LiveCode 4.6.1 message path and behaviors
keith.clarke at clarkeandclarke.co.uk
Mon May 23 14:14:09 EDT 2011
Thanks for that clarification Richard - and your message path article that helped me get underway with LiveCode.
Backward compatibility is always challenging, though perhaps slightly easier than predicting future compatibility!
On 23 May 2011, at 19:03, Richard Gaskin wrote:
> Keith Clarke wrote:
> > It's clear now that my question, which sought factual clarification,
> > has raised questions of LiveCode philosophy, developer perspective
> > and architecture. There is clearly no single correct answer, no
> > 'unified theory'.
> There is a sense of theory about LiveCode's language, but it's equally saddled by a richly varied history.
> The difference between "send" and "dispatch" is a good example:
> IMO (and perhaps most people's) the syntax for including arguments with "send" is beautiful in its simplicity:
> dispatch "SomeMessage" to "SomeObject" with tSomeVariable
> Send is much funkier, requiring you to send the variable name as part of the string being sent:
> send "SomeMessage tSomeVariable" to "SomeObject"
> The "send" command first appeared in HyperTalk more than 20 years ago, so when it was added to LiveCode the implementation remained the same for compatibility.
> But once RunRev identified a need for a different message like "dispatch", unencumbered by historic compatibility they were free to use more graceful syntax.
> It might be nice if "send" could be enhanced in a future version to use the "with" token to passing arguments, but with so many priorities to address this will likely have to wait.
> So in the meantime we have one of a few such anomalies in the language.
> Richard Gaskin
> Fourth World
> LiveCode training and consulting: http://www.fourthworld.com
> Webzine for LiveCode developers: http://www.LiveCodeJournal.com
> LiveCode Journal blog: http://LiveCodejournal.com/blog.irv
More information about the use-livecode