Open Source (was Don't you just wish Rev would do this?)

David Bovill david at openpartnership.net
Thu Jun 7 20:21:51 EDT 2007


On 08/06/07, Trevor DeVore <lists at mangomultimedia.com> wrote:
>
> On Jun 7, 2007, at 3:34 PM, David Bovill wrote:
>
> > Not sure - but whats missing from this:
> >
> >> Adobe plans to release all of the components of the Flex SDK
> >> needed to
> >> create Flex applications, including the Java source code for the
> >> ActionScript and MXML compilers, the ActionScript debugger, and the
> >> ActionScript libraries that make up the core Flex framework. Adobe
> >> Flex
> >> Builder, the Eclipse-based IDE, is not part of the open source
> >> announcement.
> >
> > Sounds pretty comprehensive to me?
>
> That is quite a bit of stuff but it is missing Flex Builder (the
> development environment),



Flex Builder is there and so are the compilers - I am not sure what the data
services are... but in general I agree that they have carefully chosen to
keep hold of some strategic parts of the picture while open sourcing enough
to keep them in the good books of the  community. I still don't quite get
what is being held onto with the latest moves though.
.

> Does this make sense?



Yes it does and I totally agree.  If the CGI engine were open you could look
at that. RunRev could retain the copyright and dual licence it, and if they
asked for the copyright on all the new submissions - then they could look to
incorporate any bits that worked for them into the standalone engine. I
remember Scott Raney saying that the engine was basically object oriented
ages ago, and that he had to drop plans to take it further as there was no
demand back then.

One way of thinking about it is to have the ability to create language
wrappers around otherwise obscure syntax of other langauges and frameworks?
I've been trying to do that with web services by creating objects and
referring to properties of the object. Having (global) objects that do not
require GUI elements would help a lot and simplify the syntax too.

Would there not be a path to do this which builds on Andres work and uses
socket or pipes to existing frameworks in the short term - perhaps using the
.NET DLR stuff to create a language parser in the longer term?



More information about the use-livecode mailing list