Y
HN Search
Hacker News Search
new
|
comments
|
top
|
jobs
agbell
searching PlanetScale…
1.
▲
2.
▲
3.
▲
4.
▲
5.
▲
6.
▲
35 ms
·
331.
▲
Facebook Prepares To Open Source A New PHP
(allfacebook.com)
2 points
by
agbell
17y ago
|
1 comments
332.
▲
by
agbell
17y ago
In short, a computer program is tactically superior, whereas a human chess player is strategically superior, making the combination of the two a completely superior chess player.
333.
▲
Advanced Chess
(en.wikipedia.org)
2 points
by
agbell
17y ago
|
1 comments
334.
▲
by
agbell
17y ago
Great Idea. Rather than re-flow, it cuts up the pages so they fit in 'landscape' mode. Reflow works well, but not with diagrams and images.
335.
▲
by
agbell
17y ago
If you are logged into Google, it sends more info using a different function For Example: return rwt(this,'','','res','1','AFQjCNEdw7LJgOmjlxPAZ-jF7e3aPy1bbQ','&sig2=D3Wg-V0MDjmRS-gqjk66rA','0CAkQFjAA')
336.
▲
by
agbell
17y ago
This is a scam, but likely not one entity. Probably an affiliate marketer found how well linking two auto-bill offers together with a fake story worked and then everybody started copying. The fake blogs and fake news sites seem to be the c
337.
▲
by
agbell
17y ago
Seems like MS has had go at this problem many times since 1990 and never cracked it. http://en.wikipedia.org/wiki/WinFS#Development
338.
▲
by
agbell
17y ago
I have heard that if you notify paypal you may see a big spike, then when it gets auto-flagged, it will quickly be un flagged when they see the note on your file.
339.
▲
by
agbell
17y ago
Yes, the deals are sometimes crazy exploitative. But at the same time, they sometimes dish out lines of credit to people just because they like them. The real problem with the show is the format. The amount of money they ask for, they need
340.
▲
by
agbell
17y ago
This article is from their speculative 'The world in 2010' publication, which attempts to forecast the coming year. That means its a bit more speculative than standard economist articles. Wall-street bailouts definitely made business less s
341.
▲
by
agbell
17y ago
I hope we don't lose Dragons’ Den. The Canadian version is great entertainment.
342.
▲
by
agbell
17y ago
"While most of the customer development/lean startup process focuses on product development you can also apply it to your marketing process." Customer Dev / Lean start up is about marketing (ie finding a market) not product development
343.
▲
by
agbell
17y ago
A pattern matching solution is much more readable and it forces you to build small functions if you are pattern matching in the function declaration.
344.
▲
by
agbell
17y ago
This is a service business. Paid contractors is to prevent cashflow issues when jobs stop.
345.
▲
by
agbell
17y ago
+1 for Church issued credit card with automatic 10% for religious tithing
346.
▲
by
agbell
17y ago
Author of HOW GOD CHANGES YOUR BRAIN
347.
▲
ManyWheels: A Lean Startup Case Study on Vetting Opportunities
(kevindewalt.com)
3 points
by
agbell
17y ago
|
0 comments
348.
▲
by
agbell
17y ago
I would generate a back or lay output signal for 'Correct score' bets for soccer matches before the game started. Input would be everything you can pull out of the the data set especially including the odds on other more standard bets like
349.
▲
by
agbell
17y ago
This is not what you are asking but a market with more inefficiencies and less transaction costs is Betfair. There is a simple api and free historical data. I spent a bunch of time with Neural Nets and SVM and that data. Sadly I didn't fi
350.
▲
by
agbell
17y ago
Video Lectures has a ton of DM and machine learning lectures up for free. http://videolectures.net/Top/Computer_Science/
351.
▲
by
agbell
17y ago
I recommend starting with weka and this great book: http://www.amazon.com/Data-Mining-Practical-Techniques-Manag...
352.
▲
by
agbell
17y ago
I think his idea of every select is a report really simplifies the way an ORM is used. Instead of worrying about lazy vs eager loading of child objects, you create a de-normalized flat reporting object that represents all the data you need
353.
▲
by
agbell
17y ago
Greg Young has some very interesting ideas surrounding explicit state transition models. The end result is a system that can be rewound and fast forwarded and that doesn't use a database in a traditional way. In fact he recommends an in-me
354.
▲
Unshackle Your Domain - Explicit State Transition Models
(infoq.com)
1 points
by
agbell
17y ago
|
2 comments
355.
▲
Probabilities in the Game of Monopoly
(tkcs-collins.com)
3 points
by
agbell
17y ago
|
0 comments
356.
▲
by
agbell
17y ago
I didn't do the poll and am INTP so I assume it is representitive
357.
▲
by
agbell
17y ago
To quote your blog :" Good ideas are a dime-a-dozen. Good execution is rare." Nicely done.
358.
▲
Expert Performance Its Structure and Acquisition
(mit.edu)
4 points
by
agbell
17y ago
|
0 comments
359.
▲
by
agbell
17y ago
This looks to be a good resource for back of the envelope style estimations
360.
▲
Art of approximation in science and engineering
(mit.edu)
42 points
by
agbell
17y ago
|
6 comments
More ›