Dan Nicholson wrote:
> On Dec 6, 2007 6:47 AM, Randy McMurchy <[EMAIL PROTECTED]> wrote:
>> As I answered before, "I suppose". It just seems weird to redundantly
>> give information out because we feel that folks won't read the book,
>> so the information needs to be repeated over and over. However, in
>> retrospect since so many have chimed in, it does indeed seem to be
>> a popular idea, that really doesn't hurt a thing.
> 
> Just to complete the thought, you are absolutely correct that people
> should read the whole book and not be surprised when things break
> because they skimmed it. But I think this is a spot where we can throw
> newbies a bone. There are pointers throughout the book back to the
> "Toolchain Technical Notes" for a similar reason.
> 
> --
> Dan

The difference is that the Toolchain Technical Notes are not absolutely 
needed to build an LFS system. That page helps explain details that 
would allow you to learn more about the process, but you don't actually 
*need* to know this in order to successfully build a system. This tends 
to be info that newbies are not necessarily expected to know so it's 
fine to toss in pointers to that page from other places, and at the same 
time it's no big deal if you skip over those pointers (or put them off 
until later) because you don't really need to know all of that to have a 
functioning system.

The instructions on page 5.1 on the other hand are absolutely required 
to have any hope of successfully building a system (if you don't know 
that you are expected to start from the source dir you'll be completely 
lost, and if you don't rm the source and build dirs after each 
installation you'll find that gcc is broken in chroot), so if *anything* 
what might need to be done is strengthen the wording of what's already 
there and make it clearer about how vitally important those instructions 
are (personally I think it's pretty clear already, but since many users 
*still* have trouble with it...). For starters, I've actually felt for 
some time (just not important enough to me to bother mentioning it, but 
since we are now discussing the issue I may as well...) that the "unpack 
as lfs user" is, at best, redundant...after all you are told to be the 
lfs user a couple pages earlier in the book, and you clearly don't 
unpack as the lfs user in chapter 6 (I think this should be pretty 
obvious, but I have seen at least one user asking whether they should 
also unpack as the lfs user in chapter 6). Additionally, specifying that 
unpacking be done as the lfs user might imply that unpacking is the 
*only* thing to be done as the lfs user...granted, I don't recall 
actually seeing any users who've ever thought that, but I just like to 
look at every possibility and remove any potential traces of ambiguity...

Similarly, the last part about the build instruction assuming the bash 
shell...this also seems redundant, since the lfs user environment is 
created with bash as its default shell. If that sentence exists at all, 
it would probably be more appropriate on the page that describes the 
creation of the lfs user env.

I'm also of the opinion that placing a pointer just in the Binutils 
installation page would cause some users to think that it applies just 
to Binutils, so it would be better to just make sure that the 
instructions concerning unpacking and removing sources are visible and 
clear that they apply to every package.

Well, there is more I'd like to say but I need to be off to my college 
classes...I'm sure some here might be surprised at my comments here 
considering the number of times in the past that I've said that the 
instructions are fine, but there almost always room for improvement...
-- 
http://linuxfromscratch.org/mailman/listinfo/lfs-support
FAQ: http://www.linuxfromscratch.org/lfs/faq.html
Unsubscribe: See the above information page

Reply via email to