Conference 7.286::digital

    I agree that quality and service are the two things that will
    be key for DECs future. I'm not sure that we are moving the
    wrong way on quality though. I sit in a manufacturing plant,
    because my engineering group needed a lot of lab space, so
    have gotten to see a lot of what goes on. I think quality
    is getting better. Not yet good enough but better none the
    less. (We're talking hardware now.)
    
    My groups job is testing cluster hardware. The cooperation
    we get from both other engineering groups and manufacturing
    groups at tracking down problems *before* customers see them
    has shown steady improvement for the past four years. Manufacturing
    management really seems to be starting to believe that quality
    matters. It's getting there.
    
    Service? Well, I don't know. In-house Field Service is generally
    outstanding. If this is what our customers are getting then
    things there have improved a great deal since I left the field.
    
    Sales? I don't see any improvement there sad to say.
    
    Software Support? Again I don't know. I'll live it to someone who
    does to say.
    
    			Alfred

re: .0

To survive we must do two things; make changes, and make them correctly.

The good news, and it really is good, is that we've recognized and are
moving on the first half of the problem.  That behind us, it is now time
to move on the second half.  It is much easier to do that than if the
company had it's head in the sand and was resisting changed completely.

We have our traditions and philosophy to help us.  DEC has invovled
the "ranks" before, and so that too paves the way for involvement
now.  It to would be harder if we were traditionally "top down", but
involving the ranks is "going with the grain, not against it".

Another note in this conference regarding an OEM issue, just had a
very positive reply from a corporate manager.  It is a good omen.
So are the comments in the DVN broadcast.  My reading of KO is that
he _does_ listen to such things.

Regarding service and quality, I am not certain that we are doing
badly there.  There are distinct improvements in manufacturing,
administration, and other areas.  We might, repeat might, have some
minor declines in some traditional areas of product quality, but
the customer looks at a much larger picture than we see in, for
example, central engineering.  We are now fixing some of the worst
aspects of our quality picture, and so the bottom line can be
positive even if one observes minor localized setbacks.  Of course
that is no reason to excuse them;  but we must learn and evolve.

Service?  I can't say much about that, other than I know that it
too is being looked at and I think some positive steps are being
taken to improve there as well.

- greg

I agree that we're still producing quality products (thank goodness).  I 
don't think we've reached a point (yet) of sacrificing quality for the 
eternal bottom line.  But my perception is that we're headed in that 
direction.

The view I have of our engineering organization is that corporate cost
cutting has left them constantly understaffed.  I believe headcount has
been sacrificed in order to cut costs and to build our huge bureaucracies
-- and I'm not pointing any fingers here -- the bureaucracies exist both at
the corporate level and out in the field.  So what I see is that, to keep
their commitment to quality, engineering is extending time to market
further and further so they can produce high quality products with fewer
people and resources. And now the pressure is on (and I'm one of the ones
screaming) to reduce time to market.  It's essential if we're to remain
competitive. Something's gotta give.  What will it be? 

Increase headcount and expenses so we can reduce time to market and still
maintain our quality standards?  Not if Mr. Osterhoff has anything to say 
about it!

Sacrifice quality so we can reduce time to market without increasing
headcount and expenses?  That's where my money is.

Or not attempt to reduce time to market at all? The heat is on, and I 
believe the powers that be have recognized the necessity of reducing our 
time to market.  I don't believe they would consider this an option.

Now admittedly I'm not in Engineering, but I deal with them very frequently 
in my job and I also see what they have to say in the DIGITAL and MARKETING 
conferences.  Are my observations off base?

							Pat
    

        You can order a copy of a Digital Video Network (DVN)
        broadcast (such as the one referenced in .0) by sending
        MAIL to EXIT26::AVINFO with this information: 
        
    -  Name and date of broadcast
    -  Your name
    -  Badge number
    -  Cost center
    -  Mailstop
    -  Phone number
    -  What size video you would like (1/2" VHS or 3/4")
    

    re:  Quality discussions in .1 - .3
    
    I find it both interesting and significant that everyone seem to
    keep mentioning hardware when talking about quality.  What about
    quality of services?  Lest we forget, take a look at the annual
    report.  Service revenues are the fastest growing segment of our
    overall business.  Does anyone doubt that service revenues will
    eventually outpace hardware revenues, especially when hardware/
    software/network platforms become commodities?
    
    I applaud the strides made in Engineering and Manufacturing
    regarding quality assurance.  But I feel that we have only just
    begun to consider improving our quality of service.  For all of
    the "surveys" that rank us "Number 1", we still have a long way
    to go before can attain the real "Number 1" spot.
    
    Quality products (hardware and services) and a quality reputation
    are our keys to the future.  They're better than a license to print
    money; we all know what a dollar is worth these days!
    
    Geoff

re: .3  "Am I off base?".

Sigh, it's really hard to tell.  In twenty years in the industry,
the claim that management is sacraficing this or that to increase
market share seems to be a constant claim.  It's always there in
one form or another.  This makes trying to pick out real instances
of it from the background difficult.

So you could be right, and I can't prove you're wrong, and there
is some evidence to support your points.  Nonetheless, I'm in
engineering with a large amount of interface to non-engineering
projects and people and I think we are ok.  We must be ever
vigilant and strive for perfection in every area regardless of
headcount pressures but at the same time we have limited resources
and many areas besides engineering that need people ... this makes
for a difficult balancing act.

My 2 cents would be that we are building too many products in
engineering; both hardware and software, and that we could regain
some effective headcount by concentrating resources on fewer
things.  Some of the recent reorganizations and cancellations
suggest others in the company may argree with that assessment.

But no, I don't think we have a product quality crisis at this
point;  I'd say that overall it is on an upswing these last
two years.

- greg

    I'm also in engineering and think we're OK. I see a lot of very
    skilled, very dedicated people doing their best to boil down
    ever-complex technologies into the right solutions. I see senior
    managers measured on both cost and quality (including making the
    right decisions). 
    
    We've done a lot of work to understand "possible futures" so that
    engineering doesn't jump to conclusions such as the view that the
    market will be only commodities or, on the other hand, that customers
    will always want "the interesting projects" at any price. Given
    this work, I believe the customer is right-on with the comment about
    change -- we must be ever-more skilled at managing for it. He may
    be in a trap, though, about commodity product and only time will
    tell.
    
    /Larry

    About 1975, after building a couple of CPUs, I came up with the
    following Golden Rule: 
    
    	Digital shall deliver no wrong answers.
    
    I intend this to apply to the area of hardware reliability. I realize
    we can't get there all at once, but I have met incredible resistance
    from hardware design engineers on even easy-to-do parts that aim in the
    right direction. Why? Because its not part of the initial product goal
    set, and it would raise costs, or slow the machine, or slip deliveries.
    Oh, I know what various managers say, but their words are not backed up
    with actions. The highest goals are performance and delivery date, not
    any form of reliability. (Need I point out the -785? Well, I just did.) 
    
    There has been some progress. We now (usually) put error checking on
    busses, and most RAMs get error checked, but there is still a long way
    to go. For example, Digital is now, more than usual, re-examining what
    CPUs we should be building. (I have to be careful here because this is
    an open conference.) We discuss delivery dates, and performance, and
    circuit technologies, and price ranges. Has anybody even heard of
    reliability goals as part of these discussions? I haven't. 
    
    I will start to believe that reliability (as a measurable form of
    quality) has become really important to our product planners when it is
    discussed and the goals are set at the same time as performance, cost,
    and delivery date. 

    Just for the record (and fyi...) the panelist for this broadcast
    were:
    
    Dr. Michael Hammer, Founder, Hammer and Company
    John Naisbitt, Author (Megatrends) and Management Consultant
    Peter J. Sprague, Chairman of the Board, National Semiconductor
    Alberto Vitale, President and CEF, Bantam, Doubleday, Dell Publishing
    Group
     ...with Ted Koppel as moderator
    
    I recommend seeing it......
    
    Vicki

Re: .8

Too right!  Engineering (for hardware) is principally measured on how low
the transfer cost is.  The fact that the product may be very expensive to 
maintain is a much lower consideration.  This expensiveness may be caused 
by many things, one of which is low reliability.

I work in Diagnostic Engineering and am involved with Design for
Testability.  The number of engineers that are concerned that making a 
design testable, and therefore cheaper to maintain, is as important as the
other aspects of design is worryingly small.

I think I'm going to ask some of the designers I am working with how their 
designs prevent undetected errors being introduced.

jb

    Not to sound terribly negative, but...since I've worked in both
    (Software) Engineering and The Field (SWS)...
    
    I think it's dangerous to pat ourselves on the back about the quality
    of our products.  Check the trade rags on our disks, or look at
    some of the metrics on software and hardware problem reports from
    Customers.  Go to DECUS or a LUG meeting and really listen.  I'm
    not saying we've got insurmountable problems, I just think some
    of this discussion is starting to sound a little smug.
    
    As for The Field -- Sales is measured on revenue, not margin, so
    how about those low profitability numbers lately, huh?  We tend
    to spend more on closing the sale than we can hope to earn.
    
    Field Service is worked to death, toward goals expressed only in
    numbers, not satisfied customers.  They've taken over S.P.S. by
    responding to software remedial problems by telephone from two time
    zones away.  Right.
    
    SWS is still trying to figure out what business they're in, and
    grooming (project) risk takers by rewarding the mediocre, preserving
    the inept and ignoring many of the truly excellent performers.
    
    And I don't even want to talk about the car plan.
    
    I guess you're right, though, things look pretty rosy...
    

    Right on target, tim...  those of us in the trenches are quite aware
    that the emphasis is being put on DOLLARS not quality, reliability
    or employee satisfaction.  Customer satisfaction at least gets some
    attention.
    
    		Dave.

re: < Note 615.12 by ODIXIE::SILVERS "INERTIAL USE ONLY?" >

>                              Customer satisfaction at least gets some
>   attention.
    
    Close, Dave.  It's customer satisfaction SURVEYS that get the
    attention.  Not quite the same thing. :-(
    
    Dave

    Take a look at Ken's comments in the first paragraph of 620.0.
    Has he been misled or ill-informed on the state of customer
    satisfaction? 

Customer sat surveys are pretty easy to understand.  Someone (VP) is 
responsible to Ken for insuring customer sat.   That person develops
a survey to gather data for his metric.  Each year, his metric has to 
improve, so that he can tell Ken he's doing a great job, and customer
sat is up, and he should get a salary increase.  After this goes on 
for a couple of years, a 5 on the survey is no longer acceptable as
"average."   Soon, (now), anything below an 8 is unacceptable.  But
our customers don't understand that...they see a scale of 0-10, and
calculate (correctly) that 5 is average.   So now, since the folks
in the field have to make THEIR customer sat metric, so the VP can
make HIS customer sat metric, and the raises can be handed out down
the line, WE have to go out and convince the customer that anything
below an 8 is BAD!!   We have to "manage" the survey.   Now, we know
the survey is bogus, the customer knows the survey is bogus, (and 
resents being told that an 8 on a scale of 0 to 10 is "average,")
but we're stuck with it, in its current form, because if we do away
with it, or change it, our friend the VP doesn't have a metric to
go to Ken with next year, and might not get his raise.

    re: .15
    
    An aside about improving metrics annually...
    
    From what I understand, many people in our SWS district are being
    targetted for a 9.0 on the customer satisfaction survey.
    
    This means, if you want to "meet job requirements" for your next
    performance appraisal, you'd better hit a 9.0!  Oh, would you like
    to be rated as "exceeds job requirements"?  Time to do some heavy-duty
    "dialing for 10's"!  You want to attempt a #1 rating???  Pencil
    in some 11's on the form!!!
    
    Maybe we could just send out surveys that look like this:
    
    	Satisfaction with Digital:	9.0   9.5   10.0
	(please circle a number)
        
    It would be closer to the point, and it would save us the hassle
    of having to "inform" the customer of the "real" scale!

    Re: Note 615.15 by MISFIT::DEEP

>Customer sat surveys are pretty easy to understand.  Someone (VP) is 
>responsible to Ken for insuring customer sat.   That person develops
>a survey to gather data for his metric.  Each year, his metric has to 
>improve, so that he can tell Ken he's doing a great job, and customer
>sat is up, and he should get a salary increase.  

    Well said!
    
    Do you think somebody should tell the emperor that he has no clothes? I
    would, but over the past five years I have been at DEC, it has been
    made clear to me on several occasions that I am not to rock the
    "customer satisfaction survey" boat.
    
    Rich

    
>    Do you think somebody should tell the emperor that he has no clothes? I
>    would, but over the past five years I have been at DEC, it has been
>    made clear to me on several occasions that I am not to rock the
>    "customer satisfaction survey" boat.

Well, Rich...   

    Decisions that could possibly be career limiting are always up to the
individual...   And such noise could be considered career limiting if it 
didn't make it to KO himself...   If it did, then you are "doing the right
thing," and would probably be ok...  But you have to weight the risks and
judge the reality vs the words...

BTW...my opinions on the survey are just that, opinions!  But that doesn't 
imply that they are incorrect, either...

Bob

        Re: .15, .16, .17, .18
        
        To answer the question asked in .14, you seem to be saying that
        Ken is misled or ill-informed on the state of customer
        satisfaction. 
        
        It sounds like someone should collect evidence that the customer
        satisfaction forms are not a good judge of customer satisfaction
        and give the evidence to Ken so he can know that the statements
        he is making are not really supported by the numbers he was
        given. 
        
        I keep hearing, year after year, in this conference and in other
        places, that those surveys are bogus.  The surveys keep being
        used in the same way, so either: 
        
        	The surveys really are good measures of customer
        	satisfaction,
        
        	The surveys are intended to measure DEC employees
        	on how well they can convince the customers they
        	support to circle high digits, or
        
        	The right person hasn't been told of the problems
        	with the surveys.
        
        If there really is a problem, will someone please raise the
        issue and keep raising it until the problem gets solved! 
        
        					B.J.

    Re: Note 615.19 by ULTRA::HERBISON

>       It sounds like someone should collect evidence that the customer
>       satisfaction forms are not a good judge of customer satisfaction
>       and give the evidence to Ken so he can know that the statements
>       he is making are not really supported by the numbers he was
>       given. 
    
    Ok. I nominate you. Anyone second the motion?  :^)

    I hadn't intended to launch this into yet another discussion of
    the surveys, and I suspect Dave probably didn't either.  But since
    we're on the subject...
    
    There is every indication that people are genuinely trying to improve
    the process, or at least that's what I've observed.  The basic problem
    is that this is a blatant case of the fox guarding the chicken coop,
    and no reasonable level of improvement can be expected so long as
    management feels we're all pure of spirit enough (or perhaps naive
    enough) to accurately report how our own customers feel about us.
    Unfortunately, this rather simple fact of human nature seems to
    elude someone up in the ranks, so we go on and on with it.  I think
    the resulting frustration is reflected in any of the several tirades
    I've seen lately on the subject, so I won't throw any more gasoline
    on the fire.
    
    Getting back to the earlier subject, in the 9+ years I've been around
    DEC, there are many ways in which things are better now than they
    have ever been.  My earlier flame was directed more toward the point
    that I agree with what I've heard K.O. say many times, especially
    lately: failure isn't the scary thing, success is.  We need to be
    careful about touting our successes, and remember how far we still
    have to go.
    
    Incidentally, on a 1 to 10, I think we're probably really are close
    to an 8.  But do our customers agree?
    
    tim
    

    Here's what I propose. Let's use the same relative scale to measure
    employees at review time that we ask our customers to use to rate
    Digital. 
    
    1   2   3   4   5   6   7   8   9   10   Customer surveys
                                ^
                                |                            
      
    This is the "average" score
                                                                  
    
    
    5         4        3        2        1   Salary review ratings
                                ^
                                |

    So this is the "average" employee rating :^)
    

    re: .22  relative scales
    
    That's the best idea I've seen in a long time.  Unfortunately, while
    the customer loses nothing by giving us a top rating on the survey,
    managers have a vested interest in seeing lots of 3's.  People who
    make 1's and 2's have a nasty habit of expecting to get paid for it ...

    
    Regards.

re: .22,.23

I was told before my review that in order to get a 2 all my customers would 
have to turn in 10s on the survey on customer satisfaction.  My big 
question at the time was what do you have to do to get a 1 - but that 
turned out to be academic.

John

    re: .24
    
    One manager told me that a 1 rating was exceptional based on
    circumstances and that, many times, it would be IMPOSSIBLE for someone
    to be rated as a 1 unless they had been placed on some extremely
    crucial project.
    
    Interesting how someone who "consistently exceeds job requirements"
    (in reality) could be reduced to an "often exceeds job requirements"
    rating just because his/her manager didn't think that the job was
    that important.  Sounds like the rating verbage should be altered
    to reflect effort AND CIRCUMSTANCE, rather than simply effort.

    We employees need an ombudsman to carry the message about the
    mismanagement of the Customer Satisfaction Surveys to KO himself. 

    I wish I were independently wealthy enough to risk doing it myself.
    But, regretfully, I am not.  I am too afraid of the consequences of
    misjudging the view from such a height - there just might be some good
    reasons for doing it the way we do - ostriches must think they are
    right in their behavior.  I am not comparing KO or anyone else
    specifically to that bird, I have to much respect for him and his
    helpers, but somewhere in the chain of command there must be one! 

    I also am I am very scared of the consequences of KO and his peers
    judging ourselves on the basis of those results.  I just read him doing
    that somewhere.  Without a direct order I would not give one of those
    Customer Satisfaction Survey Result Brochures to a customer.  I am
    ashamed of it - for my company - and have told my manager as much. 
    
    The only good that the CSS's do is alert management to problem customers.
    That's good, but there must be a better way to do it.

    The only way to do real surveys is to get it done by outsiders that
    have little stake in the results, if that is possible.    

    Register your opinion about an ombudsman in note #615. 

    Fred
    

        RE: Tom Eggers' reference to KO's comments as quoted in topic 620
        -- I don't see anywhere in topic 620 the assertion that KO looked
        at the "Customer Satisfaction Survey" results in reaching his
        conclusion that customers are satisfied with DEC.  In fact, when I
        take the quoted statements at face value, I think he's basing his
        conclusion on our success in selling our products in the face of a
        soft market for computer systems. 
        
        Which, of course, does not mean that the survey itself, or the
        process for its administration, isn't bogus. 

        Tom

    Re .28:
    
    You may be right. I assumed (perhaps incorrectly) that all the effort
    that went into customer surveys would get to Ken, particularly since
    the results seem so positive. If (if!) Ken's comments in 620.0 are the
    result of the CSS customer survey, and if the survey is badly designed
    or implemented, then Ken is being misled. But there is no way to tell
    from 620.0 what the source of Ken's information was.

    Is it possible for people with satellite dishes to receive these
    broadcasts?
    
    /p

>    Is it possible for people with satellite dishes to receive these
>    broadcasts?

Nope. At least, not without a lot of work.

DVN broadcasts are broadcast using B-MAC scrambling technique, nothing is
broadcast 'in the clear'. Most 'consumer' oriented programming is descrambled
via VC2 (VideoCipher II) format boxes, so they will do you no good for DVN
broadcasts.

Even with a B-MAC decoder you would have to get someone within Digital
to enable your decoder. This enabling information is put onto the signal at the
uplink site so that only authorized sites can receive the broadcast... i.e. so
IBM'ers can't tune their dish into our broadcasts. (Each decoder has it's own
address for just this reason.... unless your decoder gets 'enabled' from the
uplink site you'll not descramble a thing.)

I've spoken to the techno-weenie at CYO (cincinnati) that deals with the DVN
stuff and he told me that DEC does broadcast it's logo for a period before the
scheduled broadcast. This allows the DVN sites to find out which transponder the
program is going to be broadcast on (it may vary with each broadcast- we rent
slices of time and don't have a whole transponder to ourselves). You might be
able to pick that up. Shortly before the actual program begins the scrambling is
turned on.... then you'll get mush.

EXIT26::DVN notesfile might be a place to ask additional questions. I believe
it is stated in there (somewhere) what specific sat that DEC uses.

Lee

Not only that, but DVN does not even use the same frequencies that the "C-band"
broadcast TV satellites use.  You would need an entirely different antenna,
LNA, and receiver.  PLUS the descrambler.