Re: Oracle and Partner Push "Informix to Java" Migration

On Mon, 27 Jul 1998 23:52:09 GMT, Mike Segel <mikey@segel.NOSPAM-.KING.OF.MYDOMAIN.-NOSPAM.com> wrote: >The intergration of Java in to the engine seems to have taken >a back burner. I didn't go to this >years conference so I don't know if Informix did mention anything about >J-SPL in 7.X product line. Your estimates of what Informix is or isn't doing ought to be based on a little more information than your own hopes or feelings. If you had gone to the conference you could have known that they are hard at work with integrating Java into the engine. It will be in the version they now call 9.2 which is the version where the current 7.x and 9.x code lines are fully merged. I don't remember exactly when they would have the first version available, but sometimes this autumn/winter. It will be a great product will all kinds of superb facilities for runing Java in the engine both as Java objects and for creating datablades. Quite exiting because it will open a large number of new oportunities for developers. It will also not be dependent on one particular Java Virtual Machine, but can be relatively easily integrated with different ones. I wouldn't expect an option for anyone but Informix to do such an integration, at least not initially, but it does mean that as better JVM's become available, or simply new versions, it will be simple to integrate them. >Informix was doing some Java stuff early on, however, nothing was >formally stated until the >Java One conference. They did talk about the same thing on last years user conference, and they have worked with it in the meantime. The Java One announcement was only the first formalization of it. If you now claim this has taken a long time, you might want to look at what have happened to Java itself in the meantime. To come out with anything to run inside the database engine before the timeframe they now plan would have been a big mistake. Only now is the Java technology required becoming available. > Also some of those early projects appeared to have >been cut under Bob in >an effort to stem costs. If you had looked a little closer (may be on their R&D figures) you would have seen that they didn't cut much costs in their development department. Those things that where cut should where done so properly. Appart from all the work on Java inside the engine they are also doing significant efforts on JDBC drivers. Particularly type 4 JDBC drivers according to the JDBC 2.0 spesification so it also supports SQL 3 which means it will properly support the Universal Data Option. If you look back a little it wasn't particularly clear that JDBC would ever support the Universal Data Option fully. Informix therefore had to start a project for creating their own class library for this purpose. The fact that this is no longer planned to be enhanced (it's going to be supported though as long as there are users) is nothing but natural and only a result of our fast moving world, not something Informix could have done something about. If you remember something they called Data Director for Java, that product was never sold. If you ever saw a demonstration of it (I did) you would have understood why. That product is properly cut totaly from their R&D budget. To me this was the type of error that any good R&D department will do from time to time. Hardly anything to talk about. To me this is a conserted and long term planned effort within the Java arena on exactly the right types and forms of products that Informix need to come out with. Good work with other words. Nils Myklebust NM Data AS Norway E-mail: Nils.Myklebust@nmdata.com FAQ at: http://www.iiug.org/techinfo/faq/faq_top.html (Now with ODBC info under "Third party products".)