Since my start was with IBM in 1957, and with large scale 'corporate systems', as opposed to the DEC 'departmental systems' that Ken Olsen was espousing at that time. That is the difference today, as it was 40+ years ago, possibly, C. O'Reilly, overlooks the difference between departmental foresight and corporate foresight, in his evaluations. And his forecasted demise of IBM.

Just an aside,,,,,, Let's face it,,, today's 4+ Core Processors have a Multi-Giga times performance increase in the last 40+ years,,,, However, the basic challenges of complete systems analysis and detailed design, have not changed,, but they are better understood.

PS: DEC is a tiny bit of HPQ today.

Cormac has hit the nail on the head as far as we're concerned. IBM announced their 'new' prices at the start of last year and since then we've been in battle and IBM is our three letter swearword. They have been so arrogant, believing that our system was a typical legacy system (although all our code (COBOL and C) is LAN based) and that they could milk the proverbial cow, that things are in the process of a backfire. CICS was used as a link from the LAN world to DB2, by September this year we will have definately no more CICS. The second phase is the appraisel of the Batch jobs running on z/OS (with possible migration to Intel devices). The third and final phase is the appraisel of z/OS and the z9 with a view to Dinosaur extinction. Phase one is decided, phases two and three will be dependant upon the next round of negotiation with IBM. To continue with the Dinosaur world it's looking like I'm looking for a new Job. Our costs are in Euro, and with the weak Dollar a super earner for IBM, but at this rate not much longer. Before this latest price startegy we were happy with the setup and had managable prices with top reliability. The reliabilty of Intel kit has improved so dramatically over the last few years that risks of downtime versus cost is no longer a major stumbling block.