Y
HN Search
Hacker News Search
new
|
comments
|
top
|
jobs
kirk21
searching PlanetScale…
1.
▲
2.
▲
3.
▲
4.
▲
5.
▲
6.
▲
19 ms
·
211.
▲
by
kirk21
13y ago
Any alternatives for Mendeley? I've tried Colwiz and hope http://bohr.launchrock.com will be great.
212.
▲
Challenges for academic research
(bohrresearch.tumblr.com)
2 points
by
kirk21
13y ago
|
0 comments
213.
▲
by
kirk21
13y ago
Yes, check this article. Have a nice day! http://pandodaily.com/2013/04/24/examining-the-thiel-fellows...
214.
▲
by
kirk21
13y ago
The pressure on academics is indeed an issue. They have to teach, guide PhD's, research and publish (without mentioning controlling internal affairs). On a side note. Has anyone found a good alternative for Mendeley? I heard http://bohr.la
215.
▲
by
kirk21
13y ago
Cool! It is necessary to have second opinions as the recent Excel blooper shows (see http://www.washingtonpost.com/blogs/wonkblog/wp/2013/04/24/i... ) I heard that http://bohr.launchrock.com/ is working on something similar.
216.
▲
by
kirk21
13y ago
It is weird that they don't stress the reduced maintenance costs. When something breaks down it is always extremely expensive (or they are overcharging you). Consider the life cycle of a car and there is something expensive that will break
217.
▲
Mendeley alternative
2 points
by
kirk21
13y ago
|
1 comments
218.
▲
Join the academic rebel alliance
(jointherebelforces.tumblr.com)
2 points
by
kirk21
13y ago
|
0 comments
219.
▲
by
kirk21
14y ago
I admire him since he admits failures. Also from an intelectual standpoint (e.g. first principles reasoning), he dares to push the limits. There are not a lot of t-shirts available of Elon so I made one myself: https://www.teepublic.com/sh
220.
▲
by
kirk21
14y ago
* Fusion research -> We should build big experiments (eg. ITER) -> First principles: if it takes years to start the experiment, you'll lose momentum. * Construction industry -> We should reduce the number of lanes so we can wor
221.
▲
Let's reason like Elon Musk
8 points
by
kirk21
14y ago
|
2 comments
222.
▲
by
kirk21
14y ago
Some products are being tested. http://www.wired.com/magazine/2011/04/ff_vasectomy/1
223.
▲
Broadband services
1 points
by
kirk21
14y ago
|
0 comments
224.
▲
by
kirk21
14y ago
Awesome! Good luck.
225.
▲
LERN (cold fusion): is it real?
(youtube.com)
1 points
by
kirk21
14y ago
|
0 comments
226.
▲
by
kirk21
14y ago
You can do these things even if you have to pay bills. Just start small with sth you can ship quite fast (a part of a robot or a small one, a robot for a niche market). You can grow your company and eventually you will be shipping robots li
227.
▲
Blanks in human knowledge
2 points
by
kirk21
14y ago
|
4 comments
228.
▲
Nerd dreams
1 points
by
kirk21
14y ago
|
2 comments
229.
▲
by
kirk21
14y ago
Don't know if you consider it as bioinformatics but tissue 3D printing is an interesting field. http://forums.reprap.org/list.php?194 If you want to keep working on machine learning, your progress also depends on which databases you can a
230.
▲
by
kirk21
14y ago
Some random thoughts: Prepare food and sell it or help tourists out. Create a website about living with a hernia.
231.
▲
by
kirk21
14y ago
It always strikes me how much time it takes to finish your paper (eg. conference template, correcting spelling mistakes, formatting figures etc.) while you could outsource this. Student-assistants are an option Furthermore it would be nice
232.
▲
by
kirk21
14y ago
Guess one of the problems is that it takes often a decade before a technology is mature and 'domesticated' by its users. You see sth on engadget and think you'll have it in your hands the next day but it takes years before you see it again
233.
▲
by
kirk21
15y ago
At first I believed that the revenue stream was like this: 1) ppl give their e-mail address 2) you scan the attachments (eg. words, subject etc). 3) you create a database 4) you sell the database to advertisers (that already have my e-mail
234.
▲
by
kirk21
15y ago
Discover (the characteristics of) every item in the universe.