Article 118620 of comp.os.vms: I've seen WAY too many messages in this list regarding "Digital is giving up on VMS" or "Digital is giving up on NT". I thought I would give you the benefit of my specific knowledge and feelings on this, since most of what I've read in this mailing list is either emotional, party wrong, or both. This knowledge comes from a variety of sources, including the product managers for NT, VMS and UNIX; the VP who owns NT and VMS (Wes Melling); many senior engineers in all the groups; and some of the chief architects of, for instance, VMS. And by the way, this isn't NDA, but if you ask for clarification and it goes into an area of NDA, I won't break the NDA (but I'll tell you why I can't tell you). I should also explain that my position at MCI requires me to examine ALL technologies in the industry, not just Digital, VMS or WNT. As such, I've been privy to the feelings that most of the predominent systems players in the industry have regarding, especially, WNT and UNIX. So, I don't make any of the following statements lightly; rather, they're a reflection (my view of it, anyway) of trends in the industry. VMS --- Digital IS NOT giving up on VMS. In fact, they're investing as much money into it now as they ever have. The fact of the matter is, VMS sales into new accounts were up 23% last year, and are expected to rise again this year (although not as much). Doesn't sound like a "decline" to me... The recent reorg that gave Wes Melling both VMS and WNT is, I think, an important one. Wes is a man with a vision (VERY interesting to talk to if you can pin him down at DECUS), and that vision is, I think, right on track for VMS. The cold, hard, fact is: VMS is not a premier operating system right now in many people's minds (and perception is reality, to the folks writing the checks). Microsoft owns the desktop for all intents and purposes, with the exception of high-performance workstations, and UNIX is the defacto standard on those. So, where does that leave VMS? Its real strength is as a high-performance backend server. And these days of two-tier/three-tier client-server, if you can't interoperate with the desktop efficiently, you wither and die. So, position VMS where it can best show its strengths. Give it strong ties to the powerful desktop vendors. Now, given that you want to own the enterprise server, make the hardware incredibly powerful, far beyond anything available (WILDFIRE), and make your server OS the only one able to fully exploit that power (VMS - NT and UNIX have already said they can't do what VMS is in the process of doing for WILDFIRE). Give it the support of the big DBMS vendors. Throw out numbers like "1,000,000 TPC-C" (a fast number today is 30,000). Throw in a date of sometime in 1998 for the first iteration of it. Now, add in a very popular TP monitor called Tuxedo from BEA. This is a very significant development for VMS. Folks have whined that "WNT and UNIX got it first" - well, SURE they did - it was DEVELOPED ON AND FOR UNIX! As a startup company, BEA had to go where the market was. Now that they have the resources of Digital behind them (Digital will partially own BEA), they can go to the VMS platform. Instead of whining, VMS users should rejoice that we're getting the preeminent TP monitor in the world ported to VMS, not just a cheap knockoff product from somebody you never heard of. What's significant about that? For one of the first times, I can now write code on VMS or UNIX (or NT or MVS) that I can directly take from one box to another with little or no changes. That was always VMS's achilles heel - you had to change code to use it from other platforms. What this does, is makes the underlying platform moot - and NOW, Digital can sell VMS platforms by saying "you no longer have to care what's underneath - you can buy huge performance for a cheap price, and with the stability, security and reliability that VMS brings to you." But understand something: the days of VMS are limited. Those of us who lived thru the decline of the PDP-8 and COS300 and the decline of the PDP-11 and RSTS/RSX/RT11/IAS and the decline of the DECsystem10/20 and TOPS10/TOPS20 (all of which I did), know very well that things change, regardless of how much you like them, and regardless of how right you think that change is. Particularly with respect to WNT, once enterprise decision-makers believe that WNT is ready to run the enterprise (I don't think it will be for several years yet), the move to WNT for enterprise servers will wipe out many servers running other OS's. WNT --- Digital has bet the corporate family jewels on WNT. They will do anything and everything they can to or for Microsoft in order to keep WNT alive on the Alpha platform. The vision of the WNT people within Digital (and that's a VERY powerful cabal) is that some years down the road, most of what Digital will sell on any size platform will be WNT. In fact, within a couple of years, they envision that, industry-wide, sales of WNT servers will equal that of UNIX servers, and that between the two OS's, they'll own 80% of the server market. This is a fairly widely-held prediction in the industry, and one that I'm not sure I agree with, by the way; I think it's going to happen, but it's going to take a few more years than the optimists do. Microsoft, on the other hand, has Digital over a barrel. If I were Digital, I would be VERY nervous over the fact that WNT has basically been cancelled on the MIPS and PPC platforms, leaving AXP as the lone surviving RISC platform. The Digital/Microsoft alliance is, I think, and based on past history with watching how Microsoft does business, one of convenience for Microsoft. They're bleeding a lot of technology from Digital, while providing relatively little in return (I've heard the agreement with Microsoft described as a "joke" by more than one senior technical person). ------------------------------ So, where does this leave Digital? - VMS is strong, provided it's positioned correctly - and the enterprise server is the appopriate place for that. - VMS has the right man at the helm (Melling) and some guys with great technical vision looking to the future. - WILDFIRE and Galaxies will set the world on its ear. - WNT is absolutely crucial to Digital's future, to the point where if Bill Gates said "kill VMS now and put WNT on all servers", I think he would be taken quite seriously (but I'm honestly not sure if Digital would tell him at that time to take a flying leap). Whew!!! I didn't intend to write this much, but I felt some of the stuff had to be said, particularly for those folks who don't have the same access to Digital that I do. This wasn't intended to be a lecture or a scolding, so please don't take it that way... -- Dan O'Reilly MCI Telecommunications Technology Planning MS 1212/117 2424 Garden of the Gods Rd Colorado Springs, CO 80919