Transcript and Dot Notation

[Dan Shafer]

Andre......

While you're not exactly wrong here, you do miss the central
point/issue. To use your example, if I'm designing a traffic system
with lots of TrafficLight objects, I need a way to create individual
instances of that object, give them identifiers, and send messages
either to the individual instances or to the class. Doing this without
some sort of notation that makes the relationships between objects
(receivers) and methods (messages sent to those receivers) is
cumbersome at best.

I don't think your suggested notation is necessarily bad and in fact
it won't surprise me if that syntax is acceptable in an OO Transcript,
*once the object instance has been created and identified*.

The other big advantage of dot notation is that it can be held as an
alternative that nobody is required to use if they don't want to use
OO in their apps. Nothing forces it. But, having said that, I'm not
sure it is possible to create a hybrid development environment in
which OO dot notation and textual freestyle exist side by side without
introducing tremendous inefficiency into the byte-code interpreter or
other mechanism for executing the application. And that is ultimately
Mark W's biggest challenge, I suspect.

On 2/24/06, Andre Garzia <soapdog at mac.com> wrote:
> Folks,
>
> taking the risk of sounding naive, why can't we deal with objects the
> way we deal with custom props?
>
> for example imagine the following Traffic Light  object with
> properties and methods:
>
> TrafficLight.stopColor             --- Red
> TrafficLight.attentionColor     --- Yellow
> TrafficLight.goColor                --- Green!
> TrafficLight.interval                 --- the interval for the cycle
> of yellow to red, for example 10 secs.
> TrafficLight.cycleInterval        --- the period the traffic light
> stays green or red before cycling, for example 45 secs.
>
> Methods:
>
> TrafficLight.go  -- Starts with go.
> TrafficLight.stop -- Go to stop....
>
> So why can't we do transcript-ish things like:
>
> set the stopColor of TrafficLight to red
>
> set the interval of TrafficLight to 20 secs
>
> and call methods like
>
> send "go" to traffic light...
>
> This would still be verbose enough to fell like transcript and maybe
> it could address the problem of transforming transcript into a weird
> lingo like language. Although I think that the parser for those
> things would be a little hard...
>
> anyway, Mark should have better thoughts than me on this...
>
>
>
> On Feb 24, 2006, at 6:14 PM, Judy Perry wrote:
>
> > What possible competitive advantage does it offer to the company
> > for it to
> > transform Transcript into yet another bit player in a very major
> > league?
> >
> > With it being an x-Talk, it offers certain advantages, such as ease of
> > learning/reading, that are all but nonexistant in your "traditional"
> > programming languages.  As such, it is a big player in a small
> > league, but
> > it's almost completely a league of its own, a league that the
> > company has
> > reported it finds profitable.
> >
> > If, as we've often discussed, Rev is unable to compete with
> > C++/Java/dot.notation.flavor.of.the.month because of its very
> > different
> > paradigm, how would making it over into just another minor OO language
> > make it more competitive?
> >
> > I've said it before and will say it again:  If true OO is what you
> > really
> > want, why not just use one of the bazillion OO languages?  Once
> > Lingo went
> > down that route, it ceased to be a learnable language for ordinary
> > humans.
> >
> > And, as for OO being OPTIONAL in Rev, remember that it was optional in
> > Lingo, too.  Only, every single Lingo book on the market dealt in
> > dot.speak, not verbose speak.  Code fragments that floated about for
> > public consumption tended to be dot.speak, not verbose speak.
> >
> > Remember the guy who not long ago wrote to the list who had problems
> > possibly with case statements and pWhiches?  What's going to happen
> > when
> > those new users have a problem and everybody responds in dot.speak?
> >
> > OPTIONAL dot.speak I fear will end Transcript's natural-language
> > orientation.
> >
> > Judy
> >
> >>>> .this.that.thatotherthing.IsThisParticularDotSupposedToBeAMethodOrA
> >>>> nObject.Sh
> >>>> ootMeNow
> >
> > _______________________________________________
> > use-revolution mailing list
> > use-revolution at lists.runrev.com
> > Please visit this url to subscribe, unsubscribe and manage your
> > subscription preferences:
> > http://lists.runrev.com/mailman/listinfo/use-revolution
>
> _______________________________________________
> use-revolution mailing list
> use-revolution at lists.runrev.com
> Please visit this url to subscribe, unsubscribe and manage your subscription preferences:
> http://lists.runrev.com/mailman/listinfo/use-revolution
>


--
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
Dan Shafer, Information Product Consultant and Author
http://www.shafermedia.com
Get my book, "Revolution: Software at the Speed of Thought"
>From http://www.shafermediastore.com/tech_main.html