Glenn C. Everhart, PhD. Digital Equipment Corp ZKO3/4T35 110 Spit Brook Rd Nashua NH 03062 To Whom It May Concern: I have been involved with editing the VAX or VMS SIG tapes (a compilation of code contributions by DECUS VMS SIG members; the name of the SIG was originally the VAX SYSTEMS SIG, and was changed to the OpenVMS SYSTEMS SIG, SIG=Special Interest Group, a couple of years back to better match our interests) since about 1985. Since that time I have had the job of writing the overall tape abstracts, which has meant I have to look over what was presented for sharing, and have noted that there are some individuals whose contributions show particularly high levels of expertise. DECUS, and the SIG, exist to promote sharing of knowledge and to help one another solve problems. The SIG tapes have been a primary mechanism for sharing software we write since 1977, although network distribution has become more important than media exchange over the last several years. To include as many as possible, we have kept both mechanisms alive. Since 1993 or thereabouts, I have noticed that Brian Schenkenberger has made significant contributions to these tapes, and has fit the description of "expert" in writing VMS code, particularly Alpha VMS code, very well. His package entitled SYMBOL on the Spring 1994 sigtapes in directory [VMS94A.TK] was the first piece of code on the sigtapes to describe in detail, with working examples, how to set up a non - driver loadable VMS executable image and use it to communicate with other processes. (These tapes can be obtained from the DECUS library on CDs mastered at the time; SYMBOL files were actually created on April 1, 1994.) His articles on how to create loadable exec images from Digital Systems Journal appeared on the same SIG tapes, as did his tool to create permanent DCL symbols (i.e., which could not be deleted.) He also contributed his CTRL-VERIFY package to the SIG on the Fall 1994 sigtapes, a useful demo of keeping things in P1 space. (The material a set of SIG tapes comes from a period starting several months before the symposium to a short time after, but is dated with the date of the symposium for convenience. In particular, the DSJ article appeared in the January - February 1993 edition of that Journal. His article about the uses and nature of XQP locking (used for file system internals handling) appeared in the Nov-Dec 1993 issue. This expertise is necessary for dealing with cluster or network file sharing issues in VMS.) The techniques he pioneered existed within VMS itself, but it was Brian who first taught them via our software sharing mechanism to the rest of the world. These packages showed great expertise in Alpha VMS systems coding and in taking concepts and techniques from VAX VMS and implementing them on Alpha VMS, where in the cases he handled, the techniques commonly used needed to be altered. The following has to do with my acquaintance with Brian, which dates from after his first symposium tape submissions. I did not become acquainted personally with Brian until somewhat later than I "knew" him by his work. I had ported my virtual disk drivers VDdriver and FDdriver, which I wrote for VAX VMS back in the 1980s (along with a number of others including striping, shadowing, journalling, memory, and remote disks, all of which can be found on older VAX SIG tapes dating from about 1987 on) and had ported these to Alpha VMS 1.0 as part of a DECUS effort to get some DECUS code ported to Alpha by the time of introduction. When the Step 2 driver interface appeared, Digital published some preliminary porting documents on the Internet and I promised the people in VMS Engineering who I had dealt with before that I would get these drivers ported to the step 2 interface once I had access to a test machine. Unfortunately my test machine access was delayed...it was not my personal property and its owner needed to keep it at the older release ... so I used the preliminary documents and, with the help of a couple folks I asked on the net (not Brian, and I don't recall offhand who it was) who did have test machines, I was able to get VDdriver ported and working correctly on a Step 2 Alpha VMS machine. Once it was posted to the net, Brian sent some stylistic suggestions on the Macro32 net discussion group about how to make its calls to a couple VMS functions more "standard". While the driver worked correctly, I decided these stylistic changes were good to do, and incorporated them. (When I got access to a step 2 machine, with final instead of prelim manuals, I saw where these points had been clarified in the updates.) That was how I became more directly acquainted with him. I still needed to port FDdriver to keep my promise, however, and asked Brian for some information to help shorten my debug cycle (which still involved sending the code somewhere else for debug at the time.) He sent me some information which I then used to complete that port to the step 2 interface. The port could have been done with the prelim manuals, and in fact the major benefit of having more detailed information turned out to be to assure me that nothing beyond what I had designed needed to be done. The port would in any case have been done as I had promised to Digital and various colleagues, but Brian helped me (and thus indirectly the rest of the SIG) get the driver ported a bit faster. I had been aware that Brian at the time had started working for ASCI, and contacted him also to ask if ASCI might have interest in selling any of my systems code, which was so far as I know unique in the industry. (It ran on Vax only at first but I eventually bought an Alpha and ported it all; I had done a step 1 alpha port previously.) That came to nothing...I was never contacted by anyone...but we have remained acquainted since that time. I had known ASCI was a software vendor since they started selling a virtual disk package a few years after I had published mine (including sources in my publication). Their package had a couple things different from mine, but I never saw anything more about it than their advertising. Brian's help to me included not a syllable about what the ASCI products do or did, which is the way we both wanted it. (If ASCI used my code, I was unconcerned, since I had given it out to the public.) I consider it a privilege however to be able to discuss internals or software questions with knowledgeable people, and Brian has shown unusual ability, inventiveness, and skill where there are not many in the world able to perform at his level. Should there be any doubt of Brian's expertise, the SIG tapes have been distributed worldwide, coming out twice a year since their inception, and the dates of his contributions as well as their contents can be verified readily by looking at CDs from DECUS (or examining the CDs directly on net at ftp.decus.org). I continue to use his code as prime examples of how certain functions can be accomplished when someone asks me about something which can use the techniques. It is manifest that Brian was one of the best of VMS Alpha systems people already during 1994, even though the machine had not been introduced long at that time. Another systems person can see this merely by examining his work at the time. (Since it was on CDs cut at the time, there can be no question of dates having been altered; the medium cannot be written to once made.) Glenn C. Everhart, PhD Open VMS Systems SIG librarian