msellout

34 ms·

331.
by msellout 13y ago
Yeah, the comment on graph databases seemed a bit too flippant.
332.
by msellout 13y ago
Execution time, after a point, is less important than development time. NoSQL is often faster for development and refactoring, because the schema is easier to change in code than in the database. Also there's the scale issue. Append-on
333.
by msellout 13y ago
You can imagine cases where map-reduce is useful without any starting data. If you are analyzing combinations or permutations, you can create a massive amount of data in an intermediate step, even if the initial and final data sets are smal
334.
by msellout 13y ago
The stock market has performed better than real estate over the same 4 generations.
335.
by msellout 13y ago
But we're not comparing with the rest of the world.
336.
by msellout 13y ago
This could replace some very expensive programs, such as Social Security and Medicare/Medicaid.
337.
by msellout 13y ago
Since money is of diminishing returns, you'll notice your share to the extent that it's useful to you. That's better than coming up with some arbitrary threshold. Whether or not it would cause inflation is unclear, because wa
338.
by msellout 13y ago
One purpose of a lump sum distribution such as the monthly "basic income" is to counteract the regressive nature of consumption tax. For example, suppose we want to provide free mass transit. We then get terrible rush hour traffic
339.
by msellout 13y ago
One option is to plot the change itself.
340.
by msellout 13y ago
It's hard for me to see the food, since it's on newspaper.
341.
by msellout 13y ago
International law.
342.
by msellout 13y ago
Whether or not an activity is science does not affect the need to take action.
343.
by msellout 13y ago
That may be the default of a t-test. Have you considered a Bayesian analysis?
344.
by msellout 13y ago
Crime per km^2 is just as valid as crime per person.
345.
by msellout 13y ago
Feel like sharing a link?
346.
by msellout 13y ago
Yes, there are plenty of American comedy skits about unintelligible British.
347.
by msellout 13y ago
Not much yet. It's not well enough understood yet.
348.
by msellout 13y ago
The training subset for each tree can still be quite large. Note that most of the implementations failed on their 12 GB dataset. Although I'm a big believer in streaming/online machine learning, it's not necessarily the best
349.
by msellout 13y ago
1. There are far more non-programmers than programmers. Even with a correlation between being a programmer and hiring programmers, it is plausible that the vast majority of hiring managers are non-programmers. This becomes more plausible if
350.
by msellout 13y ago
Please take this as a friendly tip, rather than a mean-spirited needling: I think you mean "discreet" (quiet) rather than "discrete" (distinct). I know Taleb hates nit-picking editors, but I like my prose to mean precise
351.
by msellout 13y ago
> Python is just too slow to be used for anything but prototyping. Hardware time is often cheaper than engineer time, so Python may be faster/cheaper if you consider total time to value.
352.
by msellout 13y ago
I'm not sure that "trigger happy and nonsensical" is restricted to "INTJ" communities. As far as I've read, the Myers-Briggs personality test is as predictive of behavior as is a fortune cookie.
353.
by msellout 13y ago
genetic algorithms are just randomized hill climbing with the added assumption that variables "physically" next to each other in your data structure are correlated.
354.
by msellout 13y ago
http://en.wikipedia.org/wiki/Determining_the_number_of_clust... It's "tough luck" because the algorithms determining the number of clusters aren't magical. As with all machine learning, the results reflect the assumptions built into the a
355.
by msellout 13y ago
I think you mean k-means is "the first port of call when clustering." K-nearest neighbors is a supervised learning algorithm (for regression or classification), not a clustering algorithm.
356.
by msellout 13y ago
Price is generally a signal of quality. It's hard to infer your attitude from two sentences, but I'll go ahead and advise that you change from making suggestions to making recommendations. The difference in connotation is one of leadership.
357.
by msellout 13y ago
How would you sell the exploit? Your buyer would want to verify it works before buying, but then you've already given it away. It's not like you could enforce a contract on this.
358.
by msellout 13y ago
Perhaps they didn't think it could be weaponized, so to speak. Accessing one account at a time isn't as big a deal as systematically accessing many accounts. Even if your exploit could do so, they may have safeguards against such robotic ac
359.
by msellout 13y ago
It seems a bit lazy of Friedman to interview his daughter's roommate. Why not include other sources? At least he mentioned the relationship. I don't begrudge Sharef for using the connection. It's another reminder of how important networking
360.
by msellout 13y ago
It seems natural that methods of exchanging money are constrained by methods of property preservation. Neither money nor property exist without government. Sure, you can say that bottle caps and a fence might do the trick, but those don't h

More ›