Given the microscopic volumes, I'm not sure that I would call it "momentum".... more like the drifting of a balloon. If there is a reason, I would guess that it is because there has been a general boost in optimism since the announcements at the User's Conference about next year's product and the shipping of the 8.2 release. While there are lots of question marks, people who were starting to be pretty gloomy about the prospects for any real distributed computing solution are encouraged about the promises for a new SQL optimizer gateway to the database for JBDC and the like and the [not yet too well defined] OpenAppServers. And, with the Active-X support in 8.2, some of the issues that have been troubling people about creating high quality GUI screens can now be address, albeit with a bit of work. It is still disappointing that the Active-X support didn't entend to the point that it was supposed to as per the prior User Conference. It would have meant a lot of re-write, but one wouldn't have had such an awkward mixed solution as there is now.
Speaking of distributed computing, I recently saw a PPT presentation from PRGS on distributed computing. Near the end there is a graph showing a vertical axis of cost and a horizontal axis of capability. In the lower right hand corner were a couple of workgroup class products and in the upper right corner was Forte. Progress they placed in the space in between, citing it as a opportunity market. I certainly agree with the relative ranking on the capability axis, but I couldn't help but question the cost axis. Obviously, lowering Forte to the level of Progress or below on this chart would dramatically change the interpretation. My own calculations suggest that run-time deployment at large sites, say in the 500 user range, the need to have an AppServer license for every client to do any kind of distributed computing and the likelihood of two database connections per user, one for the AppServer and one for a second session in order to get asynchronous processing, that the cost per client on this scale could easily run way over the Forte cost, especially since the Forte technology approach allows many *fewer* database connections than users. On the development seat side, a Forte seat is probably 2-3 times the cost of a Progress seat, but the total dollars here is trivial compared with the salary cost of the developer using the seat and I have no question the productivity difference would erase the purchase cost differential within a matter of weeks. |
