To: bigredfreak who wrote (7600 ) 6/23/1998 3:07:00 AM From: Tumbleweed Read Replies (3) | Respond to of 19080
Comments on "bigredfreak" NB I work for Oracle, but I dont think you'll find my comments below unfair. I think one of the reasons Oracle thinks client server is dead is that they've never been able to successfully implement a solution for performs well ... but their competition sure has. Absolutely not true. There are plenty of large client server ORCL examples out there, and ORCL holds the record for just about every large TPC/C benchmark (which is client server) by a long way. But the Internet changes EVERYTHING. The reason client server is dying is it's too expensive to maintain. For any one company, keeping thousands of PCs up to date with hardware and software is fantastically expensive. There are various estimates of $8K - $12k per annum per PC . Now, move that model to the Internet, where its potentially 10's thousands of terminals in an Extranet, and millions on the Internet. Client Server wouldnt even begin to cope, even if you could afford it.Also, the NC architecture is actually client server ... it's just using different, but similar, technology. It'll be very interesting to see how it pans out. It could be really bad or really good for them. The NC architecture is absolutely NOT client server! "Its the Internet, stupid". (To avoid the longer description) MSFT, IBM, IFMX, and SYBS hitting them on the DBMS side and SAP, PSFT, BAANF, and JDEC hitting them on the apps side ... they have a very tough road ahead of them. MSFT may be producing pricing pressure at the low end but cannot even try and compete at the middle level and upwards. (Or at the very low end either, for example Windows CE). IBM,well, I am sure they are good competitors at the high end. IFMX and SYBS!! Give me a break. Its not even worth answering that! SAP, PSFT etc. All hard competitors, . But what business doesn't have competitors? And ORCL is number 1 or 2 in market share in all the application surveys I have seen. JoeC