> Remember, IDs can just as easily be things like "32oii", "r23t584hf",
> and so on.  Their just a hook on which to hang all the properties we
> actually care about.

Yes.

One property I care about for an item, is having a unique identifier
for it, to use in permalinks and the like, interfacing with the
outside world (external to Exhibit). The id property, or perhaps
attribute (if the name invokes unpleasant feelings), seems to me to
have the same function, inside Exhibit.

To me it feels more natural to have access to it from Exhibit's data
model, than to keep duplicates of the same bits in sync manually to be
able to generate these URLs in the HTML I use to interface with humans
and the external world.

I understand that we want to help people embrace data models that have
proven more useful than others, and having a unique identifier is
probably one of them. I do not challenge that. I also agree that it is
good practice not to mix data and system semantics, but I do not
believe that it helps any cause protecting an item identifier we force
users to keep unique, from being used, in a situation that calls for a
unique id, when we have one.

I agree that the example I gave was not a good example of one and that
it is bad practice to do what I wanted to. But I also believe that the
"your data, your mess, your business" principle is a very strong thing
to have working for us in getting users to adopt a tool like Exhibit,
and that we abandon that by locking away id:s out of reach.

My original question stands; I still do not understand why id:s are
write-only data and why we prevent introspection. I want to understand
why, if there is some reason, so I can stop feeling that it is an
oversight or a bug that ought to be fixed, as I do now. I'm too much
of a geek to settle for the suggestion I remodel my datasets to work
around the issue because some authority says I should, or avoid the
question by suggesting additional labour of mine to work around
limitations imposed by the system. I *want* to understand, if perhaps
for no other reason than to be able to pass the same information on to
others who think the way I do, running into the same issue, as they
eventually would.

I feel a slight spark of passion in this discussion. That is a good
thing and a great sign. For one thing, of that I care about Exhibit,
and using it at optimum efficiency to handle my data for me. You
should all be very proud about that, and I am sure many will follow
me.

-- 
 / Johan Sundström, http://ecmanaut.blogspot.com/

_______________________________________________
General mailing list
[email protected]
http://simile.mit.edu/mailman/listinfo/general

Reply via email to