posted in programming by Patrick Fitzsimmons on

The Ultimate Search

When I want to look up a web page I once visited, sometimes I can only remember a few words, an image, or a "feel" of the page. Unfortunately, those few words are rarely destinctive enough to find the page in a Google search. Yet to my brain the page feels very distinctive; I know exactly what I am looking for. Now imagine I had an MRI/brain scanning device as I browsed the web. Then, whenever I wanted to remember a page, I would merely have to think of it. The distinctive thought pattern would be detected by the MRI and the right page retrieved. Some day . . .

posted in programming by Patrick Fitzsimmons on

Morality, Law, and Filesharing

Over the past few weeks I have received many questions about the legality and ethics of the different filesharing questions I have worked on. Lanovision/Coffeeshop and filesharing in general. I was quoted several times in a Yale Herald article, a VC who is giving me some advice about the Yale 50K business plan contest, along with The Current Law On March 29th the Supreme Court will hear arguments in the MGM v. Grokster case. The 9th Court of Appeals held that Grokster was not illegal because they had neither centralized control over the network nor any specific knowledge of infringement. The court said that since there was a significant non-infringing use of the software, Grokster could not be held liable. This decision followed the precedent of the Sony v. Betamax case where the Supreme Court held that Betamax was not liable for the infringment of people making copies of videos. Napster and Aimster were both found liable for infringement in two separate court cases. The big difference was that they operated centralized servers that people could search to find files. What the law should be The lawsuit against the p2p companies is fascinating study of how perception can create legal reality. The entire world wide web could be divided into three basic functions: web serving, DNS, and search. The Internet is a system of computers that are all running webservers. A system of Domain Name Servers (DNS) create names by which these computers can be located on the web and can be linked to. Finally, search is what allows people to find information that could be located anywhere in the billions of pages that exist. Kazaa, Morpheus, and Grokster are very simply programs that combine these three functions into a single program. Kazaa serves files, automatically locates other computers, and then searches them. That is why ruling Kazaa illegal is so dangerous - you risk making the very software that runs the Internet illegal! I actually believe that the RIAA and MPIAA are correct in suing the individuals who share files. Imagine if someone had thousands of copyrighted mp3's on their computer, installed a webserver, registered the domain name "freemp3s.com", and then popularized their site for all to go to. It would be a no brainer that the site should be shut down. Even if it was a 12-year old girl, we would understand why she was being sued. Yet using Kazaa is the exact same thing. From a legal standpoint, the RIAA would have been much better off if it started suing individuals in mass the moment that Napster and Kazaa appeared. Of course, it wanted to try to go after the software makers first because suing ordinary people is terrible PR. Morality of Filesharing Of course today, most filesharing is no longer done using Kazaa. It is done over Bittorrent, using AIM, by using the combo of iTunes/MyTunes/OurTunes. This kind of sharing is done in a smaller network and thus legally much harder discover and prosecute. Take the typical college student Sam. Sam likes to hang out and watch a few hours of TV a day. Most of the stuff he watches is of pretty dubious quality - Pimp My Ride, Real World, Texas Hold 'em, etc. There are a few good shows out there, but mostly he is just wasting time. But one day Sam downloads the Ultimate File Sharing Program. This program allows him to watch any movie or TV Show ever made absolutely instantly. Suddenly instead of settling for the crap on TV he can watch Office Space, City of God,Shawshank Redemption, Family Guy, Seinfeld, Curb Your Enthusiasm, or any of hundreds of thousands of videos. Sam is now funnier, more cultured, and less bored than he was before. Is not this an objective improvement in his situation? No one is harmed by him downloading these movies - unlike stealing a DVD from a store, when you download a program the other person still has it. Of course, if no one paid for movies, movie companies would quickly stop producing movies. I think that the best thing to happen is that we have a very messy balance, with lots of different things tried, and new models will appear. I think that sharing in a small network such as iTunes is a good thing, after all, someone still has to buy the music originally. I also think that the media companies are very resistant to change and that file sharing is the only way to push them to offer the services that technology makes possible. Currently, 90% of the cost of music is marketing and distribution. Only 10% goes to production and to paying the artist. One of our ideas with Coffeeshop was that if listening to music was socially mapped you could analyze the popularity of new songs across social networks to detect who were going to be hit artists. The idea of paying for A&R men and expensive promotions is ridiculous in the Intenet age. With Lanovision, it would be awesome to integrate Lanovision, Bittorrent, and feedback algorithms to distribute and promote the best of originally created video. Lawrence Lessig has described how many media industries were born out of a mess of piracy and legally questionable actions. For instance, movie companies originally moved out to Hollywood in part to avoid the controls of Thomas Edison's patents. I believe that the same will happen with music and video. The Internet opens up entirely new ways of doing things. We must be open as much as possible to innovation, try many things, and have a Wild West type of attitude. The Morality of Filesharing

posted in programming by Patrick Fitzsimmons on

Business and the Academy Don't Mix

I wanted to check out some of the top business books recommended by 800CEOREAD, so I decided to search the Yale Library catalog. However, the book did not appear. So then I went to the Yale library's Borrow Direct service where I could search the combined catalogs of every single Ivy League University (except Harvard's cause they very snobbishly didn't join the consortium:-P). Only two of the top ten books at 800CEOREAD were found among the combined 30+ million volumes that these libraries contain. I guess research libraries and libraries for practical use are two different things entirely.

posted in programming by Patrick Fitzsimmons on

Lanovision Interview

PVR Blog has posted an email interview that I did with the site author about Lanovision. I have emailed a bunch of bloggers in the last couple days trying to get the word out. Now, if you search Google 3,280 hits come up for Lanovision. Unfortunately, that's less impressive than it seems, as all but about 30 of them come from random news aggregators that few people read. But I did get a few good posts in LifeHacker and TechInfo blog. Altogether I have had around 200 downloads in the past couple of days. Now I just need to get Slashdotted . . . :-)

posted in programming by Patrick Fitzsimmons on

On Writing Well

My biggest complaint about the Yale education is that my level of writing has actually deteriorated over the past two and a half years. During spring break, I picked up a book from my freshman year writing class, called On Writing Well. After reading it, I realized that a lot of junk has crept back into my writing. Sometimes I believe that I am doing negative learning by writing papers for some of my classes. The writing that is most important to me is short essays, articles, and speeches. But when I write for my classes I fall into an academic writing style that is opposed to much of what I have learned about good writing. I don't think is just me - the papers of others that I read show a similar excessive wordiness. I was hoping that this blog would give me an opportunity to do some writing the way I wanted to. Unfortunately, between trying not to fail classes and working on all my software projects, there simply hasn't been the time. I have about ten saved drafts of posts that I was not able to work on enough to publish. But that's ok. For now, programming is what I want to do. I will return to writing later in life.

posted in programming by Patrick Fitzsimmons on

Developers, Developers, Developers . . .

Last year I read a great article by John Gruber that refuted many of the myths about how Windows won the operating systems war ("The Art of the Parlay"). He argued that the reason that Apple lost was not price or their failure to license the operating system. Rather it was that Apple made Macintosh a complete break from the Apple II, while Microsoft made every effort to keep DOS programs working on Windows. It seems that Macintosh is still making this mistake. Over the past month I have been trying to port Lanovision to the Macintosh. This has been a frustrating experience. The biggest problem is that I do not have a Macintosh, and so have to beg my friends to let me borrow theirs', while giving them solemn promises that I will not harm their precious computer in any way. But it's also been frustrating simply because OSX Panther and OSX Jaguar are so incompatible. My friends are pretty much split on which system they have, so I have to build for both. A lot of the tools I have been using do not install on Jaguar. One can complain about Microsoft's slow release cycle, but I am glad that I only have one version of Lanovision for Windows and that it runs on every computer. Linux has the same problem. The wxPython download site has separate binaries for Debian, Fedora, Mandrake, and RedHat. That pretty much sucks all the fun out of creating a linux version. Both Apple and Linux providers are going to have to get their act together if they want to grab more market share. After all, as Steve Ballmer said, pleasing developers is the most important part of designing an operating system.

posted in programming by Patrick Fitzsimmons on

Yahoo! enters the social networking fray

Yahoo! has revealed its new social networking service, Yahoo! 360, to selected "influencers." Accounts can be found here and on Many-to-Many. Burham's Beat also writes about 360's impact on Friendster. It's looking more and more likely that Friendster will end up being a one-hit-wonder. MySpace and thefacebook.com have already stripped away the younger user base, and now Yahoo! will likely take away many of the older, less tech-savvy users. Yet as crowded as this social networking space is, the web is still not nearly as socially based as it potentially could be. I think that there are still many opportunities remaing, and I hope to take advantage of them.

posted in programming by Patrick Fitzsimmons on

The Mac OSX Version of Lanovision . . .

. . . is almost done. You can try out a test version for Panther (OS 10.3). I'm not sure if it actually works. I was using a computer lab Mac running in restricted mode to test it out. The computer had some random restrictions. Every time I ran a command to get the current working directory, it raised a "permision denied" error. It also denied me from spawning a new process so I had to start the Rendezvous/HTTP server seperately ("lanovision.app/Contents/MacOS/lanovision -runserver") But it might work if you try it on a Mac running with normal privileges. I think that my complaints yesterday about the MacOS operating system were unjustified. That I was able to make an OSX Panther desktop application without owning a Mac and without installing any developers tools is pretty amazing. Also, the real problem with changing an OS too often is when you can't run old tools on the new version, not when new tools won't run on the old version. I think that now that with OS X, with its unix base, is and will be a great platform to develop on.

pfitz-face-formal-croppedAbout me
I am a startup guy and software engineer.  I work my craft at HubSpot, a startup in Cambridge, MA that builds inbound marketing software.  In the HubSpot early years I had to rapidly crank out prototypes as we figured out how to make a product that customers wanted.  Now I help design larger scale systems to serve our nearly ten thousand customers.

Posts by topic

Subscribe to Email Updates