[Search for users] [Overall Top Noters] [List of all Conferences] [Download this site]

Conference 35.181::insurance

Title:Insurance Industry Conference
Moderator:ICPSRV::DOVE
Created:Thu Feb 18 1988
Last Modified:Wed Feb 05 1997
Last Successful Update:Fri Jun 06 1997
Number of topics:136
Total number of notes:551

63.0. "CONSULTANT COMMENTARY" by FOOZLE::BAKER () Wed Jan 11 1989 11:08

                            CONSULTANT COMMENTARY




This topic in the insurance VAXnotes will be a repository for pertinent 
papers by consultants (internal and external), and minutes on consultant 
sessions or presentations. We will try not to burden it with things that do 
not have a bearing on the insurance market.
    
T.RTitleUserPersonal
Name
DateLines
63.1UNIX VS SAAFOOZLE::BAKERWed Jan 11 1989 11:0953
George Schussel,  who runs a software consulting company in Andover, published
an analysis of the trade-offs between these two environments in the 9/12 issue
of COMPUTERWORLD.  Given our interest in both of these environments,  I have
summarized the key points.

.  UNIX provides access to a greater variety of hardware,  including some of
   the most advanced technology.  For example,  UNIX is the chosen OS for
   RISC products,  mini-supercomputers, and future supercomputers.  It is
   also supported by IBM and DEC hardware as well as other systems vendors.
   SAA, on the other hand only supports a subset of IBM hardware and software
   environments.  The implication is that application portability is greater
   with UNIX and the cost of hardware is much less (e.g., A 386 chip and
   UNIX provides a platform that costs only $15-20K versus systems from
   IBM/DEC that sell for $80K with similar power).

.  UNIX has application transportability today while SAA promises to deliver
   in the next 5-10 years and it may never be completed.  Today there is 90%
   application portability under UNIX and it will improve as standardization
   is stimulated by OSF and SUN/ATT.

.  UNIX will attract ISVs because the value of UNIX shipments in 1991 is
   forecasted by IDC to be about the same as S370 ships.

.  UNIX workstations will compete well against OS/2.  Today UNIX workstations
   have only 20% of DOS base but the number of UNIX users is more equal because
   there are 5-6 users/workstation with UNIX.

       -  DOS can run in compatibility mode under UNIX
       -  DOS and OS/2 run only on INTEL while UNIX runs on MOTOROLA, NATIONAL,
          MIPS etc.
       -  Key ISVs e.g., LOTUS, ASHTON TATE are porting to UNIX
       -  OS/2, unlike UNIX was not designed to support multi-user
          future multi-user versions of OS/2 will not be as good as UNIX
       -  UNIX can do today what OS/2 promises to do by 1990
       -  Apple law suit may delay availability of Presentation Manager
          in SAA.  UNIX has X windows and Open Look available today.
          IBM may be forced to use different user interface if suit
          is settled to Apple's satisfaction.

.  SAA is based upon IBM's proprietary SNA;  UNIX will leverage OSI
   and other industry standards

.  UNIX is less mature in commercial applications, particularly TP.
   Third part database vendors will help to fix this.  
   ( he may not know what AIX and ATT will do to improve commercial
     applications as well)

He concludes that UNIX will provide significant benefits to customers
who have the foresight to commit all or part of their application
development to UNIX.

    
63.2GARTNER CONFERENCE ON DECFOOZLE::BAKERWed Jan 11 1989 11:11715
                   I N T E R O F F I C E   M E M O R A N D U M

                                         Date:      9-Dec-1988 11:54am EST
                                         From:      nancy scull 
                                                    SCULL.NANCY AT A1 AT CISM AT MRO 
                                         Dept:      c&ism analyst relations
                                         Tel No:    297-6229

TO: See Distribution List

Subject: GARTNER GROUP CONFERENCE ON DIGITAL:  DETAILED REPORT


   Following is my detailed report on the Gartner Group "DEC Strategies 
   Conference" that Ken Olsen keynoted last week.  It has four sections:
   
   1.   Executive Summary
   2.   Detailed Report
   3.   Audience
   4.   Additional Information on Gartner Group
   
   This conference marks the beginning of an expansion in Gartner's Digital 
   coverage.  The plan is for a Digital service to emerge from their current 
   Small Computer Systems Strategies Service.  Gartner, in fact, privately 
   referred to this as their "First Annual" Conference on Digital, so we can 
   hopefully look forward to more.  There have been seven (7) Gartner
   Conferences on IBM.
   
   The use of non-Gartner Group speakers -- especially vendor speakers -- is 
   unusual.  Inviting Ken to keynote was a special situation, and his talk 
   was very well received.
   
   In general, Gartner is enthusiastic about our products and strategies.  To 
   provide value to their clients requires, however, that Gartner be 
   objective.  So, of course, there were some negative comments as well.
   
   Overall, the conference should generate good business leverage for both 
   Gartner and Digital.
   
   
   Executive Summary
   
   Some of the key points Gartner Group speakers made during this three-day 
   agenda include:
   
   o    "A long-popular computer industry myth is that DEC is incompetent at 
        marketing.  DEC certainly has been weak in advertising, hand-holding 
        and executive-level selling.  On the other hand, DEC has been 
        successfully segmenting its markets and tuning its products to market 
        needs since the early 1970s."
   
   o    DEC's market share growth will continue into the mid-1990s, after 
        which IBM will have the potential to stop it, but not reverse it.
   
   o    DEC will continue to take advantage of technology advances, 
        architectural extensions and clever implementation techniques to 
        ensure the viability of the VAX architecture through the 1990s.
   
   o    DEC will introduce a mainframe-class VAX during 1989, with 
        availability in the second half of the year.
   
   o    Digital's approach to the desktop environment will be a broad set of 
        "accomodation" strategies.  VMS VAXstations, Ultrix workstations, 
        Macintosh, OS/2, and MS-DOS products in that order will provide the 
        greatest level of integration and function in a VAX network.  
   
   o    Its Network Applications Support strategy will gain significant, 
        though not dominant, success.
   
   o    Digital now has the capability to be a world class leader in storage 
        products.
   
   o    The DECtp architecture is impressive, but the product set needs to be 
        enhanced to match this architecture.  DEC will be successful in OLTP 
        if:  1. it fills product holes in Intact and Rdb;  2. the development 
        and CASE environments mature; and  3. OLTP applications are 
        developed.  Gartner projects Digital's share of this market to grow 
        from 4% to 9% by 1992, primarily at the expense of IBM.
   
   o    IBM's lack of a distributed relational architecture and client server 
        model will hurt its efforts to expand its current niche to satisfy 
        "low-end" (less than 40 tps) OLTP requirements.
   
   o    Software development on a VAX is so far superior in productivity that 
        Cincom Systems uses the VAX environment to develop its commercial 
        software products for both the VAX and the IBM environments!  [Ed. 
        Note:  This discussion was an outstanding testimonial.]
   
   o    DEC has really done its homework in designing EMA as a 
        comprehensible, extensible network management architecture.
   
   o    Digital remains the leader in standards adherence and implementation 
        in its offerings.  Its challenges in the office systems marketplace 
        are two:  sufficient ISV support, which Digital appears to have; and 
        execution of the strategy, the real issue.  Pre-sales support, 
        including pre-implementation design, will be particularly critical.
   
   o    DEC has the opportunity to become dominant in manufacturing, and 
        through integration influence the rest of the enterprise.
   
   o    DEC is the IBM of technical computing; therefore, it can continue to 
        succeed even without being a technology leader.  "Relegating 
        technical computing to one slot on the last day of a three-day 
        conference devoted to Digital is a sign of just how far the premier 
        scientific computing company has come."
   
   o    For continued success, a new culture will be necessary.  DEC needs to 
        move away from offering competing product lines.  The recent 
        cancellation of the 12 VUP "Argonaut" project suggests a new, focused 
        strategy.
   
   o    If DEC is successful in introducing its more complete concept of cost 
        of ownership in the MIS environment, its user-friendly approach 
        should give it a clear advantage, but the challenge is to measure 
        those costs.
   
   o    As it moves more strongly into IBM's turf, it will also face a market 
        that is 75% leased.  In this environment, Digital will need greater 
        control of its secondary (i.e., leased and used equipment) market in 
        order to control its customer base.   U.S. computer lessors already 
        generate $.9 billion in revenues by leasing DEC equipment.  DEC could 
        capture well over half of its high end leasing market should it 
        introduce attractive financing packages of its own.
   
   o    Both Gartner analysts and customers felt that DEC's biggest challenge 
        will be to establish an organizational structure to provide 
        "IBM-like" cradle-to-grave customer support -- from pre-sales to 
        implementation.
   
   The handout material included a number of Digital architecture charts that 
   we have provided them in our briefings.
   Detailed Report
   
   Following are some more detailed comments on each of the presentations in 
   order of their delivery.  (It's always interesting to see how much "non
   public" information Gartner receives from industry sources).  Gartner 
   itself will also be publishing -- and selling -- an even more extensive 
   report sometime in January.
   
   
   1.   DEC's Strategic Challenges:  Peter S. Schay, Gartner Group
   
   o    "A long-popular computer industry myth is that DEC is incompetent at 
        marketing.  DEC certainly has been weak in advertising, hand-holding 
        and executive-level selling.  On the other hand, DEC has been 
        successfully segmenting its markets and tuning its products to market 
        needs since the early 1970s."
   
   o    Digital must emphasize growth into new markets and focus on 
        "enterprise" needs in order to achieve growth goals.
   
   o    AIA will solve Digital's SAA problem before it even becomes a 
        problem, with rollout projected to start in 1990.
   
   o    Ultrix currently has poor price/performance because VAX is a value 
        priced product, but the Unix market is a commodity-priced 
        environment.  Hence, the price of the VAX hardware causes the 
        Ultrix/VAX combination not to be price/performance competitive.
   
   o    Expect Digital to integrate the POSIX kernel into VMS in order to be 
        ready to respond to RFPs that specify UNIX interfaces.
   
   o    Digital's organizational structure is not set up to provide IBM-like 
        pre-sales support.  A poll of the audience showed that most attendees 
        (about 80%) were not satisfied with DEC sales, support, and service.  
        [Ed. Note:  Would any vendor fare better with such direct 
        questioning?]
   
   o    It will take years for DEC to overcome its cultural inertia and 
        instill a marketing orientation in every sales representative, 
        software specialist, field service engineer, order administrator, 
        marketing manager, and vice president.  But most of DEC's strategies 
        appear to be on target.  And DEC's upgrading of sales and support 
        skills will, over time, lead to increasing effectiveness at 
        enterprise-level marketing and sales.
   
   o    DEC's market share growth will continue into the mid-1990s, after 
        which IBM will have the potential to stop it, but not reverse it.
   
   
   2.   Integrating the Enterprise:  Ken Olsen, Digital Equipment Corporation
   
        Ken talked for about 40 minutes, followed by 20-30 minutes of Q&A.
   
        Topics he covered include:
   
   o    History of Digital's strategy:  interactive, end-user orientation; 
        involvement with third parties; support for standards; etc.  Digital 
        provides a unique solution, and has a long-term commitment to and 
        experience with third-party suppliers.
   
   o    PCs:  major focus is integrating PCs.  Different from IBM and others.  
        IBM tells customers what desktop PC device to use; Digital gives 
        customers the ability to choose what PC they want and know they can 
        plug it into a Digital network and it will work.
   
   o    Digital is definitely a player in the commercial market.  Today that 
        represents 50% of our business.
   
   o    UNIX and standards:  Digital has a long history of supporting 
        standards.  It takes great discipline, but it's very important.  He 
        said he has assigned some of his best software people, i.e., those 
        who were responsible for the development of VMS, to work with OSF and 
        enrich Ultrix to have similar functionality.
   
   The audience submitted well over 100 questions.  Obviously there was not 
   time to answer them all.  But they were friendly, not hostile, and 
   demonstrated a genuine desire to learn more about Digital's strategies.
   
   Overall, Ken was very enthusiastically received by both Gartner and the 
   audience.
   
   
   3.   VAX Future Directions:  Craig S. Symons, Gartner Group
   
   o    DEC will continue to take advantage of technology advances, 
        architectural extensions and clever implementation techniques to 
        ensure the viability of the VAX architecture through the 1990s.
   
   o    DEC will introduce a mainframe-class VAX during 1989, with 
        availability in the second half of the year.
   
   o    Gartner Group believes that DEC has been developing a mainframe-class 
        VAX code-named Aquarius.  It was to have been a liquid-cooled system 
        operating at 25 to 30 VUPS per processor.  They now believe that DEC 
        will introduce both Aridus -- an air-cooled, 15-VUP implementation of 
        the VAX architecture using new 10,000 gate Motorola gate arrays and 
        with an optional attached vector co-processor -- and a faster version 
        (Aridus II) that will take the place of the Aquarius processor.
   
   o    DEC will continue to drive down the cost of desktop VAX computing 
        while increasing its efforts as a server vendor to integrate 
        desktops.
   
   o    DEC will introduce a new midrange system in 1990 based on a new CMOS 
        VAX processor, which will all but eliminate the VAX 8000 product 
        line.
   
   o    Future implementations of the VAX architecture, in conjunction with 
        VAX systems software, will enable DEC to maintain its 
        price/performance advantage over IBM.
   
   o    Users should focus on applications, not on hardware architecture or 
        operating system kernels.
   
   o    The 8800 is not selling well because demand for high end systems 
        generally is low and the 8800 technology is old.
   
   
   4.   DEC and the Desktop:  Myron Kerstetter, Gartner Group
   
   o    Digital's approach to the desktop environment will be a broad set of 
        "accomodation" strategies.
   
   o    Expect several announcements in early January.  "Firefox" will be 
        multiple microVAX 3000 chips on a single board selling for $50,000 to 
        $100,000.  The new VAXstation, which will not be "in place 
        upgradable" with the VAXstation 2000, will be priced in the $20,000 
        range and will be the best Windows entry.  The Personal VAX will be 
        more of a software than a hardware product, MS-DOS compatible through 
        emulation, probably from a third party, like Phoenix Technologies, 
        and will therefore, not be a strong replacement for other PCs now on 
        the market.
   
   o    Gartner expects Digital to begin to accomodate the 3270 market 
        through an enhanced Ethernet server in the not too distant future.
   
   o    Later, in about the first half of 1990, Digital will introduce 
        another new VAXstation based on a microVAX "4000" chip, running at 
        5-6 VUPS.
   
   o    The X Windows Graphical User Interface (GUI) standard, when and if 
        defined, will be key in affecting DEC's market advantage at the host 
        level.  Although DEC has repeated that it would adopt and support 
        whatever GUI is selected by OSF, selection of something other than 
        XUI (e.g., the H-P/Microsoft Common X Interface) would significantly 
        disrupt DEC's plans.
   
   o    DEC's Network Applications Support strategy will gain significant, 
        though not dominant, success.
   
   o    VMS VAXstations, Ultrix workstations, Macintosh, OS/2, and MS-DOS 
        products in that order will provide the greatest level of integration 
        and function in a VAX network.
   
   o    Availability of entry-level VAX PC LAN server products added to 
        existing larger VAXs is a basis for a viable workgroup-to-enterprise 
        interconnectivity strategy.
   
   
   Day 1 Key Issues:  All Gartner Speakers for Day 1 (Q&A)
   
   o    Digital now has the capability to be a world class leader in storage 
        products.
   
   o    Don't try to compare Digital performance to IBM because their 
        approaches to computing are different.
   
   o    Digital will introduce fault tolerant products, probably 
        microVAX-based, but not until 1990.
   
   
   5.   DEC's On-Line Transaction Processing Initiative:  Jeff Schulman, 
        Gartner Group
   
   o    Digital's DECtp architecture is impressive, but the product set needs 
        to be enhanced to match this architecture.  DEC will be successful in 
        OLTP if:  1. it fills product holes in Intact and Rdb;  2. the 
        development and CASE environments mature; and 3. OLTP applications 
        are developed.  It will be a tough battle, but DEC has a strong 
        chance.  Gartner projects Digital's share of this market to grow from 
        4% to 9% by 1992, primarily at the expense of IBM.
   
   o    DEC must quickly transition to being a solution supplier.  DEC's 
        marketing force must begin to sell solutions, not just "iron".  
        "Benchmarks are not the key indicator; the key indicator is the 
        solution set you need."    
   
   o    DEC is making significant investments in OLTP, despite belt 
        tightening elsewhere.
   
   o    Other OLTP vendors (Tandem, Oracle, RTI, and Sybase) will exploit 
        areas where DEC currently is weak, namely fault tolerance, true 
        distributed database management, portability, tps, application 
        availability, tool sets, price/performance, and solution 
        comprehensiveness.  Further, where interoperability among disparate 
        environments is an important user criterion, DEC will have less to 
        offer than its ISV competitors.
   
   o    Requirements for success in this environment include:  fault 
        tolerance; low cost per tps; data integrity; application tool kits; 
        etc.
   
   o    DEC is working toward announcement of a fault tolerant product in the 
        early 1990s.  (Gartner defines true "fault tolerance" as no more than 
        5.25 minutes of downtime per year.)
   
   o    IBM's lack of a distributed relational architecture and client server 
        model will hurt its efforts to expand its current niche to satisfy 
        low-end OLTP requirements.  (Gartner defines low-end requirements as 
        less than 40 tps, which they believe accounts for more than 50% of 
        users' requirements.)
   
   o    DEC still needs applications and systems management to prove it can 
        really handle transaction processing.
   
   o    DEC's strengths in OLTP include:  good networking; the promise of a 
        distributed client/server architecture; scalability; low cost per 
        tps; and "TP University".
   
   o    Current DEC customers will grow into OLTP as their needs mature.  
        Acquiring new OLTP customers will be difficult in the short term.  By 
        1992, however, Gartner Group thinks that through a strategy of 
        offloading selective applications, Digital could generate as much as 
        40% of its OLTP revenues from organizations that are not now DEC 
        customers, but who see the value of distributed technology.  To 
        accomplish this, DEC must develop "CICS on a VAX", and handle 
        concerns about security, systems management, and control.
   
   o    Heretofore, IBM has controlled the data, relegating DEC and Tandem to 
        roles in the front end environment.  Controlling the data will be key 
        to future strength in OLTP, so DEC will try to force people to 
        migrate to Rdb.
   
   o    Gartner Group expects DEC will focus more on Intact than on ACMS over 
        the longer term, perhaps by having one meld into the other.  Having 
        both is "software pollution".
   
   o    DEC's strategy is twofold:  protect the backend and open up the front 
        end.
   
   o    The postponement of publishing DEC's audit report on its benchmarks 
        is "troubling", for two reasons:  IBM has published an audited report 
        that raises some questions; and DEC's numbers are very aggressive, 
        but as yet unsubstantiated by a full report.
   
   
   6.   Software Development on VAX:  Stan Sewall, Cincom Systems, Inc.
   
        Stan delivered one of the best testimonials one could ever want to 
        hear regarding the improved productivity he achieves by developing 
        software on VAXes to run in both VMS and IBM environments.  He showed 
        some extremely impressive statistics, and concluded by saying that 
        even if his internal DEC charges grew from $150 per hour to $600 in 
        the face of decreasing IBM costs, he would still do this development 
        in the VAX environment because it is still that much more productive 
        than the IBM environment.
   
   
   7.   Networking and Distributed Computing:  Steve Wendler, Gartner Group
   
   o    DEC must move into PC and enterprise markets to maintain growth.    
        Its recently announced programs for Enterprise Management Services 
        and networking will significantly expand the scope of its mission and 
        strengthen traditional weaknesses, but execution is key.  It will 
        agressively pursue strategic partnerships to address the needs of 
        corporate customers, a new approach for the networks group.
   
   o    DEC has really done its homework in designing EMA as a 
        comprehensible, extensible network management architecture, but it is 
        2 1/2 years behind IBM's NetView in network management, and offers 
        nothing new.  Therefore, it will have limited success against IBM's 
        inside track with MIS executives.
   
   o    DEC is strong in local area networking, but has weaknesses in wide 
        area networking that must be addressed.
   
   o    A show of hands indicated that about 75% of the audience would like 
        to link 3270-type devices to VAX networks, but no one knows of an 
        adequate solution.  Expect some level of 3270 support before 1991.
   
   o    DEC will introduce a DECnet backbone service for X.25, analogous to 
        IBM's XI, no later than 1990.
   
   o    DEC is unlikely to introduce a DECnet backbone service for SNA before 
        1991.
   
   o    DEC is absolutely serious about OSI, and will continue to lead the 
        vendor community with aggressive statements of direction and 
        subsequent product introductions supporting OSI.
   
   o    DEC has chosen the integrated strategy route because it will perform 
        better than gateway solutions.
   
   o    DEC will maintain a value-added connectivity advantage over IBM by 
        continuing to move the target and shifting discussion to its 
        strengths, most notably applications.
   
   o    NAS currently does not address IBM servers like the 3XXX, AS/400, or 
        OS/2 servers.
   
   o    Don't expect any DEC endorsed DECnet/OSI token ring solution anytime 
        before 1990.
   
   o    DEC will not enter the PBX business, but its programs for Computer 
        Integrated Telephony, Enterprise Management, and ISDN will lead to 
        cooperative relationships with most of the telecommunications 
        vendors, while relations with AT&T will be strained.
   
   
   8.   DEC-IBM Coexistence:  John Adourian, IBM Corporation
   
        Because investment protection is key, IBM must offer leadership 
        application solutions and a portfolio of coexistence products to be 
        successful.  As part of this presentation, IBM handed out a brochure 
        describing seven of its partners for accomplishing such coexistence 
        and interoperability.  They also described the support IBM provides 
        in the form of testing centers, design review, guidance and training, 
        etc.  The speaker concluded by saying he hoped he had convinced the 
        audience that "IBM Has It Now".  When asked when IBM will introduce 
        OSI, the answer was:  "We have our best people working on it."
   
   
   9.   Office Systems:  Beyond All-in-1:  Christine Hughes, Gartner Group
   
   o    The OIS business is going through a retrenchment from many vendors to 
        just three -- IBM, Digital, and Hewlett-Packard.
   
   o    Digital's release of its Compound Document Architecture (CDA) 
        preempts the other major OIS vendors, including IBM, and provides a 
        comprehensive environment for document handling.  Its success will be 
        highly dependent on third-party endorsement.
   
   o    Underscoring Ken's comment, there will continue to be a market for 
        terminals; perhaps even moreso as we move toward actually automating 
        executives.
   
   o    There is an accelerated move within Apple to play an integrated role 
        with Digital.  This is key to Apple.
   
   o    Digital has 94% of the business videotext market.
   
   o    Both IBM and Digital share a common objective for enterprisewide 
        computing, but Digital currently has somewhat of a lead in delivering 
        it:  CDA; ISV endorsement; and Apple/Digital joint development 
        programs.
   
   o    Digital's offerings remain the leader in standards adherence and 
        implementation.
   
   o    Digital's challenges in the office systems marketplace are two:  
        sufficient ISV support, which Digital appears to have; and execution 
        of the strategy, the real issue.  Pre-sales support, including 
        pre-implementation design, will be particularly critical.
   
   o    Question:  If OSF selects CXI (Microsoft/H-P), will this be a major 
        setback for DEC?  No.  DEC will support whatever OSF selects.
   
   
   10.  DEC's Manufacturing Market Strategy:  T. Lee Wylie, Gartner Group
   
   o    DEC has the opportunity to become dominant in manufacturing, and 
        through integration to leverage that influence across the rest of the 
        enterprise.
   
   o    There are probably more manufacturing applications written for VMS 
        than for any other operating system; therefore, VMS will be here for 
        a long time to come.
   
   o    DEC is clearly thinking ahead and moving to offer key services the 
        market will need.  Gartner Group gives Digital very high marks on its 
        introduction of Enterprise Services.
   
   o    The Allen Bradley agreement represents DEC's commitment to sell VAX 
        equipment into the manufacturing environment even though it means 
        giving up a piece of the business.
   
   
   11.  Technical Computing:  Workstations and Minisupercomputers:  George 
        Weiss, Gartner Group
   
   o    DEC is the IBM of technical computing; therefore, it can continue to 
        succeed even without being a technology leader.  "Relegating 
        technical computing to one slot on the last day of a three-day 
        conference devoted to Digital is a sign of just how far the premier 
        scientific computing company has come."
   
   o    Value pricing works.
   
   o    The technical marketplace has fragmented significantly into many 
        architectures.  This represents an opportunity for DEC, but the 
        market will have to wait until 1989/1990 for Aridus, DEC's high 
        performance, vector processing system.
   
   o    This market has no leader today, so DEC's impact will be a major 
        market boost and DEC could capture a commanding share -- despite its 
        lateness -- because buying DEC is an easy decision in this 
        environment.
   
   o    Aridus will target primarily the technical market.  DEC may even try 
        to siphon off some of the Cray applications with it.
   
   o    Competition is heavy here, but the venture capitalists have been 
        burned -- to the tune of $.75 billion -- so more new entrants are 
        unlikely.  
   
   o    Convex is currently the only solid (i.e., profitable) player in the 
        high performance market.  It will need to get as much product out as 
        possible before DEC announces Aridus because DEC's entry will freeze 
        the market.
   
   o    DEC is not likely to invest in all the latest gee whiz technologies 
        because it needs to focus on OLTP, desktop, and other new markets.
   
   o    The 8800 will probably have a short life.
   
   o    "IBM can't touch the engineering/scientific departments."  It will 
        probably need to introduce another 4381-level product with a vector 
        facility.  It will also need to buy market share.  But Gartner does 
        not project high success for that strategy.
   
   o    RISC offers some very good opportunities for DEC, who needs to be in 
        there fighting.  Followon products in 1990/91 could be very high 
        performance.  Firefox, likewise, will fill a niche.  However, neither 
        of these systems are synergistic and are typical of DEC's new 
        approach to use existing technology -- either internal or from 
        outside suppliers -- and depend on standards and recompilation to 
        weave a loosely-knit fabric.
   
   o    The 6200 is a very adequate system today, with good life left in it, 
        but Rigel will offer much better performance.
   
   o    DEC is unlikely to focus so heavily on commercial applications as the 
        big revenue generator of the 1990s as to cost it market share on the 
        technical side of the business.
   
   
   12.  DEC Secondary Market and Leasing:  Jann-Marie Halvorsen, Gartner 
        Group
   
   o    DEC, unlike IBM, subscribes to a strategy of growth through 
        expansion, not obsolescence.  As DEC moves into IBM's backyard, it 
        will continue to be able to maximize its revenue without forced 
        obsolescence through its symmetric multiprocessing capabilities.
   
   o    The 8800 is overpriced at $75,000 per VUP, and will need to be 
        discounted.
   
   
   o    For continued success, a new culture will be necessary.  DEC needs to 
        move away from offering competing product lines.  The recent 
        cancellation of the 12 VUP "Argonaut" project suggests a new, focused 
        strategy.
   
   o    If DEC is successful in introducing its more complete concept of cost 
        of ownership in the MIS environment, its user-friendly approach 
        should give it a clear advantage, but the challenge is to measure 
        those costs.
   
   o    If a user presents the used equipment alternative to his DEC 
        salesman, he can usually get an additional 10-30% discount.  Gartner 
        thinks DEC is leaving money on the table with this strategy, and will 
        need to change as it moves into the high end with Aridus.  Its 
        strategy here will need to involve heavy discounting to maintain 
        account control -- either by actual price reductions, by bundling 
        services, or by introducing attractive financing packages at low 
        rates with relatively aggressive residual assumptions.
   
   o    DEC currently represents about 6% ($.9 billion) of the 1988 revenues 
        of U.S. computer lessors, growing to 9% ($3.3 billion) by 1993.
   
   o    DEC needs more control of its secondary market in order to control 
        its customer base.  It could capture well over half of its own high 
        end leasing market.
   
   o    Comdisco, the largest leasing and used equipment supplier, is a 
        primary competitor to DEC.
   
   o    The effect of DEC's new warranty policy is a 6-9% price increase.
   
   o    As DEC moves into the high end, it faces a market that is 75% leased.  
        DEC will need to establish appropriately attractive financing 
        packages, which in turn will decrease the amount of used equipment 
        available to compete with orders for new equipment.  There is an 
        opportunity for significant revenue growth and greater this account 
        control through these measures.
   
   
   13.  Panel:  Doing Business with DEC
   
        Panelists included representatives from System One, Liberty Mutual, 
        United Technologies, and Virginia Power.
   
        All were quite candid.  The biggest issues revolved around sales and 
        sales support, particularly pre-sales.  The panelists believe that 
        DEC doesn't provide the kind of cradle-to-grave service they get from 
        IBM -- for free.  No problems with hardware maintenance, but software 
        support got mixed reviews.  All seemed to agree that while it was 
        difficult to reach an understanding of what should be provided, once 
        that hurdle was past, Digital generally delivered quality results.
   
        In the end, all stated that they have and/or would put a mission 
        critical application on Digital systems.  
   
   Audience
   
   The 300 attendees represented over 150 companies, as listed below:
   
   Advanced Micro Devices       Aetna                   Agriculture Canada
   Air Products                 Alabama Power           Alberta Gov't Tel.
   American Heart Association   American President Cos. Ameritech
   Amoco                        Applied Learning Int'l. Aristech Chemical
   Arthur Andersen              Associated Group        AARP
   ARCO                         ASB Capital             AT&T
   Baltimore Gas & Electric     Bank of Boston          Bank of New York
   Bankers Trust                Bear Stearns            Bell Atlantic
   Borden                       Bull                    Burlington Northern
   BASF                         BC Hydro                BC Systems
   Canada Post                  Capital Associates      Carolina Power&Light
   Caterpillar                  Chase Manhattan Bank    Chemical Bank
   Cincom Systems               Cognos                  Combustion Eng'g.    
   Comdisco                     Conner Peripherals      Corning Glass Works
   CNG Producing                Data General            Data Switch
   Dial                         Digital Equipment       Dow Chemical
   Duke Power                   Dun & Bradstreet        Dupont
   Eastman Kodak                Electronic Data Sys.    Eli Lilly
   Ernst & Whinney              Exec. Office of Pres.   EG&G
   Fast Food Merchandisers      Fed. Res. Bank of Minn. Fermilab
   Florida Power                Fujitsu                 FMC
   General American             General Electric        Grumman Data
   GTE                          Hardee's                Harris Trust & Sav.
   Hartford Insurance Group     Hewlett-Packard         Household Finance
   Houston Light & Power        Imperial Oil            Imperial Tobacco
   Info Builders                Inland Steel            Int'l Computers Ltd
   Integral Systems             Intel                   International Paper
   Island Creek                 IBM                     IBM Canada
   IBM Denmark                  IBM Rolm Systems        ICI Americas
   IDC Financial Services       ISI - Mars              ISTC
   Joiner Associates            Kennametal              Liberty Mutual
   Lockheed                     Lotus                   LTV Aerospace
   Mallinckrodt                 Manitoba Telephone      Mfrs. Finance
   Manufacturers Hanover        Manville Sales Corp.    McCormack & Dodge
   McDonnell Douglas            Medical Center Delaware Michelin
   Microrim                     Ministry of Tourism     Monsanto
   Morgan Guaranty              MNC Information Svcs.   Nat'l Inst Stds.&Tech
   Naval Strategic Sys. Command Navistar                Neuberger Berman
   Nova                         NCNB                    NCR
   NEC                          NYNEX                   Olivetti
   Oracle                       Paccar                  PacifiCare Health Sys
   PacifiCorp Capital           Pellerin Milnor         Penn State University
   Petro Canada                 Philip Morris           Phoenix Mutual Life
   Pioneer Data Systems         Pioneer Std. Electronic Pitney Bowes
   Price Waterhouse             Prime Computer          Procter & Gamble
   Prudential                   PRC Public Mgmt. Svcs.  Ralston Purina
   Rockwell                     Ross Systems            RJ Reynolds Tobacco
   RR Donnelley                 Schering-Plough         Scott Paper
   Seward & Kissel              Shell Canada            Siemens
   Simpact Associates           SmithKline Beckman      Software A.G.
   SoundView Financial Group    Southern Chem'l Alberta Southern Company
   Sovran Bank                  St. Paul Companies      Sun
   Sun Microsystems             Sunstrand               Syntex
   System One                   SD Warren               SNET
   T. Rowe Price                Tandem Computers        Tektronix
   Tennessee Eastman            Texaco                  Textron
   Ungermann-Bass               United Airlines         United Technologies
   United Telecommunications    University of Chicago   Univ. of Pennsylvania
   Upjohn                       UNISYS                  US Dept of Commerce
   US Leasing                   US Navy                 Virginia Power
   Volkswagen                   VM Software             Wang
   Wellington Management        Wells Fargo Leasing     Weyerhauser
   Xerox                        3M
   
   
   Additional Information About Gartner Group
   
   Gartner Group is about $40 million in revenues worldwide.  It is owned by 
   Satchi and Satchi, a billion dollar, UK-based consulting firm conglomerate 
   consisting mostly of advertising firms that it has acquired.  It is also 
   affiliated with SoundView Financial Group, Inc., alias Gartner Securities, 
   which cohabitates Gartner's offices.
   


    
63.3OPEN SYSTEMS, UNIX, & SNAKE OILFOOZLE::BAKERWed Jan 11 1989 11:2863
Subject: Marc Schulman on Open Systems, Unix and Snake Oil


From:	NAME: norm becker                   
	FUNC: corp. prod. oper.       
	TEL: 276-9018             <BECKER.NORM AT A1 at FHQ at OGO>

To:	See Below
Salomon Bros. just released Marc's publication on the facts and fallacies
of the open systems debate.  His views underscore much of what Ken intended
in his "snake oil" comment.  I have summarized his analysis below.

.  Sun and AT&T have championed open systems as a way to control the
   Unix market.  They have successfully equated open systems with
   the concept of availability from multiple vendors; hence Unix systems
   are open by definition.

.  As a corollary of the first statement, proprietary systems are by 
   definition closed.  The movement toward open systems/Unix is a movement
   away from IBM and DEC's proprietary environments and is viewed by many
   as a trend toward the commoditization of the computer industry and lower
   industry profits. (Thus explaining why computer stocks are unfavorably
   viewed by investors and why valuations do not reflect the financial results 
of industry participants)

.  OSF will redefine the concept of openess away from the narrow notion of the 
operating system such that proprietary systems will be viewed as open as long
   as they provide for application portability.  An open computing
   environment is enabled by standard interfaces and protocols and does not
   require identical hardware architectures or identical operating systems.
   (This is pretty much what IBM does with SAA for its systems and what we
    do with our "extended architecture" for VAX and industry standard
    products)

.  Many people confuse operating system portability with application
   portability.  Unix does not provide source level compatibility because
   there are multiple versions of Unix.  If there were a single Unix,
   applications would not be portable without recompile because of differences
   in hardware architectures that creates binary code incompatibility.
   Consequently,  Unix should not be mistaken as a synonym for open system.

.  Sun will lose the battle for micro based RISC platform (especially
   in commercial markets) to Motorola 8800 and to MIPS.  Technical experts
   have told him that SPARC is an inferior implementation of RISC.
   (curiously, IBM's RT PC architecture is not discussed in this context,
    although AIX is presented as a superior nucleus for a future Unix
    relative to the existing kernal from ATT)

.  It will take 5-10 years for Unix to rewritten such that it will gain
   broad acceptance in commercial markets.  Its major shortcomings are
   lack of OLTP,  poor I/O performance,  file structures inappropriate
   for commercial applications (unstructured),  random disk allocation
   methods (rather than contiguous), inability to support files spread
   across two or more disks,  poor recovery from network failures.
   OSF has a two year lead over Sun/ATT in providing a better Unix since
   AIX contains many IBM enhancements over System V.


Regards
Norm

    
63.4INSURANCE OR TECHNOLOGY?WARDER::FULTONWed Jan 25 1989 09:2315
    I don't want to seem ungrateful, BUT!
    
    The opening comment of this topic was "I won't burden you with things
    that are not specifically relevant to Insurance". Well, I think
    the last three entries, while very interesting, are not specifically
    insurance related; I've seen this stuff in several other conferences.
    what I would love to see here are INSURANCE consultants talking
    about the business issues, and how technology is relevant to these
    issues. I'd also be interested in hearing about INSURANCE issues
    even if they are not technology-related; these help us in our
    discussions with our prospects. How can we claim to be a solutions
    supplier if our Insurance notes file is full of notes on UNIX, SAA,
    etc. rather than 1992, FSA, RENET, CAMIFA, et al? Am I alone?
    
    			.....Gordon
63.5Hear, Hear!!HOCUS::OHARABob 352-2968Wed Jan 25 1989 19:451