<?xml version="1.0" encoding="utf-8"?>
<feed xmlns="http://www.w3.org/2005/Atom">
   <title>Manni&apos;s blog</title>
   <link rel="alternate" type="text/html" href="http://lxxi.org/" />
   <link rel="self" type="application/atom+xml" href="http://lxxi.org/atom.xml" />
   <id>tag:lxxi.org,2008://1</id>
   <updated>2008-04-15T10:06:41Z</updated>
   
   <generator uri="http://www.sixapart.com/movabletype/">Movable Type 3.33</generator>

<entry>
   <title>Expelled</title>
   <link rel="alternate" type="text/html" href="http://lxxi.org/2008/04/expelled.html" />
   <id>tag:lxxi.org,2008://1.16</id>
   
   <published>2008-04-15T10:05:58Z</published>
   <updated>2008-04-15T10:06:41Z</updated>
   
   <summary>Not just for the search engines: Expelled...</summary>
   <author>
      <name></name>
      
   </author>
   
   
   <content type="html" xml:lang="en" xml:base="http://lxxi.org/">
      <![CDATA[Not just for the search engines: <a href="http://expelledexposed.com/"><i>Expelled</i></a>]]>
      
   </content>
</entry>
<entry>
   <title>Five things I hate about Eclipse</title>
   <link rel="alternate" type="text/html" href="http://lxxi.org/2008/03/five_things_i_hate_about_eclipse.html" />
   <id>tag:lxxi.org,2008://1.15</id>
   
   <published>2008-03-01T12:59:42Z</published>
   <updated>2008-03-01T13:15:24Z</updated>
   
   <summary>Without further ado, here they are: 1. The acronyms. There&apos;s PDE, SDK, WST, RCP, DTP, BIRT, EMF, MDT, GMT, M2M, M2T, MDT, PTP. And the list goes on and on and on. Are these people nuts? 2. The missing plugin...</summary>
   <author>
      <name></name>
      
   </author>
         <category term="en" scheme="http://www.sixapart.com/ns/types#category" />
         <category term="software" scheme="http://www.sixapart.com/ns/types#category" />
   
   
   <content type="html" xml:lang="en" xml:base="http://lxxi.org/">
      <![CDATA[Without further ado, here they are:

1. The acronyms.

There's PDE, SDK, WST, RCP, DTP, BIRT, EMF, MDT, GMT, M2M, M2T, MDT, PTP. And the list goes on and on and on. Are these people nuts?

2. The missing plugin repository.

There is no single location where it all comes together. I'm not even asking for a central download server, even if IBM could afford that. All I want is a single instance that knows about all registered plugins. So that I could install a new plugin and if that plugin had any requirements that aren't already installed, the plugin manager could point me to the correct location.

3. The business language.

Yes, I know, Eclipse wants to be big in the corporate world. But does that mean that project and plugin descriptions have to be written in managerese? "<quote>Eclipse is an open source community whose projects are focused on building an open development platform comprised of extensible frameworks, tools and runtimes for building, deploying and managing software across the lifecycle.</quote>" Worse things can be found on various eclipse web pages, but the buzzword bingo starts right there on the front page.

4. The bug tracker.

It's awfully slow, it's awfully hard to use, and "<quote>Zarro Boogs found.</quote>" is <em>not</em> at all funny.

5. The menues.

No, I don't expect to find the "Software Updates" feature in the Help menu. No, I don't expect to find the application preferences in the Window menu. ]]>
      
   </content>
</entry>
<entry>
   <title>POPFile 1.0 is out!</title>
   <link rel="alternate" type="text/html" href="http://lxxi.org/2007/12/popfile_10_is_out.html" />
   <id>tag:lxxi.org,2007://1.14</id>
   
   <published>2007-12-21T09:49:49Z</published>
   <updated>2007-12-21T09:54:40Z</updated>
   
   <summary>John has just announced the release of POPFile 1.0. It was about time for a version 1.0. POPFile is very stable, does a great job, and has some pretty nice, well kept source code. There haven&apos;t been too many features...</summary>
   <author>
      <name></name>
      
   </author>
         <category term="en" scheme="http://www.sixapart.com/ns/types#category" />
         <category term="software" scheme="http://www.sixapart.com/ns/types#category" />
         <category term="spam" scheme="http://www.sixapart.com/ns/types#category" />
   
   
   <content type="html" xml:lang="en" xml:base="http://lxxi.org/">
      <![CDATA[<a href="http://www.jgc.org/">John</a> has just announced the release of POPFile 1.0.

It was about time for a version 1.0. POPFile is very stable, does a great job, and has some pretty nice, well kept source code. There haven't been too many features added since 0.22.5, but we all thought that POPFile was old enough to warrant a version 1.0, ending the perpetual "beta" phase.

You can find the <a href="http://sourceforge.net/forum/forum.php?forum_id=767078">changelog</a> on the POPFile forums and you can find a <a href="http://getpopfile.org/wiki/download">download link</a> on our wiki.]]>
      
   </content>
</entry>
<entry>
   <title>POPFile 1.0 veröffentlicht</title>
   <link rel="alternate" type="text/html" href="http://lxxi.org/2007/12/popfile_10_veroffentlicht.html" />
   <id>tag:lxxi.org,2007://1.13</id>
   
   <published>2007-12-21T09:44:13Z</published>
   <updated>2007-12-21T09:49:23Z</updated>
   
   <summary>Gerade hat John die Version 1.0 von POPFile freigegeben. Auch wenn der Sprung der Versionsnummer von 0..22.5 auf 1.0 recht mächtig erscheint, gibt es kaum neue Features. Dafür sind wir uns jetzt der Ausgereiftheit und Stabilität von POPFile derart sicher,...</summary>
   <author>
      <name></name>
      
   </author>
         <category term="de" scheme="http://www.sixapart.com/ns/types#category" />
         <category term="software" scheme="http://www.sixapart.com/ns/types#category" />
         <category term="spam" scheme="http://www.sixapart.com/ns/types#category" />
   
   
   <content type="html" xml:lang="en" xml:base="http://lxxi.org/">
      <![CDATA[Gerade hat <a href="http://www.jgc.org/">John</a> die Version 1.0 von POPFile freigegeben.

Auch wenn der Sprung der Versionsnummer von 0..22.5 auf 1.0 recht mächtig erscheint, gibt es kaum neue Features. Dafür sind wir uns jetzt der Ausgereiftheit und Stabilität von POPFile derart sicher, dass wir nach vielen Jahren "Beta"-Status nun beschlossen haben, dass es Zeit für eine 1.0 ist.

<ul>
<li><a href="http://sourceforge.net/forum/forum.php?forum_id=767078">Changelog</a></li>
<li><a href="http://popfile.sourceforge.net/wiki/download">Download</a></li>
</ul>
]]>
      
   </content>
</entry>
<entry>
   <title>Warum schreibt das denn nur keiner?</title>
   <link rel="alternate" type="text/html" href="http://lxxi.org/2007/11/warum_schreibt_das_denn_nur_keiner.html" />
   <id>tag:lxxi.org,2007://1.12</id>
   
   <published>2007-11-16T14:58:24Z</published>
   <updated>2007-11-16T15:19:02Z</updated>
   
   <summary>David Lynch hat also den Teufelsberg gekauft. Und zusammen mit einem nicht mehr ganz auf dem Boden der Tatsachen stehenden Menschen, der sich Guru nennt und sonst wohl Raja Schiffgens heißt, will er darauf eine Universität errichten. Nicht irgendeine Universität...</summary>
   <author>
      <name></name>
      
   </author>
         <category term="de" scheme="http://www.sixapart.com/ns/types#category" />
         <category term="ouch" scheme="http://www.sixapart.com/ns/types#category" />
   
   
   <content type="html" xml:lang="en" xml:base="http://lxxi.org/">
      <![CDATA[<a href="http://imdb.com/name/nm0000186/">David Lynch</a> hat also den <a href="http://de.wikipedia.org/wiki/Teufelsberg">Teufelsberg</a> gekauft. Und zusammen mit einem nicht mehr ganz auf dem Boden der Tatsachen stehenden Menschen, der sich Guru nennt und sonst wohl Raja Schiffgens heißt, will er darauf eine Universität errichten. Nicht irgendeine Universität - nein! - die "Universität unbesiegbares Deutschland".

Berlin will aber nicht und der Baustadtrat und der Sektenbeauftragte und blah, blah. 

Aber warum in drei Teufels Namen (haha!), schreibt denn nur keiner, was z.B. auf der oben verlinkten Wikipedia-Seite zu lesen steht? Dass der Teufelsberg nämlich aufgeschüttet wurde auf den Ruinen der "Wehrtechnischen Fakultät", die da mal stand? Dass das damit die zweite "Universität unbesiegbares Deutschland" am gleichen Standort (nur ein paar Meter weiter oben) wäre? Warum denn nur nicht?

]]>
      
   </content>
</entry>
<entry>
   <title>Cyrus + SpamAssassin + POPFile = Happiness</title>
   <link rel="alternate" type="text/html" href="http://lxxi.org/2007/08/cyrus_spamassassin_popfile_happiness.html" />
   <id>tag:lxxi.org,2007://1.11</id>
   
   <published>2007-08-10T14:17:52Z</published>
   <updated>2007-08-10T14:54:01Z</updated>
   
   <summary>Sorting mail with the POPFile IMAP module is very easy, training POPFile via IMAP is also easy. You might sometimes experience slight problems though, because your mail client will see messages in your INBOX before POPFile sees them. If these...</summary>
   <author>
      <name></name>
      
   </author>
         <category term="en" scheme="http://www.sixapart.com/ns/types#category" />
         <category term="software" scheme="http://www.sixapart.com/ns/types#category" />
         <category term="spam" scheme="http://www.sixapart.com/ns/types#category" />
   
   
   <content type="html" xml:lang="en" xml:base="http://lxxi.org/">
      <![CDATA[Sorting mail with the <a href="http://getpopfile.org/wiki/experimentalmodules:imap?s=imap">POPFile IMAP module</a> is very easy, training POPFile via IMAP is also easy. You might sometimes experience slight problems though, because your mail client will see messages in your INBOX before POPFile sees them. If these are spam, e.g., you get a new mail alert only to see POPFile move such messages to your spam folder as soon as you have a look.

But who says that new mail must be delivered to your INBOX? If your IMAP server is cyrus (or any other IMAP server that supports sieve scripts), you can change that and have your mail delivered to another folder. You then have POPFile monitor this other folder and move good messages to your INBOX while it sorts spam to your spam folder and mailing list emails to your mailing lists folder.

Here's how to do it:

1. Edit the sieve rules of your server. There are several ways to do that: from the command line using sieveshell and from the web using tools like the <a href="http://www.squirrelmail.org/">SquirrelMail</a> <a href="http://email.uoa.gr/projects/squirrelmail/avelsieve.php">avelsieve plugin</a> or <a href="http://sourceforge.net/projects/websieve">websieve</a>.

What you need to get started is one single rule that moves all incoming mail to a folder of your choice that is not the INBOX. For example, you could create a folder named 'Incoming' and have sieve deliver all messages to that folder. Here's the corresponding sieve script:

<pre>
<em>require ["fileinto"];
fileinto "Incoming"; stop;</em>
</pre>

That's pretty easy. As you will have guessed, the fileinto statement files all messages into the Incoming folder. The stop statements simply stops further rule execution.

2. Now setup POPFile to watch the Incoming folder instead of the INBOX. Simply load POPFile's Configuration tab and make the "Watched folder no1" the Incoming folder. (If you have just created that folder, POPFile might not have noticed it yet; simply click the 'Refresh folders now!' button and the folder will appear in the folders drop-down). 

3. Set the output folder of one of your buckets to 'INBOX'. 

4. Hit apply and enjoy.

Your mail client will never be distracted again by messages that it isn't supposed to see anyway.

Where does <a href="http://spamassassin.apache.org/">SpamAssassin</a> enter the picture? Well, if you server is also running SpamAssassin you can use sieve scrpting to get rid of spam immediately. You won't even have to bother POPFile with that crap. To have everything that SpamAssassin thinks is spam to a folder named 'spam', simply add a line to the above script. The result should look like this:

<pre>
require ["fileinto"];
if header :is "X-Spam-Flag" "YES" { fileinto "spam"; stop;}
fileinto "Incoming"; stop;
</pre>

Easy. 

]]>
      
   </content>
</entry>
<entry>
   <title>spamness</title>
   <link rel="alternate" type="text/html" href="http://lxxi.org/2007/05/spamness_1.html" />
   <id>tag:lxxi.org,2007://1.10</id>
   
   <published>2007-05-15T13:47:50Z</published>
   <updated>2007-05-16T13:04:32Z</updated>
   
   <summary>One of the new features in Thunderbird 2.0 is that extensions can now add their own columns to the list of messages. I always thought that this sounds good and now I&apos;ve found the first extension that makes use of...</summary>
   <author>
      <name></name>
      
   </author>
         <category term="en" scheme="http://www.sixapart.com/ns/types#category" />
         <category term="software" scheme="http://www.sixapart.com/ns/types#category" />
         <category term="software" scheme="http://www.sixapart.com/ns/types#category" />
         <category term="spam" scheme="http://www.sixapart.com/ns/types#category" />
         <category term="spam" scheme="http://www.sixapart.com/ns/types#category" />
   
   
   <content type="html" xml:lang="en" xml:base="http://lxxi.org/">
      <![CDATA[One of the new features in <a href="http://www.mozilla.com/en-US/thunderbird/">Thunderbird 2.0</a> is that extensions can now add their own columns to the list of messages. I always thought that this sounds good and now I've found the first extension that makes use of this feature. But the extension doesn't just add any old column, it actually adds a very useful column if your mail server is running <a href="http://spamassassin.apache.org/">SpamAssassin</a>. 

<a href="http://ryanlee.org/software/mozilla/thunderbird/spamness/">Spamness</a> adds a column that displays SpamAssassin's spam score in an easily scannable graphical way. <img src="/spamness.png">

After you have installed spamness, you can easily scan your spam folder for very spammy and not so spammy mails and you can sort it by spaminess. (Install the <a href="http://mnenhy.mozdev.org/">Mnenhy extension</a> if you don't want to sort all your folders by spaminess and if you don't want to display the spamness column in each folder).
]]>
      
   </content>
</entry>
<entry>
   <title>MIT Spam Conference 2007 papers</title>
   <link rel="alternate" type="text/html" href="http://lxxi.org/2007/04/mit_spam_conference_2007_paper.html" />
   <id>tag:lxxi.org,2007://1.8</id>
   
   <published>2007-04-18T11:59:03Z</published>
   <updated>2007-05-16T13:05:40Z</updated>
   
   <summary>Multimediawise, the MIT Spam Conference 2007 was a complete disaster. I didn&apos;t much like the webcasts they used in former years which required me to install the much despised real-player, but this years youtubing was a real quality blow. There...</summary>
   <author>
      <name></name>
      
   </author>
         <category term="en" scheme="http://www.sixapart.com/ns/types#category" />
         <category term="spam" scheme="http://www.sixapart.com/ns/types#category" />
         <category term="spam" scheme="http://www.sixapart.com/ns/types#category" />
   
   
   <content type="html" xml:lang="en" xml:base="http://lxxi.org/">
      <![CDATA[Multimediawise, the <a href="http://spamconference.org/">MIT Spam Conference 2007</a> was a complete disaster. I didn't much like the webcasts they used in former years which required me to install the much despised real-player, but this years youtubing was a real quality blow. There isn't much to see in those videos and there is rarely anything to hear. According to the Spam Conference website, we can thank Rob Targosz of McAfee for messing up that part.

If you are not that multimedia inclined, you should be able to resort to downloading the presentations and the accompanying papers. But that would be a bit easy, wouldn't it? To make the papers accessible only to the technically inclined, the Spam Conference eggheads provided only ISO images of something that actually contains those files. 

Fortunately, I am not on Windows and reading ISO images is no big hassle. Unpacking them, giving web-safe names to the files and transferring them to my server was no big deal either.

So here they are (don't ask me what is what, though):

<ul>
<li><a href="/files/spamcon07/azam-Feature_space_reduction.doc">Azam: Feature_space_reduction.doc</a></li>
<li><a href="/files/spamcon07/Azam-paper-presentation.ppt">Azam: paper-presentation.ppt</a></li>
<li><a href="/files/spamcon07/Bekman-SMTP_Multiplexing.pdf">Bekman: SMTP_Multiplexing.pdf</a></li>
<li><a href="/files/spamcon07/Cosoi-Abstract.pdf">Cosoi: Abstract.pdf</a></li>
<li><a href="/files/spamcon07/COSOI-COMBINING_FILTERS.ppt">COSOI: COMBINING_FILTERS.ppt</a></li>
<li><a href="/files/spamcon07/Cosoi-Relevancies.pdf">Cosoi: Relevancies.pdf</a></li>
<li><a href="/files/spamcon07/Dallmeyer-Return_to_Spamlet.pdf">Dallmeyer: Return_to_Spamlet.pdf</a></li>
<li><a href="/files/spamcon07/Dallmeyer-Return_to_Spamlet.ppt">Dallmeyer: Return_to_Spamlet.ppt</a></li>
<li><a href="/files/spamcon07/eggendorfer-paper.pdf">Eggendorfer: paper.pdf</a></li>
<li><a href="/files/spamcon07/eggendorfer-smtp_tar_pit_simulator.pdf">Eggendorfer: smtp_tar_pit_simulator.pdf</a></li>
<li><a href="/files/spamcon07/Emigh_Thomason-Blog_spam.pdf">Emigh/Thomason: Blog_spam.pdf</a></li>
<li><a href="/files/spamcon07/fumera_biggio06.pdf">Fumera: biggio06.pdf</a></li>
<li><a href="/files/spamcon07/fumera_biggio06.ppt">Fumera: biggio06.ppt</a></li>
<li><a href="/files/spamcon07/MartinMerino-ensemble_clas_spam-ver1.pdf">Martin-Merino: ensemble_clas_spam-ver1.pdf</a></li>
<li><a href="/files/spamcon07/MartinMerino-MIT.ppt">MartinMerino: MIT.ppt</a></li>
<li><a href="/files/spamcon07/McIsaac-Symmetrical_DNS_Entries_2-10-07.doc">McIsaac: Symmetrical_DNS_Entries_2-10-07.doc</a></li>
<li><a href="/files/spamcon07/McIssacs-Symmetrical_DNS.ppt">McIssacs: Symmetrical_DNS.ppt</a></li>
<li><a href="/files/spamcon07/Mujica-RM_for_All_Email_Servers.pdf">Mujica: RM_for_All_Email_Servers.pdf</a></li>
<li><a href="/files/spamcon07/Mujica-RM_for_All_Email_Servers.ppt">Mujica: RM_for_All_Email_Servers.ppt</a></li>
<li><a href="/files/spamcon07/Segal-CEAS_2007_Live_Spam_Challenge.pdf">Segal: CEAS_2007_Live_Spam_Challenge.pdf</a></li>
<li><a href="/files/spamcon07/Simpson-MailChannels-TrafficControl.ppt">Simpson: MailChannels-TrafficControl.ppt</a></li>
<li><a href="/files/spamcon07/Trevino_Ekstrom-Relay_Header_Analysis.pdf">Trevino/Ekstrom: Relay_Header_Analysis.pdf</a></li>
<li><a href="/files/spamcon07/Trevino_Ekstrom-Relay_Header_Analysis_Presentation.pdf">Trevino/Ekstrom: Relay_Header_Analysis_Presentation.pdf</a></li>
<li><a href="/files/spamcon07/Watlington-spam-mit.ppt">Watlington: spam-mit.ppt</a></li>
</ul>

According to <a href="http://www.jgc.org/blog/2007/04/no-newsletter-for-april-15-2007.html">John Graham-Cumming</a>, the reason for the ISO-image is that Bill Yerazunis wants people to glance over all the papers. Makes me wonder what he would have done if he wanted people to not read the papers while still publishing them.
]]>
      
   </content>
</entry>
<entry>
   <title>Strassen-Spam by starsbook.com</title>
   <link rel="alternate" type="text/html" href="http://lxxi.org/2007/04/strassenspam_1.html" />
   <id>tag:lxxi.org,2007://1.7</id>
   
   <published>2007-04-17T13:30:25Z</published>
   <updated>2007-05-16T13:06:38Z</updated>
   
   <summary>Als ob der ganze Link-Spam im Web selbst nicht schon nervig genug wäre, gibt es jetzt auch noch Link-Spam im echten Leben: Zettelchen, die an Bäumen und Laternenpfählen hängen und zum Besuch von Webseiten auffordern. Neben der einfachen Umweltverschandlung stösst...</summary>
   <author>
      <name></name>
      
   </author>
         <category term="de" scheme="http://www.sixapart.com/ns/types#category" />
         <category term="spam" scheme="http://www.sixapart.com/ns/types#category" />
         <category term="spam" scheme="http://www.sixapart.com/ns/types#category" />
   
   
   <content type="html" xml:lang="en" xml:base="http://lxxi.org/">
      <![CDATA[Als ob der ganze Link-Spam im Web selbst nicht schon nervig genug wäre, gibt es jetzt auch noch Link-Spam im echten Leben: Zettelchen, die an Bäumen und Laternenpfählen hängen und zum Besuch von Webseiten auffordern.

<a href="http://lxxi.org/strassen-spam.html" onclick="window.open('/strassen-spam.html','popup','width=800,height=600,scrollbars=no,resizable=no,toolbar=no,directories=no,location=no,menubar=no,status=no,left=0,top=0'); return false"><img src="/strassen-spam-tn.jpg" /></a>

Neben der einfachen Umweltverschandlung stösst einem an der Sache sauer auf, dass sich der Spam, zumindest da, <a href="http://maps.google.de/maps/ms?ie=UTF8&om=1&hl=de&z=19&ll=52.469608,13.321985&spn=0.001147,0.002159&t=k&msid=117092297805473828697.00000112004f77d4d0d08&msa=0">wo ich ihn gesehen habe</a>, um den Eingang einer Schule herum konzentriert. 

Spammer war in dem Fall starsbook.com.]]>
      
   </content>
</entry>
<entry>
   <title>Thunderbird extensions you should take a look at</title>
   <link rel="alternate" type="text/html" href="http://lxxi.org/2006/10/thunderbird_extensions_you_should_take_a_look_at.html" />
   <id>tag:lxxi.org,2006://1.6</id>
   
   <published>2006-10-12T13:59:17Z</published>
   <updated>2007-05-16T13:05:52Z</updated>
   
   <summary>Here&apos;s my list of extensions for Thunderbird that you should take a look at. Note that I&apos;m not the gatherer and hunter type when it comes to extensions. I usually prefer software that can do without extensions, plugins, skins, and...</summary>
   <author>
      <name></name>
      
   </author>
         <category term="en" scheme="http://www.sixapart.com/ns/types#category" />
         <category term="software" scheme="http://www.sixapart.com/ns/types#category" />
         <category term="software" scheme="http://www.sixapart.com/ns/types#category" />
   
   <category term="4" label="Email" scheme="http://www.sixapart.com/ns/types#tag" />
   <category term="2" label="Thunderbird" scheme="http://www.sixapart.com/ns/types#tag" />
   <category term="6" label="Tweaks" scheme="http://www.sixapart.com/ns/types#tag" />
   
   <content type="html" xml:lang="en" xml:base="http://lxxi.org/">
      <![CDATA[<p>Here's my list of extensions for Thunderbird that you should take a look at. Note that I'm not the gatherer and hunter type when it comes to extensions. I usually prefer software that can do without extensions, plugins, skins, and all that other crap that will bog down your system and waste your time when all you want to do is work. The list is in no particular order.</p>

<ul>
<li><a href="http://www.chuonthis.com/extensions/buttons.php">Buttons</a>!
<p>Buttons! will give you all those buttons that the Thunderbird developers forgot to add to the toolbar (and a couple more). The ones I like (and need!) the most are the previous and next buttons that will give you the next or previous email regardless whether it's been read or not.</p>
</li>

<li><a href="http://www.mrtech.com/extensions/#toggle">MR Tech's Toggle Preview Pane</a>
<p>Another indispensible button. Although I like to be able to do everything with the keyboard, I sometimes like to use the mouse (without finding my way through a menu first). This toolbar button will simply toggle the preview pane on and off. That's it. It's as simple as that. I have no idea why Thunderbird hasn't that one built in.</p>
</li>

<li><a href="https://addons.mozilla.org/thunderbird/348/">Quick File</a>
<p>There. I told you I like to use the keyboard. Cleaning up my Inbox is something I don't do very often. When I actually do some sorting, I like to do it fast. With Quick File, it's just a key-down, short look at preview pane, and then a Alt-Q followed by some letters that are part of the name of the target folder for that piece of mail. The fastest way to archive messages to different folders.</p></li>

<li><a href="http://minimizetotray.mozdev.org/">Minimize To Tray</a>
<p>Does what the name says. And you can even choose whether you can restore Thunderbird with a single click or with a double-click.</p></li>

<li><a href="https://addons.mozilla.org/thunderbird/594/">Virtual Identity</a>
<p>Got a catch-all email account? And you use it to give different addresses to different people and organizations? Then you will sometimes find it hard to reply to emails because you haven't got a Thunderbird identity ready that uses the address it was sent to. Virtual Identity will take a look at the headers of emails you reply to and supply the correct From-address for you. And you can use it for new emails to quickly come up with a nice, new address.</p></li>
</ul>]]>
      
   </content>
</entry>
<entry>
   <title>cvsspam</title>
   <link rel="alternate" type="text/html" href="http://lxxi.org/2006/08/cvsspam.html" />
   <id>tag:lxxi.org,2006://1.5</id>
   
   <published>2006-08-10T13:58:13Z</published>
   <updated>2007-05-16T13:06:01Z</updated>
   
   <summary>Sourceforge has it and I wanted to have it too: context diffs in CVS commit emails. Sure, it sounds nice and useful. Every time you commit something to your CVS repository, you and your team members get an email from...</summary>
   <author>
      <name></name>
      
   </author>
         <category term="en" scheme="http://www.sixapart.com/ns/types#category" />
         <category term="software" scheme="http://www.sixapart.com/ns/types#category" />
         <category term="software" scheme="http://www.sixapart.com/ns/types#category" />
   
   
   <content type="html" xml:lang="en" xml:base="http://lxxi.org/">
      Sourceforge has it and I wanted to have it too: context diffs in CVS commit emails.

Sure, it sounds nice and useful. Every time you commit something to your CVS repository, you and your team members get an email from the repository that not only includes your commit message, but also includes a complete diff for all your changes.

Here&apos;s why you shouldn&apos;t try: Because all the scripts that promise to deliver those diffs will give you noting but trouble. 

I have no trouble with scripts that are no longer maintained and no longer fit to the current versions of CVS. But all those scripts are also lousily documented. After you performed those &quot;4 easy steps&quot;, you never have a running system. Heck. You might even end up with a repository that won&apos;t let you commit anything ever again.
      
   </content>
</entry>
<entry>
   <title>Grundig: RIP</title>
   <link rel="alternate" type="text/html" href="http://lxxi.org/2005/02/grundig_rip.html" />
   <id>tag:lxxi.org,2005://1.9</id>
   
   <published>2005-02-15T10:10:50Z</published>
   <updated>2007-05-16T13:06:50Z</updated>
   
   <summary>Daß angeblich die meisten Menschen ihren Videorekorder nicht programmieren können, hielt ich bislang für ein langweiliges weil einfallsloses Klischee. Bis ich neulich das zweifelhafte Vergnügen hatte, einen Rekorder der Marke Grundig programmieren zu müssen. Mit einem Schlag fühlte ich mich...</summary>
   <author>
      <name></name>
      
   </author>
         <category term="de" scheme="http://www.sixapart.com/ns/types#category" />
         <category term="ouch" scheme="http://www.sixapart.com/ns/types#category" />
         <category term="ouch" scheme="http://www.sixapart.com/ns/types#category" />
         <category term="use-it" scheme="http://www.sixapart.com/ns/types#category" />
         <category term="use-it" scheme="http://www.sixapart.com/ns/types#category" />
   
   
   <content type="html" xml:lang="en" xml:base="http://lxxi.org/">
      <![CDATA[<p>Daß angeblich die meisten Menschen ihren Videorekorder nicht programmieren können, hielt ich bislang für ein langweiliges weil einfallsloses Klischee. Bis ich neulich das zweifelhafte Vergnügen hatte, einen Rekorder der Marke Grundig programmieren zu müssen. Mit einem Schlag fühlte ich mich als <abbr title="Dümmster anzunehmender User">DAU</abbr>, schwankte ich mit hoher Frequenz zwischen Panik und Verzweiflung und begann diesen Mythos als Tatsache anzuerkennen.</p>
	<p><a id="more-7"></a></p>
	<p>Eigentlich wollte ich nur eben einen Film aufnehmen. Also Kassette rein, Fernseher an, Videorekorder an und die Taste, die das <abbr title="On-Screen Display">OSD</abbr> auf den Schirm zaubert fand sich auch einigermaßen schnell auf der Fernbedienung. </p>
	<p>Aber dann ging es schon los. Die Fernbedienung bietet einerseits einen Block von Bedienelementen für die typischen Rekorderfunktionen, also Spulen, Abspielen, Stop, usw. Andereseits hat sie den Grundig-typischen Block von rauf-, runter-, rechts-, links-Tasten mit einer OK Taste in der Mitte. Zur Navigation im <abbr title="On-Screen Display">OSD</abbr> wird letzterer benötigt. Leider erwischte ich aber immer ersteren, weil er besser erreichbar auf der Fernbedienung sitzt, die Tasten größer sind und dieser Block ganz allgmein einen höheren Aufforderungscharakter hat. </p>
	<p>Ich konzentriere mich nun also, den richtigen Tastenblock herzunehmen und schiebe den Cursor rauf und runter um einen Speicherplatz auszuwählen, quittiere mit &#8220;OK&#8221; und kann jetzt die benötigten Daten eingeben. Naja. Was heißt eingeben? Durch das Drücken auf die rauf/runter Tasten kann ich die angezeigten Zahlen größer und kleiner machen. Angezeigt werden zunächst immer die aktuellen Daten. Also das momentan eingestellte Programm und das aktuelle Datum. Bis auf die Endzeit des Timers, hier wird nicht die aktuelle Zeit vorgegeben, sondern die Startzeit. Was bei einer typischen Spielfilmlänge von 90 Minuten natürlich eine brilliante Idee ist, denn dann kann ich um die Minuten richtig einzustellen, erst einmal dreißig Minuten rauf oder runter blättern.</p>
	<p>Der letzte einzustellende Wert entscheidet über den Aufnahmemodus: Shortplay oder Longplay. Zumindest habe ich die beiden Abkürzungen SP und LP so gedeutet. Denn obwohl das <abbr title="On-Screen Display">OSD</abbr> eine Legende bietet, wird man in der Hinsicht nicht schlauer. Irgendwer hat sich nämlich nicht geschämt in der Legende folgenden Text zu verewigen: </p>
	<blockquote><p>SP: SP LP:LP</p></blockquote>
	<p>Arg! Aber bitte. Ich kenn mich ja aus. So. Die ganze Sache noch quittieren und den Videorekorder schön ausgemacht. Da stelle ich fest, daß ich die Startzeit falsch gesetzt habe. Zurück im <abbr title="On-Screen Display">OSD</abbr> muss ich aber leider feststellen, daß man fertig programmierte Timer nicht wieder ändern kann. Ich muß also noch einmal von vorne anfangen. Nachdem ich erstmal wieder auf Play gedrückt habe, lasse ich aber auch das über mich ergehen. Jetzt ist aber gut. Videorekorder aus, Fernseher aus und raus aus der Wohnung. </p>
	<p>Am nächsten Morgen finde ich eine leere Kassette im Rekorder. Oh. Der Timer ist weg und der Rekorder gibt sich unschuldig. Also doch mal die Bedienungsanleitung lesen? Ich drücke mich davor ein paar Stunden, doch dann siegt die Neugier. Die Lösung ist schnell gefunden. Grundig hat seinem Videorekorder <em>zwei</em> Standby-Zustände spendiert. In einem nimmt er auch was auf, wenn er programmiert ist, in dem anderen unterlässt er das und ist einfach nur so aus. Arg! Und während die &#8220;Schalt mich aus, damit ich sinnlos Strom verbrauche&#8221;-Taste dort auf der Fernbedienung angebracht ist, wo man so eine Taste erwartet und noch dazu in rot gehalten ist, hat man die andere, richtige, gewünschte Taste zwischen all den anderen versteckt. Prima.</p>
	<p>Eigentlich hatte ich versucht, für diesen Artikel ein paar illustrierende Bilder aufzutreiben. Aber ich hätte es wissen sollen: Die <a href="http://www.grundig.de/">Webseite der Grundig AG</a> ist eine reine Katastrophe in allen relvanten Punkten: Usability, Accessibility, Datenschutz. Hat man das obligatorische Flash-Intro überwunden, begrüßt einen ein dreiteiliges Frameset. Links findet man das <a href="http://web.archive.org/web/20050913170120/http://www.grundig.de/navigation/produkte/bedienungsanl/navigation.html">Navigationsmenü</a>. Eigentlich enthält es nur Text, aber warum HTML wenn es auch ein GIF mit leerem ALT-Attribut und jeder Menge JavaScript drumrum sein kann? Schließlich hat man dann Kontrolle über die Größe. Und die soll möglichst eine Kleine sein. Deswegen pixeln die Buchstaben wahrscheinlich auch so schlimm.</p>
	<p>Ein Klick auf das &#8220;Bedienungsanleitungen&#8221;-GIF fördert eine recht interessante <a href="http://web.archive.org/web/20050913170120/http://www.grundig.de/produkte/bedienungsanl.html">Eingabemaske</a> zu Tage. Während der erklärende Text darüber von einer &#8220;komfortablen Suchfunktion&#8221; schwärmt, für die ich nur &#8220;die genaue Typenbezeichung des Geräts&#8221; bräuchte (widerspricht sich das nicht?), fordert mich das Formular darunter auf, meine persönlichen Daten zu offenbaren: Name, Anschrift, Email. Nicht weniger als sechs Felder soll ich ausfüllen. Warum? Achja, da steht es im letzten der erklärenden Absätze: &#8220;Für die Anzeige bzw. den Download der Bedienungsanleitungen benötigen wir von Ihnen noch ein paar Daten.&#8221; Achso. </p>
	<p>Na gut. Dann denkt man sich halt irgendwelche Daten aus, um an die Leckerli zu gelangen. Denkste. Ich gab mein Bestes, aber das Beste der Grundig-Datenabzock-Maschine ist ein Hinweis auf  die &#8220;detailierten Fehlermeldungen&#8221; die verfügbar sind. Aber diese Fehlermeldungen sind eher was für den Entwickler, weniger für den Nutzer. Lustig ist auch, was passiert, wenn man einfach nix eingibt. Dann vergisst die Maske nämlich wie ihre Felder heißen und setzt vor jedes Textfeld ein schnödes &#8220;null&#8221;. </p>
	<p>So bin ich denn froh, stolzer Besitzer eines Blaupunkt Videorekorders zu sein. Obwohl sich auch der ganz einfach verbessern ließe. Wehmütig denke ich manchmal zurück an meinen alten Videorekorder aus dem Baumarkt. Hersteller: die Weltmarke Funai. Der hatte den besten Videotext-Dekoder eingebaut, den ich jeh erlebt habe. Und Dank dieses Dekoders brauchte er auch keinen <abbr title="Electronic Programme Guide">EPG</abbr>. Man schlug einfach nur die Seite mit dem Fernsehprogramm im Videotext auf, forderte den Rekorder per Tastendruck dazu auf, sich diese Seite näher anzusehen. Und dann konnte man einen Cursor von Sendung zu Sendung schieben und schon war alles programmiert. Die Welt könnte so einfach sein. Oh Funai, Du mein Lieblingsrekorder! Ich hab Dich immer noch lieb.</p>
]]>
      
   </content>
</entry>
<entry>
   <title>Fedora Core 2</title>
   <link rel="alternate" type="text/html" href="http://lxxi.org/2004/11/fedora_core_2.html" />
   <id>tag:lxxi.org,2004://1.4</id>
   
   <published>2004-11-11T13:56:12Z</published>
   <updated>2007-05-16T13:06:11Z</updated>
   
   <summary>After all the trouble with Suse, I ordered the Fedora Core 2 CDs from LinuxCdVersand.de. Installation was quite simple and quite fast. But the problems with the boot-loader seem to be ubiquitious. This time, all I got was &quot;GRUB&quot;. I...</summary>
   <author>
      <name></name>
      
   </author>
         <category term="en" scheme="http://www.sixapart.com/ns/types#category" />
         <category term="software" scheme="http://www.sixapart.com/ns/types#category" />
         <category term="software" scheme="http://www.sixapart.com/ns/types#category" />
   
   
   <content type="html" xml:lang="en" xml:base="http://lxxi.org/">
      <![CDATA[After all the trouble with <a href="http://lxxi.org/2004/10/suse_personal_91_thanks_but_no.html">Suse</a>, I ordered the Fedora Core 2 CDs from <a href="http://www.linuxcdversand.de/">LinuxCdVersand.de</a>. 

Installation was quite simple and quite fast. But the problems with the boot-loader seem to be ubiquitious. This time, all I got was "GRUB". I had to change the 'device.map' file to make it work. Why don't the installers know about this? 

I liked what I saw after the first boot. Fedora's default desktop is Gnome and I was ok with this for a couple of days. I didn't have to do anything to share my printer over the network or to access the network. 

But when I started groping around a bit, I was a bit disappointed. The installer (or whoever is responsible) didn't automagically mount my NTFS partitions. I even had to download a driver. The menues in Gnome and KDE looked kind of strange, at least not how I expected them to look and I found it impossible to change them. 

Then it was update time and I tried to let up2date do its work. But I must say that this is one clumsy application that often gives you the impression that it has died or is caught in an infinite loop. When it's busy, the UI won't refresh for minutes. It will also happily tell you that all the updates weigh in with just 0 KB each. A little more accuracy wouldn't hurt here. 

More generally, I was surprised that package handling was so badly implemented. After all, Fedora Core is in the Red Hat tradition. Or so I thought. But clicking on a RPM file in Nautilus or Konqueror only yielded "I don't know what to do with this" messages. The "Install and Remove applications" program only knows about what is on the installation CDs. And all it does is show you package groups. You cannot even search for specific packages. When I found that other rpm installation UI thingy, I had a very hard time finding out whether missing dependencies are on the CDs or if I had to download them. Dropping to a terminal and managing stuff with rpm was really much easier. 

The configuration GUI programs offered by Fedora are generally a big disappointment. Configuring system sound events in Gnome is a nightmare. Sure, you can choose another sound if you don't like the default. But you cannot disable sounds for a given event. And the default sounds sometimes are a real <a href="http://lxxi.org/startup.mp3">disgrace</a>. Does that sound like a nice and friendly welcome? Hardly. 

Or take the bootloader configuration thing: All you can do is change the default menu entry for grub and adjust the timeout. Hardly something that deserves the name "bootloader configuration". 

The worst about my newly installed system, though, was that it behaved in unpredictable ways. I never knew whether I would have net access after the next boot or not, for example. 

And why is IPv6 enabled by default? This will only help you to a poor browsing experience. 

Fedora Core 2 will not be able to serve as a replacement for Windows for me. I found it too slow, too unpredictable, too akward to customize, and too poorly documented. 

I now ordered Mandrake. I'll keep you posted (whether anybody cares or not).]]>
      
   </content>
</entry>
<entry>
   <title>Bye-Bye, Pegasus Mail!</title>
   <link rel="alternate" type="text/html" href="http://lxxi.org/2004/11/byebye_pegasus_mail_1.html" />
   <id>tag:lxxi.org,2004://1.3</id>
   
   <published>2004-11-10T13:55:00Z</published>
   <updated>2007-05-16T13:06:21Z</updated>
   
   <summary>When I first got on the internet, we had to use huge HP-UX terminals at the universtity&apos;s computing center. I used elm as a mail client back then. Finally, they managed to hook up the offices and our personal computers....</summary>
   <author>
      <name></name>
      
   </author>
         <category term="en" scheme="http://www.sixapart.com/ns/types#category" />
         <category term="software" scheme="http://www.sixapart.com/ns/types#category" />
         <category term="software" scheme="http://www.sixapart.com/ns/types#category" />
   
   
   <content type="html" xml:lang="en" xml:base="http://lxxi.org/">
      When I first got on the internet, we had to use huge HP-UX terminals at the universtity&apos;s computing center. I used elm as a mail client back then. Finally, they managed to hook up the offices and our personal computers. That was way-back when Windows 3.1 was installed everywhere and you spend your time optimizing autoexec.bat and config.sys.

You had to jump through a couple of hoops to make Windows 3.1 internet-capable and when you finally managed to get it online, you didn&apos;t really have much choice regarding client software. It was Netscape for the web and either Netscape or Pegasus Mail for email. I went with Pegasus and was a happy camper for quite some time.

Windows 3.1 was replaced by Windows 95, Windows 95 was replaced by Windows 98 which was quickly replaced by Windows 98SE. Eventually I got Windows 2000. But Pegasus Mail was always my choice.

I don&apos;t know what exactly happened, but a couple of weeks ago I must have gotten into a mail-client changing mood. Maybe it was all the hype about Thunderbird. Maybe it was also the wish to someday get away from Windows and into Linux. Or maybe it was the fact that I made more and more use of IMAP and less and less use of POP3. Whatever it was, I have now changed to Thunderbird. 

There isn&apos;t much that I miss in Thunderbird that was available in Pegasus Mail. To all intents and purposes, I really hate the fixed window layout that you find in Netscape Mail, Mozilla Mail, Outlook, and Thunderbird. The ability of Pegasus Mail to give you MDI windows for your folders and messages is something that I really miss. I also miss the filtering abilities that Pegasus offered. I don&apos;t see why I can&apos;t share filtering rules across accounts in Thunderbird; that&apos;s really annyoing. 

But overall, Thunderbird is really the better client for me. It doesn&apos;t give me any bogus new-mail alerts, It doesn&apos;t offer default tiny columns that don&apos;t even accomodate the headings in new folder windows/views. It can copy/move from and to IMAP folders without any major problems. 

But best of all: Using Thunderbird and watching its development, you get the feeling that it will become better and better and that you don&apos;t have to wait years for an improvement to arrive. Unfortuately, the same it not true for Pegasus. Pegasus Mail is developed by one single person: David Harris. He thinks that open source software is something bad. He thinks that Pegasus eventually needs a PIM. He thinks he ought to spend his time developing yet another HTML-rendering engine. 

In effect, I say &quot;Good-bye, Pegasus Mail&quot;. We&apos;ve spent a lot of time together, wrote and read tens of thousands of emails. Watched operating systems come and go. But our time together is over. I just hope you will be open-sourced one fine day.
      
   </content>
</entry>
<entry>
   <title>Suse Personal 9.1: Thanks, but no thanks</title>
   <link rel="alternate" type="text/html" href="http://lxxi.org/2004/10/suse_personal_91_thanks_but_no.html" />
   <id>tag:lxxi.org,2004://1.2</id>
   
   <published>2004-10-26T13:53:40Z</published>
   <updated>2007-05-16T13:06:29Z</updated>
   
   <summary>I gave Linux another chance. I really want to have a nice Linux install running on my machine, but I know that I need a lot of time to set it up so that it fits my needs. Well, I...</summary>
   <author>
      <name></name>
      
   </author>
         <category term="en" scheme="http://www.sixapart.com/ns/types#category" />
         <category term="software" scheme="http://www.sixapart.com/ns/types#category" />
         <category term="software" scheme="http://www.sixapart.com/ns/types#category" />
   
   
   <content type="html" xml:lang="en" xml:base="http://lxxi.org/">
      I gave Linux another chance. I really want to have a nice Linux install running on my machine, but I know that I need a lot of time to set it up so that it fits my needs. Well, I have a lot of time now and so I went out and bought a magazine with a CD containing a free version of Suse 9.1 personal. 

I cleaned the spare harddisk I have, put in the CD and rebooted. After a couple of clicks, Suse had partionioned the disk and was copying files to it. This copying process took surprisingly long, but I didn&apos;t mind that. After it was done copying the files, I was allowed to do a few more clicks and then I needed to reboot. Suse had installed grub and I hadn&apos;t changed its configuration. So I rebooted and got: &quot;grub geom error&quot;. That was it. I now had a completely unusable machine uttering complete nonsense. Cool. 

&quot;geom&quot;, I guess is short for geometry. Which, again, is short for &quot;harddisk geometry&quot;, which is geek speak for &quot;I don&apos;t know how to boot from that disk&quot;. Duh. 

So I booted from the CD again. Fortunately, I am a seasoned Windows user. I&apos;m used to clicking on &quot;Start&quot; when I want to stop and stuff like that. Booting from the Suse CD, I had to click &quot;Install&quot; when I really wanted to boot. It wasn&apos;t obvious at all, but I finally made that thing boot into my new Linux installation where I found Yast and its boot manager configuration module. I did some editing of the grub configuration, but to no avail; the machine still wouldn&apos;t boot. So I chose the option to restore the MBR of my first harddisk and then I installed grub on the second harddisk. That worked. If that would have been my first attempt at Linux, I would have been completely lost. Searching google and the Suse support db for the error message revealed that the problem was known. Unfortunately, it didn&apos;t give any clues as to how to remove the problem and make my machine boot again. All the fine tips that Suse had required a booting system when all I had to offer was &quot;grub geom error&quot;. 

Well, &quot;never mind&quot;, I thought and started to play with the new system. It was quite nice. I was able to browse my local MS network, I was able to print with my printer, I was able to make Suse update my clock on boot, etc. All very nice. 

Then I was trying to install a few more packets over FTP, when my girl friend said that she wanted to print on my printer. I knew I had to install Samba to share the printer, and I knew that I wasn&apos;t going to do that within a few seconds, so I tried to browse the her shared drive so I could open the file myself and print it. But the local MS network support had stopped working. Konqueror told me to file a bug report. A look at the boot messages told me that the smbfs service wasn&apos;t loaded (&quot;skipped at runlevel 5&quot;) because somewhere during the boot process, someone had figured that this service was &quot;unused&quot;. Duh. 

So I asked her to mail me the file and while it printed, I did some serious googling again. Another known problem; but nobody had any clue how to use that service again. I had to give up on it and I decided that I&apos;d install Samba and see how that would help. 

But that was for later. First, I wanted to get a few developer packets. gcc, make, Python, more Perl modules, etc. They all installed pretty well, at least from what I could tell. Yast closed its Software installation module on me without any hint of a failure. Compiling &quot;Hello, World!&quot; in C worked too. So I fired up perl to install some modules. Strangely, LWP::UserAgent wasn&apos;t installed and lots of modules that were required by that module were also missing. Trying to install these didn&apos;t work. The CPAN shell reported errors with make. &quot;OK&quot;, I thought (again), &quot;this is for later&quot;. 

Now, I wanted to install gnome. I&apos;m not a big fan of KDE and I wanted to have an alternative. Again, the required packets installed just fine. But how do you get suse to offer you the choice of a desktop? I found the answer with google and was finally able to log into gnome. But all I got was icons. All the text in gnome was rendered at a font size of approximately 0 pt. Switching the screen resolution didn&apos;t help. I still couldn&apos;t see anything but icons. Duh. 

Some more googling: Yet another problem that people had come across. But no solution. That&apos;s it. That&apos;s enough of Suse for me. I will have to get another distro. Although I don&apos;t know which one. I also don&apos;t know how to get one. I had Mandrake once and it wasn&apos;t bad. But downloading three iso images? No, I can&apos;t afford that bandwidth. Ordering Mandrake? Sorry, but I&apos;m not going to pay 59 Euros either. 

So I&apos;m back on Windows, waiting for a nice Linux distro to come along. 

If all of the above wasn&apos;t enough, here are some more questions I have for Suse: 

When you install sofware from an ftp server, why do you get packets that suffer from bugs and security flaws? You install some packet, e.g. Opera, and the next thing you know is that &quot;YOU&quot; tells you that there is an update available. Well, I guess it was also available five minutes earlier. Duh. 

When you log on as root, all you get is Yast. How are you supposed to administer a system without at least a shell to work with? Duh.
      
   </content>
</entry>

</feed>
