HNHacker News
TopNewBestAskShowJobs

rfurmani

884 karma · joined December 3, 2008

Email: rfurman@alumni.stanford.edu
submissionscomments
rfurmani··on Show HN: Weekend project, fun spin on dating. Only friends edit your profile
That's one of the reasons why I've stopped development on this: nowadays the only way to get a critical mass of users on a site like this seems to be by being annoying or shady, doing things such as scraping data to build up profiles, spamming friends, etc. I have been considering though opening up the whole project on github to serve as an example and see perhaps what others can make of it.
rfurmani··on Show HN: Weekend project, fun spin on dating. Only friends edit your profile
You can set permission on who can see your account, and if one does not have permission to see you then you will not appear in the search results. In particular you can set it to be completely private.
rfurmani··on Show HN: Weekend project, fun spin on dating. Only friends edit your profile
That's really cool! I have worked along the lines of making a wiki-style profile for everyone (http://wikisapien.com/) but the whole dating angle gives this site a neat hook that should attract good users.
rfurmani··on Canada’s top science prize goes to Stephen Cook
Correction: every natural NP-complete problem will have easy cases. You can construct problems for which there is no easy case, which is a shame since I thought at one point you could use this trait to seperate P and NP
rfurmani··on Negative temperatures are not colder than absolute zero
It's fine mathematics too, it's just like the chain rule (or interpreting the d notation via infinitesimals/limits)
rfurmani··on 125th Birthday of Ramanujan
125 is a power of 5 whose digits rearrange to a power of 2, namely 512, and I imagine its the only power that works like this. And if you complain about arbitrarily using base 10, that's just 2 times 5
rfurmani··on How do we tell truths that might hurt? (E. Dijkstra, 1975)
I've done a fair bit of development inside Postscript (way back when it was one of my favorite languages) and its problem in this context is that it /is/ a programming language, and one that would have to be effectively sandboxed. HTML on the other hand is purely a markup language, is extensible, can be reflowed, and is much more natural for writing documents by hand.
rfurmani··on How do we tell truths that might hurt? (E. Dijkstra, 1975)
Just like an artist only has the techniques: move pencil horizontally, move pencil vertically, move pencil toward or away from the paper. The art comes from knowing what to draw and how to combine these steps based on a vision in your head. And the real part of pure mathematics is envisioning new structures and new relations that nobody has used before, at which point the proof itself can become a detail. IT's not through lack of trying that formal computer proof systems haven't been able to touch a milli-part of modern mathematics

This is also a point brought up in the great guide: Advice to a Young Mathematician, http://press.princeton.edu/chapters/gowers/gowers_VIII_6.pdf

> We are all taught that “proof” is the central feature of mathematics, and Euclidean geometry with its careful array of axioms and propositions has provided the essential framework for modern thought since the Renaissance. Mathematicians pride themselves on absolute certainty, in comparison with the tentative steps of natural scientists, let alone the woolly thinking of other areas.

> It is true that, since Gödel, absolute certainty has been undermined, and the more mundane assault of computer proofs of interminable length has induced some humility. Despite all this, proof retains its cardinal role in mathematics, and a serious gap in your argument will lead to your paper being rejected.

> However, it is a mistake to identify research in mathematics with the process of producing proofs. In fact, one could say that all the really creative aspects of mathematical research precede the proof stage. To take the metaphor of the “stage” further, you have to start with the idea, develop the plot, write the dialogue, and provide the theatrical instructions. The actual production can be viewed as the “proof”: the implementation of an idea.

> In mathematics, ideas and concepts come first, then come questions and problems. At this stage the search for solutions begins, one looks for a method or strategy. Once you have convinced yourself that the problem has been well-posed, and that you have the right tools for the job, you then begin to think hard about the technicalities of the proof"

rfurmani··on Modern math solves Ramanujan’s ‘vision’ - may clarify black holes
One nice example of deep insight is the Ramanujan Tau sequence, which comes from expanding out the product (1-q) (1-q^2) (1-q^3) all to the 24th power and multiplied by q, as q-24q^2+252q^3-...

Ramanujan noticed that the coefficients a_n have some remarkable arithmetic properties, namely the sequence is multiplicative: a_m a_n = a_{mn} when m and n are relatively prime. There's a more complicated formula when m and n have a common divisor, and he also conjectured that the size of a_p is upwards of 2*p^5.5 when p is prime.

This led to the beautiful study of modular forms and all of the above statements have profound explanations, the last of which was proven 58 years later by Deligne in 1974, as the Riemann Hypothesis for curves.

PS: there are plenty of random discoveries that can be made about the sequence, for example a_n is congruent mod 691 to the sum of the 11th powers of all divisors of n. This also has a good explanation that is now known. Lehmer conjectured in 1947 that a_n is never 0, which has been verified up to n=22798241520242687999, but is still an open question.

rfurmani··on Modern math solves Ramanujan’s ‘vision’ - may clarify black holes
You could say that about any field that one is unfamiliar with. Heck, the most basic art concepts could seem like so to a blind man. Mathematicians have worked for ages on simplifying the concepts so every term is there because it helps a trained mathematician understand what is going on.
rfurmani··on Modern math solves Ramanujan’s ‘vision’ - may clarify black holes
You have two functions both of which grow to infinity, one of which is much better understood than the other. It turns out that if you subtract the two functions, they balance out perfectly so you end up with something converging to 4 instead of going to infinity.
rfurmani··on Would the Last Blackberry User Please Turn Out the Lights?
RIM looks like a great investment. As mentioned in this thread, they have very long term contracts so they won't die anytime soon, and they Are trading at such a ridiculously low premium that if they don't die they're bound to go up by a factor of 5. Heck their market cap is 1% of apple's, so (very roughly speaking) if they hold on to 5% of the market you'll see a massive gain. I used to think that apple had built a fortress around the iPhone with its app ecosystem. After getting a new iPhone I realized I didn't need to reinstall most of my apps, and I stead its mostly: web, messaging, phone, music. I left blackberry way back when because of its horrid web browser, but if they fixed it then it'll probably my be more than good enough for business (many of which have the phones locked down to begin with)
rfurmani··on Google Requires Proof of Google+ Activity to Attend SFPD's Bike Theft Workshop
King James II of England, when he saw the restored and expanded St. Paul's Cathedral, called it "amusing, awful and artificial." The architect, Sir Christopher Wren, was not offended. On the contrary, he was flattered.

The king did not mean that it was funny or entertaining. He was not calling it atrocious or appalling. He did not consider it fake or pretentious.

In 17th century England, amusing meant "riveting." Awful meant "full of awe" or "awe inspiring." And artificial meant "artistic."

rfurmani··on An ABC proof too tough even for mathematicians
Right now we are waiting for his corrected paper, as the current one contradicts some known number theoretic results. Mochizuki put uP an amazing amount of scaffolding and invented a new area with new abstractions. People who I trust claim that there isn't new profound number theory in this, so then it may become a matter of whether to study an learn it for its own sake, which carries less incentive if there is no immediate application.

Plus: mathematicians really don't have spare time to learn every area that comes along. Allegedly Poincare was the last true polymath. Given how math has been developing for centuries and you never really need to throw things away it continually becomes harder to get up to speed in any particular field. There's a famous quote about how the Princeton undergraduate math program is so rigorous that it brings you up to speed, to the point of early twentieth century mathematics. In this way math is incomparable to CS.

rfurmani··on Hidden Gem of iOS6: File Uploads in Mobile Safari
Does anyone know if filepicker.io has plans to support this? Currently file uploading in mobile safari is rather underwhelming
rfurmani··on Small but significant usability sins that websites should never commit
I think that works if you follow a link from another app, but not if it is a popup from safari? Or maybe they changed this in ios 5
rfurmani··on Small but significant usability sins that websites should never commit
Very true, this is especially a bugbear of mine when browsing on iPhone. All too often I have eight tabs open that I'm meaning to get back to. Then I click on a link (or accidentally click an ad) and my heart lurches as the link opens a new tab, so Safari very helpfully destroys my oldest tab and I lose whatever was there. I've noticed quite consistently that this behavior comes with sites that just want to keep you on there, so they open external links in a new tab, usability be damned.
rfurmani··on Why I went from Python to Go (and not node.js)
Very true. One of my first forays into Haskell came during the Google Code Jam, after seeing how short and elegant Reid Barton's solutions where. I solved the problems, but the programs were horribly slow, so I checked how Reid got around it: oh, he doesn't just map things and use the standard style, he needs some special constructs from the libraries.. it looks mathematically equivalent, but it looks plausible that the old way ends up holding onto too much data in memory. I eventually stopped when I realised that it would always take more time than hammering it out in C++, even if the final product in Haskell is shorter and more elegant. In fact, outside of writing APIs and frameworks, the desire for elegance seems debilitating. This also reminds me of how when I was most seriously practicing for Topcoder I could consistently produce shorter solutions than tomek (the top ranked guy far and away at that time) he would finish more quickly, since once he types it in it's done. For the record I'm now using Python (and some R) but I think I'd like Go.
rfurmani··on Proof Claimed for Deep Connection between Prime Numbers
That link is old and was posted a few years before the final paper. The more recent discussion is here: http://quomodocumque.wordpress.com/2012/09/03/mochizuki-on-a...
rfurmani··on Reddit's database has only two tables
Umm, assuming the github repo is what they actually use (i assume so given how often it is committed to) there are two tables per object. Reddit_thing_link, reddit_data_link, reddit_rel_thing_savehide, etc etc
rfurmani··on Lessons Learned While Building Reddit to 270 Million Page Views a Month
They in an ad hoc way turned postgres into a key-value store, but in reality they have Cassandra running as a "permacache" in front of it and, in practice, everything really just hits Cassandra. It looks like postgres could at some point be phased out.

Source: ive hacked at the codebase to produce arxaliv

rfurmani··on Anic - Faster than C, Safer than Java, Simpler than *sh
Technically, for Perl you can use $ for regex backreferences. I don't remember which one is preferred, but I remember it frowning upon one of them.
rfurmani··on Drinking at work: the boredom of boozeless business
Just a nitpick, the hot coffee case wasn't as frivolous as it sounds. McDonalds had been serving its coffee unreasonably hot (190F instead of 140F, which causes third degree burns almost instantly) even though they had previously been warned of the dangers. At one point you have to put a cost to it, so it doesn't just come down to them saving money on refills.
rfurmani··on Facebook has more than 83 million 'fake' users
It's not really about "like" or "not like" but fundamental economics and mathematics: things that grow behave differently then things that don't, and are worth more. Suppose FB found a way to consistently make 1B a year, give it to the investors, and only grow as much as inflation. Then the valuation would settle on something like 10-15 billion (plus however much cash it has on hand at that time). And a solid part of the market will love it as a long-term investment and tout its value. On the other hand, to value it at 50 billion you have to assume that its earnings will at least quintuple over time, so good luck to them.
rfurmani··on Living outside America
There seems to be a strong tendency to assume that erveyone is like the speaker. It's especially funny when friends of mine mention something like that (everyone is non-religious everyone is super liberal) not yet realizing that I don't fit the bill :-) Stanford actually has very strong religious student groups.
rfurmani··on Some quick Django optimisation lessons
Thanks, I'll check that out. How does it compare to johnny-cache, which is what I've been using? It has the advantage for me that I only need to include it in settings.py, and I don't need to manually plug in a manager into my models, so it's easy to enable and disable for testing. Also, it has support for different caching backends, so I can have multiple django instances using a common memcache for the models even if the models are highly mutable.
rfurmani··on Some quick Django optimisation lessons
One option is to use the debug_toolbar package. You can set it up to display only when testing or only when an admin is logged in, and it will give you a convenient bar on the right side that shows exactly how long everything took, along with all SQL queries. This allowed me to greatly cut down the number of queries by judiciously using fetch_related.
rfurmani··on Are Stanford Students Just (Really Excellent) Sheep?
I agree that the typical person can be a sheep, however:

* the church callout is unnecessary, it can be a type of self-analysis

* Alcohol is not toxic and in reasonable quantities does not have a purely deleterious effect.

* When needing to make personal connections (which one always will since you learn a lot from others, not to mention the benefits of having connections and even just social interactions) it is good to look like you fit in, so others do not become defensive.

* there's a limited amount of time and energy available for coming up with your own things. If somebody gives me clothes that are comfortable and look good, I won't spend time thinking of what clothes would "really be me as an individual". On my last personal project I had my own design, but we ended up looking through a Web Design idea book, made something that looked rather standard, and ended up with a better reaction, as it's what was expected and hence more usable. In other words, pick your battles, save up for the big innovations, recognize the value of others, and stand on their shoulders, even if you have to conform a bit to do so

rfurmani··on Server Density removes Memcached because it's "too slow"
I had a similar issue with my Django project on Heroku. My site performance went down once I enabled memcache, almost entirely due to latency (going up to 100ms at times). This is one of the reasons why I ended up moving off of Heroku and onto an EC2 instance: now I have full control of my stack, I can run what I want locally on the server, and I'm getting much better performance (nginx+gunicorn+memcached+django+johnnycache+rds)
rfurmani··on Why Use Postgres
With a reserved instance, RDS starts with $20/month for the small, $92/month for the large, and it's easy to change the size as the need arises. Heroku Postgres starts at $200/month, which is more than I'd want to pay for an early-stage project that I'm still experimenting with.
← PreviousPage 3 of 5Next →