brendano

15 ms·

61.
by brendano 15y ago
Does wpshell have a homepage and an overview what it does, even just a README file? It's hard to blame someone for not finding prior art when this doesn't exist :)
62.
by brendano 15y ago
For profiling web requests for a WSGI app, I like to use the Scotch WSGI recorder [1], so you can set it up to record one request, do the request with your browser, then replay it over and over for profiling. (This would be equivalent to y
63.
by brendano 15y ago
I put together some free online resources for linear algebra: http://brenocon.com/blog/2011/03/poor-mans-linear-algebra-te... Everyone I talked to then recommended the Strang textbook, and also Stanford EE263, both already pointed out by
64.
by brendano 15y ago
I think they made this list before auto-suggest existed.
65.
by brendano 15y ago
Alternative: first pipe through a CSV to strict TSV converter, such that "awk -F'\t'" is correct. I do this all the time, because awk is far superior to standalone perl/python scripts for quick queries, tests, and reporting. (My converter
66.
by brendano 15y ago
I took CS 229 years ago as an undergrad and still think about things from it all the time.
67.
by brendano 15y ago
Pipe Viewer dramatically increased my productivity for large scale data processing. In particular, it lets you quickly know whether something will take 5 minutes, or 2 hours, so you can plan accordingly. It's painful watching people try t
68.
by brendano 15y ago
> Consider how ridiculous that would sound if you asked "What if one of your friends had emailed you their phone number and wanted to remove your access to it (or had emailed you and wanted to remove your knowledge of their email addres
69.
by brendano 15y ago
There is a legitimate question here -- do you own the information about your friends, or do your friends? Many things on Facebook are a shared quasi-public, quasi-private space where the ownership isn't clear, and don't fit into our usual
70.
by brendano 15y ago
This is why Asians are underrepresented in tech startups --- wrong set of cultural values and inclinations. (I am Asian)
71.
by brendano 15y ago
I think tax collection records are always one of the earliest examples of writing in a ancient language
72.
by brendano 15y ago
Everyone on the panel is quite senior -- the comments have the flavor of "darn young researchers these days." It might be interesting to hear from AI researchers under 40 about this. Second, on the question of statistics and language; ther
73.
by brendano 15y ago
I think the question is these don't look like NormalCDF(coef/se) p-values given the coef and se you report. They tend to be too small. right that's my question
74.
by brendano 15y ago
Another question, how are standard errors calculated? I assume they're not from the bootstrapping since the p-values clearly aren't from the standard errors ( +/- 1.96*se is crossing coef=0 for several cases but with small p-values). The
75.
by brendano 15y ago
They're not exactly in rank-space .. they discretize to the binary variable whether or not an article made it into the top-20, then use logistic regression to model that. so the coefficients are in log-odds space of that indicator
76.
by brendano 15y ago
Great -- I'm hoping L1-regularized logistic regression will become the standard first thing to try in these quick-n-dirty "predict response variable from text" experiments. That's our approach too. (I assume this is L1 or similar since yo
77.
by brendano 15y ago
The question of historical precedent is interesting. In her post she says she's been using "zephoria" personally since 1998; Wayback Machine shows a zephoria.org landing page in 2002 and a full website in 2003. And she had a public profil
78.
by brendano 15y ago
It's obvious there's no such thing as "true quality". But I am willing to value certain types of quality more than others. I'm just expressing one vote that I agree with some peoples' consensus, that a particular type of quality of HN has
79.
by brendano 16y ago
How about, display a discretized form of points? -- e.g. high, medium, and low, with unknown thresholds (that may be changed through time or whatever). Can still use them as a heuristic for filtering, but the details and possibly divisive a
80.
by brendano 16y ago
It's very reasonable to recognize quality. I can distinctly remember several times the comments on an article were written by absolute experts in the material, full of insight and interesting content. I could believe comments like those hav
81.
by brendano 16y ago
It's fascinating! It would be great to have this information from just normal email -- the data's been there for many years -- but does it seem like privacy invasion then? There doesn't seem to be any difference...
82.
by brendano 16y ago
The use case is a bit different, but I wrote a little converter to TSV, adding in a header. I mostly use it for input into R, but I use it a lot. It only works for fairly flat JSON objects. https://github.com/brendano/tsvutils/blob/mast
83.
by brendano 16y ago
You might be interested in: http://ditz.rubyforge.org/
84.
by brendano 16y ago
I think it's WWW-2011 (it says so in the PDF).
85.
by brendano 16y ago
This exactly summarizes why I stopped using Scala. All I wanted was a cleaned up Java, but it's much more and riskier than that.
86.
by brendano 16y ago
http://en.wikipedia.org/wiki/Deep_Springs_College
87.
by brendano 16y ago
Missing a really great utility for this: https://github.com/inky/see
88.
by brendano 16y ago
Medical electronic record systems are almost always poorly designed like this one. That's why doctors perceive such deficiencies as "natural."
89.
by brendano 16y ago
There won't be a practical successor to Prolog. Clojure is possible because Lisp already is a general purpose programming language. Prolog isn't. You couldn't do many of the projects in Practical Common Lisp with it. For example, imperative
90.
by brendano 16y ago
MSR developed a (Bayesian) enhancement to Elo that apparently is used by XBox Live: http://research.microsoft.com/en-us/projects/trueskill/

More ›