elq

23 ms·

181.
by elq 18y ago
this works on pretty much all unixes and recently saved my team's ass after a 3am finger fumble that unlinked a very important 500g file.
182.
by elq 18y ago
the general theme was amazon is cheap - cheap computers, cheap office space (poor ventilation, uncomfy chairs etc), long hours, a large amount of pager duty (1 week/month not uncommon), relatively poor pay, etc
183.
by elq 18y ago
one of the reasons you hear about more amazon folks in starups vs. MS is that MS generally treats developers very well so they tend to stay around. Amazon is not such a great company to work for, at least according to the people I know that
184.
by elq 18y ago
never.
185.
by elq 18y ago
Writing software is essentially solving problems. Some of these puzzle/problems are stupid, but if you do not know how to solve problems well and frequently you probably will not be a good (let alone excellent) software developer.
186.
by elq 18y ago
| "What do you need from me to make that project/transaction successful?" Go away and leave me alone so I can solve the problem?
187.
by elq 18y ago
re: crashing on change of color themes. he's probably using the new-ish cocoa-fied emacs in the 24 branch. It's still a bit buggy. Use the carbon build from 23 or just look for a diff if you really want to use 24.
188.
by elq 18y ago
http://millicomputing.blogspot.com/2008/09/netflix-api-and-g...
189.
by elq 18y ago
The tutorial built into emacs is actually quite good. C-h t
190.
by elq 18y ago
hrm. I'd personally prefer a phone that actually worked.
191.
by elq 18y ago
I just took a cab to the airport. The driver was scheduling other trips and giving estimates of cost by using the gmaps iphone app. It wasn't a quick trip.
192.
"I can haz compiler?" - lolcode implemented on parrot(yapc.tv)
1 points by elq 18y ago|0 comments
193.
by elq 18y ago
I cannot dispute that metadata can be useful. But it appears, at least for prediction tasks similar to the prize, that an ounce of weak or strong explicit user input is worth a ton of rich implicit data (including item metadata).
194.
by elq 18y ago
Yehuda later wrote http://glinden.blogspot.com/2008/03/using-imdb-data-for-netf... and http://hunch.net/?p=331 that using movie metadata has produced no measurable improvement in RMSE.
195.
by elq 18y ago
levenshtein distance for predictions? haha. I'd be really curious to know which teams in particular use movie metadata? Yehuda Koren (the first commenter in the blog post) has explicitly stated many times that in his humble opinion movie ti
196.
by elq 18y ago
Correct. When using a latent factor method, providing an explanation for for the prediction has been essentially intractable. A common post-processing step for many teams (particularly BellKor/KorBell) is a KNN. The KNN's are used to provid
197.
by elq 18y ago
https://dl.getdropbox.com/u/52575/desktop.jpg beware, it's big - one 30" and one 24" monitor. Emacs, preview for a couple of documents, a few shells so I can monitor progress of a training run, and some music.
198.
by elq 18y ago
I use a tiling WM and a 24" monitor and I love the "minimalism" of it. It took quite a bit of getting used to, but not constantly seeing IM notifications and new email alerts has measurably improved my productivity.
199.
by elq 18y ago
abbrev's aren't quite as "powerful" as textmate's snippets. luckily, there are many elisp packages that provide this functionality, yasnippet is what I'm using.
200.
by elq 18y ago
keyboard macros, tramp, gdba mode, ESS (emacs speaks statistics), and far more.
201.
by elq 19y ago
hrm. I do all of my real work on linux and am writing this comment on my linux workstation - so I'm far from a linux basher. But, getting fonts that don't suck ass takes a lot of dingy work on linux (fedora, ubuntu, arch, you name it - in m
202.
by elq 19y ago
bingo... I've talked to Yehuda and he does know what he's talking about. Given that he works for Bell Labs, it might be within the realm of possibility that he actually tried using movie metadata. If those students have actually come close
203.
by elq 19y ago
> So could someone here please make it? Please... Agreed. Please make this! A friend and I were talking about such an idea right around the time Friendster was making rounds... and after running through lots of issues and the one sticki
204.
by elq 19y ago
why not just download the dataset and search it locally?
205.
by elq 19y ago
The half dozen or so books about Richard Feynman life and approach to problems. (somewhat cliched but) Zen and the Art of Motorcycle Maintenance. GEB.
206.
by elq 19y ago
hrm. I've never tried that. but you know, google's mission is to "organize all of the world's information". I assume this includes HR info as well. With all of that data and all of the high-end machine learning talent that works there, you'
207.
by elq 19y ago
oddly enough, a good chunk of OS/400 was written in C++ - http://portal.acm.org/citation.cfm?id=226239.226253&coll...
208.
by elq 19y ago
And there are people like me, who get pinged via some channel (linkedin, email, phone, etc) by google recruiters ad nauseam. And every time, tell them no thank you. In the beginning you'll be polite, after three years you'll ask them polite
209.
by elq 19y ago
I spent about a year and a half doing consulting with an big architecture firm several years ago. The primary project I worked on was essentially the opposite of this guy's idea - to take the existing plans for the highly rated hotels the f
210.
by elq 19y ago
Ah. Well, keep in mind that we're splitting the documents up in an arbitrary way WRT the data, so every cluster/partition has a complete copy of the index for the documents that happen to be bucketed there. And no cluster owns a single "doc

More ›