It is tough - it's a big behavior change to go from paper to electronic records. It's less difficult when scientists are using wikis or Evernote though.
It's been built before. Many times. Mostly for chemists though. However, it's never been done in a way that's integrated with the actual tools scientists use to design experiments and analyze data, so you end up with a bunch of data silos. If you ask a scientist whether they can search all of the work and associated context/data that has ever been done in their lab from one location, the answer is almost always no. We are trying to fix that.
It's really technically challenging - life sciences is so broad so you end up with lots of surface area in your product. The "apps" we provide for manipulating the primitives in biology (e.g. DNA) are huge products just by themselves.
http://ycombinator.com/ideas.html
30. Startups for startups. The increasing number of
startups is itself an opportunity for startups. We're one;
TechCrunch is another. What other new things can you do?
If ARM's business model is around licensing their IP, wouldn't getting distribution (via Apple, etc.) be considered growth? They hit profitability in a year.
"EDIT: Same goes for id, Valve and other game studios - you see sometimes it's better to have the best engineers and researchers take their time, let them develop and mature the technology they're working on and build something long-lasting that affects millions of people."
Game studios take time because game releases used to be all or nothing. There were no second chances.
I can't comment on id, but the entire reason Valve built Steam was to speed up distribution and release updates incrementally (the previous patching system sucked). It went from idea to inception in about a year. It was also buggy as hell when it was first released. Now developers can easily release additional content and make game play changes after a game is initially launched.
Sorry I posted this before you were ready for it to be public! I was asking a friend about a license and he sent me your site and one thing lead to another...
It's a bit misleading to describe this as "algorithms in python". You'll find that the course is almost entirely theoretical (as it should be) and that there is a small implementation component for each problem set that just happens to be in Python.
I don't think I said "No one who gets below a 700 has impressive credentials". I was just trying to generalize that higher scoring students are more likely to have other impressive things about them.
The math portion of the SAT doesn't even cover precalculus (stops at Algebra II). Would you really expect someone who scored below 700 to have otherwise impressive credentials?
I'll agree that the more stellar applicants do tend to have high SAT scores but that's also in part because they are a) stellar and b) the SAT is a trivial test. I still disagree that SAT is a cutoff or benchmark. I highly doubt the admissions committee sees a 600 and throws the student out before reading the rest of their application to look for things that stand out. That said, there are tons of students with 800s that don't stand out at all and get weeded out.
Exceptional candidates will be exceptional regardless of their standardized testing scores.
Edit: Not trying to say that they admit all the exceptional people either. I'm absolutely positive plenty of people get screwed by the limited number of slots despite being more than qualified.
That's not entirely correct. You don't need near perfect SATs/grades, especially at a place like MIT. I know plenty of people (myself included) who do not fit that bill. And there is plenty of opportunity to discuss your projects. I mailed MIT a packet of all the interesting work I thought I had done and they even have an optional essay where you can talk about "something you built" in their application. I think they really care about those pet projects and they are the best differentiator they have.
This blog post makes it sound like he didn't fill out an application. However, I'm guessing he did and also sent in code as supplemental material (which is quite common).
College admissions is a game. I currently go to MIT and it's definitely more meritocratic than many other institutions of similar "prestige" but it's not an exception.
High school students today optimize their grades and SATs
and after school activities. They speak French and
Chinese, play piano and paint abstract art. They dance
around and play hockey and act like they help homeless
people.
It's ultra-competitive. That said, the number of kids who do this sort of thing is waning. I think people realize that being extremely good or having insane dedication to one thing or "theme" is key.
On the other hand this stereotype arose from an era where admissions committees were trying to create a class of well-rounded students. It still occurs to some degree. However, things are trending towards creating a well-rounded class instead (lots of highly specialized students).
From a legal aspect, how is this going to last at all? IANAL but I was under the impression most meta-search engines operated in at least a semi-grey area because they don't spoon feed illegal content to you. Most just index the big trackers and (conveniently) don't discriminate between illegal and legal content. I feel like when you are dropping torrents into buckets more specific than "video" and "audio" (e.g. linking multiple torrents to a specific movie, highlighting the HD versions) any conventional argument against aiding copyright infringement is out the window.
All of our classes are identified using numbers. The 6 corresponds to the sponsoring department (EECS) and 912 happened to be available within the department, so we snagged it.
It's been built before. Many times. Mostly for chemists though. However, it's never been done in a way that's integrated with the actual tools scientists use to design experiments and analyze data, so you end up with a bunch of data silos. If you ask a scientist whether they can search all of the work and associated context/data that has ever been done in their lab from one location, the answer is almost always no. We are trying to fix that.
It's really technically challenging - life sciences is so broad so you end up with lots of surface area in your product. The "apps" we provide for manipulating the primitives in biology (e.g. DNA) are huge products just by themselves.