Planning an overall direction for LFS

dennisjperkins at comcast.net dennisjperkins at comcast.net
Fri Feb 29 07:31:28 PST 2008


I agree with what I have been reading.  It would be a shame if people lost interest in LFS because there are few sources available for teaching how to build a complete system.  The more widely disseminated this knowledge is, the better, even if 99% of Linux users don't care.

I have been using LFS as my version of Linux for a few years.  I know others do the same.  As a result, I think a package manager would be useful, partly to remove unneeded files, but also for installing them on another computer.

The first thing I did after building LFS for the first time was to see how useful it was.  Which meant building more packages.  And more packages...  I have learned a lot, including that I don't want to wait for a distro to release a new version.  I think that a few things in BLFS might be better in LFS:  dhcp, some of the bash setup scripts.  Unless you are building an embedded system, these are very useful.

LFS concentrates very well on building, but in some areas, additional information would definitely be useful.  Not everyone wants this information, and too much might be distracting to reading how to build, so maybe links to additional information would be a good way to make this information available.  BLFS does this in places, such as wireless networking.

I would like to see UTF-8 in LFS.

I have both 32-bit and 64-bit x86 systems, and I think merging them would be a good idea.

I'm not sure I like using XML for building packages.  I prefer reading scripts.  However, I do see some benefits and I am not opposed to it.

All of this talk about direction is very interesting.  I hope to see some interesting things from all of this soon.






More information about the lfs-dev mailing list