<p>since python and perl are so similar, there&#8217;s a lot of bad blood between the fans of the two languages (programmers are weird. the more miniscule a detail is, the more rabidly they&#8217;ll argue over it). surprisingly though, everyone was quite civil.</p>

<p>Guido gave a quick introduction to python and then just answered questions for a couple hours. he spent a lot of time explaining the reasoning behind python&#8217;s approach to programming. he was very diplomatic about it, acknowledging a few areas where perl was more developed than python and very delicately pointing out areas where python had advantages. </p>

<p>since python and perl have so many things in common, it&#8217;s hard to pin down many areas where one or the other has a clear advantage. pretty much anything that you can do quickly and easily in perl, you can do quickly and easily in python, and vice versa so it comes down to which one you are more familiar with. one thing that i learned python is good for is as a wrapper for C/C++ libraries. perl has facilities for this but they are notoriously ugly and difficult. python apparently does it fairly painlessly. </p>

<p>there was also a lot of talk about python&#8217;s internals: its object model, capabilities for introspection/reflection, its security model, and its nice exception architecture (one of the things that i really missed in perl when i moved from java). since <a href="http://www.yetanother.org/dan/">dan</a> was there again, there was also a lot of talk about parrot, the bytecode language and vm that&#8217;s intended to bridge perl and python (and maybe java too).</p>

<p>all in all, i doubt i&#8217;ll give up perl anytime soon but Guido has piqued my curiousity enough that i&#8217;d like to at least become mildly competent with python. if nothing else, it looks like an excellent teaching language.</p> 

<p>at the last lecture, the professor was going over modular exponentiation, which is an algorithm for doing things like 1,456,765<small><sup>3,567,234</sup></small> mod 67 really quickly without having to actually calculate the intermediate step of 1,456,765<small><sup>3,567,234</sup></small>, which would be way too large for any computer to handle. i&#8217;ve been over it a few times before, once for my algorithms class at bates and again for discrete math at columbia but i felt like i haven&#8217;t actually <em>done</em> it in long enough that i&#8217;m rusty. since it&#8217;s so important to cryptography, i figured i should brush up on it. so i do a search on the web and find&#8230; <a href="http://a-nders.dhs.org/code/expmod.html">my own code</a> for doing it in javascript that i wrote as part of a homework assignment for my algorithms class. i actually played with it for a couple minutes and clicked on the broken &#8220;view-source&#8221; link before i noticed that it was my own work.</p>

<p>i guess it&#8217;s true that long-term memory is the second thing to go&#8230;</p> 

<p>1) why do people still use Outlook?</p>

<p>2) why do people on windows/mac not at least use anti-virus software that is updated daily?</p>

<p>3) why doesn&#8217;t microsoft fix Outlook already?</p>

<p>after watching this happen over and over again (and getting countless Word documents that should have been plain-text emails), my boss and i are both strong advocates of banning Outlook in the office. we don&#8217;t think we can do it though because too many people are attached to the stupid thing.</p> 

<p>it came out absolutely delicious. </p>

<p>unfortunately, i realized the hard way that i probably should have looked in the oven before i turned the broiler on. since i&#8217;d never even so much as opened the oven door since moving into my apartment, i had failed to notice that the previous occupants had kindly left me a backing pan complete with about half an inch of burned on&#8230; stuff&#8230; that made for a nice apartment full of smoke soon after i started broiling my mushrooms.</p>

<p>in other news, i apparently don&#8217;t have any smoke detectors in my apartment&#8230;</p> 

<p>Damian has a reputation as the &#8220;mad scientist&#8221; of the perl community, attempting (often successfully) things that no one else would ever even consider like <a href="http://cpan.valueclick.com/authors/id/DCONWAY/Lingua-Romana-Perligata-0.01.readme">programming perl in latin</a> or incorporating non-deterministic <a href="http://cpan.valueclick.com/authors/id/DCONWAY/Quantum-Superpositions-1.02.readme">Quantum Superpositioning</a> into the language. but it&#8217;s really difficult to understand just how random and disturbed his mind is until you&#8217;ve seen him speak. the lecture he gave last night was kind of like a fever dream version of G&ouml;del Escher Bach.</p>

<p>he started the talk by mentioning that he is not related to that other Conway, the cambridge mathematician who invented the <a href="http://www.bitstorm.org/gameoflife/">Game of Life</a>. then he started talking about the <span class="caps">GOF</span> for a while because he was fascinated with the crazy things people did with it. eg, even though Conway (the mathematician) mathematically proved that the <span class="caps">GOF</span> was <a href="http://www.wikipedia.com/wiki/Turing-complete">turing complete</a>, you&#8217;d think that no one would be crazy enough to actually <em>implement</em> a turing machine using it. of course <a href="http://www.rendell.uk.co/gol/tm.htm">you&#8217;d be wrong</a>. and of course Damian had to make his own contribution. so he wrote a perl module called <span class="caps">DFA</span>::Cellular to make it easy to play with any kind of cellular automata in perl. CA&#8217;s work by iteratively applying a set of rules to each cell in a grid and generating the output based on that. a problem often arises when two or more rules apply to a cell; the usual solution is to just take the first one that applied and call that good. Damian decided to instead use his Quantum Superposition code on it to allow the cells to exists in multiple states at once. an interesting idea but not earth-shattering.</p>

<p>then he started talking about languages and brought up the fact that the <a href="http://www.kli.org/">klingon language</a> had been designed to use a particularly uncommon word order, which coincidently is very similar to perl&#8217;s syntax. the obvious course of action for Damian after discovering this was to write a module similar to his Latin one that would let you write perl in klingon. Lingua::tlhInganHol::yIghun was born (well, it&#8217;s being checked over by the klingon language institute for correctness, but it will be released soon). </p>

<p>then he changes the topic entirely yet again and starts talking about demonology)including some humorously photoshopped pictures of various perl luminaries) and James Clerk Maxwell&#8217;s <a href="http://cougar.slvhs.slv.k12.ca.us/~pboomer/physicslectures/maxwell.html">demon</a>. since he&#8217;d been playing with cellular automata, he decided that it would be fun to simulate maxwell&#8217;s demon in perl which he was able to do pretty quickly with the code he&#8217;d written incorporating the quantum superpositioning. hard to explain without demonstrating but he came up with a pretty good demonstration of why maxwell&#8217;s demon isn&#8217;t as contradictory as maxwell thought it was (this was also proved a while back but in a somewhat un-intuitive manner). Damian&#8217;s demonstration basically showed that with gas densities anything like normal, even a tiny difference in density between the two chambers makes it extremely difficult to let a particle from the low-density side through without letting one or more of the particles from the high-density side through at the same time.</p>

<p>then, to tie everything together, he took the cellular automata and quantum superposition based simulation of maxwell&#8217;s demon and rewrote it in klingon.</p>

<p>i should point out that Damian is really a very good lecturer; he&#8217;s funny, articulate, and manages to work it so everything makes a sick kind of sense.</p>

<p>there were a few other people there who are active in the design and implementation of perl 6 along with Damian and they spent an hour or so after the talk answering questions about and explaining the next version of the language which promises to be either salvation for programmers or a catastrophe of biblical proportions. lots of fun things like a robust and fast threading model, byte-code compatability with java, python, ruby and other scripting languages, a cleaner object model, curried functions, a <span class="caps">DWIM</span> (do what i mean) comparison operator, and countless other changes to the syntax are all in the works.</p> 

<p>&#8220;&#8230;so, i want to do something to badly disgrace you. and i&#8217;m going to do it.&#8221; </p> 

<p>we live a mere 6 blocks apart but haven&#8217;t seen each other in over a year (she claims she was &#8220;busy&#8221; with work and stuff. excuses, excuses&#8230;). weird seeing someone you used to hang out with on a regular basis after a year apart. too much happens over a year to remember and regurgitate into conversation. i feel like i must have missed a lot of important stuff.</p>

<p>anyway, the big victory for the evening is that i got my acoustic guitar that i had lent her back. as much as i love my electric, there&#8217;s something about sitting back on the couch strumming an acoustic that comforts and relaxes unlike anything else. most guys (make that <em>all</em> guys) start playing guitar because they think it will get them chicks. eventually it becomes a close friend and confidant. you can tell your guitar things that you would never tell a living person; you can tell it the things that you don&#8217;t even understand well enough to express in words yet. a little part of me dies everytime i see a rock star smash a guitar at the end of a concert.</p> 

<p>he was talking about the functioning of our society and how he was afraid to admit that it bore a strong resemblence to a biological organism at least in terms of the potential threat of a small malignant agent in its midst (viruses, cancer, terrorism, etc). he said that he was afraid of the analogy because the sociological equivalent of an immune system to deal with these threats is not a very pleasant thought since it would tend to encourage racism, intolerance and fascism.</p>

<p>then, just when he was starting to make some sense, he proposed an alternate solution. i have to give him some credit for actually proposing <em>some</em> kind of solution; that&#8217;s more than most of us can do. his solution was a rethinking of privacy in our society. he argues that there are two kinds of privacies: the right to be left alone, and the right to not be seen. both are relatively new concepts in society and he wonders if, particularly the latter one, is a very good idea. his solution would eliminate the right to not be seen; everyone&#8217;s business would be public, making it much more difficult to coordinate an assault on society large enough to damage it as a whole.</p>

<p>i&#8217;m always suspicious of anything i see/hear/read that has been filtered through the media because i know how badly things have been distorted so    i apologize if that wasn&#8217;t a very accurate representation of what jaron was trying to advocate; it was just the impression i got from seeing the interview. anyway, that impression was of a very naive argument. </p>

<p>i would love to live in a society where privacy wasn&#8217;t necessary; that would imply that there was no stratification of power and all other freedoms were respected absolutely. unfortunately as long as somewhere, some person has authority over some other person, privacy is necessary to prevent abuse of power. Orwell&#8217;s <i>1984</i> does a much better job of explaining why this is the case than i could do. if we were ever able to reach the utopia necessary to allow us to do away with privacy, terrorism wouldn&#8217;t be a problem anyway.</p>

<p>i&#8217;m pretty sure that jaron&#8217;s a smart enough guy to not actually be advocating the elimination of privacy as a practical solution to our problems. if only for the simple fact that cryptography exists and can&#8217;t be made to just disappear as much as certain government agencies would like it to.</p> 

<p>my train got in on friday night around 10. we met up with lani&#8217;s friend amy, and a bunch of random people from richmond and went out to club chaos to watch a drag show and dance (don&#8217;t worry, i didn&#8217;t dance; it&#8217;s not in my contract). </p>

<p>on saturday we slept late then went out for coffee together. the folks from richmond left and lani, amy and i went to the Asylum for 25 cent pints. it&#8217;s a little dive bar that happens to be cool enough to have skinny puppy and minor threat on the jukebox and did i mention that there were 25 <em>cent</em> pints?</p>

<p>then we met up with lani&#8217;s housemates for kebabs and wine before returning home and opening up the bottle of Absinthe that yura had smuggled back from the Czech republic. i didn&#8217;t know that absinthe was 140 proof. i do now. pretty tasty though. we weren&#8217;t doing much after that besides sitting on the porch listening to music and talking.</p>

<p>for breakfast on sunday, lani made us her family&#8217;s traditional recipe of eggs, grits and soy sauce. then we wandered over to georgetown for coffee and snacks. my nice long black coat that i&#8217;ve worn more or less continually since i bought it in hong kong in 1997 for $20 was finally starting to fall apart so lani, kim (lani&#8217;s british housemate) and i started a hunt for a replacement. unfortunately, long coats don&#8217;t seem to be trendy this year so almost no one had anything suitable for sale. we went through about 6 shops in 20 minutes without any luck. finally, just as we were about to give up and leave, we found a nice one. i think we set a new record for efficient shopping.</p> 

<p>in an effort to speed up thraxil, it now caches most of the data in static files and updates them only when it has to. so no more 50+ database hits everytime someone loads the main page. </p>

<p>i can&#8217;t really see an appreciable difference from here (postgres is so damn fast that hitting it 50 times in a row isn&#8217;t that inefficient. the main bottleneck now is probably just in the time it takes to establish a connection, fork apache and load the perl interpreter.) but in theory it should now be much more robust in the face of a heavy load (like 500 people hitting the site all at once) and even allows thraxil to maintain limited (read-only) functionality if the database gets shut down.</p>

<p>still on my todo list are a decent interface to the archives and the return of markov. once i get those taken care of, i&#8217;ll open the source code up for everyone to pick through and start writing some alternate themes.</p> 

<< prev page 35 next >>