All is relative…

It’s a bit sad that Gentoo didn’t promote this more, but Gentoo users now have support for license-based masking.

What does this mean? Well, previously, Gentoo already supported various masking reasons (like stable versus staging – the x86 versus ~x86 saga, package.mask’ing – for security reasons or critical bugs, …). Now, a new feature is added: license masking.

By default, Portage accepts all non-EULA licenses. If a package uses a EULA license, you’ll get a failure message stating that the license is ‘masked’.

Now, what good does this do for users? Well, you can now ask Portage only to accept certain licenses (like @FSF-APPROVED, which is a list of all FSF-approved licenses) and deny the installation of others. Nice, isn’t it?

I’ve added information regarding package license states (and the global as well as per-package unmasking support through /etc/portage/package.license) to the Linux Sea document.

Sometimes you want to execute a particular command, but only when you’re at home. Examples would be running fetchmail (or fetchnews) through cron, but you don’t want this to run when you’re in the train, connected to the Internet through GPRS…

My idea here would be to create a script (say “athome.sh”) which returns 0 if you’re at home, and 1 otherwise. The key of the script is that the MAC address of your (default) gateway is unique.

#!/bin/sh

GW=$(/sbin/ip route | awk '/default/ {print $3}');
MGW=$(/sbin/arp -e | grep ${GW} | awk '{print $3}');

if [ "${MGW}" = "00:11:22:33:44:55" ]
then
  exit 0;
else
  exit 1;
fi

With this script, you can then run athome.sh && fetchmail. If you aren’t home, athome.sh will return 1 and the fetchmail command will never be executed. When you are, the command returns 0 and fetchmail is launched.

It’s finally committed: I’m going to dive into the realms of database architecture. It’s with some sentiment that I’m leaving the expertise field of Apache, J(2)EE and WebSphere, but seeing the database architecture field makes it up well. I’m starting to get acquainted with Oracle DB as first platform and IBM DB2 + Microsoft SQL Server are pending. Exciting times are coming!

A few people have contacted me if they were allowed to translate the online book I’m writing (Linux Sea). Of course they are, the license allows it. However, I recommend to wait a bit. At this moment, I’m not going to release the docbook sources (I’m not writing it in DocBook, but I’m generating from another XML into DocBook) until I’m happy with the final result.

I’m glad to see that the document is well received. There is still lots of work on it (more excercises, a thorough spelling / grammar check, elaborate on certain topics, …) so stay tuned for further updates. Why are those updates “slow”? Well, let’s say that I use a “fair share scheduling” principle on all my hobbies ;-)

A few updates have made it to the Linux Sea book:

• Information regarding ndiswrapper
• Some information about udev and the symlinks that it creates

The PDF version has been updated as well.

If you’re not up to the various free image gallery sites, you might want to try out ZenPhoto. Quite powerful, easy to use and well themeable. Requires PHP / MySQL.

I’ve added quota support information to the Linux Sea book as well as information about the eclean command for cleaning distfiles and packages. The part on building a Linux kernel has been moved into its own chapter, the chapter on hardware support now has a bit more information about dealing with sound cards (ALSA support) and will contain information about sound servers in the near future. This chapter will also be used to configure the various other hardware things as they come by (printers, scanners, …).

I’ve added a draft PDF version of my Linux Sea document. If you don’t mind the A4 papersize and the bad typesetting of the text boxes (I still have lots of overflows to correct) it is quite usable.

Having documented a lot in LaTeX (back in the old days at the university), GuideXML (Gentoo’s document markup language) and DocBook (Linux Sea) I’m now pointing my arrows at DITA, the Darwin Information Typing Architecture.

DITA “forces” the technical writer in separating the content of his document in specialized subjects: reference, task or concept, or a specialized version of any of those which you can create/define yourself.

By separating content in those three subjects, you can more easily manage your technical documentation (write concepts as individual topics, tasks as end-user procedures and references for affiliated topics or command information).

Once all these documents are written, you bind them together using a DITA map (a metadocument which holds references to all related concepts/tasks/references) et voila: your documentation is ready.

Well, not really, you need to build it to something end users can read – you can use dita-ot for that. It supports building for Eclipse Infocenter, RTF, XHTML and PDF out of the box.

My everlasting document, Linux Sea, is progressing slowely but surely. I’ve started a few new chapters and also initiated a chapter on Installing Gentoo (which is more a shortlist of tasks with pointers to earlier chapters).

I also took a different CSS (docbook.css file used by the FreeBSD handbook) as it looks a lot better.