I N T E R O F F I C E M E M O R A N D U M Date: 16-Oct-1996 01:54pm CET From: Arie De Groot GROOT.ARIE.DE Dept: OEM Business Segment Tel No: +31 30 283 3166 DTN 838 3166 TO: See Below Subject: Development of SCSI and PCI-drivers for Philips Medical Systems Gentlemen, I just returned from my holiday and am trying to catch up on my work backlog and find a lot of mails on the issue mentioned in the header of this Email and have to deal with the fact that the original assigned project leader has left the company. It seems that the development of the target mode for the KZPBA/KZPSA, running under OpenVMS caused a lot of commotion, especially since it is not clear who is who, and who is responsible for all this. So let me explain the situation and try to work put the solution with you. I will answer all your requests seperatelly as soon as I have answers to your questions. My name is Arie de Groot and I am responsible for the OEM relation with Philips Medical systems in the Netherlands. This company manufactures and sells medical equipment all over the world. In this special case we are talking about MRI-scanners (300 per year, growing to 500 in a few years) that contain several computers to perform different functions, one of them being an AlphaServer running OpenVMS. We are about to take over the position of one of the other computers. To do so we have to support initiated target mode on the KZPSA/KZPBA under OpenVMS. Since this is not a standard option we discussed this problem during the customer visit in May this year. This meeting was attended by several OpenVMS engineers, and Glenn Everhart was there especially for this issue. During the discussions he did an excellent job by pointing out this this was certainly possible, even not to difficult, by modifying one of the existing drivers and add this functionality. He also indicated that it couldn't be done by OpenVMS engineering because this fuction was not very high on the priority list for OpenVMS. We agreed that, when we created this functionality in the proper way (like OpenVMS engineering would do, using the same standards of quality) and offered it (for free) to this group it could easily be embedded in OpenVMS so it would be available for all users. This scenario was also very attractive to the storage people that could do very well with this new feature. So a lot of people would benfit from this project. Glenn offered remote support to make sure the proper procedures were followed end that the product would meet the engineering standards. Over here we started a project group with Robbert Kuppens as a knowledgeable project leader, contracted the Moscow Software Center to build this SCSI-driver, and assigned quality assurance to Jur vd Burg (acceptable for engineering). We agreed with the customer that we would build the extra functionality in the SCSI-driver for OpenVMS and hand it over to engineering so they could use it and embedded in the next planned release. We also agreed that we would build the same for Windows NT but couldn't embed it in the OS since this wasn't ours, so the maintenance responsibility was at the customers'. Added to that we agreed to build a special driver (also OpenVMS and WNT) for a PCI adapter that would interface between the 2 computers in the medical scanner configuration, but this would be handed over to the customer after acceptance. This is of no importance to the SCSI-project, but references to it might pop-up in mails on the SCSI-project, so you know about this upfront. So I am the linking pin in this project between Customer, Moscow Software Center, OpenVMS engineering and all other involved parties (at least until a new project leader is assigned). Hope this clarifies the situation and I will respond your mails soon. Kind regards and sorry for the confusion, Arie Distribution: TO: _STAR::ST_LAURENT TO: _SUBSYS::KERR CC: _MSCDEV::OSIPOV CC: Robert Boers @GEO CC: TOGT CC: _SUBSYS::BODEMANN CC: _SUBSYS::DDUVAL CC: _ROW::EVERHART CC: Arie De Groot ( GROOT.ARIE.DE )