<?xml version="1.0" encoding="utf-8"?><?xml-stylesheet title="XSL formatting" type="text/xsl" href="http://taz.placenet.org/index.php/feed/rss2/xslt" ?><rss version="2.0"
  xmlns:dc="http://purl.org/dc/elements/1.1/"
  xmlns:content="http://purl.org/rss/1.0/modules/content/"
  xmlns:atom="http://www.w3.org/2005/Atom">
<channel>
  <title>Harvard Business School of Echec  - Commentaires</title>
  <link>http://taz.placenet.org/index.php/</link>
  <atom:link href="http://taz.placenet.org/index.php/feed/rss2/comments" rel="self" type="application/rss+xml"/>
  <description></description>
  <language>fr</language>
  <pubDate>Thu, 06 Aug 2009 07:34:25 +0200</pubDate>
  <copyright></copyright>
  <docs>http://blogs.law.harvard.edu/tech/rss</docs>
  <generator>Dotclear</generator>
  
    
    
    <item>
    <title>Les plans les mieux conçus des souris et des hommes souvent ne se réalisent pas - Luc</title>
    <link>http://taz.placenet.org/index.php/post/2009/06/26/souris#c196</link>
    <guid isPermaLink="false">urn:md5:4c92af9371acf8525727496a217db39b</guid>
    <pubDate>Wed, 01 Jul 2009 20:02:46 +0200</pubDate>
    <dc:creator>Luc</dc:creator>
    
    <description>&lt;p&gt;Luc may the force be with you !&lt;/p&gt;


&lt;p&gt;bonnes vacances qd même.&lt;/p&gt;</description>
  </item>
      
    
    <item>
    <title>Should I trust Ted Ts'o ? - James</title>
    <link>http://taz.placenet.org/index.php/post/2009/04/03/I-want-to-believe#c195</link>
    <guid isPermaLink="false">urn:md5:897fed79347ed6e01b5749d7282671ae</guid>
    <pubDate>Wed, 08 Apr 2009 04:24:39 +0200</pubDate>
    <dc:creator>James</dc:creator>
    
    <description>&lt;p&gt;Antonomasia: Rename breaking hard links is sometimes desired - eg back in
the day people would copy kernel trees as hardlinks which makes diff very fast,
then when they edited a file the hardlink would break, so diff still
worked.&lt;/p&gt;</description>
  </item>
      
    
    <item>
    <title>Should I trust Ted Ts'o ? - Benoît Dejean</title>
    <link>http://taz.placenet.org/index.php/post/2009/04/03/I-want-to-believe#c194</link>
    <guid isPermaLink="false">urn:md5:fde1aa5b7dde1cad26443f1d868b83c4</guid>
    <pubDate>Sat, 04 Apr 2009 18:48:02 +0200</pubDate>
    <dc:creator>Benoît Dejean</dc:creator>
    
    <description>&lt;p&gt;tjs &amp;gt; Interesting link &lt;a href=&quot;http://git.kernel.org/?p=linux/kernel/git/tytso/ext4.git;a=commitdiff;h=dbc85aa9f11d8c13c15527d43a3def8d7beffdc8&quot; title=&quot;http://git.kernel.org/?p=linux/kernel/git/tytso/ext4.git;a=commitdiff;h=dbc85aa9f11d8c13c15527d43a3def8d7beffdc8&quot; rel=&quot;nofollow&quot;&gt;
http://git.kernel.org/?p=linux/kern...&lt;/a&gt; but that doesn't fix
emacs/vim/OOo&lt;/p&gt;</description>
  </item>
      
    
    <item>
    <title>Should I trust Ted Ts'o ? - Benoît Dejean</title>
    <link>http://taz.placenet.org/index.php/post/2009/04/03/I-want-to-believe#c192</link>
    <guid isPermaLink="false">urn:md5:fef9a4ab578766b9d0075e0f30cf5d0c</guid>
    <pubDate>Sat, 04 Apr 2009 18:29:38 +0200</pubDate>
    <dc:creator>Benoît Dejean</dc:creator>
    
    <description>&lt;p&gt;Again, /etc/passwd was a (silly) example ...&lt;/p&gt;</description>
  </item>
      
    
    <item>
    <title>Should I trust Ted Ts'o ? - tjs</title>
    <link>http://taz.placenet.org/index.php/post/2009/04/03/I-want-to-believe#c230</link>
    <guid isPermaLink="false">urn:md5:013376f3d3a1e944f112b681b2f32cb3</guid>
    <pubDate>Sat, 04 Apr 2009 10:52:38 +0200</pubDate>
    <dc:creator>tjs</dc:creator>
    
    <description>&lt;p&gt;&amp;quot;Suddenly, i don't feel like editing /etc/passwd with these text editors
(but i haven't find a replacement yet). If the server was to crash, it might
not reboot at all. Brrrr.&amp;quot;&lt;/p&gt;
&lt;p&gt;You cannot edit /etc/passwd safely even with passwd since PAM itself does
not call fsync, see &lt;a href=&quot;https://www.redhat.com/archives/pam-list/2009-March/msg00017.html&quot; title=&quot;https://www.redhat.com/archives/pam-list/2009-March/msg00017.html&quot; rel=&quot;nofollow&quot;&gt;https://www.redhat.com/archives/pam...&lt;/a&gt;&lt;/p&gt;</description>
  </item>
      
    
    <item>
    <title>Should I trust Ted Ts'o ? - Merc</title>
    <link>http://taz.placenet.org/index.php/post/2009/04/03/I-want-to-believe#c185</link>
    <guid isPermaLink="false">urn:md5:24b20ff81d03d6b74b082164da1d7fb1</guid>
    <pubDate>Sat, 04 Apr 2009 04:08:59 +0200</pubDate>
    <dc:creator>Merc</dc:creator>
    
    <description>&lt;p&gt;&amp;quot;Suddenly, i don't feel like editing /etc/passwd with these text editors
(but i haven't find a replacement yet). If the server was to crash, it might
not reboot at all. Brrrr.&amp;quot;&lt;/p&gt;
&lt;p&gt;You should never directly edit /etc/passwd or /etc/group. That's why 'vipw'
and 'vigr' exist. They have 'vi' in the name but set the EDITOR env. var. to
whatever you want.&lt;/p&gt;</description>
  </item>
      
    
    <item>
    <title>Should I trust Ted Ts'o ? - Antonomasia</title>
    <link>http://taz.placenet.org/index.php/post/2009/04/03/I-want-to-believe#c184</link>
    <guid isPermaLink="false">urn:md5:70c0d05708d62ec38ef62cb3d83bddbf</guid>
    <pubDate>Sat, 04 Apr 2009 01:33:35 +0200</pubDate>
    <dc:creator>Antonomasia</dc:creator>
    
    <description>&lt;p&gt;If the file had hard links before it should have the same hard links
afterward. That's why I don't like the rename() approach for normal editing -
but I do use it for the likes of passwd, shadow, sudoers.&lt;/p&gt;</description>
  </item>
      
    
    <item>
    <title>Should I trust Ted Ts'o ? - Benoît Dejean</title>
    <link>http://taz.placenet.org/index.php/post/2009/04/03/I-want-to-believe#c229</link>
    <guid isPermaLink="false">urn:md5:58d01642227cba84f70bf55c60e0d72b</guid>
    <pubDate>Fri, 03 Apr 2009 21:54:51 +0200</pubDate>
    <dc:creator>Benoît Dejean</dc:creator>
    
    <description>&lt;p&gt;Hello Donny, here journaling does not help (unless you have a battery-backed
memory, then you can replay 'lost' I/O on reboot). Before hiting the disk, data
stays in memory, and if you crash while writing the file to disk, you're write
is incomplete (data in memory vanishes). That's why the way
vim/emacs/openoffice behave is not safe: they make a backup of your data and
then start to overwrite your file. If your computer crashes during this
overwrite, then your file is at best truncated. Of course you can pull the
backup, but that's quite annoying ... and if the backup file has not been
synced to disk before the crash, you lost the two of them. Ted Ts'o explains
how to write robust application by writing a new file, syncing it to the disk
and then renaming it (which atomic). This way, on a crash, you either get the
old or the new file, anyway a valid file.&lt;/p&gt;
&lt;p&gt;Russ &amp;gt; yep, passwd/useradd/etc WIN. They match exactly the pattern Ted
Ts'o describes.&lt;/p&gt;</description>
  </item>
      
    
    <item>
    <title>Should I trust Ted Ts'o ? - Russ</title>
    <link>http://taz.placenet.org/index.php/post/2009/04/03/I-want-to-believe#c183</link>
    <guid isPermaLink="false">urn:md5:d92931b07c1bf7463d1803192c7bb704</guid>
    <pubDate>Fri, 03 Apr 2009 21:28:42 +0200</pubDate>
    <dc:creator>Russ</dc:creator>
    
    <description>&lt;p&gt;gpasswd WIN?&lt;/p&gt;
&lt;p&gt;open(&amp;quot;/etc/group+&amp;quot;, O_WRONLY|O_CREAT|O_TRUNC, 0666) = 5&lt;br /&gt;
umask(077) = 0777&lt;br /&gt;
fchown(5, 0, 0) = 0&lt;br /&gt;
fchmod(5, 0644) = 0&lt;br /&gt;
fstat(5, {st_mode=S_IFREG|0644, st_size=0, ...}) = 0&lt;br /&gt;
write(5, &amp;quot;root:x:0:
daemon:x:1:
bin:x:2:
sy&amp;quot;..., 1341) =1341&lt;br /&gt;
fsync(5) = 0&lt;br /&gt;
close(5) = 0&lt;br /&gt;
lstat(&amp;quot;/etc/group&amp;quot;, {st_mode=S_IFREG|0644,st_size=1335, ...}) = 0&lt;br /&gt;
rename(&amp;quot;/etc/group+&amp;quot;, &amp;quot;/etc/group&amp;quot;) = 0&lt;/p&gt;</description>
  </item>
      
    
    <item>
    <title>Should I trust Ted Ts'o ? - Donny Viszneki</title>
    <link>http://taz.placenet.org/index.php/post/2009/04/03/I-want-to-believe#c182</link>
    <guid isPermaLink="false">urn:md5:396606b6a771181a3e3c4f08354e3c3a</guid>
    <pubDate>Fri, 03 Apr 2009 20:41:19 +0200</pubDate>
    <dc:creator>Donny Viszneki</dc:creator>
    
    <description>&lt;p&gt;You said Vim can &amp;quot;corrupt&amp;quot; the old version of the file because it doesn't
fsync() the rename: is this true? Doesn't safely renaming files fall under the
jurisdiction of the filesystem? Doesn't journaling in ext3/jfs/etc. take care
of this problem?&lt;/p&gt;</description>
  </item>
      
    
    <item>
    <title>Should I trust Ted Ts'o ? - tm</title>
    <link>http://taz.placenet.org/index.php/post/2009/04/03/I-want-to-believe#c181</link>
    <guid isPermaLink="false">urn:md5:b7a76ff847443bae2f51bf1b485da4c0</guid>
    <pubDate>Fri, 03 Apr 2009 20:24:34 +0200</pubDate>
    <dc:creator>tm</dc:creator>
    
    <description>&lt;p&gt;nice, constructive post, thanks for sharing, truly in spirit of proper GNOME
developer, aye?&lt;/p&gt;
&lt;p&gt;gah.&lt;/p&gt;</description>
  </item>
      
    
    <item>
    <title>Should I trust Ted Ts'o ? - Benoît Dejean</title>
    <link>http://taz.placenet.org/index.php/post/2009/04/03/I-want-to-believe#c180</link>
    <guid isPermaLink="false">urn:md5:3e5b205492ca6974d3e4d99cbb372d63</guid>
    <pubDate>Fri, 03 Apr 2009 20:15:04 +0200</pubDate>
    <dc:creator>Benoît Dejean</dc:creator>
    
    <description>&lt;p&gt;oliv &amp;gt; This emacs option would make it as bad as vim. I don't care about
the inode, i care about the ACL. Gedit solved this problem using libacl, i
guess vim too.&lt;/p&gt;</description>
  </item>
      
    
    <item>
    <title>Should I trust Ted Ts'o ? - oliv</title>
    <link>http://taz.placenet.org/index.php/post/2009/04/03/I-want-to-believe#c179</link>
    <guid isPermaLink="false">urn:md5:45a1ec6cc2390cf420411ac41e59e8a2</guid>
    <pubDate>Fri, 03 Apr 2009 19:47:53 +0200</pubDate>
    <dc:creator>oliv</dc:creator>
    
    <description>&lt;p&gt;Emacs backup behaviour can be modified with a couple of settings.&lt;/p&gt;
&lt;p&gt;There's a backup-by-copying variable, setting it to t should avoid the
save-to-new-inode-and-lose-ACL problem. It's off by default.&lt;/p&gt;</description>
  </item>
      
    
    <item>
    <title>Should I trust Ted Ts'o ? - Benoît Dejean</title>
    <link>http://taz.placenet.org/index.php/post/2009/04/03/I-want-to-believe#c228</link>
    <guid isPermaLink="false">urn:md5:d35c91eeee93f4ebfd41a55a7da9f553</guid>
    <pubDate>Fri, 03 Apr 2009 18:52:55 +0200</pubDate>
    <dc:creator>Benoît Dejean</dc:creator>
    
    <description>&lt;p&gt;schmichael &amp;gt; Don't forget about network filesystems. There are much more
likely to fail. /etc/passwd was an example, but that kind of thing could happen
with any important text file in /etc. (/etc/init.d/sshd ? ;). And we're only
talking about text editor :)&lt;br /&gt;
But you're right: carefully written tools such as useradd are safe (Ted T'so's
pattern)&lt;/p&gt;
&lt;p&gt;Paolo &amp;gt; I knew it :) I'll check this as soon as possible.&lt;/p&gt;</description>
  </item>
      
    
    <item>
    <title>Should I trust Ted Ts'o ? - schmichael</title>
    <link>http://taz.placenet.org/index.php/post/2009/04/03/I-want-to-believe#c178</link>
    <guid isPermaLink="false">urn:md5:c33692dbdf52e24ddbc13ecae43965b1</guid>
    <pubDate>Fri, 03 Apr 2009 18:33:32 +0200</pubDate>
    <dc:creator>schmichael</dc:creator>
    
    <description>&lt;p&gt;&amp;quot;Suddenly, i don't feel like editing /etc/passwd with these text editors
(but i haven't find a replacement yet). If the server was to crash, it might
not reboot at all. Brrrr.&amp;quot;&lt;/p&gt;
&lt;p&gt;Aren't the odds of this happening insanely small? Do the servers you work on
not have a UPS and encounter Kernel oops on a regular basis?&lt;/p&gt;
&lt;p&gt;If so, perhaps you have more to worry about than the saving behavior of your
text editor.&lt;/p&gt;
&lt;p&gt;Honestly the only time I'm worried about all these fsync/saving issues I see
floating around are on my laptop which has some flaky hardware drivers and
therefore stands a realistic chance of oopsing between fsyncs on me. Even then
I don't think I've ever encountered data loss yet.&lt;/p&gt;
&lt;p&gt;And wait a second... aren't you supposed to be using tools like usermod,
useradd, etc to alter /etc/passwd?!&lt;/p&gt;
&lt;p&gt;I think this nicely highlights how user error is still the greatest threat
to system stability. ;-)&lt;/p&gt;</description>
  </item>
      
    
    <item>
    <title>Should I trust Ted Ts'o ? - pbor</title>
    <link>http://taz.placenet.org/index.php/post/2009/04/03/I-want-to-believe#c177</link>
    <guid isPermaLink="false">urn:md5:811bd0708d4016f5a4c13c0eba6ff32f</guid>
    <pubDate>Fri, 03 Apr 2009 17:41:34 +0200</pubDate>
    <dc:creator>pbor</dc:creator>
    
    <description>&lt;p&gt;you can use gedit 2.26 ;)&lt;/p&gt;</description>
  </item>
      
    
    <item>
    <title>Le LaTex vieillit mal - jeff</title>
    <link>http://taz.placenet.org/index.php/post/2008/10/15/Le-LaTex-vieillit-mal#c176</link>
    <guid isPermaLink="false">urn:md5:518b4d0175ded50c1ca847770a1136ad</guid>
    <pubDate>Fri, 17 Oct 2008 15:02:35 +0200</pubDate>
    <dc:creator>jeff</dc:creator>
    
    <description>&lt;p&gt;j'ai déjà utilisé latex quelques fois, et franchement, j'hésite toujours entre ça et OpenOffice Writer (avec des backups de versions de fichiers dans un système de versionnement de code), par exemple. Ne riez pas, c'est une question sérieuse. Je ne confierais jamais un long document à Word, mais... franchement, 100 pages dans OOo en .odt, ça doit tenir non? J'ai une tante qui a écrit un livre (non scientifique) de 250 pages en .doc avec word...&lt;/p&gt;


&lt;p&gt;Chaque fois que j'ai touché à latex, j'ai fini par hurler et revenir à OOo. Connerie qui a besoin d'un minimum d'une dizaine de modules obscurs à inclure pour fonctionner corretement, supporter les urls, les guillemets, l'unicode (what you write is what you mean? certainement pas si je dois échapper tous mes accents ¬_¬), les tas de façons différentes de citer, la stupidité de ses tableaux (ça dépasse la largeur de la page? shit, out of luck!), et ça c'est avant même de se battre avec les normes de mise en page et de citations de l'APA (je suis en psychologie). J'ai passé des jours à tenter même de faire fonctionner apacite et apa.cls, fail.&lt;/p&gt;


&lt;p&gt;Disclaimer: je veux bien être convaincu que latex est la Voie. Je n'écris pas de formules mathématiques (je suis en psychologie après tout), mon c.v. est en latex, mais franchement, je trouve que ça fait un peu anachronique quand d'un autre côté, il y a la plaine verdoyante de &amp;quot;what you type is what you get et not what you will have to fix at the compilation time&amp;quot; d'OpenOffice, combiné à zotero.org pour la gestion bibliographique et l'insertion des citations.&lt;/p&gt;</description>
  </item>
      
    
    <item>
    <title>Le LaTex vieillit mal - Thom</title>
    <link>http://taz.placenet.org/index.php/post/2008/10/15/Le-LaTex-vieillit-mal#c175</link>
    <guid isPermaLink="false">urn:md5:483d531a485e4ba354cda08948797f2e</guid>
    <pubDate>Thu, 16 Oct 2008 15:46:46 +0200</pubDate>
    <dc:creator>Thom</dc:creator>
    
    <description>&lt;p&gt;Oui enfin on pourrait imaginer un langage intermédiaire comme tbook ou encore mieux celui utilisé par les wikis ce qui simplifie considérablement l'écriture des sources.&lt;/p&gt;


&lt;p&gt;L'avantage d'un subset xml comme tbook c'est que tu n'as pas besoin de rien faire pour que ton document soit lisible dans n'importe quel navigateur (puis imprimable en pdf) mais tu es obligé de te taper les &amp;lt;p&amp;gt; et autres petits bonheurs.&lt;/p&gt;


&lt;p&gt;L'avantage du langage wiki c'est la simplicité à taper, par contre il faut processer le code en quelque chose de lisible par un navigateur web.&lt;/p&gt;


&lt;p&gt;Sinon, séparer le rendu du contenu c'est exactement le rôle de css.&lt;/p&gt;</description>
  </item>
      
    
    <item>
    <title>Le LaTex vieillit mal - Loïc d'Anterroches</title>
    <link>http://taz.placenet.org/index.php/post/2008/10/15/Le-LaTex-vieillit-mal#c227</link>
    <guid isPermaLink="false">urn:md5:456012cfd5ef529b3d7848080ea86198</guid>
    <pubDate>Thu, 16 Oct 2008 13:43:24 +0200</pubDate>
    <dc:creator>Loïc d'Anterroches</dc:creator>
    
    <description>&lt;p&gt;Le côté très pratique de LaTeX en comparaison d'un langage basé sur XML quel qu'il soit, c'est que LaTeX permet de se concentrer sur le contenu et non la manière de l'écrire. On peut utiliser le plus simple des éditeurs de texte pour faire cela, on n'a aucune obligation d'utilisateur un éditeur graphique. Si j'avais du taper tous les &amp;lt;p&amp;gt;&amp;lt;/p&amp;gt; en écrivant les 350 pages de ma thèse et bien je serais mort avant de l'avoir fini. Je n'imagine même pas la situation si j'avais du entrer les équations dans une autre syntaxe que LaTeX...&lt;/p&gt;


&lt;p&gt;Pour écrire un roman, on peut utiliser un éditeur graphique, mais pour du contenu scientifique (du moins, dès qu'un peu de maths pointe son nez), LaTeX reste roi.&lt;/p&gt;</description>
  </item>
      
    
    <item>
    <title>Le LaTex vieillit mal - Thom</title>
    <link>http://taz.placenet.org/index.php/post/2008/10/15/Le-LaTex-vieillit-mal#c174</link>
    <guid isPermaLink="false">urn:md5:a8bfde6fc50098152faa0fffbf438113</guid>
    <pubDate>Thu, 16 Oct 2008 10:46:52 +0200</pubDate>
    <dc:creator>Thom</dc:creator>
    
    <description>&lt;p&gt;&lt;a href=&quot;http://www.alistapart.com/articles/boom&quot; title=&quot;http://www.alistapart.com/articles/boom&quot; rel=&quot;nofollow&quot;&gt;http://www.alistapart.com/articles/...&lt;/a&gt;&lt;/p&gt;</description>
  </item>
      
</channel>
</rss>