Quote Originally Posted by Harry Kuhman
I saw the Slashdot article too. I found this very interesting for a slightly different reason. A while ago I make a copy of DSL as the first session of a multi-session disk. I then used DSL to write files to the second session. I did not expect to have DSL somehow find the second session automatically and load stuff the way that Puppy Linux does. The thing that I found at that time was that the DSL disk would no longer boot after I had burnt the second session. I accepted this as how el torro botable disks work. But if Puppy Linux can do it then I'm wondering what they are doing differently than I did and why Puppy can boot after a second session is written on it but DSL can not. If anyone has any insight on this issue I would appreciate them sharing it. I have bookmarked the Puppy Linux pages linked to in the slashdot article, but have not read them completely yet.
I don't know bootable discs but I have a theory anyway, based on my limited understanding of the boot process:
the code that makes a bootable disc bootable must be contained in the disc directory. When you add to a multisession disc, the burning program must of course create a new directory. As a boot sector virus prevention measure, the DSL burning program is probably designed not to include any executable code from the original directory in the new directory.

I post this purely as a learning exercise for myself, in the hope that someone who actually knows the answer will confirm, or at least correct me.