[Date Prev][Date Next][Thread Prev][Thread Next][Author Index][Date Index][Thread Index]
version compare and partial orgls
- To: <mark>
- Subject: version compare and partial orgls
- From: Eric Dean Tribble <tribble>
- Date: Sun, 3 Sep 89 13:38:09 PDT
- Cc: <michael>, <us>
- In-reply-to: <MarkS.Miller'smessageofSun>,49 PDT <8909031811.AA04333@xxxxxxxxxx>
Sorry. I didn't scope properly. I agreed with this:
i.e., one tells it more about what points to it, and the other tells
it more about what it points to.
(An interesting distinction in a
doubly-linked structure. If you turn the Ent upside-down, does
inform become unify and vice-versa?)
I will support this symmetry, however. Data-inform puts more
DataObjects into the set of objects pointed to by a Stamp.
Sharing-inform puts more Stamps in the set of objects pointed to by a
DataObject in the other direction. Upside-down or rightside-up, there
is an operation which adds objects at the bottom into sets defined by
objects at the top and vice-versa.
Mi: Does it make sense to EVER freeze a DataStamp? If someone can make
a virtual copy, why can't they make a real copy, then unify it?
What am I missing?
I now agree that it doesn't make sense to freeze a DataStamp.
However, I don't see that the actual copy example bears on the issue--
An actual copy would be a copy of just the bits. Sharing-informing
the actual copy doesn't effect the identity of the original at all.
I think the point was that making an actual copy of some stuff, then
share-informing it to the original contents should be identical to
making a virtual copy of the stuff in the first place. In both cases,
the person making the copy acknowledges the originator's right to the
material, and in neither case is any change made to the identity or
data of the original material.