incubator-flex-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Franck Wolff <franck.wo...@graniteds.org>
Subject Re: Proposal for contributing code from GraniteDS
Date Wed, 28 Nov 2012 19:35:49 GMT
Alex,

Sorry for the delay, I have been very busy last week with the new GraniteDS
release, among other things.

My answers below:

2012/11/14 Alex Harui <aharui@adobe.com>

>
>
>
> On 11/14/12 2:57 PM, "Franck Wolff" <franck.wolff@graniteds.org> wrote:
>
> > Alex,
> >
> > My answers below:
> >
> > 2012/11/14 Alex Harui <aharui@adobe.com>
> >
> >> OK, I skimmed through the files.  Code-wise, it looks very nice, so
> thank
> >> you for being willing to donate this work.
> >>
> >> One thing that was bugging me as I was looking at the code is this: Did
> you
> >> look at the Java code for the Java equivalents when writing your code,
> are
> >> any portions of your code "substantial copies" of the Java sources, and
> if
> >> so, does Java's source license allow you to leverage their code, and if
> so,
> >> without attribution?  I would think there are BigDecimal equivalents in
> >> other platforms as well so this must all be "ok", but I think we need
> to be
> >> sure.
> >>
> >
> > Good point. All classes from the org.granite.math package were written
> from
> > scratch, but part of the BigDecimal code is inspired (not copied though)
> > from the Java implementation. What do you suggest in order to make sure
> > that there is no legal issue?
> If you've never seen their code, I think we are good.
>


Well, I did see their code but I didn't copy it and I don't think this a
problem. However, you can double-check upon this potential issue.



> >
> >
> >> Some other questions:
> >> 1) Terminology/Names: I saw a reference to "Bean" in one of the files.
> >>  That
> >> is also a Java concept, but is that term trademarked?  I think
> "JavaBean"
> >> is, but I think it would be hard to trademark just "Bean".
> >>
> >
> > I don't think Bean is a trademark and we can change that term if
> necessary.
> OK, I think we are ok here as well.
> >
> >
> >> 2) Packaging:  All the packages are currently org.granite.*.  Do you
> have a
> >> large body of users who will complain loudly if you change the package
> >> name?
> >>
> >
> > I don't think so: changing package names is just a research / replace
> > operation...
> OK, we'll I'm open to suggestions for package names and the swc name.  I
> think it shouldn't have granite in it because it is your company's name. So
>     org.apache.flex.math
>     org.apache.flex.reflect
>     org.apache.flex.validation
> And for the swc, other than the Container dependency and ObjectUtil
> dependency, I didn't see a lot of other Flex dependencies so I think all of
> these classes can go in one new swc?  I don't see a reason to put it under
> "experimental".  I considered names like "asjava.swc" since these are all
> based on java features?
>


Ok, I can be whatever the community wants.



>
> >
> >
> >> 3) I noticed FormValidator had a reference to mx.core.Container.  That
> ties
> >> you to the "mx" project in SVN.  Does it work with Spark Form?
> >>
> >
> > Yes, it works with Flex 4 and Spark Form. The explicit use and import of
> > mx.core.Container could be changed to something based on reflection if
> > necessary.
> This is not important for the donation, but Container and ObjectUtil bring
> in a huge set of dependencies which will not be good for smaller apps.
>


We can get rid of this dependency later.

Franck.



> >
> >
>
>
> --
> Alex Harui
> Flex SDK Team
> Adobe Systems, Inc.
> http://blogs.adobe.com/aharui
>
>

Mime
  • Unnamed multipart/alternative (inline, None, 0 bytes)
View raw message