Conference 7.286::digital

    Re .15 The State of Texas has recently requested that Digital not use
    the title "Engineer" on any of the MCS business cards. There's some
    state law that requires all "Engineers" to be certified by a governing
    body and be registered with the state. I'm sure the intent of the law
    was to make certain that civil engineers and such were up to par. Times
    have changed and the law hasn't. The Texas district held a contest to
    come up with a new name for the MCS field folks, but I can't for the
    life of me recall the winning name.
    
    Harry
    

    re: .16
    
    > state law that requires all "Engineers" to be certified by a
    > governing body and be registered with the state
    
    this is called raising revenue and restricting competition...IMO
    
    the state gets to raise more money and the "professionals"
    get to limit who can be in the business....
    
    so if your auto mechanic/plummer/utility engineer/software
    engineer/civil engineer/etc. happens to be the best at what
    he/she does THATS tough...they can't do it unless they are
    registered with the state wo by the way will test and certify
    them and the STATE will decide if they are qualified or not...
    
    "well they're the best at what they do"
    
    state: but they didn't pass the qualification test...
    
    
    

    For Texas, change the title "engineers" to "rangers".
    That'll piss 'em off.                       

    ... or maybe "evangelists" or "warriors" ...
    
    Steve

    The use of 'Engineer' is also severely restricted in many of the
    European countries (Excluding UK).  You must have a recognised
    academic qualification to be entitled to the title.
    
    They may also have a similar restriction on 'Technician' but I
    don't know for certain.
    
    Calvin

    In N.Y.C. you need to be licensed to be a Business Administrator...now
    N.Y. knows how to get you for all your worth.

    I started this note more than 10 years ago, I hope that it wasn't a
    jinx!  I think that at the time, there was talk about closing down the
    Westfield plant (I was in the Mill at the time).  Of course, Westfield
    did eventually get closed, but not until years later.

    On the subject of engineers, yup I are one.  I have looked into getting
    a Professional Engineer certificate in Mass, but the program is really
    set up for surveyors, civil, or heavy duty mechanical or electrical (as
    in power company) engineers.  The type of records and projects that
    they require just don't exist in computer/electronic engineering, let
    alone software engineering.  They also want you to be working for a
    current Professional Engineer who will sponsor you.  Maybe it's changed
    since a couple of years ago, but it didn't seem worthwhile to persue it
    then.

    ---Phil

    I got my EIT before coming to Digital.  So, I checked into getting my
    PE.  One of the requirements was that I find other PEs that were
    intimately familiar with my work to sponsor me.  There aren't any.  
    Especially considering that with some of the stuff I was doing the only 
    people that really understood it were (external) patent lawyers and 
    only a few other internal engineers.  Then, I was told that it was 
    common practice to just find PEs and ignore the requirement that they 
    be familiar with my work.  At that point, I decided that if that's how 
    PEs work and think, I'm not interested.  
    
    I already had begun to develop a less favorable impression of PEs after
    working at a nuclear plant.  There, it was required that all official
    design documents be approved and stamped by two PEs.  I found literally
    hundreds of errors that required fixes on those documents.  The 
    approvals were virtually worthless.  They were just there for CYA.
    
    Further, I know of a university professor that is widely demanded and
    respected in a multi-state region for his expert opinion in a court of
    law.  Technically, the governments require that he be registered as a
    PE.  He hasn't bothered to get around to it.  The lawyers, judges and
    so forth don't care because his expertise is rare and firmly 
    established.  He can't keep up with demand for his services.  
    
    When I hear that some job requires that I be a registered PE I avoid
    it.  Not because I can't get a PE, but because I take it as a Bad Sign,
    indicating that it's probably not the kind of work environment I would
    want to be in.
    
    Steve

    
    
    (maybe this Engineer debate should be a different note ?)
    
    Actually I think the restriction on use of Engineer, or indeed any
    similar title, without the formal classification to support it, is not
    a bad idea.
    
    As  .20 (i think) pointed out, most European countries have
    significant restriction on the use of those terms and indeed most
    employers will ask for proof that one is "an engineer", or whatever,
    rather than the fact that someone just has the title.
    
    All in all I think it will be best for the profession when (if) that
    rule is strictly applied.

    
    Oh, yes please, a different note.  A piece of paper does not an
    engineer make.

I can remember review board certifications for engineers.  Wonder what ever 
happened to that program.

    I apologize in advance for continuing this discussion.  When the
    notes get moved ...  Just saw in my latest copy of "the institute" (a
    paper shipped with "IEEE Spectrum") a front page article entitled,
    "States consider professional competency requirement in licensing." 
    The gist of it is that when it comes to PE license renewal, 4 of the 50
    US states require some measure of continued professional competency
    (CPC).  22 states and territores are considering adding such
    requirements.  It is probably the closest thing to aknowledgement that 
    I've seen of the sometimes questionable validity of a PE certificate.
    The article mentions that many states grant an "industrial exemption"
    from licensing to engineers who are employed by large companies. 
    Emphasis for licensing is on engineers who work as consultants.  
    
    Steve

    in keeping with the engineering thread and its applicability to
    digital. i have the following observations.
    
    * the governing body at my site is the APEO - Association of
    Professional Engineers of Ontario.  prime requirements to admittance
    are:
    	- work as an engineer, under a recognized engineer, for 2 years. it
          is easy to waiver "under a recognized engineer"
    	- educated at a recognized engineering - waiverable, but VERY
    	  hard to succeed.
    	- pass a ethics and law exam.
    
    so .. i get it .. what does this mean ....
    
    * the act is primary focus to civil, mechanical etc... engineering.
    
      as a software engineer my primary interest would be in intellectual 
      property, but this is not covered under the APEO act.  (copyright 
      laws, as in the US).
    
    * the act also deals with labour laws.  in ontario, many of the labour 
      laws are suspended for three professions - lawyers, doctors and 
      engineers.  technically, an engineer (even in digital) has less
      protection.
    
      example: in ontario there are laws providing some restrictions 
      on the amount of overtime a person works.  this does not apply to
      engineers.
    
      why the exceptions?  the basis being an engineer is hired on a 
      contractual basis as a professional, and hence has the responsibility
      to control-manage these conditions.
    
      aside: there is some interpretation in digital HR that anybody with 
      a digital Job-Code with the word "engineer" falls into this exemption 
      category. hence, most-bodies working in "engineering" are not bounded 
      by most labours, ie overtime restrictions, no matter if they are a
      PEng .....  discenting opinions do exist.
    
    * if involve in a legal with a government agency, the PEng carries 
      some additional clout, but is NOT a requirement.
    
    * oh yes ... as a PEng, you can sign and notarize passport applications
      and the life of.
    
    last ... i am no lawyer and do not keep up to date of all changes, so
    some errors may exist in the above statements.

    	In NY and NJ, people can call themselves whatever engineer
    they are but title of "professional engineer" is restricted to state-
    certified people only.  To qualify for that, this person need to have:
    
    1.  Bachelor degree (minimum) from recognized colleges/universities.  
    	For example, PE for mechnaical engineer must be graduated or
    	or taking major courses from colleges approved by ASME.
    
    2.  Two days of written tests of 6 topics.  Mathematics (Calculus,
    	Differential Equations), Mechanics, Economy 101, ... 
    
    3.  4 years "REAL" working experience with Bachelor degree, 3 years with 
        Master or PhD.  
    
    4.  Recommendations from three PEs
    
    Differences between a PE and a non-PE are:
    
    1.  Governmental contracts will only offer to company with PEs.  
    
    2.  PE's approval signature will be required for licensing.
    
    My understanding is in most parts of US, requirements are very similar.

    Massachusetts is very similar to the NY/NJ situation in the
    previous note. However after completing the requirements
    mentioned there is yet another 8hr written exam using real
    life problems. The questions are engineering design projects
    - not multiple choice questions. Having been through it I can
    say the entire process is a grueling experience.
    
    I can also say I know some PE's who shouldn't be but they new
    how to take the tests.
    
    From way back I learned this; Titles in themself mean nothing.
    Proven ability, experiences, and good references make the difference.

  Three guys, a technician, an engineer, and a programmer,  were driving in a 
  car down a very steep, winding hill. As they entered some of the most 
  difficult turns, the brakes began to fail. The engineer was driving, and he 
  methodically applied the hand brake and allowed the front right wheel to 
  drag along the curb, to help gently slow the car.
  
  When the car finally came to a stop, the three got out and began to access 
  their situation. The technician immediately got down on his hands and knees, 
  inspected the brakes and said "I can fix this!"
  
  The engineer looked over the entire car, walked around it more than once, 
  thought carefully and said "Well, I think that we should call a specialist."
  
  The software programmer quickly added "Why don't we just get in and see if 
  it happens again!"

	  The term "engineer" is regulated by law in New Hampshire, also.
     The appropriate RSA states people may not call themselves engineers,
     or represent what they do as engineering, without state PE licensing.
     State PE licensing requires a fair amount of work.  When asked what I
     do for a living, I say I am a "software type," although the job code
     reads "software engineer."

	  It is true, a piece of paper does not an "engineer" make, whether
     that piece of paper comes from the state or from a school.  Similarly,
     a piece of paper does not a "medical practitioner" make.  But the dif-
     ference between a trade and a "profession" is, if you blow it in a
     "profession," you pay BIG bucks, or sometimes go to jail.  It's called
     "professional malpractice."  Ask the guys who signed off on the sky-
     bridge of the Hyatt Regency.

	  We'll know software is a profession when system crashes are
     investigated by the state Attorney General's office, or when a team
     from the FCA (Federal Computing Administration) is dispatched to the
     site.  If you aren't ready for that, you aren't ready for the title
     "engineer."

	  Software is a trade, folks.  Let's not pretend we're real
     engineers -- you know, those folk who put not just their "careers,"
     but their personal assets and personal freedom, on the line every time
     they sign a design "Joe Blow, PE."

	  Actually, we should be glad software is a trade, not a profes-
     sion.  Given the state of software, if software were a profession,
     none of us could afford the errors and omissions insurance.

> But the difference between a trade and a "profession" is, if you
> blow it in a "profession," you pay BIG bucks, or sometimes go to 
> jail.  It's called "professional malpractice."  Ask the guys who
> signed off on the sky-bridge of the Hyatt Regency.

  Actually, the engineering was done correctly. But the contractor
  deviated from the design in order to cost-reduce it.  I'll provide
  details if anyone actually cares.

                                   Atlant

	  Actually, it's arguable whether the engineering was done
     correctly, which is why the state Attorney General investigated to see
     if criminal charges against the engineers were warranted.  The
     blueprints specified an impossibility: a washer and nut half way up an
     unthreaded steel rod.  There were interpretations of the blueprints
     which would have maintained structural strength (e.g., a sleeve nut
     over two threaded rods); there were interpretations that did not
     maintain structural strength (e.g., two rods next to one another).
     The collapse is covered in some detail in "To Engineer Is Human."

				     Brian

    re: .17
    
    >     Software is a trade, folks.  Let's not pretend we're real 
    >engineers -- you know, those folk who put not just their "careers,"
    	
    FYI & FWIW, there was a programming question on my PE exam.  	

I would say that any state that tries to regulate software development and the
people who do it the way they do civil and mechanical engineering seriously
misunderstands the state of the art. Civil engineering has been around for
several millenia (were the Eqyptian pyramid designers board-certified?).
Mechanical engineering has been around at least since the beginning of the
Industrial Revolution; engineering historians would probably put it even
earlier. Software "engineering" has only been around for about a half a century,
Lady Ada Lovelace notwithstanding. Strictly speaking, it is undeserving of the
term "engineering" by the standards established in other disciplines. There are
few well-established standard practices. There is instead a lot of craftlore
that is passed on in different ways with different effectiveness. It's going to
be a long time before anyone is really able to regulate it as a standardized
pursuit. They may claim to and collect fees on that basis, but they would simply
be pulling the wool over the public's eyes. It would be just as appropriate to
require certification of professional politicians; goodness knows they have
tremendous effect on the well-being of their constituents, with grave potential
for dire consequences if they screw up.

    Old saying from a book of Murphy's Laws :
    
    "If builders built buildings the way programmers wrote code, the
    the first woodpecker that came along would probably destroy
    civilization"

    Imho, software engineeering (not to be confused with "computer
    programming") is a craft. You can try to teach it, and some people will
    learn well. Some people don't need to be taught. I tried learning to
    drive a lathe once, and then I tried welding. With the lathe I could
    get the job done right but it took forever. Welding was a hopeless
    task. In both cases I had a reasonable understanding of the processes
    involved, but there was no way I could claim to be qualified for the
    job. Is software engineering any different ?
    
    F.P. Brooks ("The Mythical Man Month") was pretty spot on on how to
    organise a software engineering project, and yet so few people bothered
    to follow his recommendations. BWK, PJP, etc, had a good pragmatic
    approach to software engineering (but now, who knows what those
    initials mean?).
    
    You can put all your Software Process Maturity Models and your Object
    Technology Models and all your other trendy experts in a room and they
    probably couldn't engineer a decent real-time system to save their
    lives. 
    
    My passport used to say "software engineer" on it. Can't remember
    what's on my current one, probably something meaningless like
    "technical consultant".
    
    see ya
    jw

    
    
    re .21
    
    Were the pyramid designers in Egypt board certified; the answer is
    probably YES.   The culture at that time was probably a highly
    structured one and could have been based on the idea on apprentiship
    and craft education.
    
    

  Engineering tends to advance in waves, punctuated by disasters.

  Bridge builders, for example, discover a new technique (such as
  the cable-suspended bridge) and build the first few examples
  very conservatively. Successive improvement then occurs, with
  each new example of the basic design being higher performance,
  lighter, more elegant, etc. Finally, they build the Tacoma
  Narrows Bridge (a.k.a., "Galloping Gertie"), the thing falls
  into the chasm, and the engineers recognize that they've reached
  the fundamental limits of that basic design.

  C++ may be another example.

                                   Atlant

    re .25
    
    daftest piece of engineering I heard of...
    
    Nuclear powered airplane. great concept, except:
    
    1) reactors meant plane was too heavy to get off ground
    2) er... made a real mess when it crashed...
    
    the real sad thing is this thing actually got built ! (..by
    americans..)
    
    Re C++. a software engineer I respect described it as 'write only'.
    Which means that once you had written it, you'd be lucky if you
    understand it the next time you read it..
    
    More seriously, 'Engineer' as a qual means little in the UK  :-(.
    Low status. Hence piss-poor state of UK industry. In Germany, ery
    different. High status. Germany power house of Europe. (right wing
    Conservatives need not reply to this note).
    
    
    AW

>    1) reactors meant plane was too heavy to get off ground

Actually, it wasn't too heavy without shielding (!) except between the reactor
and the crew compartment, which is how it was flown...

I served my time as an apprentice, took a degree and graduated. That was 20+
years ago now. I doubt if many who took their degree that long ago could
still do Laplace Transforms unless they were still in that particular branch
of Engineering. I can't.
I call myself a Consultant Engineer. I am a member of a UK Engineering 
Institute and have chartered status. This is recognised in most places except 
for Germany. This country is famous for its restrictive practices. The 
Teamsters have a lot to learn from Germany. Yes they are highly paid and
now they are losing inward investment due to the high costs imposed in working
there. 

Perhaps it is time for a proper profession to be setup for engineers involved
with Software and Computer Systems. Yes, we would probably have to pay for 
our own Professional INdemnity Insurance but this is usually tax deductable.
It would however start to get our Profession to the same level as that of 
others mentioned in this note stream, which can only be a good thing in the
long term. Why ?, well I have been interviewing some recent graduates who
proport to be Software Engineers ( or 'C++" programmers). Most are almost
devoid of experience in solving problems and cannot explain how they would
debug their code. One even said "I don't make mistakes in the code I write".
I challenged him to back that up and gave him a problem to solve using C++
on a PC. After 2 hours he had failed to write a simple program to do Mortgage
Interest calculations.  These sort of people probably have potential but need
to be put into a proper training scheme. If there were a proper Software
Engineering Profession there should also be proper training to allow people
to get into it. 

We are living in a more computerized world with software appearing everywhere
wether we like it or not and this mostly has to be BUG FREE. Would you fly 
in an A320/330/340 or Boering 777 if you knew that the fly by wire software
has inherrent bugs ? There are some probably there but I am not qualified
to say if there are any or of what type.

Stephen Davies

>Would you fly
>in an A320/330/340 or Boering 777 if you knew that the fly by wire software
>has inherrent bugs ? There are some probably there but I am not qualified
>to say if there are any or of what type.


I have and there are.

The last time I saw a problem I was sitting in 1st class (after upgrading
with frequent flyer coupons) so I saw what went on.  The A320 "crashed"
while taxiing out to the runway (better than "crashing" in the air!).  It
had to be pulled back to the gate and a technician came out and "rebooted"
the system. We then flew away without problems... :-)

				-John

    re: .26/.27 Nuclear powered airplane
    As far as I know, only a reactor testbed with fuselage was built by
    Consolidated in Fort Worth. It was not capable of flight. The testbed
    was converted at a later date for naval reactor testing.
    
    fm

RE .23 - BWK is Brian W. Kernighan, and PJP is P.J. Plauger, the co-authors of
"The Elements of Programming Style" and "Software Tools". Both excellent books.
And yes, a nice pragmatic approach to software engineering. Simplicity and
utilitarianism would be their watchwords.

The main problem with trying to bring software engineering up to snuff with
other engineering disciplines is that we don't know how! Not for lack of desire.
We just don't know yet what the best and worst practices are. We have plenty of
ideas, some well established, but this industry is still sorting itself out, and
will be for a long time. I do think that a professional development path with
something analogous to apprenticeship would be good. Many organizations do it on
their own on a small scale.

As for anyone with the chutzpah to claim the code they write doesn't contain any
mistakes, humor them. They will humble themselves soon enough. It's a popular
belief that soon vanishes with maturity (you know, after some time working in
the *real* world!). Those capable of learning that lesson are as valuable as
those who recognize it from the first. Those who are incapable of learning it
should be given jobs flipping burgers where they can't hurt anybody!

    Re: .30

>   re: .26/.27 Nuclear powered airplane
>   As far as I know, only a reactor testbed with fuselage was built by
>   Consolidated in Fort Worth. It was not capable of flight. The testbed
>   was converted at a later date for naval reactor testing.

    The Convair X-6, a NB-36H, made the first flight with an operating one
    megawatt reactor on September 17, 1955.  It was intended to evaluate the
    practicality of a 165,000 pound nuclear propulsion system.  The program
    was cancelled before the prototype was constructed.

    (Source: "The X-Planes" poster, � 1993 Smithsonian Institution.)

    Bob

    re:.32 Lyon. As you cited, "The program was cancelled before a prototype was
    constructed." The nuclear airplane's prototype reactor used the water in 
    Lake Worth as its coolant. The old facility is still visible at the
    North end of the (now) General Dyanamic Plant.
    
    fm

    re .31
    
    They're not mistakes...they're features!
    
    Rob Wall
    
    re .19 (the nut and washer on the un-threaded rod)
    Well, I guess it is a little more complicated than a nut on threaded
    rod (learn to turn clockwise) but far from impossible.
    
    Re-inforce 1/2" rod (for example) with a 1/2" inside dia. pipe, split
    lengthwise into 3 pieces, welded around the rod section to contain the nut.
    Either use a die to cut threads onto the re-inforcement section (if
    adjustment is required) or select a nut sufficiently large that you
    could drill out the threads to slip around the re-inforce and weld in
    place. The washer part should be easy.
    
    I'll always be a technician.

    
    re .34 re .19
    
    Well, yes, they could have done that, but the suspension rods in question
    were quite long (tens of feet) and forming the threads even using your
    proposed scheme would have been a nontrivial effort.  It was far easier
    to make the design modification that resulted in the failure, and the
    difference in loading with the two designs was subtle enough (i.e., the
    original design was too clever for its own good) that it was overlooked.
    
    I don't know what sort of safety margins are required in building
    construction, but the design change would have roughly doubled the load
    on the failure point; I'd have thought that a factor of less than two
    safety margin wasn't enough even for the original design.   
    
    re atomic powered plane - I think the reactor was flown as payload, and
    did not actually power the aircraft.
    
    len.
    

The test flights with the reactors were made with a conventionally
powered airplane with the reactor on board.  The nuclear-only proto
was never built (but the conventionally powered one with the reactor
on board would have been a real mess to clean up if it had crashed,
and you certainly didn't want to fly close by while it was "hot"...)

There was an investigative reports (don't remember which channel)
hour-long program on various legacies of the early nuclear era,
including various tests conducted at the National Reactor Test Center
(right name?) in Idaho, that also looked at nuclear-powered engines
that directly vented to the outside air, etc.

Re .26, .27, .30, & .32......

The NB36H most definitely DID FLY !!!  However, he did not fly under nuclear
power.  The reactor was operational while airborne but did not contribute to
aircraft propulsion.  Plans for the next phase - i.e., nuclear propulsion were
cancelled.

A more ominous project was a joint effort by Curtis, Piper, and Adolf Coors
and others to develop and build a nuclear ram jet.  Piper's contribution to the
project was at a facility in north central Pennsylvania, not far from my
favorite trout fishing grounds.  I don't know exactly what Curtis did, and I 
would hope Coors ONLY provided the refreshments.  Unlike the NB36 project, there
was no intent for containment of nuclear waste.  It was anticipated that such a
weapon would only be used only in an all out exchange, in which case, leaving a
dead zone wake was considered an added benefit.  At least one example of the
engine was built and test fired somewhere in the southwest - don't have the
details at my fingers.  There was an interesting cover article on this subject
a few years back in Smithsonian Air & Space.

Just proves engineers can be mighty damned creative even while not being
terribly intelligent.

/mj

> A more ominous project was a joint effort by Curtis, Piper, and Adolf Coors
> and others to develop and build a nuclear ram jet.  Piper's contribution to the
> project was at a facility in north central Pennsylvania, not far from my
> favorite trout fishing grounds.  I don't know exactly what Curtis did, and I 
> would hope Coors ONLY provided the refreshments.

  Coors would have provided the targets, err, destination lists.

                                   Atlant



Note to the humor-impaired: Yes, I *AM* joking.

  Re the Kansas City Hyatt:

  I've also read reports that contradict the description provided
  in "To Engineer is Human". I think th econtrary article was in
  "Science", but I can't prove it with the resources at hand.

  This article clearly described the original design as calling
  for lengths of rod, threaded at the ends, and joined by intern-
  ally threaded couplers (see "A"). The contractor replaced this
  with the deadly discontiguous-rod design (see "B").

  Whereas the original design caused the stress of the lower floors
  to pass through the coupler, coaxial with the rods, the modified
  design caused the stress to pass through the welded channel, which
  was nowhere near strong enough.

                                   Atlant


                 |///|                       |///|
                 |///|                       |///|        ___
                 |///|                       |///|     NN|///|NN
           ______|///|______             ____|///|_____NN|///|NN__
          |  ____|///|____  |           |  __|///|___|___|///|__  |
          | |    |///|    | |           | |  |///|       |///|  | |
          | |    |///|    | |           | |  |///|       |///|  | |
          | |    |///|    | |           | |  |///|       |///|  | |
          | |    |///|    | |           | |  |///|       |///|  | |
          | |    |///|    | |           | |  |///|       |///|  | |
          | |____|///|____| |           | |__|///|_______|///|__| |
          |______|///|______|           |____|///|___|___|///|____|
              |\||///||\|                  NN|///|NN     |///|
              |\||///||\|                  NN|___|NN     |///|
              |\||___||\|                                |///|
              |\||///||\|                                |///|
              |\||///||\|               "N" represents   |///|
              |_||///||_|                     Nuts       |///|
                 |///|                                   |///|
                 |///|                                   |///|


         "A" -- As designed                   "B" -- As built

    
    re .39 - close, but maybe not quite right, Atlant - my understanding
    is that the failure point was the nut/coupler supporting the top level
    bridge.  If you think about it a moment, you'll see that in the original
    design the nut/coupler for the 2nd level bridge had to support only that
    bridge; in the modified design, it had to support the weight of both
    bridges.  In both designs the support rods had to carry the weight of
    both bridges.  I don't know that the welded channel was the failure point;
    if it in fact was, that might account for the failure with only a
    doubled load; again, I don't know what overcapacity is required in
    modern civil engineering projects, but I'd guess that increasing trust
    in computer simulations and increasing economic pressures are
    conspiring to make it seem acceptable to narrow these margins.
                                         
    len.
            

Len:

  I understood the stresses and strains just fine; I m sorry
  if I didn't make that clear in my previous reply.

  The report I read clearly showed collapse of the rectangular
  section (formed from the two welded C-channels) as the failure
  mechanism. Photos of the failed sections were included. And it
  was the added stress of "coupling" the stress (from the lower
  levels) from rod-to-rod across the rectangular section (weldment)
  that caused the rectangular section to fail.

  In the intended design ("A"), the rectangular section would
  have only carried the stresses associated with its own "floor";
  the accumulated stresses from the lower floors would have been
  carried solely by the rods and the threaded couplers.

                                   Atlant

Said only half in jest. Peter G. Neumann's "Computer Related Risks" describes on
page 79 bugs in a simulation program called "Shock II". This was "used to design
nuclear reactors to withstand earthquakes". Five nuclear power plants, where
this program had been used during the design, were shut down in March, 1979
because recalculation of stresses in the secondary cooling system of one plant
were "far in excess of allowable tolerances." This was *after* the plants were
built and placed online. There's more than one way to spew nuclear material all
over the countryside. The book also describes several other incidents of faulty
simulations.

I noted this when I recognized the name of the engineering firm who designed the
plants, Stone & Webster, as the company where my uncle used to work as a piping
engineer designing catalytic crackers for petroleum and chemical plants.

    
    	Just to keep this lunacy going a while longer.
    
    	Coors was to be the "nuclear body" contractor. They were heavy into
    laminated ceramics at the time, and believed their engineering staff
    could design, and then manufacture, a ceramic containment vessel for
    the reactor itself. The idea was a weight saving of nearly 50% over
    existing metallurgy methods.
    
    	Like Digital, good ideas die hard. Coors didn't actually stop
    funding research until 1980 on that project.
    
    
    		the Greyhawk

    re: .37 It would be more accurate to say the NB36H flew with "a"
    reactor on board. The test reactor was built on the shore of Lake
    Worth, Ft Worth, TX and never flew. The facility is on the opposite 
    side of the runway from the former Carswell AFB.
    The NB36H was not the nuclear acft nor its prototype.
    
    fm

RE: .42

The roof of the Hartford Civic Center in Hartford, CT collapsed some 
years ago after a major snowstorm dropped a very heavy load of snow 
on it.  The collapse was traced back to faulty stress calculations in 
a STRUDL simulation.

--PSW

> The roof of the Hartford Civic Center in Hartford, CT collapsed some 
> years ago after a major snowstorm dropped a very heavy load of snow 
> on it.  The collapse was traced back to faulty stress calculations in 
> a STRUDL simulation.

The question must be asked:

	Was the STRUDL simulation run on an Apple?

> The question must be asked:
> 
> 	Was the STRUDL simulation run on an Apple?

  Paul, we can always count on you to ask a question that cuts
  right to the core of the matter. Or plants the seeds of doubt.
  Well, personally, I'd guess they used the wrong value for Pie.

                                   Atlant