The ability to communicate is important, sure, but is writing a fancy, bogus essay about how motivated you are really the best way to measure it?

So writing about technical matters is too hard, so we should have the students write about their feelings and ambitions instead?

Obviously you wouldn't expect them to do it as well as a seasoned veteran, but being able to quickly learn about a technical matter and explain it in a clear, concise matter to someone without prior knowledge of the subject is exactly the kind of communication they are be expected to perform during the studies.

With alot of help then :) Personally, I found the essay to be the kind of pseudo-creative look-at-me-I'm-talented type of writing that shouldn't be encouraged for an engineering school. I don't blame the writer, she did what was needed to get in and did it quite well.

Thinking too much like an engineer - you got me. The problem with many subjective measurements is that they can be gamed as well: the only reason someone would describe themselves like in the essay in the article is to appear smart, driven and ambitious; in essence gaming the system. The simpler the indirect measurements are, the harder they are to subvert.

I'm sure they have to show actual skill in essay writing, but is that really the way to measure communication skills in a would-be engineer? I don't think so. The biggest problem from the fairness pov is that these essays can be written with outside help. Maybe have them read an article on an engineering topic on site and summarize the main points. Or have them write a short paper like the ones actually done during the studies.

I'm sure that a post in slashdot isn't going to make them change their policy. Since MIT has now removed the 500 word essay, it seems that it might not be a good way to measure the "drive, ambition, etc.". It seems to me that such an essay helps just as little in determining these important attributes as the xkcd capcha in differentiating humans and computers. With a little help anyone can write 500 words of bullshit like the "great" essay in the article.

I think the most important argument against the use of subjective measurements like these is that they can be used to pervert the selection system.

Here's my sample essay (in 33 words):
My dad is the CEO of Big Corp and is willing to donate millions to the university I'll be attending. So you see that I'm a very good candidate for MIT.

I agree that skills other than pure math are important for success in a top engineering school; test for those, but don't put people ahead of others because the say they're motivated.

I'm not familiar with the details of MIT admissions, but I can comment based on the admissions in engineering universities here in Finland.

The basic problems is very similar: our equivalent of the SAT:s (nationally standardized examns at the end of highschool) are bad measurements for selecting students, because most of the would-be engineers score in the top 10% of the country in math and physics. The solution here is to hold separate entrance examns that are common for all the engineering universities. The material is basically the same (high school maths and physics / chemistry), but the difficulty is set higher: most high schools students would get no points on it, only very few can score full points, but it nicely measures the differences between the good and the best. In practice getting 50% right will get you into most programmes, 85-90% into even the most popular / exclusive.

Like Jim_v2000 said:

A very important part of a selection system is fairness: it's very hard to objectively measure differences in "Drive, ambition, ideals, character, motivation", so it's better to stick to the skills that can be measured and are relevant to the subject.

From wikipedia:

Objectives of the protection

As with other intellectual property rights, the use of geographical indications is intended to increase the income of rightholders. The preambles to the Regulations cite consumer demand for quality foodstuffs and identify a number of goals for the protection regimes:

• the promotion of products with specific characteristics, particularly those coming from less-favoured or rural areas;
• the improvement of the income of farmers, in return for a "genuine effort to improve quality";
• the retention of population in rural areas;
• the provision of clear and succinct information to consumers regarding product origin.

The provision of a recompense for efforts to improve quality and the need for consumer protection are often cited as justifications for trade mark protection in other domains, and geographical indications operate in a similar manner to trade marks.

I think the last one is very important: I want to know whether I'm eating Greek feta or other "white cheese" - very often I'll choose the cheaper, non protected option, but at least as a consumer I'm able to make an informed choice. And alot of these protected products are much more obscure parts of traditional diets, they need special treatement if we want to preserve them in our markets.

If you feel like this, you should have no problem with only Champagne region sparkling wines being called Champagne - you can just choose any sparkling wine that you prefer and ignore the name. For those of us who like to make the distinction, it's good to have the power of the EU on our side.

The point is that you ran your tool manually, with the intent of collecting all the data. Why would I or one of my friends run your tool for the benefit of an evil 3rd party?

The problem described in the article is that facebook apps like funny little quizes get way more information than they need. I propose that application developers could opt to use only limited user info and get that fact recognized by Facebook. When facebook asks you to grant the rights, apps that ask for too many rights would be suspect and get less users.

There would be a simple solution: have the apps state what rights they need through the api. Instead of:

"Do you want to give this application the right to your profile?"

Have a question:

"Do you want to give this application the right to the following information:"
- your name
- your profile picture
- list of your friends names

A quiz really doesn't need to know more than that, an photo editing application might need all your photos etc. The api could also let the application developer write a short description of why such info is needed. An application that asks for too many rights would not get as many users - at least the smart users would be protected.

Here is the problem with your argument: you are comparing really fresh organic carrots from the local farm to non-organic ones from the supermarket, which possibly travelled for days before reaching you. Of course they taste better! Why do you think that is? Because they are organic or because they are fresh? I've eaten non-organic tomatos right from the plant, they tasted so much better than store bought ones.

You should keep eating the superior, local carrots. You should not, however, use that as an argument for producing food with less efficient methods like organic farming - the key to better taste is to get the veggies locally.

You have just formulated one of the main arguments for legalization of drug use: it is the fact that drug use is illegal that causes crime, not the drug use itself. If the users were able to purchase their pot at the local drug store or 7-11, there would be no need to break the gun laws.

The other side is of course that being intoxicated leads to crime, just check any statistics on how many murders were commited while drunk - a legal drug. Still, prohibition of a substance with significant demand leads to a black market, which provides organized crime with a good income. And many people who otherwise would have no dealings with organized criminals do so, because it is the only way to get drugs. The step to harder, more dangerous substances becomes smaller, since pot smokers are already doing something illegal and have the connection to the criminals.