root servers list changing?

[Matthew Hannigan]

I've found a 3rd edition and had a reread.
I feel I should give you a reply so here goes.

Cricket Liu wrote, a long time ago:
>>
>>Going from memory ... (sorry, I don't have handy -- I've probably
>>lost it.)  It had a tendency to use assumptions
>>and jargon.  A tendency of many technical books; that's
>>nothing particular to this book really.  It's a common
>>comment about technical books that easy to follow once
>>you know the subject.
> 
> 
> Well, I'm not exactly sure what you mean.  If you mean DNS jargon,
> yeah, it absolutely uses a lot of jargon.  I would count any book on
> DNS as remiss that didn't introduce you to a lot of DNS jargon.

I think the bit of jargon I particularly didn't like is
the root cache file.  It's called a cache file but it isn't
really a cache file, as far as I can see.  Also, it has many
names:
    db.cache (in the book)
    named.root (from rs.internic.net)
    named.ca    (another variant I think I've seen)
    root.cache
    etc...

>>One particular frustration that I remember I had was that
>>it didn't have an example of the primary file for the
>>localhost/loopback domain.
> 
> 
> Like the one on p. 66 of the fourth edition?  :-)
> 
> It's been there since at least the second edition, which I have a copy
> of with me, and it's probably in the first.  If it is, it's in Chapter 4.

Yeah that one :-)   The problem with it is that it's not crystal
clear what should change in it for different machines/servers.

It's probably worthwhile describing this particular case to death.
And the thought of some remote machine machine being the authority for 
your own loopback address name is a little bizarre.  I know it's 
presented as a fallback, but still.

Matt