Re: Diversity and offensive terminology in RFCs

[Andrew Sullivan <[e-mail]>]

On Thu, Sep 20, 2018 at 02:11:03PM +0100, Anton Ivanov wrote:
> A follower is following the leader because it has agreed to be lead. As
> correctly noted by Stewart, in a master/slave hardware (or software)
> configuration nobody asks the slave about its opinion on things.

Well, ok, but in (e.g.) the DNS, the slave actually is in the position
to refuse the zone transfer on the grounds that it is malformed.  This
happens sometimes, so in fact in at least one prominent use of these
terms in the IETF the absolute power of the "master" is not true.  I
am aware of other examples, too, so if the justification is the
absolute slavishness (if you'll pardon the word) of the putative
slaves, I think the description isn't quite correct anyway.

For whatever it's worth, I have no opinion about whether this bit of
language needs to change, though I do somewhat prefer the
"primary/secondary" usage.  Note that these terms denote a logical
relationship, not a temporal or existential one, so there can be
multiple primaries (in some cases) and multiple secondaries
(frequently).  It does seem to me that, if one wishes to communicate
with others, it's often a good idea to avoid causing them offense.

Best regards,

A

-- 
Andrew Sullivan
[e-mail]