who wrote the regulations generally don't care to make them clear and concise, or sometimes, consistent. So it is easy to say that all we have to do is tear it down, write the specs, and start from scratch, but are you 100% certain that you know how many hospital days North Dakota requires to be covered after a live birth? What about after a Cesarian? What does Ohio require for IVF treatment? How does Florida's motor vehicle insurance coordinate benefits for a hospital stay after an accident where the patient was not at fault? There are a million of these questions that somebody had to get the right answer to at some point in the past, and recreating it would be exceptionally difficult.

These rules were written by politicians and regulators to be read by and used by lawyers, not developers. And big companies actually like it that way because it prevents new entrants from coming in. I cannot wait for the day when all regulations need to be written in implementable code, such that companies just use that code and know that they will be in compliance.

Having said that, every insurance company can handle OOP maxes in their system. The issue that they claim is a problem is that you have one company manage hospitals and doctors visits, and another that does pharmaceuticals (PBMs). Why not just "reinterpret" the rule as 2 separate OOP maxes, one for the major medical, and another for the drugs? Regulators make that kind of regulatory interpretation all the time so postponing the whole thing seems much more political than operational

This. I make my living doing statistics on a beefy RHEL box. There is something to be said for a system that grinds through anything I throw at it, administered by a systems idiot (me), and has only been shut down once in the last 2 years (for a memory upgrade). Even my "just works" iMac, MacBook Pro, and various other iDevices need more care and attention. Hmmm why am I not running CentOS at home???

You come back to it 3 hours later to find that the object has separated from the raft leaving you with $20 worth of extruded plastic spaghetti. But if you babysit it the success rate goes way up.

It might have something to do with the nearby body heat, or maybe a hidden camera that verifies a person is there, or just pissed off little elves that don't want to be lonely. But yes, you have to babysit it :(

I think you mean 50% of the problems you have with USB. Or is it 45%?

Human nature? Free will?

Human behavior is complex enough that our current state-of-the art is not sufficient to fully measure or model it from a reductionist viewpoint. And yet holistic explanations based on simple observation have proven fairly successful over the centuries. I don't mean to be flippant, but you will get a better understanding of the motivations of man from great literature than you will from science and economic textbooks (and this is likely to be the case for another hundred years or so).

*blind

But they *are* a part of the "experiment" that AC claims to be "a fair way to run an unbiased experiment" which is what I am disagreeing with. There is almost certainly a bias there - the children who were not immunized had parents who made the decision not to immunize them.

If you could find children who were, say, placed for adoption where the ultimate home they lived with was random, and some had been immunized and others had not, then if you had a statistically significant number of autism diagnoses from that population, then you would have a nice unbiased experiment (but even then you aren't normalizing for genetic predisposition to these diseases that they might have inherited from their birth parents). But you probably don't have enough data from adoptions so the children who did not get vaccines were raised by parents who did not give them vaccines, and that type of parenting may be strongly correlated (positively or negatively) with autism diagnoses at later ages. Hence it will be very biased, and although statistical techniques can reduce that bias somewhat, one cannot expect conclusions coming out of this data to hold the same scientific rigor as a proper double bind study.

IIRC, it was poor experiment design that started this whole thing (yes that is an understated euphemism for the improper conduct that actually occurred), so I'm not just being pedantic - there are important distinctions to be made between the scientific method versus "statistics on a bunch of data I found."

I didn't read TFA, but I'm pretty sure that the parents chose to withhold the vaccine, not that doctors randomly gave some kinds a placebo while giving the real thing to others.

Some good statistics might be able to glean some information from this (multilevel regression model or such thing), but it will not be as good as an unbiased experiment.

Yes I was thinking about the moon landing. Not to take away from the feats of previous generations, but there is a big difference between watching a handful of people get to the moon, versus the thousands, or maybe tens of thousands who have first hand experience with the change in computing technology occurring in this generation.

Similar story here. 16KB TRS-80 to 40GB workstation. I thought it would be like 10,000 times or something, but was kind of surprised to break 1,000,000 when I did the calculations. It's kind of amazing when you think about it. Did my great-grandparents experience any physical measure than grew by a factor of 1,000,000 over their lifespan?