stdbrouw

77 ms·

781.
by stdbrouw 12y ago
> But one tenant of scientific thought is that truth and factuality are not measures of an individual's or society's belief, but rather, they are discovered by experiments whose results can be replicated. Discovery and replicat
782.
by stdbrouw 12y ago
Surely having a job is not the only way to be or feel like a dignified human being? And how long will you continue to feel dignified in a job that exists for the sole purpose of making you feel dignified but could really just as well be don
783.
by stdbrouw 12y ago
As the comments to that page mention, though, it's not a representation of the central limit theorem at all. It shows how a binomial distribution "becomes" a normal distribution in the long run, but it doesn't show why t
784.
by stdbrouw 12y ago
Any idea when we might see this included in a CoreOS release?
785.
by stdbrouw 12y ago
Not experiments, but observational data is often analyzed using regression, the poor man's predictive modeling :-)
786.
by stdbrouw 12y ago
Imagine that your sample of kids in Kent are actually 5mm smaller on average than your samples from the 9 other counties. The question to then ask is: how likely is it that this would happen purely due to chance? After all, even a fair die
787.
by stdbrouw 12y ago
Power analysis and confidence intervals have their own problems, but they're already a vast improvement over p-values. No need to go all-out Bayesian straight away – though certainly, if we can get the software to a point where that wo
788.
by stdbrouw 12y ago
This defense of P-values is akin to how a certain type of computer programmer tends to defend bad design decisions, like "if you accidentally typed `rm -rf /` and it wiped out your entire system, then you're an idiot, learn t
789.
by stdbrouw 12y ago
The normal distribution approximates a series of binomial trials, and in fact that's how Abraham de Moivre originally stumbled upon it: from his study of games of chance. Think Pascal's triangle or google for "bean machine&qu
790.
by stdbrouw 12y ago
"One" is actually reasonable close, e.g. "one does not simply walk into Mordor" but it certainly lacks the informality (and therefore frequency of use) that "on" has in French or the reflexive "se" in
791.
by stdbrouw 12y ago
Even though the OP talks about learning new technology, his meaning is actually very different from yours. You're talking about understanding things. The OP is talking about getting up to speed with something and using it productively.
792.
by stdbrouw 12y ago
You're still young, so I'm guessing you don't like saying no, perhaps because you need the money. But the longer you're doing this, the easier it will get to say, "well, actually, no, I only do X" and in fact g
793.
by stdbrouw 12y ago
If you don't actually believe the tech will be useful for you outside of that one job, the solution is simple: charge them for the time you spend getting up to speed.
794.
by stdbrouw 12y ago
> CS got one thing wrong that Python got right: There are many ways to do things. I don't think this is necessarily true anymore, if it ever was. For hash-like datastructures in Python, please choose between object/dict/na
795.
by stdbrouw 12y ago
Same reason diamonds are worth more than water.
796.
by stdbrouw 12y ago
Might be true, but it's easy to misjudge this: in your 9-6 you have lots of time to process any interesting ideas that come your way and then move on to the next thing, whereas while you were a rafting guide you probably chewed on the
797.
by stdbrouw 12y ago
I have my own little blogging engine based on https://github.com/debrouwere/render and what I've found is that the bottleneck is almost always IO. Rendering templates is so fast it simply does not matter whether y
798.
by stdbrouw 12y ago
> I don't think our bodies are prepared to that sort of pattern where you stay inactive for 5 hours, while your brain is still very active, and stop to run non stop for 1 hour. Isn't that pretty similar to the lifestyle of a hu
799.
by stdbrouw 12y ago
The predictive element itself has little to do with correlation/causation, though. "If you have gray hair, you're more likely to die in the next ten years than people without gray hair" is a perfectly valid statement, an
800.
by stdbrouw 12y ago
Better PDF: http://www.csse.monash.edu.au/~korb/pubs/fallacy.pdf
801.
by stdbrouw 12y ago
The push aspect of it is interesting, as a way to create a routine, but other than that, won't this suffer from the same problems that Wikipedia's "today's featured article" and "random article" features h
802.
by stdbrouw 12y ago
You're right. I was referring specifically to the fact that in this replication study, they've actually bunched together many different studies into a single questionnaire that tests them all. So it's not a straight-up replic
803.
by stdbrouw 12y ago
I guess it's debatable. You could consider a single experimental design, administered by many different labs across many different countries, to be "many replications", or you could argue that by using the exact same question
804.
by stdbrouw 12y ago
Your equivalent title would be equally misleading. It's not because an experiment does not, in one particular attempt at replication, lead to identical conclusions, that a finding is therefore "impossible to reproduce" or bog
805.
by stdbrouw 12y ago
And another big one: the low statistical power of much psychological research, meaning that even if an effect is present, the original study or its attempted replication might not find it due to inadequate sample size.
806.
by stdbrouw 12y ago
The world would be a beautiful place if everything intentional were sane :-) It's bitten me a couple of times (e.g. when unpacking arguments in a function to feed them to another function, where one of the arguments is nil) but I can s
807.
by stdbrouw 12y ago
... though I wouldn't necessarily call Lua's way of doing things "sane". E.g. if Lua encounters a `nil` somewhere in your table, it will stop iterating.
808.
by stdbrouw 12y ago
Perhaps "debriefing" would work?
809.
by stdbrouw 12y ago
HN is not the place for contentless whining.
810.
by stdbrouw 12y ago
Here's a classifier that uses explicitly programmed instructions: def classify(word): return word[0] What you would do instead with supervised learning or other ML methods, is to give the algorithm some inputs and outp

More ›