I've had a quick look and i think should be possible to support both
approaches, i'll go give it a try. Once the sca.js approach is working and
the old scaDomain.js is used we could log a warning message saying its
scaDomain.js is deprecated.

   ...ant

On 10/25/07, Simon Nash <[EMAIL PROTECTED]> wrote:
>
> OK, this member of the community will bite :-)
>
> Now that we have released 1.0, we should not break compatibility with
> user applications or user extensions without a very good reason.
> We should always try to deprecate previously supported APIs and keep
> them working rather than disabling them.  Is there any way to
> keep the old applications using SCADomain.js running, while supporting
> and recommending the new approach using jsonrpc.js?
>
>    Simon
>
> ant elder wrote:
>
> > On 10/22/07, Luciano Resende <[EMAIL PROTECTED]> wrote:
> >
> >>The upgrade process is very harmless, two line of code, one for the js
> >>reference and another for the reference declaration. Also, seems like
> >>there are some bugs on the scaDomain.js [1] that would not happen
> >>while using the manual reference. I also think that, having the two
> >>very similar bindings will make confusion and other maintenance
> >>headaches. Have said that, and as I'm working towards getting the
> >>web2.0 References working soon, I'd like to keep this as one binding,
> >>but I'm open if the community feels otherwise.
> >
> >
> >
> > Doesn't look like Mr Community is answering...Trying to maintain
> backward
> > compatibility where possible is important. It doesn't matter that its a
> > "harmless two line change", if some guy upgrades from Tuscany 1.0 to 1.1and
> > his applications don't work any more then that is a bad user experience
> > which we should try hard to avoid. Is there a reason the scaDoamin.jscan't
> > work anymore? If there is a reason then a separate new binding seems
> better
> > to me just so we can avoid breaking anyone.
> >
> >    ...ant
> >
>
>
>
> ---------------------------------------------------------------------
> To unsubscribe, e-mail: [EMAIL PROTECTED]
> For additional commands, e-mail: [EMAIL PROTECTED]
>
>

Reply via email to