vegabook

84 ms·

751.
by vegabook 11y ago
Yet interestingly you can buy a Ferrari which goes more or less 2x as fast as, say, a VW Golf GTI and costs 10x the price, and there is burgeoning demand for the former. So if those economics apply in the air as well (speed = luxury /
752.
by vegabook 11y ago
A far better measure is the London-NY time record comparison. 777-200 did it in 5.27 hours on Jan 15 ( http://www.telegraph.co.uk/news/worldnews/northamerica/usa/1... ). Concorde record was 2.88 hours (2h
753.
by vegabook 11y ago
I will try to compile this today - appreciate that thank you. I understand that Ubuntu is making Python 3 the default from 16.04 onwards so I need to get moving. Looks like the long dusk for 2.x is finally turning to night.
754.
by vegabook 11y ago
I went to 3.3 a while ago but reverted back to 2.7 because out of the 10 or so crucial libraries I needed, one was still on 2.7. Bloomberg. ( http://www.bloomberglabs.com/api/libraries/ ). That's what the peo
755.
by vegabook 11y ago
Agreed but does this apply in Numpy as well? Will Python matrix multiplications just work on lists of lists in which case they're much slower? Sorry just asking from a 2.7 holdout here as this might cause me to move.
756.
by vegabook 11y ago
Yet meta is necessary, no? That's how the community improves in a feedback loop to increase its value? StackOverflow even created an entire Meta site for that reason, though they wisely segregated it from the main location. Such a thin
757.
by vegabook 11y ago
but (s)he's been a member for more than 2 years and has a very high karma for the time. So I'm going to take this observation on HN's tendencies seriously. Why should observations on the HN community's biases inevitably
758.
by vegabook 11y ago
> "people not dying" How inconvenient....
759.
by vegabook 11y ago
So it impairs your acuity to negative signals from others while high . This is not interesting - it's kinda obvious. It would have been much more worrying if it continued long term.
760.
by vegabook 11y ago
Ron Dennis has always wanted to diversify, to use the Mclarent know-how and brand elsewhere, and that was good when they were winning races. The embarrassing spectacle of both Mclaren cars at the back of the grid, being lapped, has to make
761.
by vegabook 11y ago
Absolutely. This is useful but it is highly unlikely that any of the languages in this chart will disappear anytime soon. As I plan my learning-time investment budget for the next few years, it's only marginally interesting to me tha
762.
by vegabook 11y ago
I'm guessing part of the appeal will also be super-wide bandwidth between the cores. Interconnect throughput and/or shared memory could make this significantly more interesting than a modular rack-based system for some workloads.
763.
by vegabook 11y ago
Obviously I agree. My simple point is that the OP is making claims about a binary outcome, which could lead one to infer that stock trading binary up/down outcomes are all that matter.
764.
by vegabook 11y ago
Trade win:lose ratio > 1 is an elementary mistake in algorithmic backtesting. Not knocking Bayes. Only saying the analysis is incomplete. Cancer/noCancer is not a good analogy.
765.
by vegabook 11y ago
That is useless in the real world. Trade loses is inevitably a larger absolute value P&L than trade wins. Trade wins frequency > trade loses frequency. Trade loses size > trade wins size. You need both and the analysis only does t
766.
by vegabook 11y ago
The analogy incorrectly compares a binary outcome (cancer / no cancer) with a continuous variable (stock price returns, which may have skewed magnitudes depending on upside or downside). So while it is nominally correct about the formu
767.
by vegabook 11y ago
I agree with this. However if you accept that the circumstances of a society's existence are highly variable too, that each of the many different genetic outcomes you speak of is relevant only under very specific circumstances, then
768.
by vegabook 11y ago
Another Python concurrent library?? To boot one which makes Python look like Java? This.that.that.this.whatever()? How is this better than a few simple lines of multiprocessing module exactly? Yes I'm being negative. People need to
769.
by vegabook 11y ago
I'm not talented enough to understand what that means... please elaborate.
770.
by vegabook 11y ago
There is a good reason why millions of years of evolution has created a minority of "talented" people. Societies need super-talented people and leaders (these are often not the same people, btw), but it also needs legions of pedes
771.
by vegabook 11y ago
Nobody is arguing that tape doesn't provide the best capacity per dollar. However I get the sense that tape hasn't kept up with spinning disks, let alone SDDs (on Moore's law). It's just not the 50x factor it once was. M
772.
by vegabook 11y ago
> when raw speed is required and the edit-compile-run cycle of binary extension modules is too heavy This is a compelling pitch in my opinion, but can anybody explain to me where Lua would be better than, say, Numpy, or Numba for that
773.
by vegabook 11y ago
Funnily enough I was just this morning googling for 10Gbit Ethernet cards and a whole bunch of "Copper wire 10Gbit will hit primetime in 2015" results game up. Totally agree. Throughput is where it's at.
774.
by vegabook 11y ago
Tape?? You can get 6TB spinning HD for 300 bucks. Call it 12 grand for 1:1 redundancy on spinning disks @128TB. Spinning disk sequential throughput is more than big enough to do this overnight every day, and probably enough to backup almost
775.
by vegabook 11y ago
Moore's law is passing the baton from GHz to the storage stack. Whereas you once had a simple RAM + HD setup, you now have a teamworking hierarchy of storage technologies: Cache / 3d stacked mem / DRAM / X-point / S
776.
You (probably) don't want an Android console(tomsguide.com)
2 points by vegabook 11y ago|0 comments
777.
by vegabook 11y ago
Absolutely. But with the current peg, where USD goes so the CNY goes. Given that only 20% of Chinese exports are to America, China is becoming uncompetitive on its trade-weighted basket and must adjust in the same direction as other currenc
778.
by vegabook 11y ago
It's pretty useful if you're ssh-ing into a machine to run R and you just want a quick and basic idea of some data.
779.
by vegabook 11y ago
R package txtplot.
780.
by vegabook 11y ago
CNYEUR (euro per cny) exchange rate past 60 months: ++-------+------+-------+------+-------+------+--+ 0.15 + ** + | * | |

More ›