Re: Proposals to vote on related to 'event': deadline Dec 14th midnight GMT

On Dec 15, 2011, at 12:19 AM, Jim McCusker wrote:

> On Wed, Dec 14, 2011 at 6:48 PM, James Cheney <jcheney@inf.ed.ac.uk> wrote:
> I think "instantaneous event" is clearest among options so far.  It may be clumsy to always say "instantaneous event" - instead it might be easiest to say up front (and wherever there is potential for confusion) that we consider events to be instantaneous. The term "durative" can be used for events that have a duration.
>  
> This is the current language, but there are a number of us that object to it, since natural events are durative.

Do you mean "some" natural events are durative or "all" natural events are durative?  In either case, that is the point of my suggestion that we say "instantaneous" whenever necessary to disambiguate.

> The recording of an instant as having marked that event might be instantaneous, but superseding "event" with this special, artificial case isn't a good thing. We can just as easily use the term "event" to denote what we now call "activity", where we say that an event is a temporally extended perdurant, which would preclude misunderstandings from the process algebra community.

I actually don't think the fact that this is the way the term is used in process algebra is a persuasive argument (for or against) using it here. 

I think the same objection (that some people use a given term, e.g. entity, activity, etc. in a different way) could be used to argue against using any of the terms we are using.  For example, the way we use term "complement" is far less connected to any common use of the term.  We should avoid completely counterintuitive terminology (like complement, IMHO) but we we shouldn't contort ourselves to avoid using terms that have multiple meanings, as long as we make clear which meaning we mean.

Why not use the term "event" for both and allow both instantaneous events and durative events (=  activities) and disambiguate (only) as necessary?  i.e. an instantaneous event is just one whose start and end time are equal?  (perhaps this runs into the granularity issue though, haven't had coffee yet.)

--James