tilly

5 ms·

1.
by tilly 17y ago
Does this fact not strike you as sad? According to http://www.time.com/time/magazine/article/0,9171,1653653,00.... if we are generous in our estimate of how much is spent on gifted programs, we spend 10 times as much to educate the menta
2.
by tilly 17y ago
When I was looking for a job I'd find a potential job, then used LinkedIn to find who at that company I had some possible connection to. That was incredibly valuable.
3.
by tilly 17y ago
I don't buy it for a simple reason. It is easy to look around the macro environment and see plenty of reasons why entropy has a long way to increase. For instance for several billion more years the Earth should be receiving energy from th
4.
by tilly 17y ago
The answer depends entirely on the job role. For instance a QA person needs attention to detail which is far more likely to be found in the academically successful person. But a senior software developer needs flashes of brilliance that y
5.
by tilly 17y ago
That sounds funny, but how true is it? Let me use http://ali.as/top100/ as a list of the top Perl modules. #3 on the list is Pod::Escapes which I suspect is not implemented in very many Lisp programs at all.
6.
by tilly 17y ago
You can still do a Y combinator in JavaScript. In http://www.mail-archive.com/boston-pm@mail.pm.org/msg02716.h... I not only show how to do it, but I show one way that someone reasonably could come up with the construct. But I agree with
7.
by tilly 17y ago
If you have them in your language the common better solutions to those use cases include: 1. Labeled loop control 2. Exceptions PHP has #2 and a bad form of #1. In PHP you get to say how many loops to break out of, but not the name of the
8.
by tilly 17y ago
He's right that the period vs comma is likely wrong. But it is a legend based on a real event. For details see http://en.wikipedia.org/wiki/Mariner_1 or http://catless.ncl.ac.uk/Risks/5.65.html#subj1 for more details. Here is a summar
9.
by tilly 17y ago
The article is fine until it starts talking about Lorentzian Relativity vs Special Relativity. Then it starts making amateur errors and misses obvious points. As a random example, the traditional exposition of special relativity depends on
10.
by tilly 17y ago
"It's like hitting on women at a funeral." Am I the only person who was reminded of Ryan O'Neil? (For those who don't know, he accidentally hit on his own daughter at his SO's funeral not long ago.) But back to the point, I agree that expl
11.
by tilly 17y ago
I agree that a given A/B test should change only one metric at a time. However I've had excellent results from running multiple A/B tests in parallel. As long as inclusion in each is independent and random, the results of each are informa
12.
by tilly 17y ago
Writing tests is only good if people actually run them. The default configuration for RubyGems doesn't run unit tests. Therefore unit tests provide a sophisticated check for, "Works on my machine" but isn't so helpful for getting good bu
13.
by tilly 17y ago
I agree with the point, but may draw different conclusions from it than most will. My point of view is informed by the book, First, Break All The Rules. That started as a study by the Gallup organization on what divides great managers from
14.
by tilly 17y ago
We can quantify this. Software Estimation by Steve McConnell on page 66 lists many factors from the Cocomo II studies with their relative impact on software development. Multi-site development on average causes things to take 1.56 times as
15.
by tilly 17y ago
My opinion is that MENSA is for people whose main accomplishment in life is doing well on an IQ test. And who are insecure enough about it to pay someone else to validate that accomplishment. This opinion is shared by a number of people t