Probably they were thinking carefully with a room full of smart people considering more than short term one-upmanship on their mobile platform. Google is not Android, it's much bigger than that. They have to fiercely defend their reputation for playing nice on many/all platforms, because many of us started using their service for that very reason. Google is supposed to be better than petty BS, they are supposed to be that more-agnostic omnipresent service that is available everywhere. IMHO anyway.

Awesome timing. I just got this e-mail a few minutes ago: adding design "feed" to AutoCAD WS.
http://www.autocadws.com/blog/introducing-the-design-feed/

Yeah. Like this? http://wiki.osx86project.org/wiki/index.php/Main_Page

I'm not a jerk, honestly. I was trying to defend an earlier poster for trying to be precise about accepted meanings in research circles about certain terms. The words have looser definitions in common English and even science broadly and are accurate enough there.

And I do know that words have more than one meaning. That's what I was trying to clarify in the first place in this thread. That when someone says "well it's not technically solved" they are not necessarily just being a dink about it, but there is a well-accepted sense in which random sampling is not enough.

Like how they Jonathan's team solved checkers. http://en.wikipedia.org/wiki/Solved_game#Solved_games It took 18 years of computation. It was a pretty big deal when it finally finished.

You cannot test every possible variation in chess, I agree. It would be neat if someone could come up with some sort of proof that certain classes of positions end a certain way under perfect play, that would save you a lot of searching. I believe they actually have done that, it's called an endgame tablebase http://en.wikipedia.org/wiki/Endgame_tablebase

Just because something is hard or even impossible doesn't mean the burden of proof should be lowered. We use precise terms for a reason, so that they can continue to mean something precise.

Confidence also is a real, well-defined term that means something, and this is not it. Confidence has to do with repeated noisy measurement and how often those measurements lead you to the correct conclusion.

Chess is a deterministic, full information game. There is nothing that is random, unlike a game like Backgammon and there is nothing that is unknown, unlike a game like Poker. You may not care about proving things in this domain, but some people do. To them it is warranted to ask a perfectly well formulated logical question like "Does there exist an opening move for player A for which player A is guaranteed to win if he plays perfectly." And by play perfectly, I mean "Whatever B does, there exists a response (that A can find) leads A to win." This would mean that from the initial state, player B cannot win. This is interesting (to some people).

There exist a large (but finite) number of board positions that can be reached from the given starting point. You can say "I checked billions of the sequences of moves, and in all of them A could win with perfect play, so my analysis of the chess opening is that A wins. I'm super confident." However, without a proof, you cannot say "I solved the problem and have determined A wins."

Sorry. You cannot publish a paper whose conclusion is "I'm pretty sure P!=NP". This is different from assuming it's true in order to make other statements, such as, "assuming P!=NP, my cool new crypto method has features people would care to have."

I think that paper would not get accepted. A probabilistic statement would indicate there is a random aspect that determines the outcome. Reviewers would probably ask what is the random event that would determine if P does or does not equal NP.

I'm definitely not making the confusion you think I am. I have studied Computer Science at the PhD level at the University of Alberta, which I believe has the strongest games research group in the world. I will admit to not being an expert in games myself, but I am quite confident that when people in this area say solved, it means something specific, something stronger than "obviously true to everyone in the world". It requires proof in the rigorous, mathematical/algorithmic sense. I'm pretty sure, anyway.

For sure! That's why I was sure to make the point about there being value in the work. Again I wanted to say a few words about how "solving" is different from "making a very convincing argument about".

The difference is that solve in this context is not a general English word but rather a specific and well defined term. I'm pretty sure the technical meaning of "solving" a game or position within a game requires a proof. The meaning of proof is somewhat stronger than overwhelming evidence. We are pretty sure P!=NP, but we don't have a proof. You cannot publish a paper or write a thesis that says "I'm pretty sure P!=NP".

Note: I'm not saying this work is uninteresting, just that those pointing out that solve is being used incorrectly are justified.

As a Canadian and a scientist, I will tell you that it seems that our current government is very anti-science.

It is my understanding that experts and scientists opinions are not respected when making policy decisions. Our scientists are frequently muzzled. The long form census was recently changed. We have new crime policy that is unsupported by experts. Environment Canada has been cut drastically. I believe that our science minister does not believe in evolution. Experts, statisticians, and scientists have spoken out, have resigned, and have protested. Nevertheless, the majority of Canadians do not seem to care.

Kyoto is only the most recent item. Whether the decision was right or wrong for the environment, the fact is that the decision was political. Canada, today, is anti-science.

I just don't understand how that is anything but a net loss to "just go". If you really need/want to leave, then that makes sense. But if you would have been satisfied to stay if the compensation was appropriate, why not have that option? Not only do you have to start over with no rapport with the rest of the office, no seniority, no earned respect, but your current employer is blindsided and almost surely in rough shape because of it.

Now I don't know the OP's situation, but I'm in a small business. I only have a handful of employees and each one requires months of training before they are effective in their roles in our workflow. In the time that they are not effective, I have to pull 80 hour weeks to do their job and mine. Further, it costs me thousands (a real percentage of our yearly revenue) to advertise, interview, hire, and retrain someone else. I don't have the resources or space to have multiple staff for each role, so if someone just quits on me, it's a disaster. Again, maybe not the OP's situation, but I'm surprised how many posts here think their employer is a faceless entity that would fire them in a second. NOT ALL BUSINESSES ARE LIKE THAT.

I'm an employer. I've invested a huge amount of time and money in my people. Times are tough, so I'm not going around offering $10k raises indiscriminately. However, they are key people who I would pay that much to keep, and there would be bad blood if those people would have stayed after a round of negotiation but decided not to give me an opportunity to consider the situation.

Us bosses are not all unreasonable. Many of us have had to make tough decisions leaving one place for a better opportunity. I've actually had this conversation with each of my employees. I tell them: "if you are ever unhappy or you start feeling like the grass is greener somewhere else, let's talk about it. If the difference between keeping you and losing you is a few dollars or benefits, I want to work with you to find something that works for both of us."

The answer is that it's being taught by Sebastian Thrun and Peter Norvig. They are each kindof a big deal.

Very low probability that the course will require you to read that entire textbook, unless it's unlike every AI course I've ever taken. Also, just watching the lectures and not doing the reading will probably be super interesting if it's like any AI course I've ever taken. (I have a Masters degree and most of a PhD specializing in AI and I think this course will be crazy interesting for most geek/nerd/tech-inclined people)