Investments in lobbyists always suggest a belief (though they don't tell us whether it's true or false) that the ROI on regulatory meddling is greater than that of other purposes to which the money could be put. What could possibly go wrong?

I suspect that they don't emphasize this as much when teaching the not-dead ones; but when I took Latin it was overtly acknowledged that this was expected to improve my knowledge of English grammar and the (very large, if rather skewed toward jargon) chunk of English vocabulary that was pulled in from Latin with varying degrees of mangling.

Both because Latin grammar is substantially different, and because technical knowledge of English grammar couldn't be assumed, they didn't try to teach according to analogy with English grammar, or otherwise do something that required a formal knowledge of it.

Because of my...rather peculiar...profile on language acquisition, I ended up scraping through in large part by inferring Latin words I didn't know from the English words those Latin words were supposed to be helping me with, which was somewhat perverse; but so it goes.

There are some ongoing differences, aside from cost: With written material, you don't get the use of tone, gesture, expression, and the various other spoken-language tricks of expression that don't directly make it to paper. It is hardly impossible to write such that the reader will (mostly) correctly infer some of them; but that's exactly the sort of thing that you have to work at, or have sufficient practice to do nearly effortlessly, that you'd get for free when speaking.

There's also the difference that most spoken communication takes place in more or less real time, which allows the other person to interject, or you to elaborate on a point if the audience appears baffled, speed through a point if they appear bored, and otherwise tailor your speech to the demands of the occasion. It will lack formality; but customization counts for a lot.

Some text communication, IM and the like, is largely the same and admits of the same sort of near-real-time course corrections; but even at the level of message board posts you really start to see the effects of delay. If I fuck this up, I can post a (hopefully) clarifying reply; but I could easily end up being misunderstood by numerous people before one of them posts something that informs me and I refresh the page and see that, and get my correction in.

The 'purists' who spend their time harping on The True Rules, or replying purely to note that somebody has used 'there' instead of 'their' or the reverse, are an utter waste of time. Spending more time thinking about communication that will be stripped of spoken and nonverbal cues and sent out into the world with a nontrivial turnaround time, though, is something that I suspect we won't escape.

I agree that logistical issues for most text have declined over time (and some things that used to be text, like 'letter writing' as an actual social institution are now largely dominated by spoken word replacements); but I would argue that they aren't gone, and that additional issues that the writer needs to consider start to crop up with surprisingly small delays.

Has it ever actually happened that a natural language has either achieved such unambiguity that reliable transmission of meaning can be expected, or such chaos as to descend into mutually unintelligible babble?

Obviously, we muddle through, so it's not as though meaning is totally impossible to convey; but even areas of (pseudo)natural language, like contract law, designed and implemented by trained experts in the hope of mutually unambiguous expression are constantly hitting the rocks. At the other end, languages can and do diverge over time if some sort of population separation occurs, and certain in-group jargons and slang can be used specifically to impede understanding by outsiders; but (as much as one or both parties might loath the encounter) languages just devolving into babble because we didn't nip slang in the bud doesn't seem to happen.

I realize that Slashdot Summaries are one of the important, protected, habitats of a mixture of questionable proofreading and overt editorializing; but isn't something important being left out here?

The scheme in question is known as 'write by reading'. This apparently boils down to 'write however you want', according to a blog post that barely touches on the matter aside from a link to a German newspaper. Is it possible that this 'write by reading' theory involves some 'reading' somewhere? Maybe the notion that children will pick up grammar by exposure to it, which would make spending the time previously allocated to Learning Your Grammar Rules Children on reading things that are both examples of good writing and also useful, interesting, or otherwise better than distilled essence of grammar a plausible alternative?

Now, I'd be the first to agree that the standards of pedagogical research are... notably tepid... and education is much ruled by fads, many with little or no basis in evidence beyond anecdotes; but can we really have a useful discussion if we are going to start from a position of such inspiring intellectual honesty?

The question: "Do children pick up grammar from exposure to well written, but not otherwise grammar focused, texts sufficiently efficiently that we are better off skipping the lessons in pure grammar in favor of receiving the grammar as a side effect of reading that will also have other uses?" is a perfectly reasonable one, and it isn't immediately obvious which side the facts would come down on, so some research would be nice; but I'm pretty sure that 'Writing by Reading' is not actually a polite expression for 'Thare iz no ruls in Sckool.'

The bigger issue is that even people who don't trust the (braindead; but too convenient to die) "Hey! Let's just trust about 150 zillion different 'secure' Certificate Authorities and if they signed the cert and it matches the domain everything must be OK!" are still pretty screwed if whatever specific certificate or certificates they are using are now also in the hands of some unknown and probably malicious 3rd party...

There's a pretty big difference between 'because the system is pretty stupid, you can generate a valid certificate for any domain by knocking over any one of an alarming number of shoddy and/or institutionally captured CAs' and 'your private key, yours specifically, can be remotely slurped out of your system and used to impersonate it exactly'.

I don't know if they care; but a uniquely serialized and network connected device running Google's software doesn't really change hands unless suitably disarmed. "Oh, software is licensed not sold, license agreement not transferable, remote wipe. Do enjoy the hardware you purchased."

I'm not sure if they are 'deaf' (and if they aren't, how much of their sensitivity-to-vibrations-in-fluid is actually 'hearing' and how much is 'touch' with their little sensory hairs) or if their range of stereotyped behaviors doesn't include predators that work by suction. There are some aquatic predators that are suction hunters, presumably becaues the fluid is denser; but nothing terrestrial comes to mind. If they evolved in absense of suction attacks, they presumably either are encountering something outside their experience, or running their protocols for high wind, when the vacuum cleaner attacks.

The little bastards are brutally well optimized; but don't seem to have much general-purpose-cognition grafted on so their utility in the face of novelty is likely very, very, low.

It would actually be interesting to see if, given suitable advances in the design and construction of biomimetic robots, low-speed-but-lightweight-and-crazy-maneuverable becomes a viable strategy (presumably as a complement to a very high speed arsenal). Something with the maneuverability of a fly(especially a fly that also has a few small jets for bursts of thrust on maneuvers that exceed what its wings can provide) might well be able to walk right past missiles designed to intercept high speed conventional aircraft unless those missiles were minimally dependent on accuracy because of large warheads with lots of shrapnel and AoE. Much less useful against more prosaic rapid-fire-guns and flack; but something designed to hunt fixed wing aircraft could be in for a real surprise.

There is the question of when we run out of work to be done that humans are capable of. I would be most surprised indeed to see the crystallization of a lovely fundamental theory of everything that ties up all the loose ends; but considerably less surprised to see the supply of "With a dash of brilliance and some exploited grad students, you can have this problem beaten and written up before you die." scale problems dwindle considerably. Depending on what team physics does, they also might end up spending a long time writing neat equations predicting what a particle collider of roughly the same diameter as the kuiper belt would find if it were funded; but not looking at particularly good odds of getting one. In something like math, it seems quite likely that the number of concise, elegant, proofs is overwhelmingly tinier than the number of inhumanly large ones. I'm not even sure we'd have any reason to suspect that the supply of possible proofs is bounded; but I imagine that people will still be disappointed when the discovery of a new proof short enough to grasp within one mathematician's lifetime is a major event and the mathematical journals are cluttered with 50,000 page machine generated results.

Have you tried discovering that the merger would cause *some sort of treknobabble apocalyptic issue involving scary RF terms and America losing its god-given right to TV*; but expressing a willingness to 'review your preliminary results in collaboration with industry experts' for a modest consulting fee?

It's not bribery, it's just a corporate person engaging in free speech. Indeed, our own dear supreme court asserts the view that this sort of activity does not even create the impression of impropriety...

It's worth noting that the Mitt Romney who was elected in Massachusetts was a politician who the Mitt Romney that ran for president was bitterly opposed to on the majority of issues, and spent a fair amount of time overtly attacking, along with the filthy liberals who had elected him.

The fact that these were the same man suggests that Mitt Romney is some sort of mendacious fuckweasel whose relationship with the truth is so complicated at this point that he probably can't even lie properly anymore; but they were practically different candidates for all operational purposes.

I imagine that they only enforce as many as it takes to discourage the moderately discontent from leaving or shopping around. If they have you on theft of some specific trade secret, copyrighted implementation, patent, whatever, the noncompete would just be gravy on top of the raft of actual charges.

Effectively, the non-compete provides them with something similar to an 'option' in finance. They are under no obligation to sue you into the ground if you leave; but the fact that they could, weighted by the odds that they will, imposes an additional cost to you leaving, which lowers the cost of retaining you.

I'm pretty sure that theft of trade secrets, violation of copyrights and patents, etc. are already covered by their own bodies of regulation. It's already illegal to run away with anything approaching the implementation of anything, so non-competes are either redundant or are deliberately intended to extend the 'any kind of IP' to basically include any and all relevant job skills (including the ones they hired you for, which you obviously didn't even acquire through experience at that job), experience, vague musings during lunch, and most of the rest of the employee's brain-meats as company property.