RE: [External] : Re: Indentation with spaces

[Drew Adams <drew.adams@oracle.com>]

> >>> Why is `indent-tabs-mode' t by default?
> >>> Tabs should not be used.
> >>
> >> The world is split between 3 factions:
> >> - those users who absolutely cannot tolerate TABs.
> >> - those users who absolutely cannot tolerate the
> >>   use of SPC instead of TAB to indent.
> >> - those users who have a life.
> >
> > Hooray! If the closed-world hypothesis applies
> > then I can now say I have a life!
> 
> I don't get it joke

If CWA holds, and we know/assume the world has
only those 3 factions, and if neither of the first
two cases holds then the third must be the case.

IOW, if it's not true that you absolutely can't
tolerate TABs, and it's not true that you
absolutely can't tolerate SPCs, then you have a
life.

On the other hand, if OWA holds, maybe something
other than those 3 cases holds...

> but
> 
>   The Closed World Assumption (CWA) is the assumption that
>   what is not known to be true must be false. The Open World
>   Assumption (OWA) is the opposite. In other words, it is the
>   assumption that what is not known to be true is
>   simply unknown. [1]
> 
> Interesting! OWA seems reasonable but how did they come up
> with CWA, when is that useful

It's simpler to reason with.  If you don't know
something to be true then you conclude that it's
false.  This is a common approach - databases,
Prolog etc.  Cf. `completing-read' with arg
REQIUIRE-MATCH = t.

Of course, one can interpret "no match" as just
"dunno" or "unproven", but often it's acted on
as "false".  And as Tomas perhaps hinted, it's
not just about classifying as true, false, and
dunno.

It's about the difference between any such
closed classification and a representation or
system that's based on the idea that both
(a) the set of stuff that's classified and
(b) the classification of that stuff both
(1) are inherently incomplete and (2) can
change.

CWA is akin to not-proven-guilty-means-innocent
(or not-proven-innocent-means-guilty).  OWA
assumes only that not-proven means not-proven.

> and what's closed about it,

It doesn't allow for the possibility that there
are unknowns.  It assumes that, at any time,
what's known to be true is all there is.

It's a useful simplification.  But it presents
difficulties wrt the nature of real knowledge
and its evolution.

> that you know what you know to be true and every thing else
> you then and by that can tell is false. So it's a complete
> state of the knowledge sphere, that's what's closed about it?

Exactly.  At any given time.  And a system that
allows for increasing or changing knowledge has
somehow to deal with non-monotonic changes in
what's known (in particular).

> [1]
> https://urldefense.com/v3/__https://www.dataversity.net/introduction-
> to-open-world-assumption-vs-closed-world-
> assumption/__;!!ACWV5N9M2RV99hQ!NmVaOzeFvXD6WCus5nkkRpZP7Fiih7FVO7UUsEX
> d-B3ZRQ6mzozldTZQvG4Re0gXUeyMdIAKE0Bpn36B$