<?xml version="1.0" encoding="UTF-8"?>
<posts type="array">
  <post>
    <body>&lt;p&gt;This can be something that can make you proud in the morning :-)&lt;/p&gt;


	&lt;p&gt;Basically, try searching on different code search engines for your name, and see the results! I&amp;#8217;m sure you will be impressed with the results&amp;#8230;&lt;/p&gt;


	&lt;p&gt;Here are my results:&lt;/p&gt;


	&lt;p&gt;&lt;strong&gt;Google Code Search&lt;/strong&gt;:
&lt;a href="http://www.google.com/codesearch?hl=en&amp;#38;lr=&amp;#38;q=%22ruben+fonseca%22&amp;#38;sbtn=Search"&gt;Search public source code for Ruben Fonseca&lt;/a&gt;&lt;/p&gt;


	&lt;p&gt;&lt;strong&gt;Koders Search&lt;/strong&gt;:
&lt;a href="http://www.koders.com/default.aspx?s=%22ruben+fonseca%22"&gt;Search public source code for Ruben Fonseca&lt;/a&gt;&lt;/p&gt;


	&lt;p&gt;&lt;strong&gt;Krugle Code Search&lt;/strong&gt;:
&lt;a href="http://www.krugle.org/kse/files?query=%22ruben%20fonseca%22"&gt;Search public source code for Ruben Fonseca&lt;/a&gt;&lt;/p&gt;


	&lt;p&gt;Show me your results :-)&lt;/p&gt;


	&lt;p&gt;(idea stoled from &lt;a href="http://coreygoldberg.blogspot.com/2008/07/searching-for-your-open-source-code.html"&gt;Corey Goldberg blog&lt;/a&gt;)&lt;/p&gt;</body>
    <excerpt nil="true"></excerpt>
    <id type="integer">109</id>
    <permalink>searching-for-your-open-source-code</permalink>
    <published-at type="datetime">2008-07-04T10:54:00+01:00</published-at>
    <title>Searching for  your Open Source code</title>
  </post>
  <post>
    <body>&lt;p&gt;Something like 2-3 years ago, I started a small project with my friends called &lt;a href="http://cshare.sourceforge.net/"&gt;CShare&lt;/a&gt;.&lt;/p&gt;


	&lt;p&gt;From the site you can read:&lt;/p&gt;


	&lt;blockquote&gt;
		&lt;p&gt;cshare (from console share) lets you share any directory through a &lt;span class="caps"&gt;HTTP&lt;/span&gt; server. it&#8217;s simply to use (just launch it from the terminal), fast (multi-threaded &lt;span class="caps"&gt;POSIX&lt;/span&gt;), personalizable (XSLT themes), and client plataform independent (you just need a browser).&lt;/p&gt;
	&lt;/blockquote&gt;

	&lt;p&gt;We started the project because we wanted to write some elegant code with a modern C style, and do some experimentations on bleeding performance under the Linux environment.&lt;/p&gt;


	&lt;p&gt;Although we wrote tons of code (it&amp;#8217;s still available on the &lt;span class="caps"&gt;SVN&lt;/span&gt; repository), the application never saw the light of the day, mainly because we felt a lack of motivation due to the slow development speed with C.&lt;/p&gt;


	&lt;p&gt;So we&amp;#8217;ve completely abandon the project. I&amp;#8217;ve also tried 2 or 3 times to completely remove the project from &lt;a href="http://sf.net"&gt;SourceForge&lt;/a&gt; with no success: the process is beyond bureaucratic&amp;#8230;&lt;/p&gt;


	&lt;p&gt;However, the reason I&amp;#8217;m writing this post is because I found today a simple way of getting the same functionality of CShare, with a &lt;a href="http://www.debianpt.org/node/1693"&gt;single line of Python&lt;/a&gt;:&lt;/p&gt;


&lt;pre&gt;
python -c "import SimpleHTTPServer;SimpleHTTPServer.test()" 
&lt;/pre&gt;
	&lt;p&gt;Even tough you will not get the same performance and concurrent access like CShare would (eventually) deliver, it&amp;#8217;s simply enough to sit at your &lt;code&gt;.bashrc&lt;/code&gt;, and it really gets the job done!&lt;/p&gt;</body>
    <excerpt nil="true"></excerpt>
    <id type="integer">108</id>
    <permalink>cshare-in-one-line</permalink>
    <published-at type="datetime">2008-06-19T18:58:00+01:00</published-at>
    <title>CShare in one line</title>
  </post>
  <post>
    <body>&lt;p&gt;My current laptop has a beautiful piece of crap of an &lt;a href="http://ati.amd.com/products/radeonx1600/index.html"&gt;ATI Mobility Radeon &lt;span class="caps"&gt;X1600&lt;/span&gt;&lt;/a&gt;. At first I installed &lt;a href="http://www.ubuntu.com"&gt;Ubuntu 8.04&lt;/a&gt; on it and immediately switched to the proprietary &lt;span class="caps"&gt;ATI&lt;/span&gt; driver (to get more features and better support for multi-head).&lt;/p&gt;


	&lt;p&gt;However, the display always seemed sluggish.. I never use 3D or Composite, it all seemed related to 2D rendering. For instance, when I switched workspaces I often had to wait 10 seconds for windows to be redrawn.&lt;/p&gt;


	&lt;p&gt;Still, I managed to work like this for a while. One day however, I decided to try the &lt;a href="http://dri.freedesktop.org/wiki/ATIRadeon"&gt;open source drivers&lt;/a&gt; (it seems they have initial support for the &lt;span class="caps"&gt;R500&lt;/span&gt; chip on this graphics board).&lt;/p&gt;


	&lt;p&gt;The good news is that is supports multi-head profiles far better than the proprietary driver (&lt;a href="http://www.x.org/wiki/Projects/XRandR"&gt;XRandR&lt;/a&gt; 1.2 &lt;span class="caps"&gt;FTW&lt;/span&gt;!). Still, the 2D sucked, got no 3D nor XVideo overlay&amp;#8230;&lt;/p&gt;


	&lt;p&gt;I even thought &amp;#8220;oh this must be an Ubuntu thing&amp;#8221;... At the beginning of this week I managed to install &lt;a href="http://www.debian.org"&gt;Debian&lt;/a&gt; Unstable on this laptop, but had the same problems using the open source &amp;#8220;radeon&amp;#8221; driver&amp;#8230;.&lt;/p&gt;


	&lt;p&gt;On a desperate move, I&amp;#8217;ve tried the latest and bleeding edge &lt;span class="caps"&gt;ATI&lt;/span&gt; proprietary driver. For my surprise, all my problems were solved and the card is working properly! So I decided to log here the steps I took to get the things properly (remember, I&amp;#8217;m running Debian unstable).&lt;/p&gt;


	&lt;h3&gt;Instructions&lt;/h3&gt;

	&lt;p&gt;First, we have to download the latest drivers from &lt;span class="caps"&gt;ATI&lt;/span&gt; from &lt;a href="http://ati.amd.com/support/drivers/linux/linux-radeon.html"&gt;their website&lt;/a&gt; Then as root run:&lt;/p&gt;


&lt;pre&gt;
./ati-driver-installer* --extract fglrx
cd fglrx
./packages/Debian/ati-packager.sh --buildpkg sid
cd ..
dpkg -i *.deb
&lt;/pre&gt;
	&lt;p&gt;This will build &lt;code&gt;deb's&lt;/code&gt; from the &lt;span class="caps"&gt;ATI&lt;/span&gt; distribution file and install them or your system. The next step is to build the actual driver, and for that you can use module-assistant like this:&lt;/p&gt;


&lt;pre&gt;
m-a prepare
m-a update
m-a a-i fglrx
&lt;/pre&gt;
	&lt;p&gt;This will build and install the &lt;code&gt;fglrx&lt;/code&gt; driver for your current installed kernel.&lt;/p&gt;


	&lt;p&gt;Then you must configure Xorg. I followed this steps:&lt;/p&gt;


&lt;pre&gt;
aticonfig --initial=dual-head --screen-layout=horizontal
&lt;/pre&gt;
	&lt;p&gt;After this I just started X and &lt;strong&gt;it worked perfectly&lt;/strong&gt;. I mean, I had horizontal BigDesktop, fast 2D rendering, 3D, XVideo overlay on both displays and suspend is working like a charm.&lt;/p&gt;


	&lt;p&gt;The only problem left is that my second display has a bigger resolution that the main one. However, the &lt;span class="caps"&gt;ATI&lt;/span&gt; driver sets the second display to the same resolution as the first.&lt;/p&gt;


	&lt;p&gt;To work around this problem I just had to manually edit &lt;code&gt;/etc/X11/xorg.conf&lt;/code&gt; and add at the &lt;code&gt;Section "Device"&lt;/code&gt; this:&lt;/p&gt;


&lt;pre&gt;
    Option "Mode2" "1680x1050" 
&lt;/pre&gt;
	&lt;p&gt;Replace the resolution with the native resolution of your second display. Restart X and &lt;span class="caps"&gt;BANG&lt;/span&gt;, it &lt;strong&gt;should&lt;/strong&gt; work :)&lt;/p&gt;


	&lt;p&gt;Hope that this instructions can help you somehow.&lt;/p&gt;</body>
    <excerpt nil="true"></excerpt>
    <id type="integer">107</id>
    <permalink>solving-2d-rendering-problems-on-an-ati-x1600</permalink>
    <published-at type="datetime">2008-05-28T19:24:00+01:00</published-at>
    <title>Solving 2D rendering problems on an ATI X1600</title>
  </post>
  <post>
    <body>&lt;p&gt;This is my side of the story about the &lt;a href="http://fct.enses.org/barcamp/app/"&gt;BarCampFCT&lt;/a&gt; event.&lt;/p&gt;


	&lt;p&gt;First of all, congratulations to the organization for selecting a great place to make the Barcamp. I never went to the &lt;a href="http://www.fct.unl.pt/"&gt;FCT &lt;span class="caps"&gt;UNL&lt;/span&gt;&lt;/a&gt; before, but I loved the peace, the view and the weather. However, next time try not to do a Barcamp on a amphitheater.. it kills networking!&lt;/p&gt;


	&lt;p&gt;The day began with a slow start&amp;#8230; The poor guy talking about &lt;a href="http://firehol.sourceforge.net/"&gt;FireHOL&lt;/a&gt; had the duty of waking people and do the kick off. Next talk about technique presentations by Jo&#227;o Rico started well, but quickly became boring. I had to get out of the room and search for some coffee&amp;#8230;&lt;/p&gt;


	&lt;p&gt;This was actually a good idea, because at the coffee break room, it was happening a lot of networking! Kudos for the cookies and the coffee, they were great! This happens to be the first time to know people and mingle around.&lt;/p&gt;


	&lt;p&gt;Next it was the lunch&amp;#8230; the lunch was &lt;strong&gt;perfect&lt;/strong&gt;. &lt;strong&gt;Perfect&lt;/strong&gt;. &lt;strong&gt;Perfect&lt;/strong&gt;. Congratulations for this!&lt;/p&gt;


	&lt;p&gt;&lt;a href="http://wizi.com/"&gt;Wizi&lt;/a&gt; arrived late, and managed twice to almost kill the barcamp. On their first talk, they asked for 5 minutes, and delivered some 20+ boring talk&amp;#8230; Can&amp;#8217;t help it&amp;#8230;&lt;/p&gt;


	&lt;p&gt;But then there was &lt;a href="http://artofsystems.blogspot.com/2007/01/halfbakedcom-entrepreneurial-improv.html"&gt;halfbaked&lt;/a&gt;! Lot&amp;#8217;s of great ideas about creating a new company, and lots of laughing too. Great time! &lt;a href="http://www.tarpipe.com"&gt;TarPipe&amp;#8217;s&lt;/a&gt; didn&amp;#8217;t disappoint, although I was expecting a more practical talk about the service (/platform/product).&lt;/p&gt;


	&lt;p&gt;After more coffee and cookies, it was time to &lt;span class="caps"&gt;IGNITE&lt;/span&gt;! First I have to say that ignite sessions should have been done earlier, and the organization should be more fair and respect the speakers. Some got 10 or 15 minutes, some got 5 with auto rotating slides (killing the presentation), and then Wizi destroyed all the ignite spirit.&lt;/p&gt;


	&lt;p&gt;I managed to get 5 minutes auto-rotating slides and talked about &amp;#8220;Debian Sucks&amp;#8221; (it doesn&amp;#8217;t, really!) and the OpenSSL fiasco.&lt;/p&gt;


	&lt;p&gt;Overall it was a great day, the organization deserve a great applause! I&amp;#8217;ve got some good conversations, new ideas, met new people, and learned how a company (Wizi) can easily generate some bad buzz at an event.&lt;/p&gt;</body>
    <excerpt nil="true"></excerpt>
    <id type="integer">106</id>
    <permalink>barcampfct-my-side-of-the-story</permalink>
    <published-at type="datetime">2008-05-26T09:13:00+01:00</published-at>
    <title>BarCampFCT, my side of the story</title>
  </post>
  <post>
    <body>&lt;p&gt;Together with &lt;a href="http://whythehype.com/"&gt;Tiago Pinto&lt;/a&gt; I&amp;#8217;ve completed a &lt;a href="http://www.tarpipe.com"&gt;TarPipe&lt;/a&gt; &lt;a href="http://tarpipe.rubyforge.org"&gt;API for Ruby&lt;/a&gt;.&lt;/p&gt;


	&lt;blockquote&gt;
		&lt;p&gt;tarpipe is a publishing mediation and distribution platform that simplifies regular upload activities&lt;/p&gt;
	&lt;/blockquote&gt;

	&lt;p&gt;The gem is available &lt;a href="http://tarpipe.rubyforge.net"&gt;here&lt;/a&gt; and the sources are on &lt;a href="http://github.com/rubenfonseca/tarpipe-ruby/tree/master"&gt;github&lt;/a&gt;. It is my first released gem so feel free to send some tips and suggestions.&lt;/p&gt;


	&lt;p&gt;TarPipe is currently evolving very fast so me and Tiago will make sure our library stays up to date. Have fun doing your uploads!&lt;/p&gt;</body>
    <excerpt nil="true"></excerpt>
    <id type="integer">105</id>
    <permalink>tarpipe-ruby-api</permalink>
    <published-at type="datetime">2008-05-20T14:21:00+01:00</published-at>
    <title>TarPipe Ruby API</title>
  </post>
</posts>
