A 1CD version is not going to happen. 1CD just doesn't have enough room for a modern GNU/Linux system and a graphical installer (which takes up a fair amount of space on its own). Personally a single DVD would my favorite option, but it will be a while before everyone in the Libranet target market has a DVD-ROM drive.
As to other suggestions:
Going back to a text-mode installer isn't going to happen. It's just not needed on modern hardware. Same for a text-only Adminmenu. As for things like remote administration, web based interfaces or X tunneling over ssh are still more desirable than a text based interface.
Copying another distribution's model isn't going to work. Libranet has been and needs to stay unique. This is not to say that Libranet can't and hasn't learned from others. I have always felt that RedHat and SuSE are Libranet's biggest competition.
Sadcat is correct in stating that it is an impossible job for a small team to control the quality of thousands of software packages. However, with the help of the free software community and projects like Debian providing a pool of readily available packages providing a large high-quality set of packages is not just do-able, but desirable.
The "one app per task" doesn't really work for me. My choice of an application to get a given task done is very often different than another persons. The "one app per task" mindset works if you are pandering to the computer illiterate, which to be frank is the biggest market out there. However, Libranet isn't targeted to the computer as a appliance, what to I click to get to the internet crowd. I don't use “illiterate” as a derogatory term in this context. It is unreasonable to expect the average Joe to learn how a computer works just to be able to “use” a computer. As the internet expands into evermore areas of our lives computers will become much more like automobiles. Almost anyone can drive a car, but very few people know or care about how their car works. Fortunately it's not possible to get injured crashing a computer Smile Anyone remember Apple's “what's a dip” TV commercials from years ago?
In my mind Libranet has always been geared to the happy middle ground: those users who want an easy to maintain working system, but don't want someone to dictate how they go about doing their work.
That being said I don't think there is a need to provide every application on the planet, only those that do a good job. Providing both Firefox and Opera as web browsers makes sense, but adding little used “project” browsers adds little value since the type of users who would be using that type of software would be able to install the software themselves. Perhaps a better example is the world of email clients. There are many, many, different email clients available. What ones really bare inclusion? In my opinion: Thunderbird, Evolution, and Kmail.
The GNU/Linux system is going through the process of maturing from a hobbiest only system to a system usable by the everyday woman in the pub. The fact that there are distributions with so many different targets available is a testament to the diversity of the computer market and the ability of GNU/Linux systems to adapt to many different areas.
GNU/Linux is moving forward at a rapid and exciting pace, and I earnestly hope Libranet will be part of the big changes that we are beginning to see. To this I will add that I make no guarantees whatsoever as to the future of Libranet. At this point it's go big or go home; the two man distribution model just doesn't work for me anymore.
What I need to do now is find partners who have an interest in seeing something amazing happen. It will take me a while to gather my thoughts and figure out how best to go about this.
Tuesday, October 11, 2005
Read my posting on the Libranet forum.