Re: SEUL: Partitioning

[Erik Walthinsen <omega@aracnet.com>]

> rather than doing an actual trial installation, which is ridden
> with technical issues, and may actually not gain us real users in the end,
> i think a productive investment of effort would be (once we have
> a product) to produce a demoreel of some sort that runs under various
> platforms, and gives a good preview of what to expect after going through
> [what the user may (despite our best efforts) view as] installation hell.
Maybe...  That has it's own issues.  That entails a complete group dedicated 
to doing evil M$ things.  There are two types of demos (of the type you're 
talking about): movie and interactive.  A movie wouldn't be hard, an 
interactive one would have to be done as a trial install.

OTOH, it might be possible to, as someone suggested before, use loadlin or 
simliar to load both the kernel and a full root image into RAM from within 
Windoze or DOS, shoot Win in the head, and boot to the root image.  You then 
have the thing run off either a live image on the CD or use a userfs-based 
program I've been thinking about.  The idea there is to "mount" a RPM or dpkg 
file as a filesystem, either mounting it using IFS or just symlink things into 
place.  That means you can run an entire system off a root image that 
virtually mounts all the packages from the original package files.

That makes the entire demo/trial thing somewhat of a moot point, since it 
really consists of only a few minor projects: live image vs. userfs package 
mounter, initrd vs. loopback from CD, and how to handle user config during 
'demo'.

> in summary, marketing wins.  i suggest doing things in the following order:
> 
> 1. make a product, and make it the real thing.
>    the top priority should be that a full-blown, fast,
>    what-linux-was-meant-to-be installation will work well.
Most definitely.  The demo/trail thing should be a very low-priroity thing, 
driven by marketing if anyone...

>    (it is our job in step 1 to make the real installation technically
>    feasible, so there is no argument for the loopback install
>    being significantly safer.)
In the overall scheme of things, doing a loopback install would be a bad idea. 
 The investment involved in doing an install to a loopback file is equivalent 
to a full install (well, almost), and if we can do a complete run-of-CD 
'demo', no one will want to do a loopback install.

> 3. extra stuff.
>    maybe some of you really dig this loopback idea, and think you can
>    code it.  if so, go for it.  more extensions to linux are always good
>    to have around.
EXACTLY!!!  Just because something may not be "official sanction" by this 
project doesn't mean it shouldn't/won't ever be done.  If you want to do 
something anyway, go for it, just keep in contact with the main group.  If you 
can "robustify" (nifty word, I invented it a few days ago... ;-) the code, it 
might indeed make it's way into SEUL eventually.

> get it integrated into distributions.
Exactly.

TTYAL,
     Omega

        Erik Walthinsen - Programmer, webmaster, 3D artist, etc.   __
  __                                                              / /\
 /  \           omega@sequent.com         Work: (503)578-5314    / /  \
|    | M E G A  omega@aracnet.com         Home: (503)281-4281   / / /\ \
_\  /_          psu12113@odin.cc.pdx.edu  Majoring in CS       / / /\ \ \
                                                              / /_/__\ \ \
Omega Station: http://www.aracnet.com/~omega/                /________\ \ \
     Info on Linux, Graphics, Descent, Laptops, etc.         \___________\/


----------------------------------------------------------------------------
Simple End User Linux Mailing list
To be removed from this mailing list send a message to majordomo@txcc.net
with the line
unsubscribe seul-project
in the body of the letter.
----------------------------------------------------------------------------