crdb

The problem is that language is much higher bandwidth than the information content of the words. Context and implicit shared knowledge play a huge part and should be taken into account when assessing someone's words. Both the quote, your original use of it, and your reply to my comment are ad hominem attacks where the attack is mostly in what is left unsaid but understood by the relatively smart, well read, culturally aware audience of HN.

I'll now explain what I mean by taking both your quote and your reply and dissecting how I interpret them:

> "childish fantasy" "lifelong obsession" "unbelievable heroes" "emotionally stunted, socially crippled adulthood" "unable to deal with the real world"

The argument here goes: people who like Ayn Rand are immature and live in a dream world, which means you should not listen to what they have to say because it bears little relation with the real world. It's pretty obvious what the ad hominem is if you use this quote straight after someone quotes Ayn Rand: it is targeted at the poster and implies he is, himself, immature, living in a dream world, meaning one should dismiss his arguments without consideration.

> I was tempted to reply with a "found the objectivist"

This one is on par with telling a new black acquaintance about your black friends, or sending them news about black people doing great things. I should not need to explain it, but I will.

First, your original comment, as explained above, painted all "Objectivists" (Ayn Rand "fans", for those unaware of her writings; she groups her ideas under the term Objectivism) as immature dreamers whose arguments should be dismissed.

Then, you hint at what is on your mind, much as the man sending "black news" to his black acquaintance is hinting pretty strongly at the only thing he can think about looking at him. "I was tempted" might work in a bar discussion as a sufficient disclaimer, but HN is sophisticated and sensitive to such "fake disclaimers" (although those who point them out can be tarred with the "SJW" brush).

Your line of argument here is "I am stating that you are an Objectivist, therefore your arguments are not worth listening to because Objectivists are immature/unable to deal with the real world/see above". This is the literal definition of ad hominem, even if you hinted instead of outright stating. The context - HN, and its sophisticated audience and level of discussion - make it intentional and the equivalent of stating it outright.

You reinforce this line with the end of the sentence:

> but I'm waiting for a build, so I have time to kill

This really means "you are not ordinarily worth engaging with, because [see above], but since I have nothing to do, I will". This is reinforcing the ad hominem by introducing superiority and dismissiveness. "Here, child, let me explain."

> The quote I used doesn't attack the author, it attacks the work

It is extremely rare to attack a work using appeal to emotion without attacking the author. Usually, the best way to clear the doubt is to spell it out. You didn't. In fact, the brevity and context of the use of the quote, coupled with the fact it explicitly attacks Ayn Rand, imply the opposite.

> If you want to criticize it, you would do better to point out that the quote I used is not constructive criticism.

I did criticise it. Pointing out that appeal to emotion is not a valid line of argument is a constructive criticism. I have expanded on the idea earlier in this reply, but the core of it is that appeal to emotion attempts to use emotion to make you skip logical steps and accept a conclusion you might not have ordinarily.

> humorous quips rarely are

Ridicule is one of the most effective appeals to emotion. I could probably point 10 examples from the first 15 minutes of the Presidential debate last week. Even the words used by both candidates are carefully chosen (see [1]). Your particular quip is only humorous if you belong to the tribe that likes to auto-ridicule Ayn Rand readers (without having ever read the books or even the arguments), and otherwise serves to silence those unwilling to take the social penalty of speaking out against it.

Lest you think it is personal, it is not. Hell, I know where you're coming from - in a previous era, I was banned for several weeks from an internet forum for calling someone an "internet redneck" for presenting Republican-favouring arguments.

I am simply fed up with the rising market share of appeal to emotion in all kinds of domains from advertising (thank you uBlock, Privacy Badger etc. for making the internet useable for me again) all the way up to politics and the legislative process. This is my little contribution to fighting against this trend. And yes, the use of the word "fighting" is itself an appeal to emotion.

[1] http://blog.dilbert.com/post/147998060931/the-dark-and-rotte... - "The Dark and Rotten Election" - Scott Adams

Responding with an ad hominem - whether at the author of the quote or the poster of it by association - is both an ineffective response (the appeal to emotion signals an inability to argue rationally) and lowers the quality of discussion (by taking it away from the point, into emotional territory).

Constant appeals to emotion and resorting to a wide array of fallacies allows for strongmen to build tribes which follow them blindly (no need to bother forming policy) and is how we end up with Trump vs Clinton in 2016 (see [1] for the impact on the legislative process). Reducing the quality of debate is therefore a harmful thing with real consequences.

One line of argument might be to criticise the paranoid nature of the quote: it is entirely possible that a mess of laws arises naturally from a mix of special interests doing some lobbying to preserve small advantages for themselves, and bad laws arising as a form of horse trading necessary to get policy enacted. Only then does this mess presents an opportunity to manipulators.

[1] http://imgur.com/a/Wmoex - "Voting Relationships between Senators in the 101st through 113th Congresses"

If you mean the fund(s), I disagree.

The strength of YC is the relatively large value they provide in exchange for your equity. Their "price" is still excellent and continues to clear the market. This becomes rapidly obvious if you attempt to raise early stage funding and start comparing options.

Aside from the actual cash and the famous network, YC and its partners have earned a reputation for sticking with their investments through tough times instead of writing them off. This seems relatively rare amongst early stage investors, especially outside the US. In my relatively few years of experience in business, I've learnt to value people in the "cooperate cooperate" quadrant of the Prisoner's Dilemma very highly, and it's not a reputation you can "buy" quickly.

HN is a place to have civilised discussions thanks to both a seed of core tech personalities for historical reasons drawing a network of quality, and thorough and active moderation. I think "self-perpetuating machine" is harsh and not quite representative of most of the value derived by those of us who are not part of the YC ecosystem and unlikely ever to be (and yes, I haven't read the original article).

PS: nice site! Light, fast and with just one tracker (but "old geeks" would crunch web logs, surely?)

This is a false equivalence. Pot smoking is victimless (at most, your landlord will need to get rid of the smell for the next non-smoking tenant). Murder has, by definition, serious consequences for someone else.

In a legal system whose premise is that one should be free to do anything so long as it does not infringe upon the rights of others, it it consistent to legalise pot and keep murder illegal.

Sites like CrowdFlower have point-and-click interfaces that allow you to just upload a CSV and drag and drop its columns as fields in your task template.

What I used to do was host the images on a server, publicly, then upload the URL of the images and use it to display the images within the task. It's OK for one-offs. You get a thousand free rows with the CrowdFlower trial [1] and that's probably the most painless way to go for a classification problem.

In our case though, everything will be written in-house. Automation is how we keep headcount low.

[1] https://www.crowdflower.com/plans/

Thank you. This is the best available solution and I didn't know about it.

We'll probably go with it for now, until we hit a scale of tasks that justifies setting up a US subsidiary and dealing with the IRS.

Hi Alex, thanks for your email. Replying here for the benefit of other readers.

We can't use your product for the same reason we can't use most of the alternatives on the market: there's no way to push tasks with custom HTML/CSS templates, as is possible with MTurk. This is important at scale.

Do let me know if this ever changes.

Thanks for offering, but I doubt this methodology will scale. We really need a stable company with a product we can trust will both be around in 5 years and that can handle large volumes of tasks.

Looks like a neat site, but it doesn't allow us to structure the task (use an HTML template). The flexibility is important.

For my clients, the standard use cases for e-commerce.

For example, you might have 100,000 products plus another 5,000 new ones listed per month. Your human error rate might be 3%. When a product arrives, its buyer (the corporate person responsible for signing on the brand) gives it a category and it shows up on the site. So 3% of the new products are erroneously categorised by their buyer.

You give X human votes per product, say, 10. "Which of these categories applies to this product?"

If the votes agree with each other, you can just input that as the category. If there's a split, you can feed those products to a more open ended question like "what category is this product" and use that as a starting point for either renaming the categories, finding a new category, or tagging: if it's a 50/50 split, you can just tag the product with both categories.

This is all automated and used to cost me around $200 per 5,000 products.

Another example might be training a data set for a machine learning algorithm. You send a data set to be trained by Turk and use it as training/test sets. I'm keeping this one deliberately vague for now as AFAIK we are the only ones in our space doing this and I don't even want to mention the space due to a relatively smart competitor.

As a hypothetical example, you might be trying to predict the category from the description, gender, picture features, buyer and brand, and the above categorisation tasks can be used to train the algorithms you're testing out.

I could, but it's time consuming and I'd have to deal with the IRS.

By not operating in either the US or the EU, I save myself a lot of administrative trouble and potential tax liabilities.

A way to use Mechanical Turk from outside the US. I've needed it with every client I ever worked with, and I need it now and know a dozen companies who do.

CrowdFlower has a de facto monopoly on the outside of the US supply and charges an enormous premium for it - enough to turn off most of them. I certainly don't want to pay several thousand a month for the right to submit jobs, although I was ok with the 25% premium in the old days.

If it's an alternative marketplace, it has to have excellent automation via API. I'm not going to use this for questionnaires, I'll be submitting thousands of jobs automatically.

If you have built this already, please email me.

It's not just comfort, it's signalling. A company with a laid back culture, open kitchens, free food, a nice expensive architect-designed office is signalling that it is prosperous ("we don't cut costs here, we don't need to"). A company getting less prosperous is also about to get less promotions, less bonuses, and maybe some layoffs.

If you are 10 years old and you see your dad replacing his BMW by a beaten up Toyota, yes, he could just have decided to stop taking on so much credit, but chances are you'll think he's not doing well, even if he reassures you it's for perfectly valid reasons. So you mentally brace for giving up on that new bike you wanted for Christmas.

"They've cut the kitchens."

"Really? We just got an extra Jura coffee machine in every conference room..."

"Interesting, sounds like your cash flow is good at the moment. Are you guys looking?"

And the earlier you leave, the less affected you are - you benefit from the still strong brand, you get decent garden leave, you have little competition on the job openings and your former colleagues will be nice to you in case they need a hand out of the sinking ship later.

If you want an example of straight bribery, which is sufficiently far in the past to be non-partisan, try Rick Santorum with Accuweather.

Accuweather employees (mostly the head and his brother) donated $10,500 to his PAC and he subsequently introduced an Act trying to wall off the National Weather Service's data from the public domain.

What surprised me was how low the price tag was. These days, that barely covers a new Submariner.

See: https://en.wikipedia.org/wiki/Rick_Santorum#National_Weather...

Technically, the relation (value) is the set of tuples (i.e. the data itself, "the true statements", the rows in the table), the relation variable is what is defined by CREATE TABLE (i.e. how the data is constrained, "what can be true") and what most people are trying to model.

The relational data model - the set of relation variables - is thus the "equation that defines what is going on in your company" and what the DBMS puts in - the set of relation values, usually abbreviated to relations - can be seen as "the history of what happened at your company".

I see it differently (and have used "true" ride sharing services in Europe, like Blablacar). There is spare capacity in cars on the road, if you fill it up, regardless of the dynamics of how it happens, it's ride sharing. I might be the one sharing my taxi ride, or it could be a driver going to work. In both cases, the effect is one less car on the road for that trip. Most importantly, in the Uber Pool case the car I've just ride-shared does not then go park in the centre of town and take up valuable space, it keeps moving.

The problem with Blablacar-type apps is flexibility, unless you're doing a very common trip you'll find yourself having to be very flexible with hours and pickup and drop-off points. On the upside, it's much cheaper than alternatives especially for trip between cities.

Uber's model of having drivers basically work at as close to full capacity as possible most of the time is excellent and a great way to reduce congestion at the heart of cities, or even just the demand for parking lots.

Vans get us into the implementation, rather than the model. The model is that if there are spare seats they should be filled. The implementation is the type of car being used. I suspect that if vans were profitable for Uber Pooling they would be much more used by now, but the optimal size seems to be the family sedan or the small SUV, at least in Singapore. They might make more sense in frequent suburb -> CBD type trips where there is a lot of demand for the same trip at the same time, although in the situations where I've needed to do such a trip, there's always been frequent express buses or direct trains provided by the city (e.g. in Sydney from the North West suburbs to Darling Harbour).

> Passengers don't share the ride with other customers

I beg to differ. I just took an UberX from Changi Airport to home, a 30 minute ride during which the driver took on and dropped off a total of 3 other passengers.

The cost was about half the taxi rate.

It won't be as pretty, but I deal with large data models all the time, and like to use SchemaSpy [1] which generates an interactive page in HTML and can be used on the command line (I guess you could always modify the CSS to make it pretty). It's literally one of the most useful tools in my life, and the output is good enough to show to clients.

If I'm designing a DB or even just an SQL example, I'll run the code on my local machine (psql + the Postgres app [2]) or if I'm lucky, the client already has a server running Postgres and I can run it there instead. All SchemaSpy then needs is access to the DB and voila, interactive example.

[1] http://schemaspy.sourceforge.net/

[2] http://postgresapp.com/

> real estate tax evasion in Vancouver

If your problem is tax evasion, the solution is enforcement. If it is a big problem, it will pay for itself when solved.

Right, then let me reword, since I used the wrong wording: if you are resident in Canada for tax purposes, wherever you are resident is tax free if you own it. Otherwise - including a single day residency a year - it is taxed.

Well, I can't speak for Canada, but in most countries I've lived with the exception of the UK (non-dom rules) global income is taxed in your primary residence. So if your main economic activity is in Singapore and you declare your residence there so as to take advantage of its lenient tax rates (for example no capital gains tax), your Canadian home is not your primary tax residence and you should get hit by the land tax.

Conversely if you declare that you live in Canada whereas you're busy making a million a year in Dubai, that million ought to be taxed according to Canadian tax law as if you resided there.

There's plenty of rules already in existence to determine what is your primary tax residence as far as taxes are concerned, precisely to stop, say, a French resident from incorporating in Switzerland, getting paid into the structure, and not paying tax on his income (he can always try not to declare it, but then he becomes a criminal hoping that the rule of law is not thorough).

This used to be my position as a committed (or idealistic, young) libertarian, but, following the principle that you should be able to do anything so long as it does not infringe another individual's rights:

- there is such a thing as the commons;

- there is a cost to the government to protect your rights which is much higher for land than "virtual" assets and the cost to investors ought to reflect it.

Commons: I like that Singapore, for example, is taxing drivers around $90,000 per 10 years for the privilege of riding a car. The air is much cleaner and streets freer of traffic. Both of these were damaged by too many car owners. Similarly, the supply of land is limited whilst population can grow ad infinitum, creating ever upwards pressure on land prices.

Cost: in terms of individual rights, the army, police and justice defend your right to own that land free of damage, spurious lawsuits to drain your resources, land grabs by powerful men, and foreign invasion forces. This forms the basis of modern society and the implementation of your individual rights. The cost is proportional to land area but not as strongly correlated with company size in the equity case (you might have a few more problems but not proportionally to market capitalisation, and they involve court cases rather than policing, with most of the high costs - lawyers - born by private parties).

In theory, any land owners ought to be taxed, however, I make an exception for one's primary home (tax residence) considering that all citizen are naturally short housing (they need a place to live) and buying a home puts them in a housing neutral position and also has many positive effects on the commons (people are more invested in the future of their country if they own their home, less likely to move somewhere else, etc.).

Vancouver is a great example of citizen who have created ideal conditions for home ownership by participating in building a peaceful, prosperous economy which respects individual rights. Foreigners have accordingly swarmed to take advantage of the fact they were not charged for the benefits arising from ownership.

This has caused citizen - beyond the first generation of land owners - to be penalised for having built their own country up, as they end up unable to afford to live in it, burning up valuable income into paying high rents and house prices. A tax - which should offset what Canadians would pay elsewhere on productive activities, like income - is one of the best way to correct this imbalance, by charging the cost of enforcing individual rights to those who benefit from it.

A step in the right direction. However, the rule ought to be pretty simple: if it's your tax residential address, it's your home. Anything else isn't.

If people want to invest in stuff, let them go for stocks, where their capital will at least be supposed to create value.

I've seen much more than 15%. The US businesses run regular discounting that can make every day, branded goods as much as 70% cheaper.

It doesn't look that way online anymore because many businesses have sprouted up with overseas operations and Australian branding. For example Galaxy Perfume, a .com.au, is owned and operated by Perfume Emporium, a US business.

Shipping takes about a week if the warehouse is in Asia, which is OK since Australian post is particularly slow and with the exception of high premium same day delivery services (as offered by, say, The Iconic in big cities) you will wait half a week anyway.

The businesses that play by the rules and have a local warehouse can be hilariously off. For example, I recently bought a bunch of shirts from a Jermyn Street shirtmaker's UK website, because it was a good 50% cheaper, even with international shipping to Sydney, than getting them off their Australian one.

Thanks, done.

For those who think this is addressing a relatively small problem, one of the major Australian banks analysed [1] the accounts of 1 million of its customers (1 in 22 Australians) in 2011 and found that 74% of e-commerce by revenue was cross-border. Three quarters! In some markets, over 90% of the revenue was from abroad.

It's not just fashion and cosmetics, things like lawn mowers were also getting imported.

Part of the problem was AUD appreciation which put the MSRP of most things a good 50% above the rest of the world. Coupled with high costs in Australia (due to tight labour legislation which helps sustain an egalitarian culture), no Australian retailer could compete. The AUD has weakened substantially since then so I wonder if the picture has changed enough to make GST matter.

Charging GST might help a little bit, but the issue is FX, high local costs, and relatively cheap international shipping. For example, ASOS offers free international shipping so long as you are ready to wait 4 weeks. This is fine with most customers who prefer buying the same jeans for $60 and wait to paying twice as much and having them instantly. They employ only a handful of people in Australia to manage inbound shipments and customs, but it is (was, 3 years ago when I last looked) a significant chunk of their global profits.

That being said, protectionism didn't work out for the Australian car industry.

[1] http://globalmarketsresearch.commbank.com.au/ReportWeb/d13de...

> our real goal is to 10x or 100x the number of startups that we help, and Fellowship in it's current form couldn't easily do that

Do you account for the effects you have on the rest of the community when calculating your impact?

For example, I suspect YC is strongly responsible for the rise of "low amounts of equity without control for a small amount of funding" seed rounds, and for the existence of accelerators, in both cases globally.

The Fellowship model of investing tiny amounts - like "sparks", lighting a fire - into an enormous number of teams with as little friction as possible was the next logical step and I was looking forward to Fellowship copies sprouting all over the world.

Unfortunately by effectively ending the program you might be sending a signal to global investors not to try it, that YC Core is the deal size floor that works...

As for how it got to promoting Mugabe into turning one of the most prosperous nations in Africa into the new Weimar Germany, one interesting reading of the situation:

The Rhodesian Selous Scouts (the new, mixed race, elite force responsible for 3/4 of the enemy casualties during the Bush Wars) made the decision to go after the "moderate" party (ZAPU, led by Nkomo), weakening his troops shortly before the final election (where all Rhodesian troops were called back to the barracks and guarded by British troops).

They correctly expected Mugabe to win the election, since he could take over the countryside with no opposition, and the plan was that he had such an awful reputation that the British would not allow the election result to stand, whereas they'd be fine with Nkomo. Unfortunately, they underestimated a. how much the British were annoyed with Ian Smith and UDI and b. how little they cared about Mugabe's true nature, and generally about the fate of the "natives" in their own colonies, just wanting to wrap up the whole thing as quickly as possible and stop being the centre of global attention.

In fact you can trace Mugabe's rise to the assassination of the moderate, educated, highly competent leader of ZAPU and later ZANU, Herbert Chitepo, a national hero and the country's first black lawyer, by a former British SAS soldier working half on contract, half out of ideology for the Rhodesians. The assassination, which was followed by Rhodesian disinformation to make it look like it was a faction war, was supposed to seed dissent within the enemy camp. Instead, Mugabe took advantage of the chaos to emerge as the leader and the rest is history. For all we know, Chitepo could have been the African Lee Kuan Yew.

The rise and fall of Rhodesia is one of the most fascinating parts of 20th century history and I really recommend getting acquainted with the period. If anybody has ZANU/ZAPU/ZANLA-side sources, I'd love to read them as I could find very little on the subject and much I heard from that point of view was from Zimbabweans whose family members were fighting on that side of the conflict.

He left out the part where one must have a ruthless commitment to the goals of the company (or less charitably, oneself) at the expense of people.

An example from Napoleon's life was Marengo [1]. One of his generals (Massena) was besieged by the Austrians in Genoa. Napoleon had to make a choice between rescuing him - as he had requested - or cutting off the main Austrian force led by von Melas which was making inroads towards France; he did not have enough men or time to do both. His decision to cut off von Melas paid off in eventual victory, the reappearance of "independent" Italy for the first time in centuries, and peace for France, but cost him the city and Massena's friendship, as the latter felt betrayed, raison d'etat be damned.

In another famous instance [2], he loaded grapeshot in canons and fired at massed French citizen in the streets of Paris to defend the Revolution from Royalists. His calculation was that if he slaughtered a few in a brutal manner, he'd shock the rest into retreat; he pulled it off, saved the revolution at the cost of only a few hundred lives, and was rapidly promoted thereafter.

In both cases, Napoleon's own interests were aligned with that of the country and its people. In other cases, such as choosing a policy of economic protectionism as a response to Britain's industrial revolution being too advanced to compete against, or reinstating the Catholic Church as the official state religion and making succession hereditary once again, one could argue they were not.

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

[2] https://en.wikipedia.org/wiki/13_Vend%C3%A9miaire

This is because there are two families of tech writing, one is concise and detail oriented, the other is long form because it contains complex ideas that cannot be summarised in bullets.

Take as an example most of Dijkstra's output. Could you reduce "On the Cruelty of Really Teaching Computer Science" [1] into bullets and still retain its message?

[1] https://www.cs.utexas.edu/users/EWD/transcriptions/EWD10xx/E...