terminological obscurity

["Martin v. Löwis"]

Arthur wrote:
> Is the random set of, say, 10 human beings homogeonous or
> hetereogenous?

If they are random, than yes, they are homogenous as they are
all human, and no, the are heterogenous, because they are
random, i.e. they share no features.

> Or is that a meaningless question, as the answer depends purely on
> context?

Not purely (as you can see above), but typically. You need some
application context to know whether a set of objects is homogenous -
with respect to properties that the application cares about.

> Are objects that have hundreds of attributes each, and that conform as
> to a single attribute, which attibute accounts for there presence
> together in a list, meaningfully homogenous - outside of the fact of
> their presence together in the list?

Likely: no. It may still be that there is external state that makes
them homogenous. E.g. they may all act as keys to some database, and
be kept in a single collection so the application can find all items
associated with the keys.

Regards,
Martin