For an individual this might be a solution (even for me! I've moved around lots too)

But for a whole society it's not. Especially when we seem to be hearing similar stories across thousands of towns and cities. Is everyone supposed to move to the bay area, NYC, Boston, Seatle? What is the housing going to look like? Does the rest of the country just empty itself?

The current world increasingly favors centralization and economies of scale, so the "optimal" spread of people gets more and more centralized. Living on the fringe, unless there's a specific economic need (we'll still need some (1%? less in the long term) people living as on site farmers) is increasingly becoming an expensive "hobby"/lifestyle choice, since you're going to get less services at a much higher cost and with less opportunities for income.

There is the minor difference of moving from one location to take up relatively lucrative, but largely unskilled, jobs and moving from one location with a low-skilled job to try to find a high-skilled job. Or even from one high-skilled job to a completely different high-skilled job.

While the "less services" part is sometimes true, in semi-rural/sparse suburban areas it's often possible to get quite a bit more for a lot less. Car maintenance, food, land, recreation, all much cheaper outside of the major urban areas and high-cost states. This is yet another obstacle people face when trying to move to a big city. They can't afford to pay twice the price for half the food.

So (in my situation) if you're living in a rural community, you're spending less on your own community (and getting less from them), and spending extra to the other communities. Which is not that nice for the economic health of your community.

There's opportunity to be had in Texas, the southeast, the northwest.

That being said, the article otherwise seems like a great introduction. Not sure why they chose that title.

After all, if Black Swans were common enough to make prediction a fool's errand most of the time, the bird of that name would never have led to the book of that name, because everyone would be predicting their failure to predict things.

In the real world there are often no controls, and complex systems can be driven by an attractor for a very long time before one morning they are not, and every rule that you have is useless (often worse than useless).

Sources of error are not equal; "Black Swan Error" is unusual in that over time it may be that this source is more important than any other source of data in your domain - the strange attractor that drove the creation of your classifier over the last 20 years may never recapture your function and if that's the case your classifier will be literally the most wrong thing you could have!

However, it is not the only type. There is also the stock market, which demonstrates major unpredictability every few years, but which can also be approximated the same way between each of the Black Swans. (And they keep being Black Swans because the gap between them is large enough for people to convince themselves that "This time it's different, this time n̵o̵b̵o̵d̵y̵ ̵w̵i̵l̵l̵ ̵h̵a̵v̵e̵ ̵t̵o̵ ̵b̵e̵ ̵n̵a̵i̵l̵e̵d̵ ̵t̵o̵ ̵a̵n̵y̵t̵h̵i̵n̵g̵ growth will be eternal!")

Edit:

Point is, it generalises as how wrong you are in your predictions, and the closer your estimate of your error rate is to your actual error rate, the better your model is.

I have to say I rather prefer the black variety over the white.

Great point. We can't know that a machine learning algorithm used to make predictions won't be wrong if the future turns out to be significantly different from the past. A swan-classifier trained on images of white swans would fail hard if given pictures of black swans.

That said, people find it useful to use machine learning algorithms to predict the future, as the future tends to be similar to the past, at least in the limited domains to which machine learning is currently applied. As compute increases and we learn how to write machine learning architectures[0], we don't need to distinguish as much between 'machine learning' and plain old 'learning' and much of what philosophers have thought over the years about the problem of induction, and relevant domains of induction, becomes relevant to the topic.

[0] Or learn them. Jeff Dean mentions experimental success learning RNN architectures: https://www.youtube.com/watch?v=vzoe2G5g-w4

Any time you extend a statistical model temporally it immediately becomes mathematically invalid since probabilistic statistics are only valid for a fixed population at a fixed moment in time.

Unfortunately business and government is rife with people predicting the future based on statistical models that have no more mathematical validity than reading tea leaves.

Statistics as we know it "works" (can be derived) under the assumptions of controlled experimental data. As a thought experiment think about the weather - we know that if we build a classifier that predicts the weather in my garden tomorrow based on the history of the weather in my garden it will do very badly. Why - well because weather is very very very complex; the range of behavior is vast. But worse, it's unstable. The weather in my garden is driven by several complex systems; the ocean, the atmosphere, the earth's orbit and sol! Statistics can't predict the future of the weather in my garden.

Statistics also can't predict other things like the future of the financial markets (not least because if you find a statistical law about that they you will act on it and then screw it up)

It's important to me to bang on about this because there are loads of people who sit through their introductory courses and read the example of predicting a biased roulette wheel. Years later they end up running the company/country/community that I live in and they have a view that they can use the same principles to do it... and this thinking leads to nasty surprises for me.

Give me hourly readings of temperature, wind speed, wind direction, precipitation, cloud cover and barometric pressure for the last 10 years and I can give you a very accurate prediction of tomorrow's weather in your garden.

Many years ago I actually seriously tried to do what you describe above, I tried out all sorts of things around seasonal analysis and other features. What kills it is the chaotic nature of UK weather due to the jetstream and NAO.

that aside you are missing a larger point. if you predict the future based on past data all you are saying is "the future will be the same as the past." you aren't predicting anything. you will be wrong every single time something novel occurs, which is pretty frequently in the real world.

I don't have the data handy but to the best of my recollection weather forecasting for high/low temperature and precipitation does pretty well for the range of 24-48 hours but declines steadily in accuracy, and is no better than random guess around 2 weeks out.

That said, you are not addressing my other point, which is that "weather prediction" is just saying "things are going to stay the same." You are always starting with a set of conditions and then looking at your records and seeing what happened in similar conditions and predicting that the same thing will happen again.

Predicting that things will stay the same may come out as better than random guess in many cases but it will still be 100% wrong in cases where something novel happens.

Obviously, this is more or less feasible in practice, depending on the phenomenon under study. Calling markets unpredictable is not evidence against the existence of rigorous frameworks for statistical prediction.

Don't let bad experiences with inexperienced and overconfident practitioners blind you to established, uncontroversial, mathematical truths.

I'm from the northeast US so I miss the cafe culture and general workaholism there. If someone could recommend cafes and cities in general that have this and are within a 6 hour train ride from Frankfurt, you'll help me find my next vacation location.

Meetup.com already provides a way to filter events by location, but the problem is, people need to pay to register events, plus sometimes you are looking for just one or two other people and dont need a whole group.

discussion was here: https://news.ycombinator.com/item?id=13555124

Harder to justify when so many of them are directly related to tech and the SV startup scene.

I think a lot of members are feeling some degree of politics exhaustion. There's still plenty of discussion going on.

Realize it is likely you feel this way for a reason, and it is a sign of health. Do an inventory of the main pieces of your life and note which situations are causing you pain:

- work life

- living/housing situation

- physical health (including movement, sunlight, eating habits and sleep)

- general social life (do you have spontaneous positive interactions with acquaintances, coworkers, baristas etc.)

- personal social life (relationships with partners/family and close friends)

If you are feeling depressed I can bet you more than one thing on this list can deserve attention. Pick one you can change for the better immediately and work on it. For instance, if you are lacking social interaction in general, go pick up a coffee someplace that is not too busy and either smile and make brief small talk with someone working there or another customer. If you don't like your apartment, put effort into finding a new one. Talk to your partner or get relationship counseling. Things are not going to magically become perfect overnight, but small things every day add up over time.

If it is someone you love: Listen if they are talking, talk to them if they are quiet. Generally be supportive. Search for a positive thing they are doing or want to do and give them positive reinforcement for it. They are stuck in a negative thought cycle and try to encourage something in them that is working. Here is an example: Friend: "my life sucks. I hate my job, my expenses are too much for what I make, and on top of all that, I just found out that my girlfriend is selling herself to strangers on Craigslist. I already confronted her about it, but she just says to accept her entrepreneurial polyamory and get over it. At this rate, I am never going to get a phd in deep learning or do a startup or have a stable family like I want"

You: "cool you want to get a phd? yeah I've heard about deep learning it sounds cool... which programs would you want to do/ what projects do you want to work on etc."

Remember that scene in fight club where Durden orders the clerk to follow his dream of going to vet school? Something like that. Listen to them to figure out where they want to be headed, and encourage that.

It is notoriously difficult to work in, with way too many open spaces (open in the sense of being public, as well as open in the sense of having random 3-story-high atriums of air above you in some places for no discernable reason) and not enough quiet nooks that are conductive to focused work.

It shows every sign of being a two for one own goal in diminishing MIT's future ability to "be MIT".

But then, just because you're paranoid doesn't mean they're not out to get you.

The incubator that was Building 20 was is gone, as far as I know. Many many good things came out of it, some of which got very big, starting with the Research Lab For Electronics right after WWII (it was previously the main building for the "MIT Radiation Lab", i.e. a place to prototype and in some cases build small lots of RADARs). As of the '80s, some of its 3 phase ceiling busbars were still live, one of my favorite professor's coffepot ran straight off one.

That's a relatively subtle thing, but to the extent MIT becomes stolid and not doing new things, becomes more of a place run by consensus and the conventional wisdom, that would be an important thing to look at.

The second is of course the Stata Building. It will absolutely diminish the useful output of what's now CSAIL, if for no other reason than those who need to be able to work without distraction by and large won't be able to in the building. It's the sort of thing that can put a ceiling on the difficultly of projects people will be able to do, if they can't get into or stay long enough in flow.

Obviously people will work around this, using e.g. laptops in quiet places, but collaboration will then seriously suffer if past research on the subject is of any validity (e.g. > 30 feet? or a stairway makes a big difference).

I don't know if anyone was out to get them, it's just that a) people who had weird and very wrong ideas about how to do things, like the head of the LCS, who among many other things tried to physically enshrine a permanent disconnection between CS and AI, b) people who simply didn't give a shit about those who really were going to use it, and/or c) people who were grossly incompetent, were in charge of getting it designed and built, but not the actual "users".

And the screwups are legion. Tom Knight, #1 Lisp Machine designer back in the day and someone who did a lot of important stuff, now a "synthetic biologist" AKA into artificial life, asked for one thing, a floor that sloped to a drain. Which he didn't get. And I now notice as of 2008 he's doing this at a company he co-founded.

I suspect those two fact are not entirely coincidental. And potentially a grave loss to MIT, that part of the future just might not be done there, losing "The Godfather of Synthetic Biology" per Wikipedia is not a small thing.

And I have the tiniest window into CSAIL, and can speak freely about it (my one theoretically vulnerable source isn't really, and hasn't told me anything I can't repeat), there's I'm sure much much more to be told, and sifted through by historians or archaeologists of science someday.

6.001 is a rather different thing, after the dot.com crash, which resulted in a crash of EECS undergraduate enrollment, less than half after being 40% of the student body for decades, people panicked, those with an agenda against Scheme/LISP used the opening, but primarily, MIT decided an MIT EECS degree was going to be a fundamentally different thing.

That doesn't mean they're wrong, just that if you're looking a focus on the 6.001/SICP sort of thing, you might want to look elsewhere, like CMU I think. And maybe this was inevitable, a department like MIT's or UCB's is simply never going to be the same as a CS focused department that didn't grow out of an EE one. And Sussman at least always thought the 2 subjects were both of great importance and should be taught together, he himself just never figured out a way to really make that work, including at least one experimental one year long course covering both prior to the development of SICP and the 4 6.001-4 courses.

But in short, you're saying that MIT may no longer be "where the future happens."

That's quite possible. But at the moment, there's still an abundance of talent there. OTOH, the lack of an incubator is a problem.

I guess what I'm saying is that you might be right, ut the future is pretty hard to predict.

Or maybe MIT is now starting to seriously revert to the mean, some of the future will most certainly happen there, just potentially quite a bit less.

I may have seen some of that in the '80s, certainly the midnight execution of the Applied Biology department by envious chemistry and biology professor administrators was a sign of some sort, especially since the fate of the latter should have been much worse than having their administrative careers ended. The faculty as a whole certainly viewed it as a betrayal, which is was in spirit and "law" (MIT has the institution of the Visiting Committee to keep units of it on the straight and narrow, they were of course ignored in this).

And this is part of a bigger trend, pure administrators are growing in vast numbers in US higher education, their costs are one of the biggest drivers in this clearly unsustainable trend, they're funded by ever more Federal dollars (even if laundered as loans, which the Feds entirely took over in 2010), and from money comes power, they're taking over US universities from the faculty. Who themselves are getting segregated into small numbers of high cost tenured and large quantities of low cost associates who care barely make ends meet, plus the old bane of graduate student instructors who aren't good at it.

MIT, at least for now, is not going to succumb to some of these trends, as long as associate professors are rare exceptions that prove the rule, like SF author Joe Haldeman, for classes are otherwise taught by tenured or tenure track faculty, and adequate teaching ability is required for tenure, as well as the minor detail of being #1 or #2 in your subfield (as judged, in part, by those Visiting Committees).

Similarly, like CalTech, there's a high floor on admitting undergraduates, they've got be able to do one term of the calculus beyond the AP BC sequence, and calculus based mechanics and physics (and chemistry and maybe biology).

But....