crdb

> I've found that executives will often try to poke holes in the predictions, especially if the ML solution is risky or potentially threatening to them

In particular, automation that reduces headcount reduces their justifiable budget and therefore power within the firm, salary and benefits, and external status. For an example of the latter, Havard Business School asks you how many people you are currently managing when you apply for an MBA.

This creates a strong incentive to block any attempts at automation or increased efficiency, especially when said inefficiency is not reflected on the KPIs used to gauge the executive's performance. Customer satisfaction and error rates are rarely measured well, nor with a refresh rate sufficiently high to be such a KPI. Blocking is easiest to do by seeding mistrust in the person attempting to build the automation, and in the automation itself.

Part of being an effective data scientist/big data engineer/whatever the buzzword du jour is consists of figuring out what KPIs the executive wants to maximise and sell him on that instead. The good old "work on making your boss look good".

> Conservative ideology is focused around defining progress [...] as a return to a previous state of affairs

> To a conservative ideologue, the truth lies behind us.

Given the context, I disagree with your interpretation of the "conservative" label.

It is obvious (to me) that what the author means in this particular article, and the grandparent, is "Conservatism as defined in United States political ideology" [1], perhaps of the Barry Goldwater flavour, and not your broader and in this context misleading definition, which coupled to some of the more vocal people calling themselves conservatives today leads to a form of association fallacy I've observed some university departments use precisely to reject "conservatives".

This particular definition of conservatism is compatible with scientific progress and new thinking, as shown by the evidence of decades of both happening in politically conservative university departments (in the US and abroad) in the 20th century.

[1] https://en.wikipedia.org/wiki/Conservatism_in_the_United_Sta...

We were building a price comparison company. I was trying to figure out ways to make money after we were turned down by YC and unsatisfied with 4 months of fundraising. Thought we'd try and sell our price lists to businesses who wanted to be competitive.

I tracked down the biggest in the region, found someone there who had upvoted a couple answers of mine on Quora, pinged him there, and he replied. Said it was interesting, but he had no idea what price comparison was, so forwarded me on to the COO.

I met the COO at Starbucks, we talked for a couple hours. Initially, he said it was a very interesting concept, but he didn't understand how to integrate our information into his product. I said it would be easy to integrate it in his data warehouse. He asked me what that was. I explained, he asked me how much we'd want to build it for them part time. The contract was signed the next day after a 30 second meeting with the CEO, and so we were in the data business.

I didn't get on well with the CEO, in great part because I was still inexperienced at the political side of sales and would say what I thought in emails, particularly about their existing BI efforts. So we parted ways after 6 months, but they had a data warehouse and automated reports by that point, which afaik are still running there under a new team.

To this day it is my only cold pitch conversion. Every other client I ever had came through connections (making me wonder the value of spending time on pitching). Sometimes as random as - I met a guy in a conference I talked at, he invites me to his next government event, he introduces me to his friend who left the civil service for a VC, the VC likes what I'm saying and introduces me to her portfolio companies, one of them signs.

I'll paraphrase Facebook themselves [1] back at the statement of yours that I disagree with ("His logic suggests so"): (A => B) !=> (B => A).

That Viaweb was competitive because of blub, does not imply that blub is a necessary condition for any company to be competitive, or that not-blub would impede Facebook from becoming successful.

For example, if fast new features are only possible with blub, and new features are not as essential to Facebook's success as they were to Viaweb's, Facebook's use of not-blub (or perhaps, non-use of blub) would not materially impact Facebook's success.

One way in which this might happen is: Viaweb is B2B, and its customers value new features; whilst Facebook is B2C and a social network i.e. the value is in the number of people on the platform, not its new features.

Another explanation might actually just be that there was no high enough level language in Viaweb's days to compete with Lispers, and plenty in the PHP era. To an extent we witness this today with the myriad frameworks available which in theory eliminate any edge for fast prototyping except in safety-critical or performance-critical applications.

[1] https://facebook.github.io/react/blog/2016/01/08/A-implies-B...

In theory, you can watch anybody, but for this to work, you'd need someone to actually look at the evidence i.e. suspect the person in the first place. Quis custodiet ipsos custodes... And in this case it's the guardians of the guardians: who are you going to trust with watching the head of MI6's counter-intelligence division 24/7?

As the CIA found in its paranoid early days, over-suspicion is highly counter-productive (see e.g. [1]). You just HAVE to trust your managers. Even if all services have some kind of internal intelligence setup to watch the watchers, it's focused downwards where the risks are higher and cost of suspicion lower, and it has to be run by people you trust the most. You balance the cost of a mole, versus the opportunity cost of turning down genuine defectors and trusting your own staff.

This is why (according to [2]) many European services are run by people from wealthy, old families, preferably with at least a few members who have served. It's expected that on average, these people have profound ties to their country - if nothing else, in the form of immense property portfolios - and thus are much less likely to defect. You can't bribe them, you can't offer them a better life [3], what's left? Ideology, and that can be somewhat watched for. Of course, you deny yourself some great talent that way but as with managing a hedge fund, you want to limit downside rather than maximize upside. Nationalism and dynastic politics are a form of defense against foreign enemies.

Hence Philby. He was trusted by design, not because MI6 were "bloody incestuous fools" or whatever, and no amount of surveillance would have changed that.

[1] http://www.amazon.com/The-Main-Enemy-Inside-Showdown/dp/0345...

[2] http://www.amazon.fr/Missions-methodes-techniques-speciales-...

[3] Kalugin recalls in http://www.amazon.com/First-Directorate-Intelligence-Espiona... that he was somewhat worried about Philby's living conditions in the Soviet Union (unheated flat, alcoholism, etc.). It's been a while since I read the book but I recall he or someone else decided to push for him to have a good quality of life as an advertisement to other potential defectors; only then was Philby put up in proper accommodation and welcomed by the Soviet system.

Oh, I don't mind, personally, and I agree with you. I also think security on these apps is done by the dissuasive power of knowing that you are live-tracked and almost certain to be caught, rather than any pre filtering (which is the taxi industry's MO); this is a feature.

This is what Uber checks for in California: https://newsroom.uber.com/details-on-safety/

Basically any lifetime sex offense, or terrorist register; and anything violent, or DUIs in the last 7 years.

Both of my drivers would not have passed the last 7 years bit (from what I deduced from our conversation) so the standards are clearly different abroad.

Background checks? I had 2 UberX/Grab (the Singaporean Lyft) drivers who openly admitted to being ex-cons - and that in Singapore too. One even laughed that British prisons were like a 5* hotel compared to Asian ones.

If these companies perform background checks, what are they checking for?

> my dogs eat better food than the majority of the world

This immediately reminded me of Viktor Belenko [1] who flew his MiG-25 to Japan and defected. He wrote (I think, in his autobiography, it could have been an interview) of the amazement that dawned upon him when he entered a Western supermarket for the first time.

edit: found it.

"Once I bought a can which said "dinner." I cooked it with potatoes, onions, and garlic-it was delicious. Next morning my friends ask me, "Viktor, did you buy a cat?" It was a can of chicken-based cat food. But it was delicious! It was better than canned food for people in Russia today. And I did test it. Last year I brought four people from Russia for commercial project, and I set them up. I bought nibble sized human food. I installed a pâté, and it was cat food. I put it on crackers. And they did consume it, and they liked it. So the taste has not changed." [2]

[1] https://en.wikipedia.org/wiki/Viktor_Belenko

[2] http://www.videofact.com/english/defectors2_4en.html

> My suggestion to reduce property speculation is to reduce income taxes drastically (1/2 to 1/3), raise property taxes to at least 1.5% (5x) and provide basic tax assessment increase controls for owner-occupied dwellings, up to say $1m in value.

Exactly. Unfortunately, the short term effect in most countries would be a brutal recession, probably some banks going under/being nationalised, a flight of capital out of the country, and so on - enough to dissuade any administration to try.

It does not help that most politicians are from the middle upper class and own large amounts of property. Few will have the courage to devalue their own net worth by double digit percent.

Thus I think it exceedingly unlikely that we will ever see a shift from production taxes and consumption taxes to "dead" asset taxes. In fact, the historical precedent is the other way round.

There are many others, but France's ISF (wealth tax, of 1.5% of the government-assessed asset value per year) is one of the reasons you can buy a large castle there for less than a studio in Sydney.

> it is tough to side with capitalism on this subject

Assuming you mean "with the philosophy of free markets and individual rights" by the catch all "capitalism", it's not.

The foreign buyer is benefiting from state-provided services, such as the police providing some degree of safety from robbery and damage, safety from invasion and confiscation by the armed forces, and a sound legal framework and justice system that protects their property right.

Citizen pay for this via income taxes and their employer's corporate tax, as well as value-added taxes on products and services purchased (since there is no modern country that generally allocates a certain type of tax income to a certain type of spending). Non-resident owners do not. They get a free ride.

Thus, the free market, moral, individual rights protecting solution that might be Barry Goldwater approved is to charge the non-resident owners a yearly tax that covers the cost of services provided.

And again this evening! New publication date for Death's End in English: October 2016... it better be very, very good.

I think the rules for the sophons are not entirely clear, because he set them up as a plot device for a few of the things he needed (FTL, no more science, omniscience of the enemy) in order to create the special conditions that force humanity to defend itself by creative thinking and this is also why I didn't spend too long cross-referencing their viability. Nevertheless I vaguely remember them being some kind of giant PCB when unfolded in 2D with a fairly high level of abstraction and autonomy without being strong AI (and therefore a new civilisation). It's a given in the plot anyway that you don't want to create a higher civilisation, because of the DF... well, I'm not going to spoil it, but the Trisolarians have very good reason not to create real intelligence capable of resetting its goals.

The translation is as good as it's going to get. The main issue is that Liu is from the PRC and spent most of his life there. It took me years in Asia to slowly realise how differently the Chinese think from Westerners (hell, Westerners themselves think very differently just across borders - compare the worldview of a Brit to that of a Nordic or a Frenchman). If you compare Strugatsky and Asimov, you also get this huge cultural difference that is untranslatable because of the large amount of metadata bagage that Russians would already have coming into Strugatsky's work after growing up in Russia.

The book is "very Chinese", if that makes sense, which I think is what confuses people (and to me is a feature). A quick intro to what that means might be watching a few episodes of If You Are The One (非诚勿扰), a slightly crazy dating show that since the second season has had a government censor added to the panel to moderate the gold diggers; an Australian TV channel has started subtitling a bunch of episodes, although I unfortunately could never find the infamous season 1 ("I'd rather cry at the back of a BMW than smile on a bicycle"). You both get a sample of PRC normal citizen thinking and a taste of how the government is setting the tone culturally.

> I'm not sure if I buy the alien-culture, though. The Trisolarians seemed just as human as ... humans, especially given their introduction in the titular video game, and the Sophons barely seem to have a culture at all.

Did you really think so? I can't remember where DF started, so I'm going to have to be careful with spoilers but in my view, the "game characters" were either humans (professor Ye if I remember well), or weak AI game characters programmed by the Trisolarians in an attempt to map their propaganda to humans. The game itself is an information operation: it is designed to cultivate a resistance within humanity and disinform humans psychologically in order to gain yet another edge over them militarily.

However Trisolarians themselves are a "superior civilisation" which is both helped and hindered by a particular feature of theirs which we don't have (related to the Tines, actually) and which becomes the core plot point enabling most of DF to play out (like the Wall Facer project). Sophons are just sophisticated communication devices, there to send back information to Trisolarians so they can better understand, hinder and later destroy Earth; they are not themselves strong AI (and therefore not a civilisation).

Thus what makes it interesting is the game theoretic aspect of it, and especially the treatment of information flow. Trisolarians know everything except the thoughts of humans (and, initially, they also lack the information that humans are capable of private thought).

It is not what they say (which is purposely adapted to look human) but how they react to learning bits of information that makes them a fully fleshed out alien culture (such as learning abstract concepts like deception, then figuring out the game theory of deception and altering their conversation onto the ETO accordingly).

The human fifth column goes from being a convenient tool of further aggression, to a core path to avoiding the one realistic chance humans have against the invasion. The ability of the Trisolarians to create a high quality "virtual machine of human minds" within their civilisation capable of such manipulation makes them a convincing "higher civilisation" (which is what will make the third book interesting, I bet, especially after the closure brought by DF).

I liked this essay on plot holes (and more generally good writing) despite the all caps and rather long winded style: http://birthmoviesdeath.com/2012/10/30/film-crit-hulk-smash-...

Personally, I found the strength of TBP and DF to be the depth that Liu achieves with his plots at the highest level, and it's important to take a step back and not judge him as pure hard SF (as it seems to have been sold to you, based on your review) but as a writer of literature using SF as his medium (as might have Heinlein, PKD, or Asimov who is his clear and direct inspiration).

The science "plot holes", the suspension of disbelief were I think fair play as a base to deploy his explanation of the Fermi paradox (which is to an extent the point of the whole series). Like psychohistory in Asimov's Foundation or Batman's seeming invincibility in Nolan and Ledger's masterpiece, it helps plot development and the greater ideas about human nature, politics, information operations and so on that are put forward by the book.

The analyses of human nature, particularly of bureaucracies and destruction via inaction (to quote Admiral Rickover, "the person who disclaims responsibility is correct: by taking this way out he is truly not responsible; he is irresponsible").

What makes a good book/movie/narrative work of art? To me it is increasingly about the plot and, to quote FCH, about how long the work can make you think about it after you're done reading it. It's clearly working since it's been weeks since I've read the books and I still couldn't resist making the quip despite the increasing amount of karma it is costing me for being off-topic.

Compared to another famous recent Hugo winner, Rainbows End, I thought Liu's work was much more enjoyable, because it was technically unpolished but had a great, interesting, innovative plot (and I'm sure half of the polish went missing in translation), whilst the latter was technically impeccable if now a little dated, but had little in the way of plot; the plot was almost a frame in which Vinge was developing his ideas about wearable tech and information operations' logical conclusion. It's like Esterhase mysteriously putting on a bad Hungarian accent in Smiley's People when he was the epitome of the try-too-hard British arriviste in TTSP - you can forgive them the glitch since the core of the writing is so exceptional and both Esterhases were fabulous characters with great depth.

Liu's ability to toy with the reader and take unconventional paths is rather rare in today's crop of writers. I would draw a parallel there with many of Charlie Stross' books; the polish is not always there, but he keeps reinventing himself and (at least for me) each foray into a new genre or avenue of thought is a success (and a source of great movies and books to explore).

As a side note, I thought it was one of the few successful depictions of a truly alien culture (what you call near-Strong-AI?) in SF today (and no, I didn't think Vinge's FOTD's Hugo winning hive-dogs were truly alien: put human heads instead, set it in medieval "Ireland-Spain", and you get Game of Thrones, down to an epic fire battle before the walls of King's Landing).

As a second side note, isn't it fun to draw parallels between TBP and the real world? How do you, today, as the PLA, deal with the fact the NSA can hear and read everything except your thoughts when planning defense strategy? Suddenly the sophons don't seem so alien... I don't like tin foil hats, so I didn't try too hard to put myself in the Chinese position, but it was definitely an interesting line of thought.

Finishing Dark Forest made me really wish I could speak and read fluent mandarin, so that I wouldn't have to wait another 4 months for the final volume! They pushed the publication date back again...

The Trisolarians are coming!

> As such, Oracle frequently oversold and underdelivered, but eventually grew big enough that it didn't matter

This is an extraordinarily common pattern in so many industries, when incentives are not aligned and it's OPM. And the worst thing is, if everybody else is doing it, you also have to do it.

For example, bank traders could (and in some places, it seems, still can) just lever up risk since if things go well, they outperformed their peers (riskier stocks go up - and down - faster) and made a great bonus; if a crash made things thornier, the bank would be in bad shape anyway and nobody would pick you out. I remember greatly admiring a famous "value" long/short equity fund in 2006 and a friend who was a more senior and experienced trader pointing out that they were "just levered beta"; three years later that fund was down a good 50%, proving him right.

I was reading, back in 2008 or 2009, the annual reports of a bank I was interviewing at. The letter from the CEO was noting, through the 2000s, the increase in risk taking in the mortgage securitisation space with increasing worry, and reassuring shareholders that no, at Bank X, this was not done because the ROI was not there to justify the risk.

Then a couple years before Lehman, there was a change in CEO due to "relative underperformance" of the stock, and the letter said something along the lines of "I know we're a bit late to the game but we are investing a large amount in that space and hired some great people from the competition".

Naturally that bank then needed a particularly large government bailout a couple years later, although the chairman of the board during the period famously retired with a very large pension (in fact his car was stoned for it by activists), showing that this strategy is still a great way to improve one's personal situation.

> I feel bad for a lot of the good employees who get caught up in the mess.

It seems to be standard in the job market to use the same strategy: overpromise and push the cost of an employee down as a result - and HR is actually rewarded for this.

I've been approached by many large tech companies for jobs and when talking with the managers there, the valuation they put on the options part of the package can get REALLY ridiculous (in some cases, 4-5x what the market thinks, let alone savvy investors). Older developers are wise enough to counter these offers, but I see a lot of younger folks enthusiastically buy into the myth and quit burnt out a couple years later.

I took an HR guy from a very famous and large Bay Area company that is mentioned here often for drinks and he admitted after about 7 whiskies that most of the people working there - including top, APAC-region managers - were paid about half what they could get in more normal companies, and that this was by design and a target. To be fair, I can't completely criticise them for doing it, since there was real value in working there for a while, as all the other companies in Singapore would then bid higher for your output detecting a "competence signal" in the American brand.

> people take out tens of thousands or even hundreds of thousands in loans to go to college

I also do not understand this when there are many cheap or free options [1] in the US, and most private colleges have excellent scholarship programs (i.e. free rides) for those smart enough, but not rich enough to attend (much better than anything in Europe). Especially when the cost of education has skyrocketed and its payoff is so low - hell, I usually ignore the education part of the CV when hiring.

Having the right to live and study in the US, being eligible for these opportunities is a treasure by global standards (even if some European countries have it a bit better). But that's a separate discussion.

> It annoys me that you are focusing on the things she did "wrong" rather than asking whether Yelp is culpable for some of the blame here.

This position you implied for me, but it was not what I was stating. I am attempting to understand the reasons behind the displayed pattern of decision making (and you indeed provided a plausible explanation), not passing any kind of moral judgment.

[1] quick Google: http://www.thebestschools.org/magazine/tuition-free-colleges...

> while being pitched ads for my brand new vocal studio in the South Bay

> took out a loan with 300% interest to move to SF to begin training December 1st at Yelp

> pays $2500 in rent (I make 2k a month so YES, my family has to help and I am so sorry to burden them), pays $1200 a month in daycare

I do not understand.

Why would someone take a loan (at a ridiculous interest rate) to move to a situation with higher rent (let alone cost of living) than salary, especially with kids in tow, especially in a country with at-will employment, especially having just started a new business? What was the upside that would be worth so much stress and risk?

I'm big specifically on the 787 and the A380 in this order (perhaps reversed if I can get on upper deck on the A380, which almost never happens) because of the noise levels, humidity and pressure, which are MUCH better than the 777 and 747 and earlier aircraft, even renovated.

Most airlines flying into SIN have old IFE on the older aircraft which means 1 in 2 seats has a massive metal box taking away half your foot space, and my knees and the sides of my legs really do not appreciate this. Next generation (e.g. new BA 773s) is a flat thin black box, or nothing visible. For some reason, a 3-4-3 777 is also a guaranteed trolley-shoulder bash every time they serve food, another good reason to avoid airlines that like to fly it (including SQ); my shoulder doesn't stick out as much from a newer aircraft seat but this could be psychological. Also, the A380 upper deck window seat has massive amounts of space on the window side which feels somewhat more private (you can practically stash your bag there).

One side effect of how quiet the 787 and A380 are is that you can hear babies from much further - perhaps 10 rows vs 2. So, it can be harder to sleep if you're seated too close to a baby cluster in Y.

I don't mind taking ages to deplane or board. It's a fraction of the flight anyway, and the actual walking is quick and the rest seated. I used to care about the crowd/volume when flying to SYD because SYD customs used to be highly inefficient, with LHR level of queues for foreigners. They made recent changes and now it's faster than for Australians using e-gates, so I'm not as bothered by a slow deplaning for that either.

I do mind less noise in my ears for the 22h of flying from Australia to Europe on 3 legs. I'll fly budget over SQ if the budget airline has a 787... sandwiches are cheaper than lost sleep!

My wife is Australian, and when we arrived in London from Sydney, I'd pop through the e-gates with my biometric Shenghen passport, and then settle down in the cafe opposite arrivals, and took my time reading the paper and enjoying a (not so) fine English breakfast. Usually had time for a third of a book as well. Then she'd finally get out.

I never understood why LHR had to have such long immigration queues. The solution is pretty simple: add more officers. Those booths were never fully staffed. I suspect some kind of labour dispute is involved.

I fly yearly or more to many of these airports.

LHR is down to what you are flying. If you are flying BA, you land and take off in T5 and it will be quick and painless, leaving you time for a nice meal at Plane Food. In one instance, I turned left out of the aircraft, went through empty security (as the only transferring passenger) and had a quick chat with the immigration officer about whether Plane Food was open yet (no :( ), and literally 3 minutes after deplaning, I was next to my gate. Make that 20 minutes if you have to go back to A gates (or whatever the ones with the restaurants are). Last time I got bored and decided to walk between A and C gates, which was a fun if eery experience down an empty tunnel several km long; sort of beats the crowded gate train.

If you are flying not-BA, then, you're not LHR's priority and it might suck, especially with a terminal change. I haven't done this in years for precisely that reason, but I remember cramped terminals, impossibly complicated options to get between them, and generally bad service throughout. Somehow there are always works at LHR, just like with the tube in London, where you check not to see which lines are down but which are up today. Last Christmas, the inter-terminal transfers were done via a free shuttle bus that used the motorway...

I hear you on the FRA walks. But Lufthansa reliably flies A380s internationally, even if the food sucks. And that's a guaranteed quiet flight vs the 50% chance of a crummy old 747 on BA... still, it must be at least 1-2km! And this time you have no time for the much better food options, and a lot of people are walking with you.

ZRH is reliably halfway between those two - the big plane and small plane buildings are separated by a medium length walkway which is always the same. However food is so damn expensive!

I would fly AF if I could just for the better food (and free flow champagne, even if it's bottom shelf) and guaranteed A380 again. But every single of my CDG experiences has indeed been a horror story. The place is like a study on how to screw up an airport. In one case, the only airside toilet was blocked for cleaning, which wasn't ongoing because the cleaning lady felt like taking an extended break, reading her paper and insulting those like me enquiring about whether we could possibly use the toilet before immigration queues...

For Americans living in New York or Washington, I'd recommend having a look at GVA. It's a small, very efficient airport with a lot of connections all over Europe and the Middle East (including cheap ones as an EasyJet hub) and has decent lounges. I think both United and Delta fly direct from these cities. Plus, if your layover is long, you can always get out and take the (free) 6 minute train to the city.

The biggest benefit for me internationally (flying long haul once or twice a month from SIN) is absolutely being able to take a shower between two long-haul flights and, if the lounge is not airside, shave (e.g. The Haven by JetQuay at SIN).

Peace and quiet is a second benefit. If you're waiting a couple of hours whilst sleep deprived (often the case for business travel or when booking budget flights that leave at 1:30am), those loud, constant announcements are very annoying. Bonus points if the airport has terrible seating and the lounge has comfortable sofas. If you need to work for a bit, it's a great place to do so, especially in airports that don't have free WiFi (increasingly rare these days).

The food is rarely good in Asia (at least at those lounges I have access to). Reheated noodles and some meat if you're lucky and flying at the right time. It's convenient, at most, saving you the trouble.

It really varies by city. HNL and NRT have terrible lounges for PP card holders - basically a room where you can sit for a while, sometimes with a free beer (but there is a great sushi-ya just opposite the NRT lounge). GVA has my favorite ones - nice view on mountains, usually empty (except if Aeroflot is flying soon...) and a decent breakfast with muesli and fresh bread and pastries. SIN lounges often have free massage chairs. Which are available for free in T3 as well, but it saves having to take the monorail there.

On timing, if you are flying long haul you can face serious delays (like a huge queue at check-in and security) and need to pad it by a couple hours. Most of the time, you don't face these delays and have a spare hour to kill, and in most airports the lounge is a better place to do so. Sort of applies to short haul too; I once missed LHR-GVA check-in on BA by a single minute and they were very annoying about it despite the empty airport (had to book another flight).

It comes down to the marginal utility of your money vs. the upgrade.

For example, I fly Singapore to Tokyo about once or twice a year, sometimes more often, always on Scoot, a budget airline run by Singapore Airlines. Upgrading from economy to ScootBiz costs as little as $80 with their "bid for biz" program. I always "bid for biz" and it's absolutely worth it: early boarding, free food, much more space (for yourself and your carry on), etc. So the marginal utility of the upgrade for me is worth more than $80.

At some level of net worth, flying First on a legacy airline to Tokyo, maybe Japan Airlines which I had to take in economy once for $1,500 vs the $200 or so Scoot can go down to, would have more utility to me than the $10,000 extra I'd have to fork out. I don't know what that net worth would be (probably north of 50 million), but I know that past perhaps a few million dollars in liquid assets, I would always fly business even at 3x the cost, and I know plenty of people in that situation doing exactly that. And there's people for whom flying SQ Suites [1] for $30,000 one way is a "discount" from flying the same route privately.

[1] reposting because such a fun read: http://dereklow.co/what-its-like-to-fly-the-23000-singapore-...

There actually is a formal version of "extreme status" on BA, in the form of a Black Card [1] [2]. Basically, you need to be something like "the guy responsible for the travel spend at big corp" and each membership is approved by the board.

Benefits supposedly include delaying an entire airliner by up to half an hour if you are running late, and it's so rare many lounge staff will deny you entry, not understanding what the card is.

[1] http://www.flyertalk.com/forum/british-airways-executive-clu...

[2] http://www.airliners.net/aviation-forums/general_aviation/re...

For our last 2 clients (with both, offering a "data team as a service"): the reporting tools were written in Haskell, using Servant [1] which got started trying to do the same at Zalora to set up a customer service management tool. The motivation was that it made painful stuff much simpler and faster to write (performance was a bonus).

It's also used for a bunch of tricky data manipulation problems that were harder to solve using pure SQL. Example: identify a unique customer based on attributes (email, phone, address, etc.) one or more of which can change when a customer tries to create a "new" profile to grab the $10 new customer signup voucher. This ran recursively on the entire customer dataset (12 countries, some customers had created as many as 250 profiles) in about 4 seconds on a m3.medium instance.

We don't/didn't write blog posts or tweets about it though. I think the set of people who write blog posts (i.e. both have free time to do so, the talent, and the motivation) is relatively small, and the number of blog posts and other visible stuff being published is proportional to the size of the community, so you'll see many more Node.js/RoR posts than you will on using set theory to reduce a 5,000-table flawed data model systematically or the kind of everyday stuff Haskellers are doing everywhere. Also, personally at least, I don't feel like I know enough to have much to offer by writing a blog, so I stick to making these semi-anonymous HN comments.

Interviewing Haskellers, I found a lot of examples of similar work - CRUD tasks, web services, etc. built in Haskell within a larger organisation and running quietly in the background. Something small and modular that can be tacked on quietly.

One of the creators of Servant moved on to Tweag [2], a French big data company, where he's working on PB-size distributed machine learning projects for large corporate clients, which I think is where Haskell really shines today. I suspect NDAs will stop them from talking much about it... I've always wanted to do similar work for clients but so far, no dataset was "big" enough to justify moving away from well established R libraries.

[1] https://github.com/haskell-servant/servant

[2] http://www.tweag.io/

Because angel investing is a numbers game, and extending funding beyond those who traditionally have access to it has always been YC's edge (whether or not they saw it as such). You can fund 1,000 companies for 30k and if one becomes Uber [3] and all the others fail your 7% stake might be worth as much as 145x your initial investment. 30 million is only a half dozen Yahoo parties, many countries have wasted multiples of that on fancy "entrepreneurship encouraging" offices.

When Georges Doriot financed DEC in 1957 [1], he got 70% for his $70,000 ($600,000 in 2016 [2]). These were proven scientists with a track record of successful research. It was considered a landmark deal because it was not blue chip, and the founders were not connected. How many smart people languished in corporate or government labs instead of starting another DEC or Microsoft? What was the opportunity cost of the lack of capital availability?

Then you have YC v1.0, which invested $20k for 7%... which we easily forget today was a non-controlling stake, and thus groundbreaking (or at least trend-setting). YC set the terms for the rest of the world to follow (slowly - I still meet, here in Singapore, angels who want complete control from the seed round onwards). Founders could actually still control their company after taking outside money!

Of course there were the ancillary benefits of the network and so on, but to me the really groundbreaking idea was also that significant expansion of the number of companies that could get funded. Working class scholar at MIT whose daddy doesn't know Tim Draper? No worries, you just need to fill in the form, YC will get in touch.

Competition, and inflation (both general inflation, and of opportunity cost from the two tech booms) drove that up to $120k, still for 7% - which even today is a very good deal for a pre-product company, and allowed thousands of founders to have access to a much better deal than they would otherwise.

Who is left out at this point? Pre-product companies that need money to get the MVP done, and foreign companies where seed investment is done on much worse terms than in the US for a variety of reasons that are regularly discussed here. They don't need much; as has been amply written about you can get ramen profitable on 20-50k with many ideas, and this is literally years of living costs in much of the world. But that capital is just not available because wealth is concentrated in (often foreign) land, bonds and blue chip companies. 1.5% is laughably small, doubly so when you consider this is basically free money unless you become a fairly decent success.

YC is not the only company who understands this. The very idea of 500 Startups - "if you have an angel, you're eligible" - is about numbers over any attempt at reading the future.

I can't wait to see how it works out. And how people will copy it. And I love the philosophy of it: YC is expressly saying that no, not all talent is concentrated at "Stanford" - that there are many variables in life (such as geographical location or social background) that will impede a large number of talented people from having access to the resources necessary to capitalise on that talent.

[1] https://en.wikipedia.org/wiki/Digital_Equipment_Corporation#...

[2] http://www.dollartimes.com/inflation/inflation.php?amount=70...

[3] http://www.bloomberg.com/news/articles/2015-12-03/uber-raise...

Related anecdote:

I had "computer science" classes in middle school in France. They consisted of typing classes. I still remember sheets of "QSDF JKLM" or whatever it was that had to be typed x times before moving on to a slightly more complex pattern. I think the class was graded based on how far you got down the sheet by the end of the hour.

The modern way to get untraceable money (according to [1]) is to set up (legitimate) small companies, then wildly overcharge the government for services.

Looks like your standard government incompetence and passing cash to well connected friends, and doesn't show up in any sort of government black budgets accounting.

The author estimates that one Western intelligence agency has over 2,000 such companies.

Fun fact: should one of these companies actually end up profitable in the course of its legitimate activities, it then gets overcharged by the unprofitable ones.

[1] http://www.amazon.fr/Missions-methodes-techniques-speciales-...

Yes indeed and too young to live through that amazing era of aeronautics, starting with Kelly Johnson's Shooting Star (I think I read in his bio that he was literally standing in the cockpit, canopy off, during tests), when people using slide rules pushed things like the SR-71 [1] from paper to 80,000 ft in 3 years, and everything was being "discovered" by incredibly creative trial and error. Or the insane USAF pilots taking Thunderchiefs dive-bombing (itself a controversial design certainly not made for THAT) over Vietnam with a 50% survival rate over 100 missions. Or, well, anything to do with the B-58. Yet it was that era that got me to study engineering in the first place and I wasn't alone.

As with [0] I actually sailed in Sydney Harbour in 2012 with a chap who flew Sea Furies into Korea (!!!) - maybe from the same carrier? He was ancient and I never learnt his name, but he was quite a character. Got shot down over water and swam to safety, apparently. As ever those stories were told second hand, as that generation doesn't like showing off.

Today's CAD-designed aircrafts (CAS apart) don't really fly, they just deliver an electronics package somewhere in the general vicinity of the combat zone and hang around until the computer decides what to destroy and gets on with it. And many of them don't even have a pilot anymore... still amazing technology, but not quite the same spirit. Certainly not when they also become covert welfare programs to buy votes in 49 states and international favours in 8 nations. Kelly Johnson claimed he could deliver Oxcarts for $3m per unit...

Those of my friends who stayed in the industry are now building rockets. In some ways, today's space race is yesterday's air - the rules aren't set, the market doesn't exist, and the politicians haven't done a land grab yet.

[1] And as was his habit:

> In 1968, Secretary of Defense Robert McNamara canceled the F-12 interceptor program; the specialized tooling used to manufacture both the YF-12 and the SR-71 was also ordered destroyed.

https://en.wikipedia.org/wiki/Lockheed_SR-71_Blackbird#SR-71