On Sun, Dec 21, 2008 at 12:26 PM, Michael McGrady <mmcgr...@topiatechnology.com> wrote: > > On Dec 20, 2008, at 8:16 PM, Niclas Hedhman wrote: > >> On Sun, Dec 21, 2008 at 12:33 AM, Michael McGrady >> >>> I would suggest it is a very bad state of affairs when the lookup code >>> for >>> services must be included in the services. This is an architectural >>> howler >>> of the first order in my opinion. If you decide to do this, it will >>> never >>> work in modern architectures, such as SOA. This idea should be viewed >>> clearly and rejected resoundingly. >> >> And that seems to show that you don't differ between API and >> implementation, or at least argue in the wrong manner. > > Frankly, at this stage I haven't even bothered to look at the > implementation. I cannot see how your note relates to what I said, Niclas. > If there is anything in the world that I believe I know the difference > between it is API and implementation.
But you claim that JavaSpaces API depend on Jini (which most people here equates with Discovery, Join and Lookup specs)... So, then please point me to exact spot where JavaSpaces API requires "Jini" ? And Entry, Lease and Transaction are not it, since they are also independent of Jini (as defined). If it is not an API/Impl issue, is it then that the word "Jini" appears on every specification and one easily get the impression that all those specs together, and no subset, is Jini? Well, that is historical; There were no Lease specification out there. IIRC, not even JTA existed, so the Transaction spec was also written (and I happen to think it makes better sense than the JTA/JTS one)... Cheers Niclas
