Conference 7.286::digital

    
    The conference has never lacked for suggestions for how to get Digital
    back on track.  It's also had a lot of clear warnings about stupid
    decisions long before the numbers have come back in with the proof.  If
    some of the tone borders on cynical pot-shots, it's probably because
    the suggestions tend to fall on deaf ears.  And the best efforts that
    the folks in the trench's make come to nothing when decisions are made
    that undermine those efforts, and make us all look like fools.
    
    Wanna suggestion?  Pump some money and focus back into OpenVMS.  It's
    what is paying the bills, and while it may not be the magic bullet to
    fix our problems, it still carries the weight of this $13b company, and
    can continue to carry it for a while.  It *is* our core competency.  It
    *is* our customer base.  Secure it, feed it, and *then* grow other
    competencies leveraging what we are successful at.  OpenVMS fits the
    bill in the vision of the future where the Internet is the system, and
    standards like Java make the actual O/S transparent.  Where big Oracle
    back room servers need 24x365 disaster tolerance.  Where PCs stay where
    they belong - on the desktop, and fewer of those - as PDA's replace
    them as cheaper to operate.
    
    Ooops.  Sorry.  I'm a tree hugger.  A tree hugger in the group paying
    the bills, with all the customers, but nonetheless, a tree hugger that
    is not part of the solution that ain't working.
    
    
    

	I understand the base noter's sentiment and I am at least
humble enough to realize that I would be woefully inadequate
at trying to run a company this size let alone a business 
operation within it but... I just can't help feeling that we could
do better and, sad to say, the market feels the same way.

Dennis 

| Is there a conference out there that is concerned with helping Digital get
| back on track? This one especially seems to smell of nothing but bashers
| and pot-shot artists.

This is one of the best notesfiles in this company. You can really sense the
pulse. I have followed it as long as I can remember. It does not lack
suggestions nor snide remarks when appropriate. It certainly has its own
style. But it is a definite goldmine.

To my knowledge, the vast majority of VPs and above at Digital do not
partipate in notesfiles. This notesfile is definitely the right place to go
if you want unbiased information about events, products and rumors. It also
the unofficial shoulder to cry on when you feel down.

If you want good news, read Digital Today. If you want to hear things the
way they are, come here. I am sure you can stimulate creative discussions
about getting Digital out of trouble.

>Per 

>Note 5200.0                        I like DEC                          3 replies
>Is there a conference out there that is concerned with helping Digital get
    
    'nuff said...write it up that wall a 100 times...DIGITAL...DIGITAL...
    
    And use a spell checking brush...just to be sure or the name police
    will pull you by thee ears. 

    re:.1
    	
    	While at a recent Windows NT conference, I talked to a number of 
    individuals working in the VMS team, and they too stated your sentiments 
    "Pump some money and Focus back into OpenVMS."  You also added the 
    sentiments - Secure it, Feed it, and Grow it.   These sentiments bring
    up a lot of questions that I would love to hear the answers to; however, 
    getting answers to these questions are difficult due to my work locality. 
    Fred, if you have some of the answers to the questions below, please send 
    them to me.  I do not hear much about VMS futures on the West Coast.  
    
    		
    	Has there been a formal statement made about the future of VMS, 
    why its funding is cut, and  is it headed for maintenance mode?   Do
    the VMS engineers get this type of information? 
    
    	Fred, do VMS engineers believe that if the VMS UI and command line
    syntax was updated to be current with Today's technology, that it will
    be a viable product?  If so, were there any marketing studies done to
    examine this? Is the problem with VMS is that it does not run on a pentium?
    Is VMS too costly to further develop for the return on investment in 
    today's market?    I think today's customer wants inexpensive platforms 
    capable of running multiple operating systems.  Some such platforms use
    the pentium.  
    
    	I'm curious how an new version of VMS, running on a pentium,
    written in C++, with a windows-style GUI, would sell.   The big issue is
    what return on its investment would Digital get if it built this new
    version of VMS.   
    
    	Any answers?
    
          Earlier in this conference it was stated that sentiments have
    fallen on death ears.  Are the ears really death, or just the mouth
    silent? I do not know if management is hearing your concerns or not.
    Perhaps management can address this issue with a response to this
    note.
                                                                                    

I realized how I felt about this notes conference a few days ago when I got
"login info invalid at server node" or somesuch message over a fairly lengthy
period when trying to access it. I imagined that the brass had finally decided
to silence the employees and that all I would ever learn in the future would
come from PR releases, position statements, and DVNs. I was surprised at how
visceral my reaction was and how greatly relieved I felt when I was later able
to regain access.

That said, I too get annoyed at the endless whining and gratuitous insults that
are all too often heard in these precincts. And any change in Company policy is
invariably labeled an assault on the employees regardless of the merits. The
change in the retiree health-care eligibility is a case in point. On the other
hand, many valid objections have been posted about other personnel policy
changes.

I was rather annoyed (but kept silent) the other day when what I thought was a
good management statement in respose to the HP-MS alliance was posted and the
first reply was what I considered a knee-jerk insult of management. I did not
reply because I didn't have the expertise to evaluate the merits of the
statement, but I felt badly that the manager who had written the statement was
dealt with summarily with an insult. (Admittedly the manager didn't post the
statement so it was not as if another noter had been dismissed with an insult.) 

Management had prepared a comprehensive  statement and was ready to respond to
the "alliance" as soon as it happened. If management had made no reply, the
howls of protest would have come swiftly. So damned if they do, damned if they
don't.

It is very enjoyable to dump on, ridicule, and otherwise get off by jabbing at
our highly paid managers who are continually criticized here. Unfortunately,
some, not most, of such postings are akin to "so's your mother." I think
it is those types of entries that we'd be better off without, as much fun as
they are.

On the other hand, when someone asked here "who is the architect of our
exchange mail messaging policy?" the answer that came back was "Bill Gates" and
I found that bitter irony perfectly appropriate and rather funny. Too, the
constant reminders about our marketing problems are, IMO, appropriate when not
totally unconstructive. 

As others have stated, there is no shortage of good analysis, criticism, and
proposal in these notes files and I do get the impression that much of this
does filter up the chain even if the VPs don't read it directly. If someone had
the time, a book could be written about his company, its problems and
opportunities just from this very rich source. I think the empty insults and
knee-jerk negativism only diminishes the effectiveness of the many thoughtful
postings here, postings that go to some trouble to diagnose this Company's very
serious problems and to recommend solutions. I'd like to thank those who take
the time and trouble to write such notes. 

Several weeks ago when I asked here why the stockholders were not screaming
bloody murder at our giving IBM the credit for AltaVista, I was deluded enough
to think I might have had some influence when the ad stopped appearing very
shortly thereafter and the topic was never mentioned again. Of course, the huge
chorus of protest is probably what brought whoever allowed that fiasco to his
senses. The length of time it took to get that resolved was indeed
discouraging.)

Sorry for the length of this ramble. I can now expect to be thoroughly beat up
for trying to suppress free expression, no doubt. :%)

>                     <<< Note 5200.5 by DECWET::B_LEAHY >>>
>                    -< re :-1  My questions on VMS future >-

    re:.5
        	
>    	While at a recent Windows NT conference, I talked to a number of 
>    individuals working in the VMS team, and they too stated your sentiments 
>    "Pump some money and Focus back into OpenVMS."  You also added the 
>    sentiments - Secure it, Feed it, and Grow it.   These sentiments bring
>    up a lot of questions that I would love to hear the answers to; however, 
>    getting answers to these questions are difficult due to my work locality. 
>    Fred, if you have some of the answers to the questions below, please send 
>    them to me.  I do not hear much about VMS futures on the West Coast.  
    
    First of all.. We have been and currently are an OpenVMS company.
    
    It pays the bills, brings in the cash flow, and generates most of the
    profits that take us to the break even point.
    
    Digital Unix and WNT are growing rapidly but it's a percentage increase
    and not an actual numbers being measured.. 
    
    OpenVMS and it's customers drive Digital and have for well over 10 years.
    
    The OpenVMS futures are discussed at DECUS, and with some of the largest
    customers (Like the Executive Session held in Arizona last Month) in the 
    world who need 24x7x365 computing, regardless of the costs.  
    
    Unix, even with it's "Clustering" doesn't begin to approach this level 
    of computing and there is a very large business need for this type of 
    computing.  We need to take this fact to heart and stop fooling
    ourselves that Unix will Catch up in the next release or some nifty
    new Datawarehouse software will mask the limitations Unix has in 
    the 24x7x365 space.  We need to Sell and Market OpenVMS as THE critical
    computing solutions.  Today even the NT people believe that with file
    system failover and application restart technologies they will be 
    playing in this market in the near future. 
    
    
    The yearly 6 billion Dollar OpenVMS and 30 billion Dollar MVS marketplace 
    typifies this type of computer systems and these customers will not
    accept Unix or WNT into this type of environment until those O/S have
    been proven effective and have a 3-5 year track record with large
    Datacenter Management.  
    
    Even with a more and more distributed processing world with objects and
    middleware, larger, and faster, centralized repositories and transaction 
    systems will still be required.
    
    We have a unique technology with OpenVMS that represents the highest
    availablity in the computer industry, even higher than fault tolerance
    when properly configured and we are not marketing or trying to sell 
    it anymore.  Without NEW customers, a marketplace slowly dies... 
    
    That is where we are right now with OpenVMS...
    
    
       		
>    	Has there been a formal statement made about the future of VMS,
    
    yes, it was hastily added to our vision statements when it was pointed
    out that OpenVMS wasn't mentioned...
     
>    why its funding is cut, and  is it headed for maintenance mode?   Do
>    the VMS engineers get this type of information? 
    
    OpenVMS's engineering budget has been shrinking over the past several
    years but it's not at maintenance mode yet.  The OpenVMS teams are aware 
    of these numbers but it's not something we've been enjoined to talk about
    since the Budget "Parity" with Unix R&D a couple years back.  
    
    It also hasn't been a good thing to talk to our customers about since the 
    budget has been contracting.
    
    The story goes that OpenVMS is a mature product requiring less R&D
    to keep it on parity with the rest of the industry..
    
    And that's true to a point...
    
    
>    	Fred, do VMS engineers believe that if the VMS UI and command line
>    syntax was updated to be current with Today's technology, that it will
>    be a viable product?  
    
    OPenVMS already supports CDA and can be managed via the OpenVMS
    management workstation (On WNT), UI is pretty pictures not what we're 
    talking about for Large Scale Computing, having a W95 interface isn't 
    going to sell a 2 million dollar server...Otherthings will
    
    
    >If so, were there any marketing studies done to
>    examine this? Is the problem with VMS is that it does not run on a pentium?
 
    Pentium is a DEAD architecture.  Mercede has been blessed to take it's 
    place, so Intel is just marking time here and we have an opportunity to
    sell and shore up our place in the server marketplace.
    
    Why would we port OpenVMS to a Dead Architecture when it already is 
    running (in 64-bit mode) on Alpha and is available on 32-bit VAXes?
    People don't care about the chips inside if their applications run...
    
    OpenVMS runs on Alpha and could run on the lowest cost Personal Alpha
    workstations if it was politically acceptable at Digital to do so.
    
>    Is VMS too costly to further develop for the return on investment in 
>    today's market?    I think today's customer wants inexpensive platforms 
>    capable of running multiple operating systems.  Some such platforms use
>    the pentium.  
    
    OpenVMS is cheap to develope compaired to the potiential return...
    IBM paid cash for Lotus with profits made from MVS in a single year...
    
    Customers want inexpensive workstations that will run every application
    ever made.  OS has little to do with it. The costs of Servers is 
    usually a minor part of a customer solution. Minor, but very important
    if there are consequences to having downtime, or lost data.
    
    The OpenVMS marketplace currently makes a substantial profit for the 
    company but if we don't feed the cash-cow a little, it'll drop dead
    even as we try to milk it one last time.
    
    Please internalize this because it's important.  Almost everything Digital
    Sells is a commodity product and is a repack of some other companies wares
    with a very low profit margin per unit.  OpenVMS has a high profit margin 
    and we should be fighting to keep it as a viable product for as long as 
    we can.
    
    
    >Then who pays for the development of WNT and Digital Unix?
    
    Digital Does, from profits that OpenVMS gives back to the company.
    
       
    >	I'm curious how an new version of VMS, running on a pentium,
    >written in C++, with a windows-style GUI, would sell.   The big issue is
    >what return on its investment would Digital get if it built this new
    >version of VMS.   
    
    You're hung up on this Pentium thing... It's not our marketplace, 
    We shouldn't be in the portablity business... We should be in the 
    UNIQUE BUSINESS FEATURES business doing things that no other O/S
    can do economically, like Nonstop computing, Multi-Terabyte
    Datawarehouses, Disaster resiliant clusters up to 500+ miles apart,
    with extreame security...
    
    >	Any answers?
    
    Yes but who's listening who doesn't believe that Digital will 
    reshape it's self into a commodity reseller and still stay viable.
    
    You sell commodities you become CompUSA, you sell complex tools and 
    solutions and you remain Digital Equipment.
    
>          Earlier in this conference it was stated that sentiments have
>    fallen on death ears.  Are the ears really death, or just the mouth
>    silent? I do not know if management is hearing your concerns or not.
>    Perhaps management can address this issue with a response to this
>    note.
    
    The OpenVMS gets retold ever couple months but never gets down to the
    sales channels or rest of the company...
                                                                                    

    If OpenVMS is in such sad shape, spin it off like we just did to 
    Altavista and DECmessageQ and ObjectBroker... Spin it off with it's 
    own sales force and engineering and watch OpenVMS become viable, 
    sellable, growable and very profitable in a very short period of 
    time.
    
    I would join such a company spin off immediately, even if I had to 
    turn in my 10 year old DEC badge.
    
    JMHO
    
    John Wisniewski
    OpenVMS Ambassador 
    (1 of 119 world-wide who support and defend the OpenVMS Business
    in addition to helping sell Digital's other products.)

    re: .5
    
    Hi Barb.
    
    I can't pretend to speak for OpenVMS, nor do I set strategy for OpenVMS
    or the company, so I can only offer my opinions on many of the questions.
    Here is my off-the-cuff ramble, feel free to NEXT/UNSEEN.
    
    VMS has had a specific strategy over the last few years, it was to sell
    the idea of a OpenVMS/NT Affinity/Connectivity.  Some of it has
    delivered, some of it hasn't.  Some of it was marketing, some of it was
    technology.  The latest sign that this has finally moved into better
    focus is the fact that we are now managed by the same VP (Wes Melling).
    
    OpenVMS has been squeezed, in my opinion, because we have been given
    the job to engineer the NT Affinity, without additional resources.  The
    people at the top who make strategy have not believed in OpenVMS as
    part of their vision for a long time, and as a result we are viewed as
    simply a source of cash until one of their strategies bears fruit.  As
    a result, we have participated in the company wide shrinking in both
    budget, and headcount.
    
    Because of the stigma placed on OpenVMS within the company over the
    last half-dozen years, many engineers have left.  Some have gone on to
    pursue where they believe their skills will be more valuable in the
    future (many, like you, have left for Seattle, many as the first step
    to going to Microsoft).  The pay freezes, and uncompetetive pay for
    many of the highly skilled people in VMS has also taken it's toll. 
    Many have walked across the parking lot to Oracle for anythere from 25%
    to 100% increases in pay.  The company places no great reward for
    staying in, or loyal to OpenVMS.
    
    I think it's wrong to focus on command line interfaces versus a GUI. 
    The primary interface to OpenVMS should be considered other computers. 
    PCs on the desktops, other servers, workstations.  DCOM, CORBA, and
    other client/server distributed object interfaces is where it's at.  As
    is the ability to interoperate with all of the internet applications
    that are proliferating.  And that's where the Affinity program needs to
    succeed.  My humble opinion is that we should *really* be focused on
    interoperating with Win95 clients, and less on NT servers.
    
    OpenVMS, and it's layered products, is the result of thousands of man
    years over nearly 20 years.  The idea that it could easily be rewritten
    in C++ and preserve binary, operational, and bug compatability with
    it's application base is doubtful.  Nor does simply moving to a Intel
    base imply that we will be able to bring the OpenVMS customer base
    along with it.  The Alpha experience has shown how difficult it is to
    move to a new architecture.  In retrospect, the Alpha architecture
    should have included binary execution / emulation of the VAX
    architecture, and truly allowed the entire VAX base to pick up and move
    to Alpha. Performance be damned.  It was the combination of *price* and
    performance that killed the VAX market.  Intel proved that RISC wasn't
    the key to success, they reproved the point OpenVMS had already made
    over the life of the VAX - it's binary COMPATABILITY across a range of
    price/performance and over time.  As a contrarian, I believe that a
    $2500 OpenVMS system with a $500 OpenVMS license would sell like
    hotcakes, and so would a PCI-based, 200 VUP VAX.
    
    As you know, we had several A/D and research efforts to move OpenVMS
    into the next century.  NT is the result of one of them, because we
    foolishly thought we could protect the cash cow.  Could we put OpenVMS
    onto Intel?  Yes, at the source level for most applications I think the
    answer is yes.  But it woudn't be OpenVMS, it might be it's successor.
    
    The base O/S still has some life kicking, and if all goes well, the
    "Galaxy" project that has been the buzz since we first suggested it,
    may show just how much life we have in us.  The intent being to
    leverage those thousands of man years of effort in making OpenVMS a
    relaiable, robust, scalable architecture - to create a server with
    features, and performance that will continue to put OpenVMS years ahead
    in technology *and* the ability to exploit the technology.
    
    I'm not dimwitted enough to believe that OpenVMS, will ever kill the
    Wintel machine, take over the desktop, or ever again be what it once
    was.  Frankly, it was all the thousands of people that wrote layered
    products for OpenVMS who we have laid off, sold off, killed off, that
    we have no hope of ever rebuilding - that made OpenVMS great.  We are
    a shell of that company now, and it started to happen long before the
    change of presidents, when we were sucked into the "open" systems smoke
    and mirrors game.
    
    BUT:  Take a cue from IBM.  Refocus some energy, committment, and
    visibilty on OpenVMS - and our customers.  It can pay the bills a
    while longer.  Don't abandon all the other investments, but pick one
    or two that have potential, say the Internet, JAVA, StrongARM, whatever.
    Heck, apply the Alpha technologies to building a new VAX - do you have
    any idea how big the VAX market still is?  Look for synergies with
    what we have today, don't just look at the market and play me-too.
    We'll end up UNISYS, or worse, DG.
    
    We are *still* a $13b company.  Why be ashamed of the past that got us
    here?  And that is still keeping us afloat?
    
    

    Re: .0
    
    As I've asked before, how can one weight the significance of the views
    expressed in here when the sample population (of DIGITAL staff
    participating in this Conference) is so tiny, hmmmm..??
    
    Re: .1,.8
    
    Well Said Fred!
    
    btw, where do I find the numbers which support your thesis that OpenVMS
    is our major profit source?
    
    Thanks,
    John.

    FWIW..
    
    How about a typical large Customer perspective ?
    
    We have a large Customer who had approx 100 OpenVMS systems 2 years
    ago. At that time, they decided NT was the future, so they migrated
    only their file/print services from PathWorks to NT. They bought
    approx 35 Alpha 2100 systems to deploy NT across the country. Their
    financial and bread-n-butter applic's remained on OpenVMS until they
    could become more familiar with NT.
    
    Now, fast forward to the last few months. As with many Customers, this
    Customer is going through a massive re-centralization effort as they
    now recognize the perils and hidden support cost's with big
    decentralized environments. They are also faced with reality that along
    with re-centralization comes a big requirement for high end reliability
    and availability.
    
    They had to decide whether to start moving their OpenVMS app's to NT 
    or continue with what they have.
    
    Their decision was very easy. They have just purchased a bunch of
    4100's and 8400's to run OpenVMS 7.1. Their file and print environment
    will continue with NT as they do like NT, but they now realize after
    2 years experience with NT, just what the differences in the two OS's
    are.
    
    The real "kicker". The number of OpenVMS systems in this Customer
    environment will likely drop from a high of 100 or so a few years ago
    to probably about 25 or less in the upcoming year.
    
    Now, the marketing folks will say "see, there's a company
    de-emphasizing OpenVMS as they have dropped from 100 to 25 systems!"
    
    HOWEVER, this does not take into account that these 25 systems are much
    larger systems and they now have much higher availability requirements
    than the systems 2 years ago. A single outage now impacts a much larger
    number of people than before.
    
    So, yes the total number of OpenVMS systems has decreased in this
    environment, but the requirement for a high end reliable platform has
    drastically increased.
    
    On a side note - the gui management side of OpenVMS is a very important
    piece to the future of OpenVMS, because we need to remember that while
    the system manager set's up the environment, it is the operations staff
    in the computer room that manages the day to day activities - the same
    ones who know windows and find DCL to be as cryptic as UNIX.
    
    Note - I like NT and in fact is a very big part of what I do these
    days, but I would suggest that while NT is eating Novell, Banyan and the
    MAC's market share, the question needs to be asked "why is MVS, HP-UX
    and OpenVMS type systems continuing to do much better in the datacenter
    environment ?"
    
    One suggestion - re-centralization and it's higher availability
    requirements.
    
    Enough of my rambling ..
    
    Regards,
    
    / Kerry
    

RE: .5 -< re :-1  My questions on VMS future >-

I can't add much to the brilliant analyses done by John W and Fred K, but
I can say "Amen, Brother".  OpenVMS is keeping this company afloat while the
1-3.5-9 strategy is taking effect, and will continue to do so until either
the other areas of the company (Digital UNIX, NT, Internet, Intranet, etc)
start generating actual profits, as opposed to just smaller and smaller 
losses over the succeeding years.

As for where are we getting our numbers?  Oh, little things like the
numbers of AlphaServers sold with each O/S, the number of MCS engineers 
who are dedicated to supporting the AlphaServers that are in the field
which are running each of the O/S's, the number of Sales Reps who make
their budget year after year on each of the O/S's, etc.  These numbers
are frequently very tightly held, because they state something that is
totally politically unacceptable in this company: OpenVMS is a good thing.

>    	I'm curious how an new version of VMS, running on a pentium,
>    written in C++, with a windows-style GUI, would sell.   The big issue is
>    what return on its investment would Digital get if it built this new
>    version of VMS.   

Well, I think that we already have an example of this: an operating system
backed by a major player in the industry, running on commodity hardware,
with a full object-based GUI.  It's called OS/2, and it has totally failed
to make any impact against the Wintel monopoly.

As much as it pains me, I am forced to agree that OpenVMS will never take
over the desktop.  But I don't think UNIX (any flavor, whether Solaris,
HP/UX, AIX, SGI, or Digital UNIX) will take over the destop either.  Look
at the last years workstation sales numbers: across the board every vendor's
workstation sales units dropped 20% from the previous year (some more, some
less, but the workstation market as a whole dropped 20%). 

OS/2 is a joke, Apple continues to lay off people, OpenVMS does not even
show up as a blip in the workstation market no matter how closely you look, 
the entire UNIX workstation market drops 20%, Compaq has the best year in
its history, Digital PCs show major growth, HP sells a huge number of PCs,
and Sun is feeling sufficiently panicked that they try an end-run with
Java.  What does all this tell you?

It says that if you are counting on a non-Microsoft desktop to pay the
bills, then you better get familiar with the Chapter-11 laws, because you
are going to need them real soon now.  Microsoft owns the desktop, and
people had just better get used to that for at least the next 5 years.

So what does this mean for Digital?  Give Bruce Claflin whatever he wants,
because he is in a position to give us some profitable business.  But it
also means that Digital UNIX and OpenVMS had better get used to being in the
server market, and had better interoperate *real well* with Microsoft
desktops.  Ie, All-Connect and Affinity.

With all respect to Digital UNIX, it is simply not ready to play in the
24x365 never *EVER* go down not even for hardware failure or operating
system upgrades even if the entire site is destroyed by fire, arena.  It
compares very favorably against other UNIX high availability offerings,
and I personally beat HP and Sun with a 99.99999% up-time requirement
by selling Digital UNIX and TruCluster (3 deals so far, and I am working
on 2 more this week).  But it simply does not yet (note that I said *yet*) 
compare to VMSclusters, and every un-biased observer (like Gartner and IDC 
and D.H. Brown) agrees with that statement.

So as my brethren said, let us love our OpenVMS customers (can you say
customer satisfaction surveys, which the SLT is now being measured on?),
while growing our other lines of business.  

We can do this, if we choose to do so.  

-- Ken Moreau

    re: .5
    
    Sorry Fred for jumping in and answering (point by point) Barb's 
    question question to you... 
    
    This HP/MS thing has me jumpier than the budwieser frogs at an
    east Texas Distributor... and my keyboard fingers are staying
    a might itchy too..:-)
    
    
    re: .11
    
    Amen Brother Ken Amen...
    
    It's never a sin to sell a Digital Product to a customer.. Only if 
    it's the wrong one for the Job at hand...
    
    --
    
    I've said it before and I'll say it again, I have to live with my
    customers even if Digital destroys itself in the marketplace, the 
    only thing I have to trade is my honesty, integrity, and technical
    correctness, with or without Digital.
    
    I've helped sell Trueclusters, WNTclusters, and been selling REAL
    CLUSTERS on OpenVMS for over 15 years, when asked by a customer 
    for my opinion I will explain (in a positive way) the functions, 
    features and differences, recommending the closest Digital product
    that meets their availablity, reliablity, and business needs.
    
    What I find more and more that I can't displace FUD our company 
    has planted about OpenVMS even to sell another Digital product...
    
    Something has to change...
    
    John Wisniewski
    
    
     

    RE .9:
    
    About 4-5 months back Wes Melling spoke to the VMS organization;
    one of his more memorable comments was along the lines of:
    
    	Of the $400 million profit Digital posted last quarter,
    	VMS was responsible for $600 million.
    
    This was Q4 FY96, I believe (the one where we took a big hit for
    restructuring).
    
    Fred might remember the numbers better than I, but I believe that
    $600 million was at least a 10:1 return on the money spent on VMS
    engineering, marketing, etc.

    I strongly agree with several of the above noters about the continuing
    importance and potential of OpenVMS. 

    The report referred to in note 5182.13 is an interesting read. Gartner
    reports a higher ISV interest in OpenVMS Alpha than either UNIX or WNT,
    medium vs low and low. It seems we should be able to do something with
    with this if only a little reemphasis was put on OpenVMS, externally
    and internally. My theory is that the high emphasis on WNT in our ads
    has caused committment to OpenVMS to be questioned. 

    - Bob

    re: the base note 
     Yes I like DEC too
    
    >>If you're not part of the solution, you're part of the problem.
    
    The comments many people in here make are about getting Digital back on
    track. The come from dealing with customers every day and seeing that
    we have lost contact. Plain speaking is needed, it is part of the
    solution. Smart companies welcome it, encourage it, listen to it, build
    quality on it.
    
    ..Kevin..

    re: .6  (Steve Z)
    
>I was rather annoyed (but kept silent) the other day when what I thought was a
>good management statement in respose to the HP-MS alliance was posted and the
>first reply was what I considered a knee-jerk insult of management. I did not
>reply because I didn't have the expertise to evaluate the merits of the
>statement, but I felt badly that the manager who had written the statement was
>dealt with summarily with an insult. (Admittedly the manager didn't post the
>statement so it was not as if another noter had been dismissed with an insult.) 

>Management had prepared a comprehensive  statement and was ready to respond to
>the "alliance" as soon as it happened. If management had made no reply, the
>howls of protest would have come swiftly. So damned if they do, damned if they
>don't.
    
    I, for one, will admit that it was good to see management with a ready
    response to the issue.  Clap, clap...
    
    But that's NOT the issue, unfortunately.
    
    The problem is that we are in a position WE SHOULD NOT BE IN!  We KNEW
    that eventually Microsoft WOULD act in its own best interest -- they
    always do!  We KNEW that HP could NOT ignore WNT if it began making
    serious inroads into the marketplace.  But, yet, we systematically
    go about selling off our products and destroying our crediblility in
    all market segments except WNT.  Then, BOOM!  HP drops a big one into
    the center of our one last center of effort.  And WE KNEW it HAD to
    happen!
    
    Our word-smithed attack on the HP-MS announcement fails for one reason:
    HP succeeded in a major marketing victory -- and we let it happen by
    leaving our flank exposed.
    
    Out here in the field, it seems they want just about everyone to
    become Microsoft junkies.  I never seem to get any communications about
    making sure we have enough Digital Unix or OpenVMS skills around.  But,
    we need to take the vast army of Microsoft Certified people and
    increase the number by some huge percentage!  I don't know if they're
    just positioning NSIS for a sell-off, but the REAL messages that come
    down the ladder look an awful lot like "One egg -- one basket:
    Microsoft."  Yes, we see the fluff messages around 1-3-9, but the only
    calls to _action_ I ever seem to see are based on Microsoft WNT,
    Exchange, W95, etc.
    
    I repeat: a Microsoft strategy is fine, but a "Microsoft ONLY" strategy
    is suicide (for everyone but Digital Semiconductor, at least).
    
    -- Russ

Re: .7

>>    The yearly 6 billion Dollar OpenVMS and 30 billion Dollar MVS marketplace 
>>    typifies this type of computer systems and these customers will not
>>    accept Unix or WNT into this type of environment until those O/S have
>>    been proven effective and have a 3-5 year track record with large
>>    Datacenter Management.  

John, You forgot to mention the $15B AS/400 market IBM created from nearly 
nothing during the same years we shrank our VMS business.  I read recently that 
AS/400 was  SAP's fastest growing platform!  (There are a lot of statistics 
around and this may be one that deserves significant skepticism, on the other 
hand it may be accurate and a useful guide for us.)  Contrast openness, 
functionality, and cost-of-ownership between VMS and AS/400.  We win.  Somehow 
these advantages were not translated into market share.  It may be too late to 
create a $15B business out of VMS, but if we were to try I suspect it would begin 
with attracting solutions partners to the VMS platform.  There apparently is a 
marketplace for whole solutions on "proprietary architectures".

Best Regards,
Dennis

    Yep, I was talking to an ISV who sells into OpenVMS and Digital UNIX
    (most of their sales are on OpenVMS).  I asked what is the platform
    that you get the most requests for that you would like to port to...  
    They answered AS/400 followed by AIX, and then by Windows NT.
    
    				-John

How did we a search for a positive note, which is what I read in .0, 
turned into a "VMS vs. the world" note"?

Nobody here holds the holy grail, but people read these notes and put a 
lot of belief in them, but if you read this conference what you come out 
is plenty (not all but enough) of negativism. And negative attitude is 
never going to get this company out of the hole, believe me.

I remember a sign I read in a mountain cabin in the late 60's:

"If you're so smart how come you're not rich!" 

(I'm neither, and with a daughter looking at an ivy league college, fat 
chance!)

BTW, all that profit that people are talking about came from servicing 
the installed base (yes, it is mostly VMS) not from selling new  
systems, be them VMS, UNIX or NT. So saying that VMS pays the bills, is 
half true, but is a little like living in the past, it is definitely not 
where we are going if we look at current trend in system sales. I would 
venture to say that the SBU is definitely not making any money nowadays.

And even then, service (break/fix) revenue is declining since customers 
are replacing those VMS systems and our contract penetration is lower in 
non-proprietary operating systems.

    re: .19
    
    When I service an account, do an upgrade, include more Digital 
    Content into a solution for a currently Digital shop it's still a 
    win... and it increases our profit because we've not had to invest 
    to develop a new customer (New customers are very expensive in 
    terms of sales to find, sell, and close) vs selling into a current
    customer into a different department...
    
    Both strategies need to be employed for a company to remain successful.
    
    Servicing a 6 billion dollar a year marketplace isn't a bad thing... 
    
    Not growing and defending that 6 billion dollar revenue stream is 
    a very bad thing...
    
    IMHO..
    John W..
    
    This will not change until we incent the sales people to sell
    everything that we sell and reward them appropriately for bringing
    a new fortune 1500 customer into the fold...
    
    Let's see... Pay salespeople to bring in new customers??? Nah...
    too easy a way to increase business and our customer base..
    
    JRW.

    
    
    re .0
    
    see note 5193. 
    

> "If you're so smart how come you're not rich!" 

  Um, because richness is more a measure of greed than intelligence...?

  Sorry, couldn't resist giving you an answer from the 60s.   8)
  As for the remark about tuition bills, I [am about to] feel your pain.

  RE: the remark about systems trends...  It wasn't the crux of your post, I
  recognize, but it sparked this reaction:  To echo others' comments in
  other strings, I hope DIGITAL's strategy from here on will be founded on
  vision, grown from talking to customers and employees who deal directly
  with customers (with some consideration that a customer & a DIGITAL
  planner each may recognize something about the future of a particular
  market that the other doesn't), and not on abstracted "trends".

  As far as my personal 2 cents, I vote we should proclaim a 1-4-16, if you
  will, that reinstates OpenVMS to the company of DIGITAL UNIX, NT, and
  networking, and in a way that attracts as much attention as possible to
  our reassertion of pride in who we (still) are and the solutions we offer.
  (I'll spare everyone my ideas for the "16"...)

  My sense of this string is that noone's advocating OpenVMS to the
  detriment of the others, or any potential future 5th or 6th horsemen, (I'm
  a user & enthusiastic booster of each of the 4, truly), just trying to
  listen to our customers and offer a heterogeneous range of solutions that
  differentiate us in multiple systems market(s).

  "Trends" don't assert anything but today's data...  And certainly have
  less sway on what actually moves future events & decisions than clear &
  communicated vision.  - Jay

I was in no way putting down VMS, (I sold my share of VMS, and RSTS, and 
TOPS-20, systems back in the early times (3 DEC100 and one DECathlon). I 
am all for pushing VMS more strongly than we are today.

The main point I was trying to make is that the general tone today is 
very negative (with some justification), but being/acting negative is not 
conductive to any positive results.

And yes, trends can be changed.

    I like DEC
    I like Alpha
    I like my colleagues here
    I like winning at SPEC95 - see for example note 4700 in this conference,
    or see http://tlg-www.zko.dec.com/~henning/raw466.html
    
    But we don't win all the time.  For example in the above-mentioned web
    pointer, the AlphaServer in question came in first place for only 5 out
    of 8 categories.  
    
    True, it's hard to wait for EV6.  Things will get worse before they get
    better (i.e. until EV6 is in volume).  Hang on 'til then and don't let
    the nay-sayers sink the boat.
    
    For a long-term view of one of Digital's strengths, see the graphs in
    http://www.digital.com/alphaserver/performance/vups_297.html
    
    /john

    In the spirit of .o ie positive DIGITAL news : The following is an
    extract from the most recent Windows NT online magazine newsletter 
    (they state that their newsletter now has 15,000 subscribers ..)
    
    Reference: http://www.winntmag.com/update 
    
    "Dear NTer's,
    
    This week's issue includes an item about Digital's new 533MHz Alpha
    processor. This item is incredibly interesting to me-especially on the
    heels of Microsoft's announcement that it won't wait for Intel's P7 chip
    to release a 64-bit version of NT 5.0. I've been an Alpha fanatic since
    the first day I learned about them. I think Digital's marketing slogan
    of "fastest chip on the planet" went straight to my head-I bought two
    150MHz systems the following week. That was in late 1994 when the
    fastest Intel you could buy was a 60MHz Pentium, complete with floating
    point bug. Today, I still own both of those Alpha systems, and they're
    still running strong. They've been performing grueling tasks in file
    service and Web service since day one, and not a single piece of those
    boxes has ever failed-not even the mouse (that sound you hear is me
    knocking on wood).
      In my opinion, Digital deserves the chance from Microsoft to be the
    first 64-bit NT platform-Digital has earned this opportunity through
    hard work and excellent engineering. Granted, in earlier days, locating
    software that ran on the Alpha was tough-but today that's not quite so
    true. After we all get our hot little hands on a 64-bit version of NT, I
    think you'll see the vast majority of vendors producing native Alpha
    code. And I'd venture to guess that you'll also see major computer
    vendors developing new systems founded on Alpha architecture as well.
    Well that's my two cents, enjoy this week's UPDATE!
    
    Have a great week!
    
    Sincerely,
    Mark Joseph Edwards 
    [email protected]
    

    re .19
    
    The base note was a put down, and a challenge.  It summed up the
    content of this conference as rampant negativism, and asked for
    positive suggestions for putting Digital back on track.  My response,
    which fed the responses, was to make a positive suggestion.  It
    happened to be a suggestion to refocus some energy on OpenVMS.
    
    Make your suggestion.  Let's debate it.  It might not make any
    difference (i.e. nodbody may be listening), but you never know.
    I threw out an idea, Barb asked some specific questions to probe the
    foundation of my suggestion, John responded based on his experience in
    the field, others related their experience, I rounded out the basis for
    my suggestion and responded to some of Barbs assumptions.  I enjoyed
    it.  It was positive.  It did not tear down Digital.  I think it called
    into question *if* we are heading down a path that can support *this*
    company?
    
    I've concluded that the people who make the decisions aren't any
    smarter that the rest of us, and don't have better ideas.  At best they
    have better information, and access to smart people with ideas.  
    Maybe those smart people may find a glimmer of an idea in here, or a
    warning to heed.
    
    In response to the question about the proposition that OpenVMS pays the
    bills.  If I had direct access to such information, I could not put it
    into a file like this.  My information derives from public statements,
    from individuals with access to such information, and from limited
    knowledge of some specific data.
    
    

    I think there is little doubt that our OpenVMS customer base and
    revenue stream is larger than that of our Digital Unix and Windows
    NT/Alpha, combined.  That should be fairly obvious.  It is also obvious
    that this does not reflect the current market trend, which is strongly
    towards NT, and has been strongly towards some generalized Unix
    platform for quite some time now.  It will never be strongly directed
    towards OpenVMS, although in some niche markets and installed base
    markets it will continue to thrive for quite some time.
    
    While NT and Unix continue to mature into viable commercial platforms,
    we can differentiate OpenVMS as a solid, reliable alternative, but only
    for so long.  This is particularly the case since we also market our
    own versions of both NT and  Unix, and a strategy of differentiation
    based on reliability over our other products looks rather foolish.
    Still, we have something like a half million OpenVMS customers.
    
    Although NT Affinity is a viable and useful program, it probably
    doesn't go far enough, in some senses.  Working in OpenVMS Network
    Engineering, I can atest to the fact that OpenVMS can and will improve
    in networking technology, not only as it applies to  NT Affinity, but
    in the overall networking arena as our strategy shifts from OSI towards
    IP.
    
    As for positive suggestions, two come to mind: 
    
    a.) We need more technical market analysis - marketing as a research
    vehicle that provides specific technical market trends which OpenVMS
    and our other products can use to formulate product strategies.  We
    still don't have sufficient marketing support in this context, and
    marketing is chartered to support Field Sales and Advertising much more
    so than market research.  This is a serious product weakness.  It is
    very likely the reason for some of our biggest failures in the past,
    including (nearly?) missing the entire PC market and the Unix market as
    well.  I mean, we were a major player in the development of Unix, but
    we still missed the boat in a very big way.  Digital Unix is a
    wonderful product - we should have been on top.
    
    b.) We should strive to overcome the natural tendency to remain loyal
    to a product even when everything tells us that the product, like all
    products in all businesses, has a limited lifespan.  What can we do to
    prolong the revenue stream by meeting our customer's needs?  What are the
    big questions with which our OpenVMS Customers wrestle every day?
    
    NT or Unix.
    
    In the long run, we can continue to incrementally improve OpenVMS
    through NT Affinity and better networking technologies that make it
    interoperate more smoothly in the evolving customer environments. 
    We know we cannot stop the trends toward NT and Unix, both of which run
    on our existing Alpha hardware.  We can also improve our NT and Unix
    products with what we have learned in our OpenVMS products, e.g.
    clusters.
    
    If we could offer our customers the ability to run 80% of their Alpha
    applications on either NT or Unix, with binary compatibility, we could
    make a fortune.  I think that's where we should be in five years.
    
    
    tim
    

|    b.) We should strive to overcome the natural tendency to remain loyal
|    to a product even when everything tells us that the product, like all
|    products in all businesses, has a limited lifespan.  What can we do to
|    prolong the revenue stream by meeting our customer's needs?  What are the
|    big questions with which our OpenVMS Customers wrestle every day?

    I thought the pendulum had swung the other way at Digital. That is,
    anything made by us is inherently bad and proprietary in the negative
    sense. Therefore, lets drop most of what we do and try to meakly copy
    Microsoft or whoever else is hot in whatever area. We have become these
    middle men, without a real vision.

    My suggestion would be to have a little bit more confidence in our own
    ideas and be bold and try to persue them. This does not have to mean,
    throw out Microsoft. But it does mean that if we think we can make more
    money with VMS, what is wrong with that? Persue it. By putting all eggs
    in the Microsoft basket, we are just another me-too company. Incidentally,
    I am not pooh pooh-ing the Affinity stuff. I think that makes sense to
    do. 

    And, lets not kid ourselves, Microsoft is not our friend. They are
    simply a business partner who'd partner with anybody worth their
    time. Let us create our own stuff that will pull Microsoft to us
    rather than the other way around.

    >Per

    >a.) We need more technical market analysis - marketing as a research
    >vehicle that provides specific technical market trends which OpenVMS
    
    Interesting.  Did you know that we have done market research, which
    we then ignore because it doesn't match our strategy?  Seriously.  I
    once attended a meeting regarding low-end servers, the meeting turned
    out to be a presentation of the findings of a outside research company.
    Their research showed a bunch of interesting things, and focused on
    customer perceptions of Digital, Alpha, and Servers...  the analysis?
    We needed to leverage our great OpenVMS base and the perception of
    Alpha.  That we could not be anything but a marginal UNIX player, and
    that the NT server business would be owned by Intel.
    
    As this guy was talking, a marketing/business type said: "I don't want
    to hear this.  OpenVMS has no future.  We've already made that
    decision.  Tell me how to sell NT servers."
    
    I had heard similar stories about a lot of this type of information
    comming in that was being ignored.  It was the first time I actually
    saw in action that all we really wanted was someone to confirm our
    decisions.
    

    >  <<< Note 5200.27 by UCXAXP.UCX.LKG.DEC.COM::GRADY "Squash that bug!
    (tm)" >>>
    
     >   I think there is little doubt that our OpenVMS customer base and
     >   revenue stream is larger than that of our Digital Unix and Windows
     >   NT/Alpha, combined.  That should be fairly obvious.  It is also
    obvious
     >   that this does not reflect the current market trend, which is
    strongly
     >    towards NT, and has been strongly towards some generalized Unix
     >   platform for quite some time now.  It will never be strongly
    directed
     >   towards OpenVMS, although in some niche markets and installed base
     >   markets it will continue to thrive for quite some time.
     
    
    OpenVMS has 5-7% of the server market today... By Gartner's estimates
    nt and Unix may divide 80% of the market between themsleves in 
    some form or fashion...
    
    That leaves 20% of the server marketplace that we already have a 
    presense in for OpenVMS to domininate in the "OTHER" catagory of 
    servers.
    
    Current market trends twords NT were current market trends twords
    unix not even two years ago, we need to differenciate and add value,
    in all three market segments... Not just the commodity worlds of 
    Unix and WNT...
    
    IBM, Amdal, Unisys, Hitachi, and a whole host of other companies are making
    a good living on profits from that 20% of the market WNT and Unix can't 
    touch.. OpenVMS should be taking it's unfair share of that market place
    and tailoring itself for sales there in the forseeable future...
    
    JMHO
    
    John Wisniewski 

    >>If you're not part of the solution, you're part of the problem.
    
    What about "If you don't understand the problem, your solution won't
    work."
    
    If you want to know what the problems are tune in here. If you want
    candid, and sometimes personal and emotional, suggestions of how to fix
    the problems, tune in here. OR Hum the tune "Don't worry, be happy",
    ignore the natterings of negativity and continue to cheer for solutions
    designed by those who don't know what the problem is.

re: .28
    
|    My suggestion would be to have a little bit more confidence in our own
|    ideas and be bold and try to persue them. This does not have to mean,
|    throw out Microsoft. But it does mean that if we think we can make more
|    money with VMS, what is wrong with that? Persue it. By putting all eggs
|    in the Microsoft basket, we are just another me-too company. Incidentally,
|    I am not pooh pooh-ing the Affinity stuff. I think that makes sense to
|    do. 

    There is, of course, plenty of money still to be made with VMS, but in
    the long term, i.e. five years and beyond, there is limited lifespan to
    that particular product.  It will be a miracle if we have half of the
    current OpenVMS customer base five years hence.   I'm not trying to be
    negative, but realistic.  We should do everything we can to maintain
    this installed base, but we should strive for realism.  Lots of our
    customers are just sitting back, waiting for NT and  Unix to prove
    themselves in commercial settings.  Five years, give or take, ought to
    bring a significant shift in that direction, IMHO.  This is a market
    trend that we simply do not have the power to overcome.  Confidence in
    our own ideas, yes, but not to the point of not-invented-here.  My
    thoughts are not based upon a lack of confidence at all. 
    
    Re: .30
    
|    Interesting.  Did you know that we have done market research, which
|    we then ignore because it doesn't match our strategy?  Seriously.  I
|    once attended a meeting regarding low-end servers, the meeting turned
|    out to be a presentation of the findings of a outside research company.
|    Their research showed a bunch of interesting things, and focused on
|    customer perceptions of Digital, Alpha, and Servers...  the analysis?
|    We needed to leverage our great OpenVMS base and the perception of
|    Alpha.  That we could not be anything but a marginal UNIX player, and
|    that the NT server business would be owned by Intel.
    
|    As this guy was talking, a marketing/business type said: "I don't want
|    to hear this.  OpenVMS has no future.  We've already made that
|    decision.  Tell me how to sell NT servers."
    
|    I had heard similar stories about a lot of this type of information
|    comming in that was being ignored.  It was the first time I actually
|    saw in action that all we really wanted was someone to confirm our
|    decisions.
    
    We do this all the time.  It's a shame, really, because it's so myopic.
    Our lack of market research isn't just an organizational quirk, but an
    indemic failure to respect the need for such information in the first
    place.  "I don't want to hear this" says it all.  More than any other
    element of our product strategy, it is this attitude, combined with the
    natural tendency toward not-invented-here, that will destroy this
    company if we don't learn to overcome them.  There is no profit in
    arrogance.
    
|    OpenVMS has 5-7% of the server market today... By Gartner's estimates
|    nt and Unix may divide 80% of the market between themsleves in 
|    some form or fashion...
    
    Growth in NT and Unix market share to 80% will surely reduce the
    OpenVMS market share substantially.  This has been going on for some
    time now already, just with Unix.  NT and  Unix can offer things with
    which OpenVMS, for all of its value and grace, can never compete.  The
    biggest of these is not a feature or anything about OpenVMS itself at
    all.  
    
    It's applications.  NT and Unix have more applications, and always
    will.  That means they'll sell more systems.  We want to sell more
    systems.
    
    We're a hardware company.  We sell systems.  Software, along with a
    number of other components, make our hardware valuable to potential
    buyers.  Software IS applications.  If OpenVMS is to exist in any sense
    whatsoever in, say, 2007, then we have to be very creative about how to
    do so.  Moreover, we have to focus on what our customers themselves
    want, not simply our own desire to see OpenVMS last forever.  We have
    plenty of research to indicate our customers love OpenVMS...but there's
    always a "but..." haunting that loyalty of late.  We need to address
    that anxiety and see it as an opportunity for the future.
    
|    That leaves 20% of the server marketplace that we already have a 
|    presense in for OpenVMS to domininate in the "OTHER" catagory of 
|    servers.
    
|    Current market trends twords NT were current market trends twords
|    unix not even two years ago, we need to differenciate and add value,
|    in all three market segments... Not just the commodity worlds of 
|    Unix and WNT...
    
    I agree, but that is a short-term strategy.  I've seen those reports
    too.  We should also prepare for the long term.  Sharing only 20% of
    the market with a dozen other vendors is going to be a very, very
    crowded niche.  I think we should shoot for domination of the 80%. 
    We're perfectly capbable of it.  That is anything but a lack of
    confidence.  In fact, it is only our lack of confidence that keeps us
    from this.
    
    The percentage of Unix market share for servers will probably shrink to
    40%, but the sheer number of systems will grow dramatically.  Both
    percentage and total systems market share for NT will also grow
    dramatically.  Why should we resign ourselves to a portion of the
    market that is, at best, shrinking percentage-wise and maybe remaining
    static in total systems?  What if we throw the weight of our OpenVMS
    installed base into growing the other portions of our business, namely
    NT/Alpha and Digital Unix?  Anyone looking at NT or Unix has to think
    about Intel vs. Alpha.  What if we were to give them a good reason to
    choose Alpha first, namely their legacy OpenVMS applications?  If
    they're looking at this already, they're more than likely looking at
    Intel.  
    
|    IBM, Amdal, Unisys, Hitachi, and a whole host of other companies are making
|    a good living on profits from that 20% of the market WNT and Unix can't 
|    touch.. OpenVMS should be taking it's unfair share of that market place
|    and tailoring itself for sales there in the forseeable future...
    
    The 20% will segment into niche markets based largely upon installed
    base applications.  As long-term strategies go, this segment is
    probably the hardest one to penetrate...  If these largely MVS-based
    applications couldn't or wouldn't move to NT or Unix, why should they
    move to OpenVMS?  As the applications market focuses on NT and Unix,
    what portion of this segment will new, non-installed base systems
    occupy?  What will it take to compete?  A lot.
    
    I'm far more inclined to embrace the market trends that we know
    already, and prepare to dominate that market, and make a bundle in the
    process.  If we shoot for some small piece of the 20%, I think we're doomed.
    
    tim
    

    critique (no matter how it is expressed - emotional, rational,
    negative, positive etc.) is always to be taken serious. Only if a
    person has no hope for the better he/she stops even critisizing. 
    
    I still like DIGITAL aka DEC, I still have hope, faith and I'm fully
    committed to do everthing I can (which might not be enough - or I 
    would have left some years ago. 
    
    
    - no spell check done sorry for any inconvenience this mighth cause -
                    

    There are a number of areas in which OpenVMS can conceivably be
    marketed to considerable advantage which have been neglected.
    
    We already market robustness (7X24X365...).
    We don't market security (an area becoming more and more important in
    	the networked world...)
    We don't market manageability much (for all the limitations, OpenVMS
    	can be managed in clusters, usually is, and a cluster can cover
    	an AWFUL lot of users and of storage of all kinds)
    
    Galaxies can prove that the OpenVMS paradigm is scalable to higher
    sizes than anything else, in useful ways even where a lot of I/O is
    going on.
    
    However, unless a marketing presence educates the market to the
    existence and usefulness of OpenVMS, in every area it has special
    competence in, sales folks will fear to sell it, customers will fear to
    buy it and nobody will bother figuring how to use or program for it.
    
    Unless/until this happens, the technical excellence that has been built
    over the last 20 years and more is of impact only to those few who are
    able to be convinced by word of mouth.
    
    Amen too, Fred.
    

From the market research perspective, and for the record, I've seen the 
40-40-20 quoted many times using Gartner Group sources.  Enough, already!
While Gartner Group does more qualitative analysis, the market sizings in 
the industry are usually based on quantitative data from analysts like 
IDC and Dataquest.  In the final analysis, the point is that most IT 
organizations will be operating in a mixed environment.

For the sizing specifics, let's look at the IDC baseline, which is available 
internally via the PM&D Workbench (http://pmd-wb.zko.dec.com).  Market sizing 
is generally done two ways, sized by unit shipments and sized by dollars of 
revenue.  For the server and workstation categories, from calendar years
95 and forecasted through calendar year 98, the baseline data says:

			CY95		CY96		CY97		CY98
UNIX Server units	413,256		473,906		548,043		612,067
UNIX Workstation units	713,292		787,256		845,865		902,577
   Total Unix Units   1,126,548	      1,261,162	      1,393,908	      1,514,644

NT Server units		200,013		406,914		594,534		758,421
NT Workstation units	468,792	      2,678,747	      6,823,068      10,320,877
   Total NT Units	668,805       3,085,661	      7,417,602	     11,079,298

Non-Unix/Non-NT Srv Uni 896,435	      1,034,721       1,227,291       1,348,900
 minus Netware Srv Unit 539,864		687,088		893,368	      1,037,025
 Proprietary Srv units  356,571         347,633         333,923         311,875
 Propretary Workstation  20,574          17,151          13,830          10,978
   Total Propri. Units  377,145         364,784         347,753         322,853

Total Srv mkt (units) 1,509,704	      1,915,540	      2,369,868       2,719,388
Total Workstation mkt 1,202,860       3,483,153       7,682,763      11,234,432
   Total Market Units 2,712,564       5,398,693      10,052,631      13,953,820

So, based on these unit sizings, and since the data is projected only to CY98,
we see that by CY98, the entire server and workstation market, relative to 
UNIT SHIPMENTS, is segmented into buckets 10.86% Unix / 79.4% NT / 2.32% propri-
etary.  But SERVER UNITS only looks like 23% Unix / 28% NT / 11% propri-
etary and 38% NetWare.  And WORKSTATION UNITS fall into buckets of 8% Unix /
92% NT.

It's a very different story when the market is sized by revenue.  $ are in 
millions.

			CY95		CY96		CY97		CY98
UNIX Server $		15,367		17,877		20,937		24,665
UNIX Workstation $	11,594		12,433		13,123		13,703
   Total Unix $		26,961		30,310		34,060		38,368

NT Server $		 1,617		 3,649		 5,854		 8,751
NT Workstation $	 2,212	        10,531	        23,327		33,141
   Total NT $		 3,829		14,180	  	29,181		41,892

Non-Unix/Non-NT Srv $   36,826		37,802		38,581		36,960
 minus Netware Srv $	 3,691		 4,654		 5,874		 6,681
 Proprietary Srv $	33,135		33,148		32,707		30,279
 Propretary Workstation    353		   259		   204		   160
   Total Propri. $	33,488		33,407		32,911		30,439

Total Srv mkt in $	53,810		59,327		65,372		70,377
Total Workstation mkt   14,159		23,223		37,654		47,005
   Total Market $	67,969		82,550	       103,026	       117,382

So, going by the server and workstation market size in revenue by the end
of calendar year 1998, we have partitions of 33% Unix / 36% NT / 26% propri-
etary (remainder Netware).  For SERVERS, it's 35% Unix / 12% NT/ 43% propri-
etary / 9% Netware.  For Workstations, it's 29% Unix / 71% NT.

In the end, a loyal customer in any size market stays loyal when treated
well.  Given that the average tenure of an IT manager is 20 months, this
risk adverse gang will only be loyal to vendors that help to minimize their
risk.  Platform longevity, safe passage through architectural migrations,
and advantageous cost of ownership stories help alot.  The market is infinitely
bigger than the high-performance niche that Digital has tailored its messaging
to in the past several years.  There is lots of opportunity out there.  It's
ours to lose.

    
    Thank you Fred, and others, for the responses and information.  Working
    here in Washington, we hear little if any news on VMS.  So your
    response gives me a broader perspective of the industry.    
    
    As stated in note .6, yes "I'm stuck on a pentium."  This is due to my
    work environment on the Microsoft campus where I see lots of Intel
    boxes from various vendors.  Thus my computing focus has changed from
    big server orientation of VMS to the desktop and server issues related to
    Microsoft products.   
    
    	I asked the questions in note 5 to re-gain a broader perspective of 
    the environment, and also, to let others know the perspective of someone 
    who works with lots of desktop products.  Again, thank you.
    
    
      
     
    
     
    Aside: (Yes, I know the named changed to OpenVMS, but when I worked in 
    the VMS group it was called VMS. Having been a bit twidler, I'm stuck in 
    the mind set -VMS bits are still VMS bits even though the textual name 
    change. )   
    
     

    
    Barb, VMS = OpenVMS.  I've "trained" myself to always type OpenVMS 
    instead of VMS, even though in haste I still backslide.  How are Lee
    and the kids?  Don't you love Seattle?
    
    Let me pose something that I've suggested at least once before in this
    conference.  There are two disparate futures being predicted:  The
    "Microsoft/Intel" future continues to push the PC on every desk, and
    lot's of small distributed servers.  The "Oracle/Sun" view is of large
    powerful servers, connected to cheap network appliances.
    
    In the Wintel senerio, everyone runs a homegenious combination of
    compatable hardware and software.  Both need this because they are
    driven by volume commodity concerns.  In a few years, they will have
    systems large enough to handle that mythical 3rd tier - where Alpha
    currently is positioned.
    
    In the Oracle/Sun senerio, data is centralized into large data servers,
    with cheap front ends (deja vu, it's terminals & timesharing) with the
    Internet and standards making both the server and front-ends
    indifference to the underlying hardware and O/S.
    
    The Wintel choice is a low-margin business, with only software makers
    making any real money.  The OraSun choice makes the front-end a
    commodity (Strong-Arm, PCs, etc), and keeps the backroom a high-margin
    business dominated by robustness, availability, and performance.
    
    Which business should *we* be in?
    
    What O/S is best positioned in terms of robustness, scaleability,
    availability, disaster tolerance, and performance?
    
    

    
    
    FWIW, I, being a Field Servant would like to submitt the following
    observation. 
    
    I maintain 2 highend (24X7, Mission Critical, OpenVMS) CI based
    VMSClusters. The revenue (hardware and software) from these two sites 
    is in excess of $150,000.00 per month for service. It takes about
    5 engineers servicing the "non-production" world to equal the cash
    flow from these 2 sites. Please note that this is not a full time job.
    I have several other tasks to perform since I at most spend 1 day a
    week servicing these sites. Hardware failures are rare, software issues
    are resolved remotely (generally).
    
    In the production world it has been my experience that since a customer
    has bet his business on the solution he selects he is disinterested
    in the cheapest commodity solution available. Lots o money is spent on
    redundancy that is never used. (so we load balance)
    
    I myself prefer to be in the HIGH END, HIGH MARGIN, HIGH RISK, HIGH
    VISIBILITY, and high stress arena.
    
    I have a worry to insert here in addition to my comment. One of my
    sites, Kaiser, has recently purchased about $40,000,000.00 worth of
    Hitachi Data Systems Iron to displace the Blue stuff in the room.
    Every time I look at it I can get the word Toyota out of my mind.
    
    On a good note we just replaced 16 DataGeneral systems with 8 (6 and
    2 spares) AS1000's. These buggers run Magic. More regions to follow. 
    
    Regards;
    
    Bob

    
    Based on my experience and actual Customer buying decisions of recent
    months, it appears many readers are misinterpreting the recent
    popularity of NT as being in all 3 tiers ie. tier 1 (desktop), tier 2
    (deparmental office server) and tier 3 (enterprise server).
    
    This is absolutely incorrect. NT is doing very well at the tier 1
    and tier 2 level, but is failing miserably at the tier 3 level. This
    is why MVS, HP-UX, DIGITAL UNIX and yes even OpenVMS (as evidenced
    by one our local big NT Customers just buying a whack of
    4100/8200/8400's to recentralize and deploy OpenVMS 7.1 / Oracle) 
    are doing so well these days.
    
    So, how can we leverage OpenVMS ? 
    
    First, we should only acknowledge that OpenVMS is dead if we also 
    state that MVS is dead. Anyone seen recent big MVS sales increases?
    
    Second, as previously stated by someone in this thread, applications
    and features are what drive the platform. So a suggestion:
    
    Just as Oracle shook the db world with it's huge TPC numbers 64
    bit AlphaUNIX  platform, what is needed for OpenVMS is a "killer"
    app to rekindle the interest of developers. How about the following
    suggestions:
    
    	- Team up (even pay majority of cost if required) with Netscape
          to develop killer Netscape high availability server platform.
          Reason is that Internet availability is becoming extremely
          high requirement right now as evidenced by recent AOL issues.
          All the focus on thin client fat servers will help to justify
    	  this as well.
    
    	- In conjunction with previous point, develop OpenVMS as the leader
          in IPV6 (now in beta around the world including DIGITAL UNIX).
    	  This has huge potential as IPV6 not only has complete
    	  compatibility with IPV4 (current TCP/IP), but also has autoconfig
    	  capabilities. Imagine talking to Customers and stating that they
    	  can start to de-emphasize DHCP, WINS, DNS and all the assorted
          TCP/IP address management  issues.... Add to this the added 
          security in IPV6 and you start to see some very convincing
    	  possibilities.
    
    p.s. btw - IPV6 was designed to be aligned on 64bit boundaries - does
    this not cound like great Alpha potential ?
    
    Given that OpenVMS is now almost completely managageable from a windows
    gui (backup as well shortly from what I have heard), then all of this
    could be a very nice package.
    
    Lets see now - OpenVMS 84xx cluster with memory channel and linked to
    datacenter in remote location, Netscape Server with Commerce Server
    pkg, IPV6 integration all managed by nice windows based gui.
    
    Whats not to like for all those big ISP's?
    
    :-)
    
    Regards,
    

RE: .32

>    There is, of course, plenty of money still to be made with VMS, but in
>    the long term, i.e. five years and beyond, there is limited lifespan to
>    that particular product.  

In other words, because the product will not be the premier platform that
our great-great-grandchildren use on their systems, we should ignore it
(and all those profits) today?

Of *course* OpenVMS has a limited lifespan.  The exact same thing can be
said of every single product and technology in the world, past, present
and future.  Windows 3.1 had a limited lifespan, but that didn't stop
Microsoft from pushing the heck out of it, and making $Billions from it.
And even if we take your number of 5 years as gospel (and I think it is 
too low), that still means that OpenVMS is going to be around for nearly
the entire market life of Windows 3.1, which had some limited success
in the marketplace, and made just a few bucks for its parent company. :^)

I have noticed an odd trend in the notes which advocate abandoning
OpenVMS, to wit: 

>    I agree, but that is a short-term strategy.  I've seen those reports
>    too.  We should also prepare for the long term.  Sharing only 20% of
>    the market with a dozen other vendors is going to be a very, very
>    crowded niche.  I think we should shoot for domination of the 80%. 

Why can't we do both?  Around 2 years ago Microsoft had 3 operating systems:
Windows 3.1, Windows NT, and this new thing in development called 95.  They
were perfectly comfortable selling 3.1, while field testing its eventual
replacement, 95.  Digital seems convinced (or determined, like the stories
told about the management studies in the last few notes) that NT can replace 
OpenVMS, (though a good case has been presented for why this is not true),
so why can't we be equally comfortable selling both?

Operating systems are not spouses, religions, or political entities such 
as countries, which demand unswerving loyalty and total rejection of all
other allegiances.  It is possible to believe in more than one operating
system at a time, and to actually acknowledge that different operating 
systems have different strengths which makes them appropriate for different
situations.

IMHO, every person in this string and others who has advocated OpenVMS as
a viable choice has stated it to be *PART OF* Digital's strategy, *ONE OF*
our offerings, and have stated that we should sell the customer whatever
they want that will solve the problem.

Again IMHO, almost every person in this string and others who has opposed
that position has stated it as "We should focus on 'foo', and therefore
we should not spend any effort/time/money/attention on OpenVMS, because
to do so would distract us from 'foo'".

Well that is just silly.  That is like saying that since we sell 15" 
monitors then we should not sell 17" monitors, because it might distract
us from the 15" monitor market.  Or that since we sell WORM optical drives
then we should not sell R/W optical drives, since the two technologies
compete against one another.  Or that now that we have a 4.3GB disk drive
we should eliminate our 2.1GB disk drive products, because our customers
might become confused with our two product lines.  Every one of those
things do in fact compete with each other, and requires careful thought
from Marketing, Sales, Technical Support, Engineering, and all other
areas of the company in order to properly position them as appropriate
for each customer situation.  But we do that every day, as a standard
part of our jobs.  It isn't that hard... :-)

We are a big company, with many bright and dedicated people.  Some of them 
can concentrate on one market, and some of them on another.  As long as 
they all make money, or have solid plans to grow into a profitable business,
there is nothing wrong with being in many lines of business.

As Tim said:

>    We're perfectly capable of it.  That is anything but a lack of
>    confidence.  In fact, it is only our lack of confidence that keeps us
>    from this.
 
-- Ken Moreau

    Greetings all,
    
    Our customer had a meeting with a VP concerning the future
    of VMS. So the person came down and told them in firm convincing 
    language that VMS is like very important to us. Also he introduced
    himself asbeing chartered to grow the VMS buisness. The customer
    asked me and why the fellow never sent some sort of letter 
    as a summation of his presentation. It took a day or two but I
    came up with a (what I think of as a clever remark) clever remark...
    "The guy's job is to grow VMS, he doesn't have time to write
    letters!"
    
    Fred

	He should make the time.

							mike

re .41:

You were lucky to get a VP who "told them in firm convincing language that VMS 
is like very important to us"

Wes Melling told a customer "If I were in [customer' name deleted] shoes, I
wouldn't bet my job on the VMS answer".

    I hope that you are not implying that we place any blame on Wes Melling
    for telling the truth.

When will some of you people realize that losing a VMS customer is *not*
equivalent to gaining an NT or UNIX cutomer? What makes you think that
these loyal VMS customers (sorry, how do you refer to them? legacy, 
tree-hugging luddites?) will stay with DIGITAL after that very same
DIGITAL has forced a total revamp of their IT strategy upon them. Yes I 
did say "forced"! It looks like DIGITAL itself is the only one that is
killing VMS by publicly abandoning it. 

Why isn't IBM embarassed to take money of their AS/400 customers (and 
they mostly program in RPG for Pete's sake!) Happy customer's -> Bigger 
profit -> Bigger dividend to shareholders. How old-fashioned!

It seems people like to speak of trends and that the latest trend is NT.
I see no problem with this! Get out there and sell NT to the masses but
*STOP* raping, pillaging and plundering the VMS client base to make up
the numbers to satisfy your theory on trends!

Anyway the base noter was looking for "good news" and here it is.
Regardless of what my or anyone else's views of the world are, if you
want facts just look at the number of VMS jobs that are out there at
the moment. And before anyone jumps up an says "That's just because
of legacy Y2K issues" take another look and you'll see that the
majority is *new* development! (I know, my contract is up and I've
been paying particular attention :-)

The second bit of good news is that DIGITAL still has the best
engineers that keep delivering the goods version after version.
It's because of these people that VMS *will* survive DIGITAL's version
of Pol Pot's "Year Zero" programme, well into the next century. VMS
*is* that much better and customers know it!

On the issue of positive suggestions here's two:-

1) DECdtm The sleeping giant!

   This single piece of functionality is capable of securing the current
   VMS installed base. All you have to do is document existing functionality.
   Now new development or testing costs just give customers what they have
   been asking for many times in both the VMS and Rdb notes conferences.
   No official statement of course, but the noises from Rdb engineering 
   indicate that the branch managment services are supported. 

2) The Affinity(Migration) program.

   Publish the names of the people that hijacked the Affinity Program and
   deliberately turned it into the Migration Program. A true Affinity 
   program is a good thing with new software moving to VMS but dedicating
   your life to getting VMS developers onto NT is not so smart.
   
Regards Richard Maher.

    re .41
    >  "The guy's job is to grow VMS, he doesn't have time to write
    >   letters!"
    
    
    grew by +1 VMS->WNT
    
    Sorry....  :-)

    There's a limited lifetime in OVMS primarily because we don't seem to
    tell anyone what it can do. How many in the market and press and board
    rooms think of VMS as something that runs (ran?) on old slow VAXes and
    haven't heard about it on Alphas? Or how many have seen it even
    mentioned in the last, say, 5 years? I run into people constantly that
    don't have a clue what it is, and others who think it's for VAX only
    and whose information is 10 years old. 
    
    In the software business you HAVE to keep telling folks what you have
    or they assume (usually with justification) that you're out of
    business. If you keep not telling them, you will be, soon.
    
    It is, but doesn't have to be, this way...
    

    
    FWIW, perhaps worth repeating - NT is doing miserably in the tier3
    environment. HP-UX and MVS are doing very well right now.
    
    Why is it so difficult to understand that there are two entirely
    different markets for tier 2 platforms and tier 3 ?
    
    At the risk of starting a rat hole, I suspect it is because our
    current marketing organization have primarily LAN backgrounds and
    continue to believe what PC mag's are saying.
    
    As an example, NT now has clusters - right? Therefore, what advantage
    does OpenVMS have ?
    
    Those that understand what real big limitations NT Clusters have in
    relation to OpenVMS clusters are likely not marketing types ie. the
    marketing types would not be able to differentiate these issues to
    either upper mgmt, other vendors or Customers.
    
    To bad, because MVS, HP-UX are now growing their business very well
    at a time where we continue to say "NT will replace OpenVMS .."
    
    Sigh, I like NT a great deal, but I certainly would not stand in 
    front of a Customer and state that it is a reliable, highly available
    platform to place my bread-n-butter app's on. It is a tier 1 (desktop)
    or tier 2 (dept) server - period.
    
    Digital marketing and mgmt currently has some serious misunderstandings 
    of what is happening in todays large systems environment. We are focussing 
    on tier 1 and tier 2 so much, that we have forgotten to do any marketing 
    with tier 3.
    
    :-(
    
    Regards,
    
    / Kerry
    

There are a lot of good things that VMS can do, but i wonder to what extent
VMS product management shot itself in the foot over the last few years with
things like:
	1.  no binary compatibility on Alpha (vs. the PDP-11 compatibility
		on the VAX)
	2.  trying to force customers to go to DECnet/OSI when many would
		have preferred to stay with Phase IV
	3,  keeping UNIX and PC interoperability/intercommunications
		functions as add-on products rather than integrated in the OS
I almost think we would have been better off pushing Alpha as a UNIX hot box
and bringing VMS into the mix more slowly, since the VAX business didn't have
to dry up overnight the way it did.  Then maybe we could've credibly pointed
toward a future RISC migration path while thinking through the issues that
led to the above 3 gaffes.

In short, i don't think it was all someone on high deciding to downgrade
VMS.  The VMS folks also painted themselves into a bit of a corner.

- paul

Talking about how we never promote our VMS offerings - a wonderful exception 
to this is the "inform" magazine, "the bi-monthly news magazine for Alpha, VAX 
and Intel users."

This customer mag has consistently backed VMS and its applications as well as 
the 1-3-9 stuff, and it makes a refeshing change. Front-page story in the 
April 97 (European) issue is headlined "Worldwide, DIGITAL keeps enterprises 
running." Sub-head: OpenVMS: Nothing stops it.

There's also a pull-out on Messaging solutions which covers ALL-IN-1 and 
MailWorksa AS WELL AS the Alliance stuff.

It seems some parts of the company are trying.

grahame

    .49  I believe that VMS itself is culpable, but not with the examples
         given.
    
    I believe VMS is to blame for not proactively engineering it's
    successor, and for helping to block next generation work.  This mostly
    occurred back when VMS was still successful.  One result was the
    shutdown of Mica/Prism, which lead to the development of NT.  I also
    blame VMS for it's NIH attitude when it was on top, and how they rolled
    over people and groups, creating a lot of ill will throughout other
    parts of the engineering groups.
     
    Example #1, the lack of VAX compatability on Alpha, came quite a bit
    down the road.  By that time, UNIX was the top dog.  Raw hot-box
    performance was all anyone wanted, and keeping with RISC concepts,
    software would deal with compatabily issues (VEST, etc).  The real
    problem here was that development on migration tools like VEST were
    turned off, and refocused to the NT side, too early on the VMS side. 
    Consider what a FX!32 equivalent for VAX/Alpha VMS would provide.  In
    addition, we failed to make sure to port ALL the infrastructure to
    Alpha.  Layered products, libraries, etc.  So customers were literally
    stranded.  I agree that if they had asked *my* opinion, I would have
    said that VAX binary compatability was a requirement... 
    
    Example #2, well, we haven't tried to force the change until V7.1.  But
    I agree that we should simply have made OSI a optional saveset, and
    declared victory.  You will find very few people on the fourth floor wh
    use it, except by accident the first time they installed V7.1.
    
    Example #3, I agree, but can't say it has contributed to the ultimate
    decline of VMS, except perhaps that because they are/were not
    integrated in the O/S, they didn't become fundamental to how the O/S is
    used, like RMS, or DECnet became.
    
    

Attached is a copy of a letter from Wes Melling which he has asked me
to post on his behalf.

This is a standard letter that goes out to customers who ask about
the future of OpenVMS or similar OpenVMS policy.  I apologize for any format 
issues  as it was saved in txt from a .doc file.  By Monday 3/31 I will 
place the.doc file on:

    		http://visavi.zko.dec.com   
    		click on Customer & Field Engagement

Sue Skonetski
OpenVMS Engineering


1997

An open letter from Wes Melling, Vice President,
Windows NT and OpenVMS Systems Group




The OpenVMS operating system environment holds a special place in the
computer industry. It was the centerpiece of the minicomputer
revolution, the first operating system to prove that scaling from
desktop to data center was practical, and the first to demonstrate that
clustered systems could achieve levels of availability well beyond
mainframes or "fault tolerant" systems.  It was, and continues to be, a
huge market success.

As we come to the turn of the millennium, we are pleased to see that
OpenVMS is retaining its place in the industry as a viable, profitable
business for Digital. Much of the OpenVMS stability comes from the fact
that its technology was so advanced that it has been easy to keep it in
a leadership position.  OpenVMS was one of the first operating systems
to offer DCE, the first with a CORBA-compliant object broker, and the
first to have three-tier client/server computing as a mainstream
paradigm (OpenVMS has over 12,000 three-tier production systems today,
while most operating systems are still working on their first 100). 

OpenVMS is still the fastest operating system environment in the world,
as recent TPC-C benchmarks prove, and it is still the most highly
available system on the market, as the recent fire at Credit Lyonnais
demonstrated. Because of OpenVMS' high reliability, it sells strongly
not only into the Digital customer base, but off-base as well.  Of the
first 1,000 Turbolasers (8400/8200) shipped, 480 ran OpenVMS, and 120 of
these went to customers who had never bought from Digital before. Their
reason for purchase was almost exclusively reliability.  (It is an
interesting note that the vast majority of the world's computer chips,
including Intel, Intel clones and Power PCs, are built on fabrication
lines that are run by OpenVMS, because of its reliability.)

In March of 1995, we announced an architecture which enables the
integration of both OpenVMS and Windows NT in a three-tier application.
The architecture leverages the known affinity between the two operating
systems (they had the same architect and are so similar in concept and 
operation that learning, using and managing both is simple).  The synergy 
offers the application base of Windows NT with the scaling, integrity, 
security and availability of OpenVMS. As part of the architectural 
announcement, we re-positioned OpenVMS as the ultimate high end for 
Windows NT.  Response has been enthusiastic, so much so that sales of 
new OpenVMS systems are stable (in a market segment that contains MVS, 
OS/400 and Tandem Guardian and is generally declining).  Because of 
clustering, OpenVMS customers tend to displace less than they install,
which means that stable new sales actually grow the accumulated base, 
with a healthy impact on our add-on, upgrade and service businesses.

In short, we have a thriving, multi-billion dollar business which, as a
percentage, is one of the most profitable systems businesses in the
industry.  We intend to invest to nurture it for our own benefit, and
that same investment protects our customers' investments and their
relationship with us.

1. 	First, we will continue to invest to make OpenVMS the ultimate high
end for Windows NT. To that end, we have announced full 64-bit
capability for OpenVMS as part of the Version 7.0 release in January
1995,  the largest single increment in functionality in OpenVMS history,
and we have added a panoply of middleware and application development
tools to support the three-tier, heterogeneous environment.  A strong
list of sophisticated new applications is flowing to this environment.

2. 	Second, we are extending OpenVMS into the Internet market. All our
future systems will ship Web-ready, and we have provided the software to
retro-fit our customers' installed systems.  The architecture we have
used for Windows NT integration is fully compatible with Oracle's
Network Computing Architecture, and we will be among the first to offer
web servers, mail systems and commerce systems in that architecture.  We
have also announced for the OpenVMS and Windows NT environments the only
solution currently available for carrying full transactional integrity
across the public Internet, and that facility is an easy extension for
our ACMS customers.

3. 	Finally, we are investing to raise the bar in availability and
scalability, since we see a dramatic increase in the number of systems
being developed which will demand large 24 x 365 and disaster-tolerant
production platforms.  We believe we can sustain the leadership OpenVMS
currently enjoys in this arena.

The enhancements we provide for OpenVMS will generally apply to both the
VAX and Alpha platforms, except where 64-bit capability is inherent to
the enhancement.

As with all software products, OpenVMS will someday enter an End-of-Life
phase, and you should know what our common practice is at that time.
When a software product enters End-of-Life, and OpenVMS has not, 
software support functions are phased out over time, based on the 
following hierarchy: 
		
		SOFTWARE SUPPORT  FUNCTIONS

			Add Major Functionality
			
			Add Minor Functionality
			
			Support New Standards
			
			Enhance Performance/Reliability
				
			Support New Hardware
			
			Fix Defects

As you can see, the first layer to be peeled off is "addition of major
functionality".  Second, is "addition of minor functionality".  Next, we'd 
stop supporting any new standards.  After that, no further enhancements 
to reliability and/or performance, support of new hardware, and finally 
fixing any defects in the software.  Past history suggests the wind-down 
strategy can take a decade; the "fix-defects" stage can go on long after 
that.  As you may realize, we have not begun executing this hierarchy 
strategy with OpenVMS.  We have no plans to do so in the near future, 
and are still adding significant new major functionality to OpenVMS.

With respect to VAX hardware, Digital is fully committed to the large
VAX installed base for sales and ongoing support for VAX systems.
Digital introduced three new VAX systems in October, 1996.  In addition,
Digital continues to focus on new features and enhancements while
improving investment protection with OpenVMS 5.5-2 compatibility, new
option qualification, and lower cost of ownership.  This is confirmation
of Digital's commitment to support our loyal customers.  Digital will
continue to meet the demand for new VAX systems, through roughly the
year 2000.  For new hardware acquisitions beyond that point, you should
be planning to use OpenVMS on Alpha or Windows NT.

We anticipate strong demand for new OpenVMS systems well into the next
decade.  Beyond that, it is hard to predict anything in this industry,
but one thing is clear.  Even if we ignore new systems business,
software maintenance is in itself a profitable activity for us.  With
the number of new mission-critical systems being installed this year on
OpenVMS, it would be reasonable to expect that we would still have
substantial software maintenance revenues ten years from now and beyond.
 While Digital's standard software maintenance commitments are for a
relatively short term, reflecting the rate of change in the industry,
broad market and sales trends have always been a better predictor of an
operating system's longevity.  The size (and loyalty) of the OpenVMS
installed base, combined with its observable resilience in adapting to
new markets, are, in my judgment a strong foundation for continuing
investment on both your part and mine.
 
Your local DIGITAL representatives and DIGITAL Business Partners would
be happy to discuss this further with you.  We can send you further
information on our most recent OpenVMS announcements which point to the
engineering investments we are making right now for product delivery in
the next year.  We believe those investments give some insight into the
way we think about the product.


Yours sincerely,





Wes Melling
Vice President, 
Windows NT and OpenVMS Systems Group
Digital Equipment Corporation



    

    
    Great !
    
    Now if we could only arrange to get that letter in the hands
    of our VARS, distributors and Customers ... Perhaps as an insert 
    to the FORUM newsletter mentioned here earlier ? Perhaps via the 
    DECUS mailing lists ? Perhaps as a selection on the 
    www.openvms.digital.com home page ?
    
    Regards,
    
    / Kerry
    
    

    Thanks for the posting Sue .... frankly I'd love to have a copy of this
    on Digital letterhead or as a glossy to leave with every  one of my
    OVMS customers!

> and it is still the most highly
> available system on the market, as the recent fire at Credit Lyonnais
> demonstrated.

For those of us who have never heard of a recent fire at Credit Lyonnais, what 
does this refer to?

PJDM

    Re .55, Credit Lyonnais - a major French Bank - had a fire that gutted
    its headquarter.  Thanks to their purchase and deployment of Digital's
    Business Recovery Server (aka Disaster-Tolerant VMScluster) prior to
    this fire, the customer was able to regain their normal operations
    within hours.
    
    --svb

    Great!  I needed a page 2 story for the DFWLUG Netsletter and Wes's
    Letter was wonderful...
    
    Thanks Sue,
    
    John W.

    Ditto for the SEWLUG (SouthEastern Wisconsin) DECUS mailing going out
    in the next couple of days. And for the next meeting I've got Randy
    Barth giving an update on the progress of UNIX and NT clusters trying
    to catch up to OVMS. That should warm the VMS folks hearts...
    
    Don Evenson

    As my chemistry proffesor would say, "if you are not part of the
    solution you are just sediment...", and I think there is way too much
    sediment in Digital right now.
    
    Steve
    

>    As my chemistry proffesor would say, "if you are not part of the
>    solution you are just sediment...", and I think there is way too much
>    sediment in Digital right now.
    
I thought this string was about negative sentiment about DEC, not excessive
sediment about DEC.

    
    	I heard it as "If you're not part of the solution you're part of
    	the precipitate."
    

    
    
    Now I'm confused again, just 2 days ago, I thought there was a thread
    about how expensive our memory is against the commodity market, as a
    result of our "qualification".  
    
    I was sure that note was in this conference.  Was it.
    
    If so, why did it "disappear"!
    
    /art

>I was sure that note was in this conference.  Was it.

It started around 5218.22