LFS-5.1pre2 space figures (ch6+)

Matthew Burgess matthew at linuxfromscratch.org
Sun May 9 03:22:53 PDT 2004

On Sun, 9 May 2004 10:39:39 +0100
Matthew Burgess <matthew at linuxfromscratch.org> wrote:

> On Sun, 9 May 2004 02:01:25 +0100 (BST)
> Ken Moffat <ken at kenmoffat.uklinux.net> wrote:
> <lots of numbers>
> Thanks again Ken,
> I'll just add a couple of comments:
> > There's a further issue in the presentation - at the start of
> > chapter 6 it looks as if anything under 1MB is shown in exact Kb,
> > then up to 10MB in units of 0.1MB, then in whole (or nearest) MB. 
> > Later, packages under 1MB start appearing in units of 0.1MB, while
> > e2fsprogs stakes a claim to 48.4MB.  Plenty of scope for editorial
> > judgement here ;)
> My editorial judgement tells me that consistency is more important
> here than nit-picking accuracy - it is after all only a guide.

To this end, I think I've decided to use the following conventions:

For packages >= 1MB we'll use whole MB measurements
For packages < 1MB we'll use measurements to the nearest .1 MB

This way all sizes are specified in MBs to maintain consistency.

Not wanting to start any flamewars, but should we be using the MB symbol
in this case, or should we bow down to the IEC's SI standard, which
specifies MiB for binary measurements? (see



More information about the lfs-dev mailing list