Strange contents of long name

[Richard Gaskin]

Peter Haworth wrote:

> There was a thread a couple of days ago about changes in the way text
> properties were handled in a recent release.  That's not three but my point
> was that the "backwards compatible" argument isn't a hard and fast rule, as
> I think you're saying with reference to the shared group message change.

But at least that brings us up to three:  text properties, shared group 
messages, and now the soon-to-be-implemented change to make long IDs 
more consistent. :)

Still, with more than 3,000 tokens in the language and a history going 
back to 1992, I think it's safe to say that RunRev does a reasonably 
good job in trying to maintain backward compatibility where practical.

In fact, the change to text property handling is an excellent example in 
that regard:

In the olden days, text properties were bound together in such a way 
that changing any single attribute would cause the others to be reset to 
their defaults.

It had been that way since the early days, for reasons known only to Dr. 
Raney (probably having to do with an unusually compact property record 
structure).

Whatever the reason, it was darned annoying to have to write code to 
deal with it, and true, now that it's fixed that code no longer serves a 
purpose and in some case may need to be changed.

But the benefits of separating text properties so outweigh the one-time 
cost of script revision that I don't know anyone who would argue that we 
should go back to the old way of handling that.

There will almost certainly be other changes down the road as well, but 
I'd be surprised if our community ever became as angst-ridden over 
version changes as Python, Drupal, and other users are.


> As you say the fix doesn't really help as an overall solution, so I am left
> with little choice but to either check version numbers and do things
> differently accordingly or use a method that will work in all versions.
>
> Judging by the response to this thread, I doubt my tool is the only one to
> be affected by this!

True, many had discovered this, but I'm not sure how many have written 
code that depends on the old way of working with IDs.   There are so 
many ways to address objects within one's own stuff that this sort of 
issue is likely to affect only generalized developer tools like yours.

--
  Richard Gaskin
  Fourth World
  LiveCode training and consulting: http://www.fourthworld.com
  Webzine for LiveCode developers: http://www.LiveCodeJournal.com
  Follow me on Twitter:  http://twitter.com/FourthWorldSys