tst

11 ms·

1.
by tst 12y ago
Atul Gawande wrote about 'non-human' tools in medicine [0]. He advocated using checklists but he has also written about other tools. One I remember was a flow chart to predict some sickness. It outperformed humans solidly. The big
2.
by tst 12y ago
I'm also recovering from a depression which lasted for quite a while. It absolutely sucks because you think you're worthless, nobody loves you, you can't get anything right and the best would be if you just wouldn't exis
3.
by tst 12y ago
I think the analogy to Wozniak is great. I remember an article some months ago in which he basically said the same: He wants to tinker and hack. I remember watching some parts of notch's livestreams. I loved the enthusiasm he had. He w
4.
by tst 12y ago
About 7 years ago I wrote my first line of Python and I loved it. Before that I had some experience with C and PHP and the clarity of code and the great documentation was mind blowing. The slogan 'batteries included' was actually
5.
by tst 12y ago
You can check out Machine Learning: An Algorithmic Perspective by Stephen Marsland[0] which takes a less math-driven approach to ML. The code is available online if you want to take a look at it (it's written in Python)[1]. [0]: http:
6.
by tst 12y ago
I can recommend scrapy[0] if you work on a bit bigger problem. But even then if you familiar with scrapy it's incredible fast to write a simple scraper with your data neatly exported in .json. [0]: http://scrapy.org/
7.
by tst 12y ago
I can give you some feedback if you want. In the past I have searched freelancers in the web, app and design space. For each position I maybe look at ~200 freelancers. That means a freelancer is out on average in maybe 10 - 20 seconds. What
8.
Software Engineering, Not Computer Science [2004](stevemcconnell.com)
1 points by tst 12y ago|0 comments
9.
by tst 12y ago
As someone with no experience in computer graphics it's always insane to see demos like this. Especially using just around 250 lines of javascript. Really impressive. Also on the topic, the demo scene stuff is mind blowing, too. [0] [0
10.
by tst 12y ago
Here's a pretty good talk about restricted Boltzmann machines by Geoffrey Hinton. He explains the concepts and problems very well and basically without maths: https://www.youtube.com/watch?v=AyzOUbkUf3M
11.
by tst 12y ago
> I believe in the real world, you are best represented by a body of work and/or the ability to demonstrate that you are capable of tackling the problems associated with the job in which you are considering. This is such a good advi
12.
by tst 12y ago
I would even go further. The biggest problem isn't just using p values and R^2. I think the biggest problem is that a lot of people didn't learn statistics properly. Properly is a vague term. So what do I mean? Instead of obsessin
13.
by tst 12y ago
I completely agree on the package manager issue. I switched to OS X about 6 years ago. Before that I mainly used Arch Linux with wmii. My switch from Arch to OS X took about 6 months in which I slowly migrated. At first, it was pretty great
14.
by tst 12y ago
I would be interested to see how this effect affects the conversion rate. I could imagine that it could work very well on physical products.
15.
by tst 14y ago
Just a small tip which may ease the search for methods. The general term for "on the fly" learning is online learning [1]. The rest depends on your problem but there are often online variants of offline methods, e.g. when you work with Gaus
16.
by tst 14y ago
I think these numbers are based on one blog post the company made[1]. They wrote: "[...] defined that an app has to hold a position for at least 7 days to be considered as "ranked". That was the case for 265,959 apps from the 18th of July
17.
by tst 15y ago
There was a site called hearablog.com ( https://twitter.com/#!/hearablog ) which narrated some of the more famous tech blogs. Jason discussed it on This Week in Startups: http://www.youtube.com/watch?v=gTn5qNeHny4 I personally think that
18.
by tst 15y ago
The data is pretty easily accessible as JSON - however, here's the data I used: http://pastebin.com/nRYv40U8 Firstly, a boxplot with the quotient of entertainment contributions to entertainment & internet contributions. http://i.imgu
19.
by tst 15y ago
Here's the full report at Science: http://www.sciencemag.org/content/334/6061/1413.full
20.
by tst 15y ago
Yes: http://thenextweb.com/eu/2011/01/13/german-google-analytics-... However, I've never heard of someone paying this fine.
21.
by tst 15y ago
Probably Kuka: http://drstienecker.com/tech-332/11-the-kuka-robot-programmi...
22.
by tst 15y ago
Reminds me of the infamous Rapidshare Captcha with cats. [1] There were hard to solve. However, also for humans. [1]: http://www.labnol.org/internet/favorites/cats-inside-rapidsh...
23.
by tst 15y ago
That doesn't mean so much if you don't know the effect size and variance. In the end you have to look at the confidence intervals.
24.
by tst 16y ago
Under the law of demand, i.e. rising prices decrease demand, it is overpriced when lowering the price increases total revenue. Reading the comments on his post, it could be overpriced.
25.
by tst 16y ago
There's a really good book about this (sadly only in German). However, the author[1] writes about his first business which was basically a mail order company for Darjeeling tea. He imports the tea, a company packs and ships it. Almost all o
26.
by tst 16y ago
The comment by Myzplk is quite insightful: They're profit margin is impressive, but there are quite a few companies in the same range - http://ycharts.com/calculations/rankings/sectors/Technology/...
27.
by tst 16y ago
Which part of Finance? Banks? Trade platforms? Funds?
28.
by tst 16y ago
Also there's an older thread on HN: http://news.ycombinator.com/item?id=568556
29.
by tst 16y ago
Depends on your product and customers. http://news.ycombinator.com/item?id=1907332 http://news.ycombinator.com/item?id=335514 http://news.ycombinator.com/item?id=1787533 http://news.ycombinator.com/item?id=302202 http://news.ycombi
30.
by tst 16y ago
You're welcome! I think I see your problem in 3, (2 isn't too hard, you can buy your DNA sequence for ~$200). There are a lot of logistics companies which may have valuable data but can't mine them properly (especially the smaller, not Fede

More ›