Anne, I was thinking in terms of the basic middleware providing cross-enterprise connectivity, but as you observe it does not really affect your answer. Now if one were to use a language-agnostic implementation of tuple spaces....
Gervas --- In [email protected], "Anne Thomas Manes" <[EMAIL PROTECTED]> wrote: > > Gervas, > > [What do you mean by "ESB"? Are you talking about an infrastructure > comprising many products that supports SODA? Or are you talking about a > specific product? My answer is the same either way, but it helps to be clear > on this point.] > > An ESB should support interactions using J/JS-like technologies, but I would > not recommend building an infrastructure that uses J/JS as its core > foundation. From my perspective, J/JS is too Java-centric to form the basis > of an open, platform-neutral integration bus. That said, I think that JMS is > an inappropriate basis for an ESB for the same reason. > > Anne > > On 3/13/06, Gervas Douglas <[EMAIL PROTECTED]> wrote: > > > > Anne, > > > > If someone were to design an ESB from scratch, do you think J/JS with > > its decoupled capabilities would make a good basis? > > > > Gervas > > > > --- In [email protected], "Anne Thomas > > Manes" <atmanes@> wrote: > > > > > > Eric, > > > > > > My comment was to William rather than to IONA. There's a lot of > > interesting > > > technology in Jini/JavaSpaces, and anyone that desires to be an > > expert in > > > distributed computing technologies should at least learn the basics > > of the > > > system. Therefore I recommend to William that he spend some time > > learning > > > about it. > > > > > > From IONA's pespective, I don't think you should invest development > > dollars > > > into developing a J/JS plug-in for Artix unless you have customers > > that are > > > asking for it. > > > > > > As for Celtix, if anyone needs/wants a J/JS plug-in, they can build > > it -- > > > such is the beauty of open source. > > > > > > Anne > > > > > > On 3/5/06, Eric Newcomer <e_newcomer@> wrote: > > > > > > > > Hi Anne, > > > > > > > > I think you're familiar with what we're doing with our > > > > open source ESB, Celtix, or at least I'm sure you are > > > > familiar with our architectural approach. > > > > > > > > Are you suggesting it might be of interest to us to > > > > investigate Jini/JS in the context of developing a > > > > plug in for it? > > > > > > > > We are doing a CORBA plug in for Celtix at the moment > > > > but of course we are always interested in > > > > recommendations for other plug ins to consider. > > > > > > > > Thanks, > > > > > > > > Eric > > > > > > > > > > > > --- Anne Thomas Manes <atmanes@> wrote: > > > > > > > > > Jini/JavaSpaces is quite different from CORBA. It's > > > > > not at all a > > > > > language-specific reinvention of the same old thing. > > > > > It's worth your time to > > > > > investigate it. > > > > > > > > > > Anne > > > > > > > > > > On 3/3/06, William Henry <william_henry@> > > > > > wrote: > > > > > > > > > > > > Thanks Gregg. I'll have a deeper look at that > > > > > site. At first glance > > > > > > this looks like a good environment for a > > > > > technology like Jini. I have > > > > > > a friend that will be very interested in this > > > > > article. What's > > > > > > interesting is the real time (ish) performance. My > > > > > friend has been in > > > > > > this industry (high performance car data > > > > > collection. monitoring and > > > > > > test and configuration etc) for years and can't > > > > > imagine moving from > > > > > > embedded C to Java. I'll have to have him look at > > > > > this - he'll be > > > > > > interested. I'd also be interested in performance > > > > > metrics and > > > > > > performance satisfaction metrics for this example. > > > > > > > > > > > > > I continue to be saddened by the loud voices of > > > > > ignorance that push > > > > > > > away > > > > > > > technologies such as Jini which have so much to > > > > > offer for those > > > > > > > motivated enough > > > > > > > to look. The problems of SOA and related > > > > > distributed computing are > > > > > > > not simple. > > > > > > > The Jini platform was put together by people > > > > > who understand the > > > > > > > science. > > > > > > > > > > > > BTW many of us non-Jini-user folks understand good > > > > > science too. I'm > > > > > > not pushing it away - I just think that it has > > > > > it's place and limits > > > > > > like all technologies. > > > > > > > > > > > > > Your interest in native only integration is a > > > > > little > > > > > > > alarming. To me, it says that your systems are > > > > > living on the edge of > > > > > > > performance, or software development > > > > > platforms/skills are limiting > > > > > > > your choices. > > > > > > > > > > > > Money, licensing, existing incumbant technology, > > > > > existing high > > > > > > performance, mission critical and > > > > > leave-well-enough alone > > > > > > applications are often limiting my customers > > > > > choices. My customers > > > > > > can't just throw out everything they have and > > > > > rewrite their > > > > > > application in the next new fad of the month. > > > > > They've already > > > > > > invested millions in these applications and are > > > > > doing $billions of > > > > > > business with them. Nor can they insert a wrapper > > > > > layer that will > > > > > > impact their high performance mission critical > > > > > systems - so they want > > > > > > native integration to a standards-based outside > > > > > world. In other words > > > > > > they don't want service enabling technology to > > > > > have a negative impact > > > > > > on their service. (Hence why my customers also > > > > > tend to look more for > > > > > > smart end-point integration than having to use hub > > > > > technology - this > > > > > > may seem contradictory but it is not). > > > > > > > > > > > > Thanks for the link. I promise to make myself less > > > > > ignorant of Jini. > > > > > > I have to say this: from my first look at Jini - > > > > > it looks a lot like > > > > > > a reinvention of CORBA but without the language > > > > > independence > > > > > > obviously. In fact it looks entirely like a Java > > > > > specific reinvention > > > > > > of CORBA. (Which already has a Java language > > > > > binding. Why reinvent > > > > > > this???) > > > > > > > > > > > > In fact when I go back to the F1 example I wonder > > > > > why someone > > > > > > wouldn't use an embedded C CORBA ORB to solve the > > > > > problem? Like other > > > > > > part so the automotive industry are now doing. ?? > > > > > > > > > > > > William > > > > > > > > > > > > On Mar 2, 2006, at 10:30 AM, Gregg Wonderly wrote: > > > > > > > > > > > > > William Henry wrote: > > > > > > >> Okay has anyone else had enough of this? Is > > > > > this the Jini list or the > > > > > > >> SOA list? > > > > > > > > > > > > > > William, I'm sorry to strike a sour note with my > > > > > reponses. I've > > > > > > > sat back and > > > > > > > watched this list for some time and have seen, > > > > > as Gervas pointed > > > > > > > out, many > > > > > > > different themes passing through. I am > > > > > particularly concerned > > > > > > > about the > > > > > > > re-creation of technologies by the ignorant. > > > > > So, I am sitting here > > > > > > > trying to > > > > > > > communicate which "features of desire" for the > > > > > week, are already in > > > > > > > place in the > > > > > > > Jini platform. I've probably (some of the > > > > > responses to yours say > > > > > > > clearly) done > > > > > > > a horrible job of providing a convincing > > > > > argument. > > > > > > > > > > > > > > I continue to be saddened by the loud voices of > > > > > ignorance that push > > > > > > > away > > > > > > > technologies such as Jini which have so much to > > > > > offer for those > > > > > > > motivated enough > > > > > > > to look. The problems of SOA and related > > > > > distributed computing are > > > > > > > not simple. > > > > > > > The Jini platform was put together by people > > > > > who understand the > > > > > > > science. The > > > > > > > fact that the platform doesn't come with a web > > > > > services endpoint, > > > > > > > and that there > > > > > > > is not one visible in the Jini community, should > > > > > speak to where the > > > > > > > real needs > > > > > > > are. There are, of course other technologies > > > > > that are starting to > > > > > > > reveal an > > > > > > > understanding of the science, and the needs of > > > > > the users. > > > > > > > > > > > > > > Jini has no lack of ability to provide > > > > > performant solutions that > > > > > > > interface to a > > > > > > > wide range of entities. Your interest in native > > > > > only integration > > > > > > > is a little > > > > > > > alarming. To me, it says that your systems are > > > > > living on the edge of > > > > > > > performance, or software development > > > > > platforms/skills are limiting > > > > > > > your choices. > > > > > > > > > > > > > > Look at > > > > > > > > > > > > > > > > <http://developers.sun.com/learning/javaoneonline/2005/coolstuff/ > > > > > > > TS-1915.html> > > > > > > > for an example of a performant system created to > > > > > interface with low > > > > > > > level > > > > > > > devices. It provided a winning advantage to the > > > > > F1 team. > > > > > > > > > > > > > > Gregg Wonderly > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > Yahoo! Groups Links > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > Yahoo! Groups Links > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > === message truncated === > > > > > > > > > > > > __________________________________________________ > > > > Do You Yahoo!? > > > > Tired of spam? Yahoo! Mail has the best spam protection around > > > > http://mail.yahoo.com > > > > > > > > > > > > > > > > > > > > > > > > Yahoo! Groups Links > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > Yahoo! Groups Links > > > > > > > > > > > > > > > Yahoo! Groups Links <*> To visit your group on the web, go to: http://groups.yahoo.com/group/service-orientated-architecture/ <*> To unsubscribe from this group, send an email to: [EMAIL PROTECTED] <*> Your use of Yahoo! Groups is subject to: http://docs.yahoo.com/info/terms/
