A bullpen makes a lot of sense to me. If coworkers doing exactly the kind of work I do were colocated, we'd all learn quickly what forms of interaction were productive and preferred, so we could avoid getting on each others' nerves. I used to work in a suite containing only team members. We all preferred it greatly to our present cubicle farm.

The principal problem with open offices are the disturbances arising from non-team members, especially the rude few who won't consider the harm their noise inflicts on neighbors.

Etzioni's point is a good one. To date, all AI apps have been designed to passively sit and do nothing until given a specific task. Only then do they act. For Hawking to be proved right, AIs must take the initiative, to choose their own goals. That's a horse of an entirely different color.

Of course, there's no reason why AI agents could not become more autonomous, eventually. Future task specs might become more vague while AIs are likely to become more multipurpose. Given enough time, I'm sure we'll have mobile robots able to do more than sweep floors in a random pattern. But Commander Data is a long way from an iRobot Roomba or Rethink's Baxter, both of which are dumber than my phone.

In the real world, autonomous robots are not going to arise for decades. And when they do, if they drive on the same streets or share the same office spaces, they too will have to obey the same rules of conduct as the rest of us. You won't get special privileges just because your brain is made of silicon.

Editing the historical record sounds awfully like hiding your past. Why isn't this like pretending the Holocaust or Stalins purges just never happened? Wouldn't IBM like to assert (without contradiction) that it never assisted the Nazis in the Death Camps?

This is an initiative only a corporate tool could love.

That's where US R&D dollars have gone. Camelot!

If you read the court case mentioned, the supreme court ruled that a search warrant was required before police could access the defendant's computer, which they did not do.

Anonymity was tangential to the case at best.

Exactly. This announcement marks the official start of the Obama presidency Reality Distortion Phase, where the lame duck madly fabricates an idyllic legacy that he gave a damn for liberté, égalité, fraternité.

Just like Nixon.

Right. With luck this kind of exurban facility will make good use of selective dispensations from the MI DMV to extend their trials off premises and onto roads like you describe.

Dialing up the real world noise is essential to bring these cars up to speed -- missing or obstructed lane markers and signs, poorly marked or uneven road edges, and the introduction of noise like leaves, snow banks, and pools of accumulated rainwater all need to be mastered before automation has any business driving cars, buses, trucks, or passengers in the many parts of the country like the Michigan coutnryside where often "the sun don't shine".

Exactly. Ann Arbor has persistent winter snow and occasional sleet, heavy rain, tornados, and even flooding. Its weather is often a perfect storm for drivers and a far cry from the ideal idyllic settings used so far to test automatic cars.

A2 is the real world. And its mix of academia and auto company proximity make it ideal for this role. Seems like a perfect marriage.

If we were serious about ending criminal acts in the US government, we would:

1) create a fully independent office inside the government to investigate and prosecute wrongdoers, with powers no less than congress' Special Prosecutor (i.e. equal to the presidency)

2) offer whistleblowers generous retirement benefits for life (to escape retribution)

3) give them blanket immunity from prosecution

4) prosecute the gov't wrongdoers all the way up the chain of command, *starting* at top executive levels

But the US government does the opposite. That's the very definition of racketeering and organized crime.

With a theoretical PhD, if you're applying for non-research jobs, you're probably seen as overqualified and suited to the wrong mix of skills. If the years of study toward your PhD don't translate to a capability that the employer values, then they're likely to see it as irrelevant, and see you as having "The Wrong Stuff".

Try describing your PhD research in some way that's more relevant to the company you're applying to. If it's mathematical, describe it as "analytic" or "data intensive" and not "theoretic" or "provably valid". Data mining and machine learning and AI and big data are hot right now. Make your skills sexy.

And be sure to write a cover letter that's tailored to the job, the industry, and the employer. These days, mismatched or over-general applications get tossed almost immediately.

For any R&D company that has a lot of in-house raw data, the Watson Discovery Advisor is likley to generate a lot of interest.

Imagine you're an executive VP in R&D in a board meeting. You receive this challenge from the CEO who hates your guts: "Our R&D productivity continues to decline. What're you doing about this? How are you extracting every last bit of value from our data? Our major competitors are using tools like Watson. Why aren't we?" You damned well better have an answer.

I work for a Fortune 100 R&D company that is *very* interested in improving its R&D ROI. I know for a fact that any opportunity to reevaluate our data to derive additional value (e.g. new prospects) will set off bells among the C suiters. IMHO Watson, and especially Discovery Advisor, is the first system I've seen with that potential.

Of course, IBM is going to have to step up its game in loading and tagging all that data. I suspect that's where most of its new Watson staff will work. I suspect the most fruitful features in data are not readable in natural language (English). Much has been summarized in graphs, or lies in tables, or in addenda. Or it's buried deep in old screening results stored in flat files that were long ago archived to tape. And it's certainly not present in easy-to-access content like online research paper abstracts.

But all it takes is one or two significant new leads to make the millions you spent hiring Watson look like money *very* well spent. And personally, I think that scenario is entirely plausible.

This pattern of ebb and flow in the tech world is nothing new. Every decade brings a computing novelty which invites revolutionaries who rethink the user experience. The universe seems to expand. All the developers get excited, jump on the bandwagon, and revel in the myriad possibilities -- for as long as the high lasts. Now the latest bandwagon has slowed and the Next Next Big Thing seems far far away...

1977 brought us the personal computer. 1984 was GUIs and WYSIWYG computing. 1988 was the network (and email and AOL and Usenet). 1994 was the web. 2008 was smartphones. 2010 was social computing. 2012 was The Cloud.

But 2014 is... dullsville. Sherlock is bored. Get used to it. This is the game that never ends.

I suspect most events like you describe probably would occur too quickly for conventional cameras to capture, but I see your point. It seems to me that kind of mition would have to take the form of a percussive force that arises without visible warning -- like the launch of an explosive powered bullet, and unlike the launch of a golfball being struck by a moving golf club that rapidly approached the stationary ball before making contact. And as you suggest, I doubt the firing of a bullet is the kind of motion seen most often in videos.

I suspect the vast majority of conventional motion sequences follow a path of 1) slow accel followed by slow decel (providing little clue as to directon of time), or 2) slow accel followed by fast decel (something that occurs often in forward moving time sequences, as a moving object is stopped suddenly by an impact). Thus path #2 is probably frequent enough and visible enough to be the anomaly that lets Freeman's group recognize the backward passage of time.

Dr Freeman spoke about this work at CVPR this week. In the videos I saw he identified small markers of temporal transition as indicative of moving forward or backward. Those they labeled as backward appeared to recognize asymmetric movement -- as in gradual acceleration followed by sudden deceleration as uniquely forward flow (as when a hand swings down and strickes a table top) -- an asymmetry that cannot occur in reverse (as in sudden acceleration followed by gradual deceleration).

Dr Freeman did not propose this as the causal phenomenon in question, but that made the most sense to me in light of the motions he identified as evidence for backward motion.

FISA is strictly a federal warrant court. Local police and prosecutions don't use it. This ruling applies principally to local police conduct and evidence, secondarily to federal police conduct and evidence.

Yes, the FBI could still rely on FISA's rubber stamp. But county mounties can't. And it's the sheer number of the latter which pose the greater threat.