Fede_V

37 ms·

331.
by Fede_V 12y ago
I really like those reviews micheal, but I wanted to comment on a point you made - about White Bear and White Christmas, there is a lot of very deep and interesting philosophy about what it means to be alive, and what it means to be someone
332.
by Fede_V 12y ago
I thought 15 million points was by far the best. TV shows like Big Brother, X-factor, etc, all pray on the vulnerability that certain people have for wanting to seek out fame and notoriety. Incidentally, Sony had a recent patent for intera
333.
The two cultures of mathematics and biology(liorpachter.wordpress.com)
126 points by Fede_V 12y ago|69 comments
334.
by Fede_V 12y ago
I'm not sure how people make it to the end of that article without being scared shitless about the future of newspapers. Newspapers have an incredibly important social function as muckrackers, exposing corruption, explaining politics,
335.
by Fede_V 12y ago
Yeah - this is a sample of the 'gold standard' of psychology papers, and only 10/13 could be reproduced. The reasons for shoddy reproducibility are p-value hacking, intense pressure to publish at all costs, and a premium on &
336.
by Fede_V 12y ago
There was a very interesting design discussion by JMW on the julia-dev forum about nullable arrays, and column dtypes: https://groups.google.com/forum/#!topic/julia-dev/hS1DAUciv3... Even so - right now Panda
337.
by Fede_V 12y ago
There was a good blog post this by Armin. Basically, if you have to concatenate strings, don't do so via += but use "".join([]).
338.
by Fede_V 12y ago
Wow, I'm really surprised. Guido is usually incredibly nice, but he comes across as very condescending in that comment. I love Python, it's my favorite language by far - but I love all those different tools (numba, pypy, hope, nu
339.
by Fede_V 12y ago
Even with the high pay and the intellectually challenging work, I've never found finance the slightest bit appealing. Somehow, in finance, the impression I always got was that the engineers/algorithm guys were always second fiddle
340.
by Fede_V 12y ago
The closed form solution is not that difficult to derive at all if you put the problem as a reoccurrence equation with boundary conditions.
341.
by Fede_V 12y ago
Hey, not really a question, but.. thanks for making sure that all previous content stays up. I can't claim to have been an avid reader, but I've read several excellent articles on your mag, and knowing that it will always be avai
342.
by Fede_V 12y ago
I completely agree. It's also a consequence of trying to compete with outlets like Buzzfeed that produce a huge amount of low quality stories on a shoestring budget, then optimize the hell out of SEO to get very high traffic.
343.
by Fede_V 12y ago
This is the really dangerous consequence of chasing after stories that have nice narratives. If a story fits your ideological narrative perfectly, you should be twice as skeptical - real life stories are messy, imperfect, and very rarely p
344.
by Fede_V 12y ago
Oh, absolutely. I'm just saying, even if everyone was doing everything within the law, you'd still get suboptimal outcomes because of how the incentives line up for all the players involved.
345.
by Fede_V 12y ago
Not completely in this case. But, take Lockheed, for instance. They are notorious for running supply chains that stretch across as many states/districts as possible so that whenever there are any possible budget cuts, their lobbyists
346.
by Fede_V 12y ago
Unfortunately, this is a huge failing of large federally funded projects. Contractors (and federal agencies too, to a large extent) have long since learned that the best way to get a big contract from the US government is to spread yourself
347.
Spanish publishers move to halt Google News closure(arstechnica.com)
3 points by Fede_V 12y ago|1 comments
348.
Spain wants Google News back(worldbulletin.net)
2 points by Fede_V 12y ago|0 comments
349.
by Fede_V 12y ago
This is an absolutely awesome answer, and in the best tradition of old school hackers. This is why open source rules. If you have some time - would you mind writing a high level overview of architecture/design choices that go into wri
350.
by Fede_V 12y ago
No. This is a package from one of the core scikit-learn developers focused on prediction in cases where the output data isn't just a scalar (regression) or a category (classification), but the output has a structure.
351.
by Fede_V 12y ago
I think the increase in fraudulent papers getting published is a symptom, not the disease. The disease is that there are too many scientists doing research for a given research budget. Due to the ferocious competition for grant money, peop
352.
by Fede_V 12y ago
I'm sure technology can improve things - for example, the NYT is doing a much better job now of marketing its (excellent) cooking section. My impression though, is that what makes great journalism is reporting. Almost all the time, w
353.
by Fede_V 12y ago
My distinct impression is that the limiting factor in quality journalism is gum-shoe reporting, not incredibly powerful CMS. Buzzfeed has the most advanced CMS, but their reporting pales with respect to the NYT. When they do bother to do p
354.
by Fede_V 12y ago
I am not a lawyer, and have absolutely no meaningful opinion on the merits of this lawsuit. However, Uber has absolutely not helped their case by having executives act like frat bros. As much as we'd like to think that the law and pub
355.
by Fede_V 12y ago
I want to add to this post - in my experience, a crappy PhD project with a great advisor is orders of magnitude better than a great PhD project with a shitty advisor. A shitty advisor will make your life horrific - will pile a toxic combina
356.
by Fede_V 12y ago
Doesn't this sort of confirm that 'ideas are cheap, execution is everything'? The abstract idea of 'an online virtual currency' is garbage without the brilliant crypto implementation. I admittedly have a very stron
357.
by Fede_V 12y ago
I'm not really sure what the point of this post is. Julia is a fun language to play with, it has a great and very helpful community and, in its niche, it's very good. Its creators have been very good OSS citizens and joined forc
358.
by Fede_V 12y ago
Because Julia has been carefully designed to allow functions to compile down to very fast machine code. There are a few important design choices that are necessary to make it possible to do this (type stability, etc) - there are a few talk
359.
by Fede_V 12y ago
Leaving aside the pro-Uber aspect of the piece, the criticism of the existing solutions was spot on. If you add a huge amount of friction before you can offer feedback, the only people to do will either be normal people who had truly horrif
360.
by Fede_V 12y ago
There was a great Ben Horowitz blog about exactly why this is incredibly poisonous. If you force people to negotiate for every raise instead of having a fair, scheduled process in place, you create a horrendous atmosphere where people feel

More ›