[DE | EN | FR | JA | ES]

After finally setting up the network in my new apartement I would like to welcome you to another issue of Georg's Brave GNU World. I hope to have found an interesting mix of technical, political and philosophical topics. Let's begin with sad news from France.

France and Free Software

Since noone is being excluded from using/developing Free Software and it has been proven several times that commercial Free Software is possible, the "threat to free competition" is obviously a hoax. It seems that this proposal tries to keep the backdoor open for proprietary software while pretending to further Free Software. This is one of the reasons why the French Free Software organisation APRIL opposes the proposal [6]. Apparently some French politicians have realized the importance of Free Software and demand access to the source code for themselves while keeping this freedom and security from the public they should represent. This is why I want you to oppose this proposal and get the general public interested because this situation is simply not acceptable.

But fortunately there are also good news from France. David Mentre told me about a study that deals with the applicability of the GNU General Public License in France [7]. The study basically comes to the conclusion that the GPL can be used in France without major problems.

Now I would like to get to the technical part and the first feature is about a remarkable word-processor project.

e:doc

e:doc is a WYSIWYM-frontend for an unlimited number of backends; where backend in this context means an output format like LaTeX or HTML for which the backend are already included in the distribution.

The user enters the text, structures and formats it in e:doc and then outputs it in the desired format. The text is saved in an e:doc internal format which preserves all of the extended information which includes possible interactions between tags like citations and the bibliography or references and labels.

What makes e:doc so powerful is its modular concept. The definition of the internal document structure is independent of the backends and can be expanded without paying attention to the possible output format. This makes it very easy to give the documents special traits that can be transformed for the output formats in the best possible way. Of course this also allows adding other output formats easily without having to modify the documents. Operating system specific options and personal settings are also configured in separate places which allows for easy and comfortable personalization.

Fortunately e:doc runs under Unix as well as Win32 - files can easily be exchanged. This makes it perfect for the use in professional areas where a common appearance is necessary but most of the people are afraid of LaTeX. The author himself sees publishing companies, newspapers and scientists as the most important target groups.

It is planned to make e:doc an official GNU Project as soon as its functionality is a little more well-rounded which shouldn't take too long. But since it is distributed under the GNU General Public License this is not too much of a step as far as the user is concerned.

Thomas Schmickl still has a lot of plans for e:doc but as he earns his money as a biologist working on honey-bee experiments he suffers from massive lack of time. Lately Tim Stevens started helping him but there is still too much work to do and I've been asked to ask for more developers. Especially "Perl programmers, LaTeX gurus, HTML designers, native speakers" (quote) are needed. Should you belong to one of these categories and are interested in participating in the development, you should directly send mail to Thomas [9].

GNU Parted

This makes it possible to get rid of proprietary Windows programs like "Partition Magic" if the goal is to reorganize the harddisk to make space for another GNU/Linux distribution. But Parted can also do the so-called "disk imaging." The idea is to preinstall one machine with the operating system plus software and all configuration and then write the partition on one or more CDs. After putting a minimal GNU/Linux plus GNU Parted on the CD and making it bootable you get a very easy and comfortable possibility to transfer this setup to another machine; of course it also offers the possibility to restore the original configuration in case it got lost somehow.

Since there are better programs under GNU/Linux for this purpose, GNU Parted mainly fills a gap with Windows installations which posed an ethical dilemma for the author. Should he really help people installing Windows? His opinion on this is that it makes it easier for Windows users to switch over to Free Software step by step - so his answer was yes. Especially with big firms this is probably a realistic standpoint.

Also quite interesting is the connection between GNU Parted and the programs DiskDrake and ext2resize. DiskDrake is a spin-off of an early version of GNU Parted that lacks some of the functionality but it does have a nice interface; the lack of which is one of the main problems of GNU Parted according to Andrew Clausen. The link to ext2resize is even tighter because the projects are sharing their Ext2 code - although ext2resize does have some additional features like resizing mounted partitions.

All the functionality of GNU Parted is implemented in the libparted library - so the biggest weakness (a missing GUI) should be rather easy to fix. Besides this the current tasks are stabilizing the current development version and including support for Macintosh, Sun & BSD disk labels as well as optimizing the disk imaging support.

Direct GNU Sponsoring

Companies take part in the GNU Project by paying developers to work on special GNU Projects. This frees the developers of the need to earn money otherwise; the company has direct control where the money goes and does get the best man/woman for the job while directly supporting the GNU Project and Free Software.

Since this is also a service to the Free Software community I have been thinking about a "green list" of firms that operate in this way. With the help of this list end-users can prefer those firms to indirectly strengthen the community again. Ideas, questions and comments about this would be very welcome [1].

By the way: you do not have to go to Australia to find such a firm. Since the beginning of June I am getting paid by the iD-Pro Germany GmbH to work on my latest Project GNU AWACS while finishing my degree. Only because of this I am able to work on GNU AWACS at all. It is great that the amount of firms working this way is continuously increasing.

And now there is one thing OKUJI Yoshinori asked me to mention.

Why Texinfo?

The idea behind Texinfo is to have just a single file in the Texinfo format - similarities of this file with TeX are (obviously) wanted. The user only has to maintain a single file which saves a lot of effort and time. This file can then be used to create different output formats - possible are HTML, plain ASCII, PDF, ROff, DVI (and hence Postscript) as well as info. This means the documentation can easily be distributed over the next in the best suitable form and is also available on the computer for online browsing.

Although some users at first think the info files are somewhat strange, they offer the big advantage of being structured and allowing jumping between topics via links. By the way: to view them you do not need to use the "info" program. The Emacs does have a great info mode and programs like pinfo, tkinfo or the GNOME help browser are info-capable.

The mentioned advantages have made Texinfo the GNU documentation standard and even if people get discouraged by the TeX-like structure the time to learn it is easily saved later in most cases.

Before I finish this issue I would now like to discuss a question that has been raised by Thomas Pollak from Austria.

Forging of the Copyright?

Let me first tackle the question why the GPL does not have an explicit paragraph about this issue: the generally accepted copyright law makes it illegal to spread someone else's work under your own name and the GPL uses this copyright law to anchor the license in the code. Additionally a license could hardly talk about everything you could or should not do - so it is usually centered around what one is allowed to do.

Now about the possible danger of "copyright forging." When dealing with this one has to raise a few questions. First of all the direct monetary gain for producing Free Software is rather small - most of the money is raised with support and the copyright-question is completely irrelevant for this. So why should anyone want to forge a copyright? One might think that the possible prestige of having written a program is worth the risk.

What would happen if someone did that? Since the program has been written by someone else the "forger" had to find it somewhere. Possibilities are CD-collections or the internet. But although it is easy to spread a program it is impossible to delete all old copies from CD-collections, or Web-/FTP-Server backups - the magic term here is "widespread distribution." So it can easily be proven that a prior version of the program did exist under another copyright.

This will most certainly mean the end of the forgers reputation and it is highly questionable whether someone would take such a risk; which is why I think this risk is neglectably small.

...now onto my bike