[lsb-discuss] LSB conference call agenda (Tuesday, January 30, 11am ET) (and LSB 3.1 Update 1 status update)

[Wichmann, Mats D]

>I have looked at it and I'm a bit confused...
>
>Do I understand correctly that the errata refer to the LSB 
>specification?

Yes

>The problem is that the LSB specification, as published on the 
>web, does not contain any documentation which seems to be 
>referred to by the errata. The LSB specs contains only lists
>of functions without any explanation, only with 
>references to SUS, POSIX, and other specs, but no 
>documentation (which, by 
>the way, means it is unusable by itself unless I buy all the 
>fat books of POSIX, SuS, etc. :-( ).
>
>But the errata document seems to contain fixes for the 
>documentation. So, is it in fact meant as errata to the 
>POSIX or SUS documentation?

It is only errata to the LSB specification text.

The LSB does not duplicate text that exists in other
specifications.  I agree with you that this makes it
somewhat harder to use, but there are copyright issues
along with many others (the amount of work, etc.) that
prevent just wholesale importing from other published
works.

However, there is not always another specification to
refer to, and in this case the master specification is
"LSB" and in this case the text is inline; and in other
cases a function may have changes from the base 
specification, in which case that delta is inline in
the LSB.  If you look a little deeper I think you will
see this text.

For example, look at this section and click the next
button a few times:
http://refspecs.freestandards.org/LSB_3.1.0/LSB-Core-generic/LSB-Core-ge
neric/libcman.html

The errata text is generally constructed by comparing
the published text version of the spec with a newly
constructed one with the changes applied, so that may
be a more interesting reference:

http://refspecs.freestandards.org/LSB_3.1.0/LSB-Core-generic/LSB-Core-ge
neric.txt