loud thinking by david heinemeier hansson

loud thinking by david heinemeier hansson me about gallery company girl projects ruby on rails basecamp highrise backpack campfire ta-da list writeboard more feed archives october 17, 2007 13:28 swearing at work boosts morale speaking of fuck, researchers find it to boost morale at work: regular swearing at work can help boost team spirit among staff, allowing them to express better their feelings as well as develop social relationships, according to a study by researchers."we hope that this study will serve not only to acknowledge the part that swearing plays in our work and our lives, but also to indicate that leaders sometimes need to 'think differently' and be open to intriguing ideas." october 12, 2007 14:16 kathy's zones of emotion i miss kathy's blogging. don't you? october 11, 2007 16:08 potty mouths i'm allergic to people who willingly and without irony use the term "potty mouth" in adult conversation. the notion that a word like fuck can make your brain curl up and cry like a toddler is so pathetically disturbing that it makes my skin crawl. it has the plastic smell of a barbie playhouse and the repressed insecurities of casual friday khakis. i can't fucking stand it. but at least the potty mouth reaction is a useful leading indicator for personality fits. it's almost as good as the f-bomb reaction. both are fake euphemisms that are actually much worse than the honest words they're trying to put a fig leaf to. and if you're serious about using them, i'm serious about thinking you're too fucking lame to bother further debate with. which of course is ironic. since the whole potty mouth fuzz is about distancing yourself from that foul person on the basis of words. heh. but at least us sailors recognize that the sea flows both ways. that by using wonderfully flexible joker words like fuck, we're sending a signal of distance to the inevitable crowd who takes offense from that. the potty mouth crowd seems shocked — shocked — that their language could have a similar effect in the opposite direction. anyway, fuck it (how great is this word? i seriously considered naming this post "fuck: the mother word" but in the end my despise of the potty mouth term won out just slightly over my affection for the word fuck). if you want to dive deeper into the wonderful world of cursing, i highly recommend what the fuck: why we curse by steven pinker p.s.: see also fuck, a marvelous tour of the utility of this word. september 24, 2007 11:37 the language of bias if you don't like something new that's getting a lot of attention, you call it out as being all hype propelled by fanboys enamoured by an immature solution. if you like that something new, you say it has momentum that's being accelerated by passionate advocates of a fresh approach. september 23, 2007 18:58 thinking about the big rewrite? rewriting existing applications from scratch in a big bang-style is rarely a good idea and often ends in failure. chad fowler wrote a good post summarizing some of the reasons why a year ago: the big rewrite. september 19, 2007 16:35 good times at railsconf europe it's hard to fathom that we've just finished our fourth railsconf in just 15 months. the first one already seems so long ago, which just underlines how fast things have been moving in the rails world. it's even more impressive to think of all the people we've had go through the conferences. more than three thousands attendees combined and some of the most interesting speakers in technology. this year i was especially pleased to hear roy fielding speak and to have the opportunity to chat with him afterwards. the past year and a half been about trying to understand and implement the ideas and the specifications that roy has been involved with for more than a decade. it was very pleasing to hear him talk so warmly of our progress so far. sun continued to surprise positively (and still slightly puzzling), but i had so much to say about that, i went ahead and wrote it up separately in sun surprises at railsconf europe 2007. as i process all the inputs from the conference, i'm sure i'll have more to say. but for now i want to make sure to thank everyone involved. ruby central and o'reilly did a spectacular job organizing everything. and i had a great time speaking to so many different users of rails. the most amazing things are being done with this framework we're building. p.s.: my keynote slides were few and almost entirely without meaning out of context, but casper fabricius did a good job with his summary. september 19, 2007 15:23 sun surprises at railsconf europe 2007 the biggest surprise at railsconf europe 2007 was hearing craig mcclanahan from sun speak. craig is the creator of struts, the original blockbuster web framework for java, and more recently java server faces. not exactly an obvious advocate for ruby on rails, but he sure played the part well. and not just on a superficial level because sun was a diamond sponsor of the conference, but from months of experience developing rails applications using the latest techniques and frameworks (like pushing the envelope with active resource). he even went as far as to say that developing web applications in java after working with rails would probably not be a particularly pleasant experience for him. wow. sun's support for rails goes even further, though. their work on making netbeans a kick-ass rails ide seems to be progressing at a rapid clip (and already winning fans). yet it pales in comparison to the rise of jruby as driven by thomas, charles, and ola. these guys are improving their jvm implementation of ruby at a truly insane speed. jruby on rails is already in production at a fair number of companies. the rails is just a war argument is certainly making it easier for companies like thoughtworks to push rails deep into the enterprise. throw the "integrate with your java legacy" argument on top (where you can even go as far as work with ejbs!) and it's easy to see why the enterprise is getting all giddy. "why are they doing all this?", that's a common concern with most ruby folks (and a question i must have asked them personally at least half a dozen times ;)). apparently, they see this as something quite simple: a sun that's heavily involved with rails on the software side is a sun that's much better positioned to sell loads of hardware to run all these new rails applications in both the enterprise and to start-ups. it seems to be working too. joyent is running all their new gear with sun badges. and they got some serious applications cooking there, including twitter. so it's kinda hard to argue with their good deeds. which is a peculiar feeling for a rails person to have towards a company that used to be regarded solely as the high church of enterprisey. i guess that they've accepted that java (the language) is no longer always the answer regardless of the question, as it used to be. yet there's still that nagging doomsday image of a mystery man sitting somewhere inside of sun petting a black cat with a hollow laugh just waiting to reveal his evil master plan as the reason for all this. but the exceptional work and high character of people like tim bray, nick sieger, thomas enebo, charles nutter, and now even craig mcclanahan championing ruby and rails inside sun, i think i'll have to waive goodbye to the last sliver of that suspicion very soon. july 30, 2007 12:34 three years with ruby on rails we've just passed the three year anniversary of ruby on rails' first release to the public. which in turn means that i've been working on the framework for more than four years. wild. it's been an incredibly rewarding experience. i've met so many great people. worked with so many talented programmers. seen so many amazing sites and applications be launched off rails. to image that this whirlwind tour all started because i wasn't happy working with the mainstream environments and decided to give ruby a try. and to image the impact ruby on rails have had on the industry with none of the big-company backing that traditionally is needed to make waves there. what's even more rewarding is knowing that we're not done. that the work to make web-application development ever more pleasant continues every day. we might have taken the pace of radical changes in the core framework down a few notches (remember when we did multiple releases per month?), but the steady stream of improvement and refinement continues relentlessly. so cheers, rails. july 21, 2007 09:34 a billion dollars doesn't buy much these days just days after microsoft announced spending a billion dollars on extending the xbox 360 warranty, my own console surrended to the dreaded three flashing red lights. now you'd think a billion dollars would buy microsoft some premium, grade-a service and expedition to make short order of such a widespread problem, no? think again. it took three separate tries to even get through on their support line. the final attempt required about an hour on the line. all that to get a paper box shipped as the return package using ups 3-day service (unlike, say, apple's overnight delivery). but much worse, expected service time is 4-6 weeks! between the attempts at calling support and shipping boxes back and forth with snail service, i'm looking at a two-month turn-around from problem to resolution. yikes. how am i supposed to cope with no forza motorsports for that along!? it's inhuman, i tell you. june 15, 2007 15:13 registration open for railsconf europe 2007 i've actually never been to berlin before, so i'm excited to get the chance to not only see one of the great cities of europe, but at the same time share in a meeting with the rails community in europe and around for railsconf europe 2007. the doors for registration have just been opened and until august 6th, the price is €645 (after that, it jumps to €795). i really had a smashing time last year in london. there was somehow more time to get into more discussions with people than the more hectic version in the us. and i got to premiere a bunch of new rails features i had been working on over the summer. so i hope to see a lot of the familiar faces from last year and new ones too when we convene from september 17th through 19th in germany. june 06, 2007 11:28 multi-core hysteria and the thread confusion new cpus are growing in cores and not in ghz. that's a tough problem for applications that have been traditionally single-threaded, like games. they have to learn all new techniques and rework their thinking to get the most out of the next-generation platforms. but the fear of that transition has bled into places where it's largely not relevant, like web-application development. which has caused quite a few folks to pontificate that the sky is falling for rails because we're not big on using threads. it isn't. multiple cores are laughably easy to utilize for web applications because our problems are rarely in the speed of serving 1 request. the problem is in serving thousands or tens or hundreds of thousands of requests. preferably per second. threads are not the only way to do that. processes do the job nearly as well with a drop of the complexity. and that's exactly how rails is scaling to use all the cores you can throw at it. the 37signals suite is currently using some ~25 cores for the application servers that all the applications have dips on. we'd welcome a 64-core chip any day. read more: a good summary of a discussion on multi-core programming in general. june 05, 2007 13:39 why there's no rails inc idc predicts the market for open-source software will reach some six billion dollars in 2011. no wonder the vcs are getting anxious to play on that roulette. which brings me to why there's no rails inc. it certainly isn't for lack of vcs wanting to fund. i've had more than a handful conversations with various outfits eager to pour big money into such an operation, but i'm just not interested. there are many reasons not to be interested in vc money these days, but let's just give two specific ones for rails. first, rails is not my job. i don't want it to be my job. the best frameworks are in my opinion extracted, not envisioned. and the best way to extract is first to actually do. that's really hard if your full-time job is just the extraction part since you now have to come up with contrived examples or merely live off the short bursts of consulting. for some that might work, but i find that all my best ideas and apis come from working on a real project for a sustained period of time. second, the growth of the rails ecosystem has been staggering. there are so many shops out there offering rails consulting and training. i believe part of that proliferation is due to the fact that there's no core-group monopoly that can dominate the market. i believe a rails inc consisting of a large group of core committers would have an unfair advantage in the training and consulting space — easily siphoning off all the best juice and leaving little for anything else. there are plenty of examples in our industry of that happening around open source tools. it's much more satisfying to see a broader pool of companies all competing on a level playing field. may 28, 2007 18:22 digesting railsconf 2007 going from five to sixteen hundred people is a big risk for a conference. there's so much to lose: the atmosphere, the coherence of content, and the interestingness of the people. but in my mind we didn't, railsconf 2007 was a roaring success. there were so many great debates going on, so much fascinating work happening, and so extraordinary tales of adoption. it was wonderful to meet up with people like martin fowler, ward cunningham, tim bray, dave thomas, robert martin, and other industry leaders. but in many ways, even more wonderful was the level of involvement from everyone else. i remember rubyconf '03 when we just had a couple of people doing professional ruby work. this year at railsconf more than half the room raised their hand when i asked how many were working professionally with rails. what a leap. so many people doing applications in all niches and of all shades. plenty of startups, naturally, but also plenty of so-called enterprise operations. from banks to insurance companies. thoughtworks announcing that 40% of all new business in the us is ruby on rails projects. wow. i loved the fact that it wasn't all about the nitty gritty stuff either. we had an extra action marching band that pushed the comfort level of many on the fun side of things. and on the more serious side, alan francis explored the similarities between the rails and xp movements on a higher plane of approach, angry teenager-tendencies, and peaks. i also much enjoyed the fact that it was broader than just ruby and rails circle. that we had avi bryant talk to us about this magical parallel universe of smalltalk. and that we attracted people like scott hanselman from the .net world (and that he posed plenty of opposing opinion that we sorta captured in a podcast with martin fowler and me). all in all, a spectacular extended weekend. it made me all the more excited for turning another chapter in the conference book in berlin come late september with railsconf europe. all photos by the always awesome james duncan davidson may 28, 2007 14:28 cranking up the machinery so i finally had a few spare moments to work on the loud thinking machine again. instead of going with one of the million packages out there, i decided to eat some dog food and just roll my own. yes, yes, terribly inefficient from a productivity perspective, but i indulged myself with a learning experience on how it feels to setup a small rails application from scratch using ubuntu fiesty, nginx, mongrel, and sqlite3. as a side-effect, i haven't bothered implementing comments for my little machine just yet. and i'm thinking that's actually a blessing in part disguise. i think i'll be happy with the tranquillity for a while. may 18, 2007 09:55 while we're hoping that the backups are good loud thinking is down for the count while the server that held it up for the past six years is in icu. hopefully it'll make it through and allow the archives to appear unscathed. but if not, we'll deal with that too. currently i have from june '05 and backwards recovered.. nothing like a little forced spring cleaning to get the cruft moved out. but if you sent me any email to the loudthinking address over the past 3-4 days, it might not have made it. so if it was important, please do shoot it over again. april 17, 2007 13:14 getting back to the foil i used to fence in high school. it's such an elegant sport. yes, stamina, agility, and even strength are elements, but fencing foil is so much more about technique. a small rotation of the wrist is all it takes to parry an attack and you're all set for riposte. i relearned that last wednesday when i went my first fencing class in more than a decade. most of the footwork was still imprinted in my memory, but that was the easy part. the hard part was realizing just how much technique i had lost as i was getting schooled by a 7 year-old girl. which is of course also one of the wonders of fencing. it allows for such a wide range of physical attributes to enjoy it together. in the small group that was at the session that night, it ranged from alexa, 7 to a 50-something argentinean. with a few teenagers and me in between. so lots of fun, but i'm also grateful that there's a full week between sessions on the beginner's team. it's taken at least half that to be able to walk without pain again. nothing like exercising muscles that have laid dormant since jurassic park was a box office hit. april 15, 2007 13:15 scaling to multiple databases with rails remember that point about rails lacking an easy-to-use way of dealing with multiple read/write databases? strike that. nic williams has released magic multi-connections. it makes it dead easy to use a cluster of databases to scale read and write speeds higher than a single connection would ever allow. that in itself is wonderful. williams let code be his reply to the discussion of twitter's woes on scaling the database. i would of course rather have seen this work come out of twitter, but i'm happy that they got a free offering handed to them regardless. they didn't even have to pass step 1 in brian mccallister's road map for getting stuff fixed in open source. and the turn-around time was within the same day of this whole thing blowing up. now how could this be. how could nic fix such an apparent "critical flaw", as others have billed the lack of this facility in rails, in such a short time? simple, he did it in less than 75 lines of ruby as a plugin for rails. less than 75 lines. in my mind, that's the crux of the story. that extending rails to do what you want is often much simpler than you think. that you can't compare extending a high-level framework written in a language like ruby to, say, patching apache or mysql. the barriers of entry are simply not in the same sport. so let's use this occasion to celebrate the wonders of open source ("some times you can just ask and you will receive"), but at the same time keep the effort involved in this example as a guidance for the future ("maybe next time, i could just have a look at how hard it would be to fix myself"). and of course, a big thanks to nic williams to making a big fuss a non-issue.

loud thinking by david heinemeier hansson  Précédent 614  Précédent 613  Précédent 612  Précédent 611  Précédent 610  Précédent 609  Précédent 608  Précédent 607  Précédent 606  Précédent 605  Précédent 604  Précédent 603  Précédent 602  Précédent 601  Précédent 600  Précédent 599  Précédent 598  Précédent 597  Précédent 596  Précédent 595  Précédent 594  Précédent 593  Précédent 592  Précédent 591  Précédent 590  Précédent 589  Précédent 588  Précédent 587  Précédent 586  Précédent 585  Suivant 616  Suivant 617  Suivant 618  Suivant 619  Suivant 620  Suivant 621  Suivant 622  Suivant 623  Suivant 624  Suivant 625  Suivant 626  Suivant 627  Suivant 628  Suivant 629  Suivant 630  Suivant 631  Suivant 632  Suivant 633  Suivant 634  Suivant 635  Suivant 636  Suivant 637  Suivant 638  Suivant 639  Suivant 640  Suivant 641  Suivant 642  Suivant 643  Suivant 644  Suivant 645