<?xml version="1.0"?>
<rss version="2.0">
  <channel>
    <title>Advogato blog for softkid</title>
    <link>http://www.advogato.org/person/softkid/</link>
    <description>Advogato blog for softkid</description>
    <language>en-us</language>
    <generator>mod_virgule</generator>
    <pubDate>Sun, 12 Oct 2008 18:07:22 GMT</pubDate>
    <item>
      <pubDate>Thu, 9 Oct 2008 15:08:51 GMT</pubDate>
      <title>I need to read documentation</title>
      <link>http://www.advogato.org/person/softkid/diary.html?start=160</link>
      <guid>http://perso.hirlimann.net/~ludo/blog/archives/2008/10/i-need-to-read-documentation.html</guid>
      <description>&lt;p&gt;Yesterday I started installing FreeBSD, on an old Mac-mini G4. The install broke just after partitioning the drive with the following error :&lt;br /&gt;
&lt;code&gt;&lt;br /&gt;
Error mounting /dev/ad0s3 /mnt Operation Not Permited&lt;br /&gt;
&lt;/code&gt;&lt;br /&gt;
And I could not get past that step in the install. After fumbling a bit - I got back at reading the docs and discovering that the partitioner did not really work for the PowerPC port - and that booting directly in FreeBSD was not possible :-( That made FreeBSD/PowerPC  something unusable for what I had in mind. I might If I ever get such a machine at home , and find the time try to fix that but .....&lt;/p&gt;

&lt;p&gt;A few minutes after that I went to ask a question to a colleague about the ads network and the delivery of ads - to which he replied that I should have carefully read the email he had sent that same morning. I had read the email up to 66% of it and the information I was asking for was lying in the last 30% of the email.&lt;/p&gt;

&lt;p&gt;When I start anything by reading the documentation I often get bored and confused - but I think that after those two "issues" in the same day I will read some stuff more carefully from now on.&lt;/p&gt;</description>
    </item>
    <item>
      <pubDate>Wed, 8 Oct 2008 15:07:48 GMT</pubDate>
      <title>Data consistency</title>
      <link>http://www.advogato.org/person/softkid/diary.html?start=159</link>
      <guid>http://perso.hirlimann.net/~ludo/blog/archives/2008/10/data-consistency.html</guid>
      <description>&lt;p&gt;I'm still playing with data and metadata - I like using last.fm ever since I discovered they knew nothing about the rock album I brought back from Greenland with me. So I've inserted the needed data into the musicbrainz database a few weeks ago, and as the band is knew I tend to listen to it's tunes a bit more than the other artists I usually listen to.&lt;br /&gt;
So my funny founding of the day are :&lt;br /&gt;
* search for "Unnuaq Band" in music or group - and no result pop up.&lt;br /&gt;
* there is a page for the artist which contains a picture and a bio of the band.&lt;br /&gt;
* the page contains a 'Top Albums' item, If you look in the album tab of that group it's empty.&lt;/p&gt;

&lt;p&gt;I find it very funny that last.fm knows about the band - but searches fails. One album seems to be known but not completely. I think last.fm did not synch with musicbrainz yet. Or that the data I've added to musicbrainz is not available to last.fm, anyhow I find it funny that the data shown is not consitant at all from one page to another.&lt;/p&gt;</description>
    </item>
    <item>
      <pubDate>Wed, 1 Oct 2008 09:13:35 GMT</pubDate>
      <title>Html composer project</title>
      <link>http://www.advogato.org/person/softkid/diary.html?start=158</link>
      <guid>http://perso.hirlimann.net/~ludo/blog/archives/2008/10/html-composer-project.html</guid>
      <description>&lt;p&gt;Nvu/Kompozer now has a successor, the name is &lt;a href="http://bluegriffon.org/" &gt;bluegriffon&lt;/a&gt; or so it seems. Some source code is available, so those of you who know how to build mozilla should be able to try it.&lt;/p&gt;</description>
    </item>
    <item>
      <pubDate>Fri, 26 Sep 2008 12:11:35 GMT</pubDate>
      <title>Registration for Eurobsd con is open</title>
      <link>http://www.advogato.org/person/softkid/diary.html?start=157</link>
      <guid>http://perso.hirlimann.net/~ludo/blog/archives/2008/09/registration-for-eurobsd-con-is-open.html</guid>
      <description>&lt;p&gt;&lt;a href="http://eurobsdcon2008.eventbrite.com/" &gt;Registration&lt;/a&gt;for EuroBSDCon 2008 in Strasbourg, is finally open .&lt;/p&gt;

&lt;p&gt;The talks schedule is available &lt;a href="http://2008.eurobsdcon.org/talks.html" &gt;here&lt;/a&gt;, the tutorials schedule is available &lt;a href="http://2008.eurobsdcon.org/tutorials.html" &gt;here&lt;/a&gt;.&lt;/p&gt;

&lt;p&gt;There is a 20% discount for students (use discount code &lt;strong&gt;Student&lt;/strong&gt; when&lt;br /&gt;
registering).&lt;/p&gt;

&lt;p&gt;&lt;br /&gt;
&lt;/p&gt;</description>
    </item>
    <item>
      <pubDate>Tue, 23 Sep 2008 07:07:24 GMT</pubDate>
      <title>Fixing the CD lookup function in Musicbrainz's Picard</title>
      <link>http://www.advogato.org/person/softkid/diary.html?start=156</link>
      <guid>http://perso.hirlimann.net/~ludo/blog/archives/2008/09/fixing-the-cd-lookup-function-in-musicbr.html</guid>
      <description>&lt;p&gt;If Like me you want to use MusicBrainz Picard metadata tagger, and are running Mac OS X, you'll probably encounter ticket &lt;a href="http://bugs.musicbrainz.org/ticket/3532" &gt;3532&lt;/a&gt;, which displays this nice little message box : &lt;br /&gt;
&lt;a href="http://www.flickr.com/photos/lhirlimann/2880861851/" title="Picard CD error by Ludovic Hirlimann, on Flickr" &gt;&lt;img src="http://farm3.static.flickr.com/2215/2880861851_b26fc5db47_o.jpg" width="420" height="192" alt="Error opening library: dlopen(libdiscid.0.dylib, 6): image not found " /&gt;&lt;/a&gt; which says : "Error opening library: dlopen(libdiscid.0.dylib, 6): image not found ".&lt;/p&gt;

&lt;p&gt;Now if you read the ticket you'll see that the issue is fixed but that a new version of &lt;a href="http://musicbrainz.org/doc/PicardTagger" &gt;Picard &lt;/a&gt;as not been released yet. Fixing the issue is easy.&lt;/p&gt;

&lt;p&gt;First you need to have the developer tools installed , which you can doanload from &lt;a href="http://connect.apple.com." &gt;connect.apple.com&lt;/a&gt;. Once they are installed do the following in a Terminal window "&lt;br /&gt;
&lt;code&gt;&lt;br /&gt;
curl -O http://users.musicbrainz.org/~matt/libdiscid-0.2.2.tar.gz&lt;br /&gt;
tar zxvf libdiscid-0.2.2.tar.gz&lt;br /&gt;
 cd libdiscid-0.2.2&lt;br /&gt;
./configure&lt;br /&gt;
make&lt;br /&gt;
sudo make install&lt;br /&gt;
&lt;/code&gt;&lt;br /&gt;
After the sudo command you will be prompted for a password - enter your password. &lt;br /&gt;
This will fix the issue, you'll need to start restart Picard - but picard was probably already closed when you started doing this.&lt;/p&gt;

&lt;p&gt;On a side note if you are running a PowerPc machine (G4/G5 based macs) the musicbrainzfolk are looking for &lt;a href="http://blog.musicbrainz.org/?p=335" &gt;feedback on the PPC build of picard&lt;/a&gt;.&lt;br /&gt;
&lt;/p&gt;</description>
    </item>
    <item>
      <pubDate>Sat, 20 Sep 2008 20:06:32 GMT</pubDate>
      <title>Nice skype trick</title>
      <link>http://www.advogato.org/person/softkid/diary.html?start=155</link>
      <guid>http://perso.hirlimann.net/~ludo/blog/archives/2008/09/nice-skype-trick.html</guid>
      <description>&lt;p&gt;I've just found out the best feature about skype's messaging system. In recent version of skype you can edit old messages using the mouse and a right click on the timestamp of the message - that's boring an slowwwwwwwwww.&lt;br /&gt;
Today I wanted to correct muself, as I tend to be unable to write anything properly in any chat window I'm involved in and did it irc style using :&lt;br /&gt;
&lt;code&gt;&lt;br /&gt;
s/lov/love/&lt;br /&gt;
&lt;code&gt;&lt;br /&gt;
after typing enter I did not see a new message as I expected , but my previous message was edited and the mistake corrected. That is so cool.&lt;/p&gt;</description>
    </item>
    <item>
      <pubDate>Thu, 18 Sep 2008 15:06:08 GMT</pubDate>
      <title>Feed moved to atom.xml</title>
      <link>http://www.advogato.org/person/softkid/diary.html?start=154</link>
      <guid>http://perso.hirlimann.net/~ludo/blog/archives/2008/09/feed-moved-to-atomxml.html</guid>
      <description>&lt;p&gt;My rss feed is not updated anymore. Please update your bookmarks to point to http://perso.hirlimann.net/~ludo/blog/atom.xml.&lt;/p&gt;</description>
    </item>
    <item>
      <pubDate>Wed, 10 Sep 2008 13:10:14 GMT</pubDate>
      <title>Mashing Pictures through metadata</title>
      <link>http://www.advogato.org/person/softkid/diary.html?start=153</link>
      <guid>http://perso.hirlimann.net/~ludo/blog/archives/2008/09/mashing-pictures-through-metadata.html</guid>
      <description>&lt;p&gt;As a flickr user I add &lt;a href="http://www.amazon.com/gp/redirect.html?ie=UTF8&amp;location=http%3A%2F%2Fwww.amazon.com%2FMetadata-Marcia-Lei-Zeng%2Fdp%2F1555706355%3Fie%3DUTF8%26s%3Dbooks%26qid%3D1221049665%26sr%3D8-1&amp;tag=softkidshomep-20&amp;linkCode=ur2&amp;camp=1789&amp;creative=9325" &gt;metadata&lt;/a&gt;&lt;img src="http://www.assoc-amazon.com/e/ir?t=softkidshomep-20&amp;amp;l=ur2&amp;amp;o=1" width="1" height="1" border="0" alt="" style="border:none !important; margin:0px !important;" /&gt; to my pictures as much as I can. There's three way to achieve that on flickr (well four when you include geotagging). Using a title, adding some description on what the picture is about, and &lt;a href="http://flickr.com/photos/lhirlimann/tags/" &gt;tagging&lt;/a&gt;. Tagging is the easiest way to describe a picture with a few simple word. You add keywords to you pictures when you tag them. Some of these tags are a bit specila as they may link your pictures with other websites or with data that does not reside on Flickr itself. Flickr uses special tags called machine tags - with these you can link to events that are posted on the upcoming service by yahoo or with concerts that are annonced on last.fm, &lt;a href="http://www.amazon.com/gp/redirect.html?ie=UTF8&amp;location=http%3A%2F%2Fwww.amazon.com%2FTagging-People-powered-Metadata-Social-Voices%2Fdp%2F0321529170%3Fie%3DUTF8%26s%3Dbooks%26qid%3D1221049665%26sr%3D8-9&amp;tag=softkidshomep-20&amp;linkCode=ur2&amp;camp=1789&amp;creative=9325" &gt;tagging&lt;/a&gt;&lt;img src="http://www.assoc-amazon.com/e/ir?t=softkidshomep-20&amp;amp;l=ur2&amp;amp;o=1" width="1" height="1" border="0" alt="" style="border:none !important; margin:0px !important;" /&gt; enhances the content and makes it findable. There is probably more use with other sites but I don't know it.&lt;/p&gt;

&lt;p&gt;&lt;br /&gt;
While I was on vacation in greenland and ended up in a concert in a bar in Ilulissat.I took a few pictures and really liked how the guys were performing. So I asked them if I could buy a CD to which the answer I was given a small piece of paper was an URL. Url Which I ended visiting and buying the their unique album which to my surprise dates from the beginning of the 90's.&lt;/p&gt;

&lt;p&gt;&lt;br /&gt;
I imported the CD into my music collection and started playing it - as I'm a last.fm user I had a look at the data that last.fm maintained on &lt;a href="http://www.last.fm/music/Unnuaq+Band" &gt;"Unnuaq Band"&lt;/a&gt;. It was quite empty so I made the first two versions of the wiki page added a scan of the group's logo that I took from the album. Actually I needed to add wiki information before the upload of the scan was taken into account by last.fm's software. Then I made sure that the group was connected with the Greenland group.&lt;/p&gt;

&lt;p&gt;After two weeks of playing the band's song I looked at my &lt;a href="http://www.last.fm/user/ludovic" &gt;stats&lt;/a&gt; on last.fm and was suprise not to see the album name in what I was playing. So I kindly &lt;a href="http://www.last.fm/forum/21713/_/453443" &gt;ask on the forums&lt;/a&gt; where the data came from, and the answer was musicbrainz. From time to time, last.fm synchs it's data from Musicbrainz. So I created an account and started adding data about the album. Using the taggers provided by last.fm did not do much because the music was unavailable on musicip. So I gave musicip a look and would have done sampling with Music IP if their software supported my os :&lt;br /&gt;
&lt;em&gt;Thank you for visiting musicip.com. A Mac version of the iTunes plug-in is&lt;br /&gt;
planned but not currently in development. If you sign up for our beta testing&lt;br /&gt;
program you can be one of the first to try out the new plug-in.&lt;/p&gt;

&lt;p&gt;http://www.musicip.com/beta-join.jsp&lt;/p&gt;

&lt;p&gt;Let me know if you have any further questions.&lt;/em&gt;&lt;/p&gt;

&lt;p&gt;So I can't provide &lt;a href="http://musicbrainz.org/show/artist/details.html?artistid=554427" &gt;all the data/metadata&lt;/a&gt;, and need to do that by hand, at the moment but will probably do.&lt;/p&gt;

&lt;p&gt;So I'm having fun these days figuring out how I will tag my picture when I'm going to upload it to flickr. Shall I use machine tags like &lt;a href="http://www.last.fm/group/Flickr/forum/19687/_/453458" &gt;lastfm:band=&lt;/a&gt; or should I create musixbranz:MBID=, shall I use both ? I would rather use the later, so all sites using musicbrainz's data would benefit from my tagging and could, if wanted show my flickr pictures on their website.&lt;br /&gt;
Now I don't mind adding more than one tag, but was wondering what the best approach would be. Ideas, comments welcome.&lt;/p&gt;</description>
    </item>
    <item>
      <pubDate>Fri, 5 Sep 2008 08:06:58 GMT</pubDate>
      <title>V8</title>
      <link>http://www.advogato.org/person/softkid/diary.html?start=152</link>
      <guid>http://perso.hirlimann.net/~ludo/blog/archives/2008/09/v8.html</guid>
      <description>&lt;p&gt;Continuing &lt;a href="http://dev.chromium.org/developers/how-tos/getting-around-the-chrome-source-code" &gt;looking at the code&lt;/a&gt; which Chrome is built upon, I had a look at &lt;a href="http://code.google.com/apis/v8/intro.html" &gt;V8&lt;/a&gt; the javascript engine that compiles js code to native code.&lt;br /&gt;
In particular  at cpu-ia32.{h|cc} and cpu-arm.{h|cc}. So the good news here is that v8 is available for &lt;a href="http://en.wikipedia.org/wiki/ARM_architecture" &gt;ARM&lt;/a&gt; on top of x86. It seems that adding PowerPC support is doable. I was looking for a cool project to work on during my free time as I have not been involved in computer stuff outside my work for quite a while now. Just unsure I have the time required to do that. I'm quite puzzled by the lack of ia-64 support too.&lt;/p&gt;</description>
    </item>
    <item>
      <pubDate>Thu, 4 Sep 2008 20:07:02 GMT</pubDate>
      <title>Chrome and 404's</title>
      <link>http://www.advogato.org/person/softkid/diary.html?start=151</link>
      <guid>http://perso.hirlimann.net/~ludo/blog/archives/2008/09/chrome-and-404s.html</guid>
      <description>&lt;p&gt;So yesterday I bitched about the fact that Chrome, replaced &lt;a href="http://en.wikipedia.org/wiki/HTTP_404" &gt;404 pages&lt;/a&gt; with a input box that would let user perform google searches. I did not like the idea , that 404 where replaced by a search box.&lt;br /&gt;
Today I downloaded the source of Chrome and had a look inside and found this interesting comment :&lt;br /&gt;
&lt;code&gt;&lt;br /&gt;
  // When the frame request first 404's, chrome may replace it with the alternate&lt;br /&gt;
  // 404 page's contents. It does this using substitute data in the document &lt;br /&gt;
  // loader, so the original response and url of the request can be preserved.&lt;br /&gt;
  // We need to avoid replacing the current page, if it has already been&lt;br /&gt;
  // replaced (otherwise could loop on setting alt-404 page!)&lt;br /&gt;
  bool is_substitute_data = loader-&gt;substituteData().isValid();&lt;/p&gt;

&lt;p&gt;  // If it's a 404 page, we wait until we get 512 bytes of data before trying&lt;br /&gt;
  // to load the document.  This allows us to put up an alternate 404 page if&lt;br /&gt;
  // there's short text.&lt;br /&gt;
&lt;code&gt;&lt;br /&gt;
in &lt;a href="http://src.chromium.org/viewvc/chrome/trunk/src/webkit/glue/webframeloaderclient_impl.cc?view=markup&amp;pathrev=1359" &gt;http://src.chromium.org/viewvc/chrome/trunk/src/webkit/glue/webframeloaderclient_impl.cc?view=markup&amp;pathrev=1359&lt;/a&gt; around line 252. &lt;/p&gt;

&lt;p&gt;This simply means that if you want to keep your 404 pages they need to be 513 byte in length. If they are 513 Chrome considers that the page contains interesting bits of information for the user and does not replace it with a google crafted page.&lt;/p&gt;</description>
    </item>
  </channel>
</rss>
