Groups | Search | Server Info | Keyboard shortcuts | Login | Register [http] [https] [nntp] [nntps]

Groups > sci.electronics.design > #445922 > unrolled thread

OT: in favour of gradualism

Back to article view | Back to sci.electronics.design

Contents

  OT: in favour of gradualism bill.sloman@ieee.org - 2017-02-03 03:39 -0800
    Re: OT: in favour of gradualism Winfield Hill <hill@rowland.harvard.edu> - 2017-02-03 04:21 -0800
      Re: OT: in favour of gradualism bill.sloman@ieee.org - 2017-02-03 04:57 -0800
        Re: OT: in favour of gradualism Jon Elson <elson@pico-systems.com> - 2017-02-03 11:38 -0600
          Re: OT: in favour of gradualism Joerg <news@analogconsultants.com> - 2017-02-04 07:21 -0800
            Re: OT: in favour of gradualism krw@notreal.com - 2017-02-04 10:42 -0500
              Re: OT: in favour of gradualism Joerg <news@analogconsultants.com> - 2017-02-04 08:15 -0800
                Re: OT: in favour of gradualism krw@notreal.com - 2017-02-04 11:36 -0500
                  Re: OT: in favour of gradualism Joerg <news@analogconsultants.com> - 2017-02-04 09:40 -0800
                    Re: OT: in favour of gradualism krw@notreal.com - 2017-02-04 13:52 -0500
                      Re: OT: in favour of gradualism Joerg <news@analogconsultants.com> - 2017-02-04 11:50 -0800
                        Re: OT: in favour of gradualism krw@notreal.com - 2017-02-04 20:31 -0500
              Re: OT: in favour of gradualism Jon Elson <elson@pico-systems.com> - 2017-02-04 18:04 -0600
                Re: OT: in favour of gradualism krw@notreal.com - 2017-02-04 20:33 -0500
      Re: OT: in favour of gradualism jurb6006@gmail.com - 2017-02-03 05:00 -0800
        Re: OT: in favour of gradualism bill.sloman@ieee.org - 2017-02-03 09:03 -0800
      Re: OT: in favour of gradualism Don Y <blockedofcourse@foo.invalid> - 2017-02-03 09:35 -0700
        Re: OT: in favour of gradualism bill.sloman@ieee.org - 2017-02-03 09:10 -0800
        Re: OT: in favour of gradualism tabbypurr@gmail.com - 2017-02-03 17:18 -0800
        Re: OT: in favour of gradualism Joerg <news@analogconsultants.com> - 2017-02-04 07:24 -0800
          Re: OT: in favour of gradualism Don Y <blockedofcourse@foo.invalid> - 2017-02-04 12:34 -0700
            Re: OT: in favour of gradualism tabbypurr@gmail.com - 2017-02-04 12:04 -0800
            Re: OT: in favour of gradualism Joerg <news@analogconsultants.com> - 2017-02-04 12:07 -0800
              Re: OT: in favour of gradualism Don Y <blockedofcourse@foo.invalid> - 2017-02-04 14:15 -0700
                Re: OT: in favour of gradualism John Larkin <jjlarkin@highlandtechnology.com> - 2017-02-04 14:02 -0800
                  Re: OT: in favour of gradualism Don Y <blockedofcourse@foo.invalid> - 2017-02-04 15:19 -0700
                    Re: OT: in favour of gradualism "tom" <tmiller11147@verizon.net> - 2017-02-04 17:36 -0500
                      Re: OT: in favour of gradualism Don Y <blockedofcourse@foo.invalid> - 2017-02-04 15:46 -0700
                        Re: OT: in favour of gradualism Lasse Langwadt Christensen <langwadt@fonz.dk> - 2017-02-04 14:56 -0800
                          Re: OT: in favour of gradualism Don Y <blockedofcourse@foo.invalid> - 2017-02-04 16:17 -0700
                        Re: OT: in favour of gradualism "tom" <tmiller11147@verizon.net> - 2017-02-04 18:22 -0500
                          Re: OT: in favour of gradualism Don Y <blockedofcourse@foo.invalid> - 2017-02-04 16:55 -0700
                          Re: OT: in favour of gradualism John Robertson <spam@flippers.com> - 2017-02-04 18:22 -0800
                            Re: OT: in favour of gradualism Don Y <blockedofcourse@foo.invalid> - 2017-02-04 21:26 -0700
                  Re: OT: in favour of gradualism "tom" <tmiller11147@verizon.net> - 2017-02-04 17:35 -0500
                    Re: OT: in favour of gradualism Don Y <blockedofcourse@foo.invalid> - 2017-02-04 15:44 -0700
                      Re: OT: in favour of gradualism "dcaster@krl.org" <dcaster@krl.org> - 2017-02-04 17:42 -0800
                      Re: OT: in favour of gradualism krw@notreal.com - 2017-02-04 20:56 -0500
                Re: OT: in favour of gradualism Joerg <news@analogconsultants.com> - 2017-02-05 08:14 -0800
                  Re: OT: in favour of gradualism Don Y <blockedofcourse@foo.invalid> - 2017-02-05 11:56 -0700
                    Re: OT: in favour of gradualism Joerg <news@analogconsultants.com> - 2017-02-06 14:35 -0800
                      Re: OT: in favour of gradualism Don Y <blockedofcourse@foo.invalid> - 2017-02-06 16:43 -0700
                        Re: OT: in favour of gradualism Joerg <news@analogconsultants.com> - 2017-02-06 17:12 -0800
                          Re: OT: in favour of gradualism Don Y <blockedofcourse@foo.invalid> - 2017-02-06 18:59 -0700
                            Re: OT: in favour of gradualism Joerg <news@analogconsultants.com> - 2017-02-07 07:40 -0800
                          Re: OT: in favour of gradualism tabbypurr@gmail.com - 2017-02-07 12:16 -0800
                      Re: OT: in favour of gradualism krw@notreal.com - 2017-02-06 21:11 -0500
                  Re: OT: in favour of gradualism tabbypurr@gmail.com - 2017-02-05 12:43 -0800
            Re: OT: in favour of gradualism krw@notreal.com - 2017-02-04 20:42 -0500
        Re: OT: in favour of gradualism John Larkin <jjlarkin@highlandtechnology.com> - 2017-02-04 10:05 -0800
          Re: OT: in favour of gradualism Joerg <news@analogconsultants.com> - 2017-02-04 10:24 -0800
          Re: OT: in favour of gradualism Don Y <blockedofcourse@foo.invalid> - 2017-02-04 12:45 -0700
            Re: OT: in favour of gradualism "dcaster@krl.org" <dcaster@krl.org> - 2017-02-04 17:54 -0800
              Re: OT: in favour of gradualism Don Y <blockedofcourse@foo.invalid> - 2017-02-04 21:45 -0700

Page 2 of 3 — ← Prev page 1 [2] 3  Next page →

#446135

On 2/4/2017 8:24 AM, Joerg wrote:
> On 2017-02-03 08:35, Don Y wrote:
>> On 2/3/2017 5:21 AM, Winfield Hill wrote:
>>> bill.sloman@ieee.org wrote...
>>>>
>>>> Atul Gawande has written another impressive piece for the New Yorker
>>>> http://www.newyorker.com/magazine/2017/01/23/the-heroism-of-incremental-care
>>>>
>>>>
>>>> It is mainly a plea for better primary-care medicine - what your
>>>> own doctor does for you at your local practice - but it includes
>>>> a reference to the 1967 collapse of the Silver Bridge across the
>>>> Ohio River, which was due to a failure in inspection and maintenance,
>>>> making the point that while doing heroic operations and building
>>>> new bridges are both extremely glamorous, the cost-benefit analysis
>>>> says that it's better to spend time on making sure that what's
>>>> working stays working, and perhaps improve it a bit from time to
>>>> time.  Development rather than a new design from scratch.
>>>
>>>  Would that we could snap our fingers and get a doctor like that.
>>
>> Why do we EXPECT medical professionals to be more "dedicated"
>> than any other profession?
>
> Because we as engineers are dedicated?

You've never met a "9-to-5" engineer?

I had a (regular) job, many years ago.  Bringing up the prototype
I naturally was excited and "didn't want the day to end" (until
seeing it run).

Next day, boss called me in his office and casually mentioned:
    "I see you kept the guys here until 8PM last night..."
(keep in mind, *I* am not the project manager.  nor have any
other "special responsibilities" as far as the project was
concerned -- Just Another Engineer with my little portion of
the pie)

I was startled by the comment -- as if it was an accusation.

He went on to say that he couldn't recall the last time ANY of
"the guys" had stayed past 5PM.

I did a gig at an IBM facility where the shifts were staggered
in 6 minute intervals.  I'd watch the engineers that I was
working with diligently gather up their belongings to ensure
they'd be leaving in their assigned 6 minute interval -- while
I'd typically spend much of the night continuing my tasks.

>> Next time you "cut a corner", think about the likelihood that
>> your healthcare providers are behaving similarly -- do you think
>> YOU cut that corner out of malice, laziness, ignorance, etc.?
>
> Some of us don't cut corners.

Every engineering decision cuts a corner -- everything is a tradeoff.
*You* decided which way to cut the corner (in favor of X over Y).

Do you strive for -- and achieve -- "perfection" with every design?

If you leave the MD's office in better condition than when you
arrived, has he *not* cut a corner?  Could you have left in
EVEN BETTER condition?  Should he hound you AT HOME to follow
his prescribed treatment plan?  How *dedicated* is he -- not?

[toc] | [prev] | [next] | [standalone]

#446142

On Saturday, 4 February 2017 19:34:52 UTC, Don Y  wrote:
> On 2/4/2017 8:24 AM, Joerg wrote:
> > On 2017-02-03 08:35, Don Y wrote:

> > Some of us don't cut corners.
> 
> Every engineering decision cuts a corner -- everything is a tradeoff.
> *You* decided which way to cut the corner (in favor of X over Y).

nod

> Do you strive for -- and achieve -- "perfection" with every design?
> 
> If you leave the MD's office in better condition than when you
> arrived, has he *not* cut a corner?  Could you have left in
> EVEN BETTER condition?  Should he hound you AT HOME to follow
> his prescribed treatment plan?  How *dedicated* is he -- not?

Most medical decisions are corner cutting, ie less good than is possible. The world has limited resources and medical care is expensive.


NT

[toc] | [prev] | [next] | [standalone]

#446143

On 2017-02-04 11:34, Don Y wrote:
> On 2/4/2017 8:24 AM, Joerg wrote:
>> On 2017-02-03 08:35, Don Y wrote:
>>> On 2/3/2017 5:21 AM, Winfield Hill wrote:
>>>> bill.sloman@ieee.org wrote...
>>>>>
>>>>> Atul Gawande has written another impressive piece for the New Yorker
>>>>> http://www.newyorker.com/magazine/2017/01/23/the-heroism-of-incremental-care
>>>>>
>>>>>
>>>>>
>>>>> It is mainly a plea for better primary-care medicine - what your
>>>>> own doctor does for you at your local practice - but it includes
>>>>> a reference to the 1967 collapse of the Silver Bridge across the
>>>>> Ohio River, which was due to a failure in inspection and maintenance,
>>>>> making the point that while doing heroic operations and building
>>>>> new bridges are both extremely glamorous, the cost-benefit analysis
>>>>> says that it's better to spend time on making sure that what's
>>>>> working stays working, and perhaps improve it a bit from time to
>>>>> time.  Development rather than a new design from scratch.
>>>>
>>>>  Would that we could snap our fingers and get a doctor like that.
>>>
>>> Why do we EXPECT medical professionals to be more "dedicated"
>>> than any other profession?
>>
>> Because we as engineers are dedicated?
>
> You've never met a "9-to-5" engineer?
>

Yes, but usually in government positions. Very occasionally in private 
industry but then those were very bright folks.


> I had a (regular) job, many years ago.  Bringing up the prototype
> I naturally was excited and "didn't want the day to end" (until
> seeing it run).
>
> Next day, boss called me in his office and casually mentioned:
>     "I see you kept the guys here until 8PM last night..."
> (keep in mind, *I* am not the project manager.  nor have any
> other "special responsibilities" as far as the project was
> concerned -- Just Another Engineer with my little portion of
> the pie)
>
> I was startled by the comment -- as if it was an accusation.
>

If it was a union shop it might have been.


> He went on to say that he couldn't recall the last time ANY of
> "the guys" had stayed past 5PM.
>

That doesn't sound like a place I'd want to work at. Don't get me wrong, 
I strongly believe in balancing one's private life with work and pulling 
all-nighters year round isn't good. Neither is living off Big Macs and 
128oz caffeine-laced sodas with tons of sugar. People make too many 
mistakes then anyhow and they should have a life. Even engineers. 
However, dropping the pencil at 5:00:00 no matter what is not cool. 
There has to be some flexibility. In my consulting agreement I have a 
clause that if I have to work after hours or on holidays because the 
situation demands it there will never be an upcharge (which some other 
professions tend to do).


> I did a gig at an IBM facility where the shifts were staggered
> in 6 minute intervals.  I'd watch the engineers that I was
> working with diligently gather up their belongings to ensure
> they'd be leaving in their assigned 6 minute interval -- while
> I'd typically spend much of the night continuing my tasks.
>

That would drive me crazy.


>>> Next time you "cut a corner", think about the likelihood that
>>> your healthcare providers are behaving similarly -- do you think
>>> YOU cut that corner out of malice, laziness, ignorance, etc.?
>>
>> Some of us don't cut corners.
>
> Every engineering decision cuts a corner -- everything is a tradeoff.
> *You* decided which way to cut the corner (in favor of X over Y).
>
> Do you strive for -- and achieve -- "perfection" with every design?
>

Generally yes. Perfection to me and in the eyes of my clients is when 
the design is reliable, easily produceable and below cost target.


> If you leave the MD's office in better condition than when you
> arrived, has he *not* cut a corner?  Could you have left in
> EVEN BETTER condition?  Should he hound you AT HOME to follow
> his prescribed treatment plan?  How *dedicated* is he -- not?


What is usually meant by cutting corners is not small compromises. It's 
the bigger stuff, things that can have dire consequences. Example: I had 
major back issues in the past. What did most of the medical pros do? The 
usual, inject cortizone, prescribe pain pills, prescribe muscle 
relaxant. _That_ is cutting a corner. Finally another professional (not 
an MD but a CPA!) told me I might have a magnesium deficiency and it 
couldn't hurt to take supplements for a while. Half a year later I felt 
a lot better and have never had a debilitating back pain attack ever 
since. It has been over a decade now so I can be quite sure that the 
medical diagnosis of the CPA was better than that of several MDs.

-- 
Regards, Joerg

http://www.analogconsultants.com/

[toc] | [prev] | [next] | [standalone]

#446158

On 2/4/2017 1:07 PM, Joerg wrote:
> On 2017-02-04 11:34, Don Y wrote:
>> On 2/4/2017 8:24 AM, Joerg wrote:
>>> On 2017-02-03 08:35, Don Y wrote:
>>>> On 2/3/2017 5:21 AM, Winfield Hill wrote:
>>>>> bill.sloman@ieee.org wrote...
>>>>>>
>>>>>> Atul Gawande has written another impressive piece for the New Yorker
>>>>>> http://www.newyorker.com/magazine/2017/01/23/the-heroism-of-incremental-care
>>>>>>
>>>>>> It is mainly a plea for better primary-care medicine - what your
>>>>>> own doctor does for you at your local practice - but it includes
>>>>>> a reference to the 1967 collapse of the Silver Bridge across the
>>>>>> Ohio River, which was due to a failure in inspection and maintenance,
>>>>>> making the point that while doing heroic operations and building
>>>>>> new bridges are both extremely glamorous, the cost-benefit analysis
>>>>>> says that it's better to spend time on making sure that what's
>>>>>> working stays working, and perhaps improve it a bit from time to
>>>>>> time.  Development rather than a new design from scratch.
>>>>>
>>>>>  Would that we could snap our fingers and get a doctor like that.
>>>>
>>>> Why do we EXPECT medical professionals to be more "dedicated"
>>>> than any other profession?
>>>
>>> Because we as engineers are dedicated?
>>
>> You've never met a "9-to-5" engineer?
>
> Yes, but usually in government positions. Very occasionally in private industry
> but then those were very bright folks.

I've experience/exposure to gummit labs (i.e., all of their funding
comes from the taxpayer), large multinationals (i.e., tens of thousands
of employees worldwide), large corporations (thousands of employees),
startups (one to four souls) and struggling wannabes (~100 bodies).

In virtually all of those scenarios, it was rare to find folks
EAGER to spend THEIR TIME pursuing the corporate goal.  The few
cases I was able to single out were typically very bright/inquisitive
folks who were completely engaged in the CHALLENGE before them.

>> I had a (regular) job, many years ago.  Bringing up the prototype
>> I naturally was excited and "didn't want the day to end" (until
>> seeing it run).
>>
>> Next day, boss called me in his office and casually mentioned:
>>     "I see you kept the guys here until 8PM last night..."
>> (keep in mind, *I* am not the project manager.  nor have any
>> other "special responsibilities" as far as the project was
>> concerned -- Just Another Engineer with my little portion of
>> the pie)
>>
>> I was startled by the comment -- as if it was an accusation.
>
> If it was a union shop it might have been.

Nope.  Just a big Fortune 500.  Staff wasn't used to "working late"
("The problem will still be here, waiting for us, Monday morning.
If the company can afford the time for these meetings and other
cruft, they surely can afford to wait until then!") and management
didn't expect them to!

>> He went on to say that he couldn't recall the last time ANY of
>> "the guys" had stayed past 5PM.
>
> That doesn't sound like a place I'd want to work at. Don't get me wrong, I
> strongly believe in balancing one's private life with work and pulling
> all-nighters year round isn't good. Neither is living off Big Macs and 128oz
> caffeine-laced sodas with tons of sugar. People make too many mistakes then
> anyhow and they should have a life. Even engineers. However, dropping the
> pencil at 5:00:00 no matter what is not cool. There has to be some flexibility.

Why?  Do you expect your dentist to be willing to work late as often as
folks show up WANTING him to work late (on their problems)?  Or, your
plumber?  Or, the landscaper?  Or...

Just because you have *one* project (or ten!), doesn't mean you should
be slave to it.  What cost should your wife/kids bear for your
"dedication"?

> In my consulting agreement I have a clause that if I have to work after hours
> or on holidays because the situation demands it there will never be an upcharge
> (which some other professions tend to do).

IME, that's usually a result of someone's unrealistic expectations/demands
or poor planning.

When I was young, bosses could brow beat me into agreeing to unrealistic
schedules.  I've literally run out of the shop with THE prototype in my hand
as I tried to finish wiring it before field trials.

With time and experience, I learned that folks will ALWAYS want more.  Why
not?  *I* would want more!  Its up to me to discipline the use of my time
in a manner with which I'm comfortable -- NOT my employer/client.

I recall a 5PM progress meeting with a client on a Friday.  After discussing
some changes to the design, the question was asked:  "So, do you think we can
see something MONDAY?"  I worked hard at keeping my expression "cool" and
replied:
    "Well, as it's now 6:30PM on /FRIDAY/, I don't see how that is possible..."
I.e., "No, I'm not keen on giving up MY weekend any more than I see YOU
willing to give up YOURS!"

>> I did a gig at an IBM facility where the shifts were staggered
>> in 6 minute intervals.  I'd watch the engineers that I was
>> working with diligently gather up their belongings to ensure
>> they'd be leaving in their assigned 6 minute interval -- while
>> I'd typically spend much of the night continuing my tasks.
>
> That would drive me crazy.

Apparently, they found the benefits and demands (lack thereof)
more than compensatory!

I'm lucky in that I can work when I want.  And, largely, under the
circumstances of my own choosing.  So, if I'm engrossed in a
particular aspect of a problem, I can continue to work on it
regardless of the time of day or other potential commitments
("Why don't you get yourself something to eat as I'm going to be
busy with this for a while...")

But, I *love* my work.  Few other folks that I've met can say
that -- even if they revel in  the challenges presented.

>>>> Next time you "cut a corner", think about the likelihood that
>>>> your healthcare providers are behaving similarly -- do you think
>>>> YOU cut that corner out of malice, laziness, ignorance, etc.?
>>>
>>> Some of us don't cut corners.
>>
>> Every engineering decision cuts a corner -- everything is a tradeoff.
>> *You* decided which way to cut the corner (in favor of X over Y).
>>
>> Do you strive for -- and achieve -- "perfection" with every design?
>
> Generally yes. Perfection to me and in the eyes of my clients is when the
> design is reliable, easily produceable and below cost target.

To the healthcare provider, we KNOW the patient WILL DIE.
So, as long as he gets to live another day, is that "perfection"?
If you could have eeked out a few more cents on the product cost...
or, increased the MTBF or decreased the MTTR by a bit more, isn't
that MORE perfect?  In the health care analogy, if you could
ensure the patients chances of surviving for N years were increased
by X%, wouldn't that be a more dedicated professional?

>> If you leave the MD's office in better condition than when you
>> arrived, has he *not* cut a corner?  Could you have left in
>> EVEN BETTER condition?  Should he hound you AT HOME to follow
>> his prescribed treatment plan?  How *dedicated* is he -- not?
>
> What is usually meant by cutting corners is not small compromises. It's the
> bigger stuff, things that can have dire consequences. Example: I had major back
> issues in the past. What did most of the medical pros do? The usual, inject
> cortizone, prescribe pain pills, prescribe muscle relaxant. _That_ is cutting a
> corner. Finally another professional (not an MD but a CPA!) told me I might
> have a magnesium deficiency and it couldn't hurt to take supplements for a
> while. Half a year later I felt a lot better and have never had a debilitating
> back pain attack ever since. It has been over a decade now so I can be quite
> sure that the medical diagnosis of the CPA was better than that of several MDs.

You spec a 2000hr part in a power supply.  It meets the design specifications.
It meets the expected lifespan of the product (e.g., clears warranty).  But,
realistically, you KNOW that this will result in the $200 product being
shit-canned in 2-3 years (of normal use).

Have you cut a corner?  Have you achieved perfection?  Or, is it someone
else's "problem"/concern cuz that's what they asked for?

[toc] | [prev] | [next] | [standalone]

#446165

On Sat, 4 Feb 2017 14:15:33 -0700, Don Y <blockedofcourse@foo.invalid>
wrote:

>On 2/4/2017 1:07 PM, Joerg wrote:
>> On 2017-02-04 11:34, Don Y wrote:
>>> On 2/4/2017 8:24 AM, Joerg wrote:
>>>> On 2017-02-03 08:35, Don Y wrote:
>>>>> On 2/3/2017 5:21 AM, Winfield Hill wrote:
>>>>>> bill.sloman@ieee.org wrote...
>>>>>>>
>>>>>>> Atul Gawande has written another impressive piece for the New Yorker
>>>>>>> http://www.newyorker.com/magazine/2017/01/23/the-heroism-of-incremental-care
>>>>>>>
>>>>>>> It is mainly a plea for better primary-care medicine - what your
>>>>>>> own doctor does for you at your local practice - but it includes
>>>>>>> a reference to the 1967 collapse of the Silver Bridge across the
>>>>>>> Ohio River, which was due to a failure in inspection and maintenance,
>>>>>>> making the point that while doing heroic operations and building
>>>>>>> new bridges are both extremely glamorous, the cost-benefit analysis
>>>>>>> says that it's better to spend time on making sure that what's
>>>>>>> working stays working, and perhaps improve it a bit from time to
>>>>>>> time.  Development rather than a new design from scratch.
>>>>>>
>>>>>>  Would that we could snap our fingers and get a doctor like that.
>>>>>
>>>>> Why do we EXPECT medical professionals to be more "dedicated"
>>>>> than any other profession?
>>>>
>>>> Because we as engineers are dedicated?
>>>
>>> You've never met a "9-to-5" engineer?
>>
>> Yes, but usually in government positions. Very occasionally in private industry
>> but then those were very bright folks.
>
>I've experience/exposure to gummit labs (i.e., all of their funding
>comes from the taxpayer), large multinationals (i.e., tens of thousands
>of employees worldwide), large corporations (thousands of employees),
>startups (one to four souls) and struggling wannabes (~100 bodies).
>
>In virtually all of those scenarios, it was rare to find folks
>EAGER to spend THEIR TIME pursuing the corporate goal.  The few
>cases I was able to single out were typically very bright/inquisitive
>folks who were completely engaged in the CHALLENGE before them.
>
>>> I had a (regular) job, many years ago.  Bringing up the prototype
>>> I naturally was excited and "didn't want the day to end" (until
>>> seeing it run).
>>>
>>> Next day, boss called me in his office and casually mentioned:
>>>     "I see you kept the guys here until 8PM last night..."
>>> (keep in mind, *I* am not the project manager.  nor have any
>>> other "special responsibilities" as far as the project was
>>> concerned -- Just Another Engineer with my little portion of
>>> the pie)
>>>
>>> I was startled by the comment -- as if it was an accusation.
>>
>> If it was a union shop it might have been.
>
>Nope.  Just a big Fortune 500.  Staff wasn't used to "working late"
>("The problem will still be here, waiting for us, Monday morning.
>If the company can afford the time for these meetings and other
>cruft, they surely can afford to wait until then!") and management
>didn't expect them to!
>
>>> He went on to say that he couldn't recall the last time ANY of
>>> "the guys" had stayed past 5PM.
>>
>> That doesn't sound like a place I'd want to work at. Don't get me wrong, I
>> strongly believe in balancing one's private life with work and pulling
>> all-nighters year round isn't good. Neither is living off Big Macs and 128oz
>> caffeine-laced sodas with tons of sugar. People make too many mistakes then
>> anyhow and they should have a life. Even engineers. However, dropping the
>> pencil at 5:00:00 no matter what is not cool. There has to be some flexibility.
>
>Why?  Do you expect your dentist to be willing to work late as often as
>folks show up WANTING him to work late (on their problems)?  Or, your
>plumber?  Or, the landscaper?  Or...
>
>Just because you have *one* project (or ten!), doesn't mean you should
>be slave to it.  What cost should your wife/kids bear for your
>"dedication"?
>
>> In my consulting agreement I have a clause that if I have to work after hours
>> or on holidays because the situation demands it there will never be an upcharge
>> (which some other professions tend to do).
>
>IME, that's usually a result of someone's unrealistic expectations/demands
>or poor planning.
>
>When I was young, bosses could brow beat me into agreeing to unrealistic
>schedules.  I've literally run out of the shop with THE prototype in my hand
>as I tried to finish wiring it before field trials.
>
>With time and experience, I learned that folks will ALWAYS want more.  Why
>not?  *I* would want more!  Its up to me to discipline the use of my time
>in a manner with which I'm comfortable -- NOT my employer/client.
>
>I recall a 5PM progress meeting with a client on a Friday.  After discussing
>some changes to the design, the question was asked:  "So, do you think we can
>see something MONDAY?"  I worked hard at keeping my expression "cool" and
>replied:
>    "Well, as it's now 6:30PM on /FRIDAY/, I don't see how that is possible..."
>I.e., "No, I'm not keen on giving up MY weekend any more than I see YOU
>willing to give up YOURS!"
>
>>> I did a gig at an IBM facility where the shifts were staggered
>>> in 6 minute intervals.  I'd watch the engineers that I was
>>> working with diligently gather up their belongings to ensure
>>> they'd be leaving in their assigned 6 minute interval -- while
>>> I'd typically spend much of the night continuing my tasks.
>>
>> That would drive me crazy.
>
>Apparently, they found the benefits and demands (lack thereof)
>more than compensatory!
>
>I'm lucky in that I can work when I want.  And, largely, under the
>circumstances of my own choosing.  So, if I'm engrossed in a
>particular aspect of a problem, I can continue to work on it
>regardless of the time of day or other potential commitments
>("Why don't you get yourself something to eat as I'm going to be
>busy with this for a while...")
>
>But, I *love* my work.  Few other folks that I've met can say
>that -- even if they revel in  the challenges presented.
>
>>>>> Next time you "cut a corner", think about the likelihood that
>>>>> your healthcare providers are behaving similarly -- do you think
>>>>> YOU cut that corner out of malice, laziness, ignorance, etc.?
>>>>
>>>> Some of us don't cut corners.
>>>
>>> Every engineering decision cuts a corner -- everything is a tradeoff.
>>> *You* decided which way to cut the corner (in favor of X over Y).
>>>
>>> Do you strive for -- and achieve -- "perfection" with every design?
>>
>> Generally yes. Perfection to me and in the eyes of my clients is when the
>> design is reliable, easily produceable and below cost target.
>
>To the healthcare provider, we KNOW the patient WILL DIE.
>So, as long as he gets to live another day, is that "perfection"?
>If you could have eeked out a few more cents on the product cost...
>or, increased the MTBF or decreased the MTTR by a bit more, isn't
>that MORE perfect?  In the health care analogy, if you could
>ensure the patients chances of surviving for N years were increased
>by X%, wouldn't that be a more dedicated professional?
>
>>> If you leave the MD's office in better condition than when you
>>> arrived, has he *not* cut a corner?  Could you have left in
>>> EVEN BETTER condition?  Should he hound you AT HOME to follow
>>> his prescribed treatment plan?  How *dedicated* is he -- not?
>>
>> What is usually meant by cutting corners is not small compromises. It's the
>> bigger stuff, things that can have dire consequences. Example: I had major back
>> issues in the past. What did most of the medical pros do? The usual, inject
>> cortizone, prescribe pain pills, prescribe muscle relaxant. _That_ is cutting a
>> corner. Finally another professional (not an MD but a CPA!) told me I might
>> have a magnesium deficiency and it couldn't hurt to take supplements for a
>> while. Half a year later I felt a lot better and have never had a debilitating
>> back pain attack ever since. It has been over a decade now so I can be quite
>> sure that the medical diagnosis of the CPA was better than that of several MDs.
>
>You spec a 2000hr part in a power supply.  It meets the design specifications.
>It meets the expected lifespan of the product (e.g., clears warranty).  But,
>realistically, you KNOW that this will result in the $200 product being
>shit-canned in 2-3 years (of normal use).
>

A typical aluminum electrolytic cap will have a rated lifetime of 2000
hours, and a shelf life the same. That's 80 days, totally insane. Does
Digikey stock them for hours, like dougnnuts? They actually work for
decades.


-- 

John Larkin         Highland Technology, Inc

lunatic fringe electronics 

[toc] | [prev] | [next] | [standalone]

#446167

On 2/4/2017 3:02 PM, John Larkin wrote:
>> You spec a 2000hr part in a power supply.  It meets the design specifications.
>> It meets the expected lifespan of the product (e.g., clears warranty).  But,
>> realistically, you KNOW that this will result in the $200 product being
>> shit-canned in 2-3 years (of normal use).
>
> A typical aluminum electrolytic cap will have a rated lifetime of 2000
> hours, and a shelf life the same. That's 80 days, totally insane. Does
> Digikey stock them for hours, like dougnnuts? They actually work for
> decades.

The hour rating on an ecap indicates lifetime to a particular loss of
capacitance AT WVDC AND OPERATING TEMPERATURE.

So, spec a 105C cap in a device that is operating at an ambient of 25C
with virtually no self-heating from ripple current and it will degrade
at a rate ~250 times slower than that exact same component operating
at a combined ambient + self-heating temperature of 100C.

E.g., in a power supply where ripple currents tend to be high AND
ambient temperature elevated (because other nearby components are
dissipating power) and you WILL notice a marked difference in
lifetimes between 2000hr components and 5000hr devices.

Given that the cost (labor) of repairing damn near anything makes
a $100-200 item effectively disposable, then specing a 2000hr part
when you could just as easily spec a 5000hr part (without incurring
HUGE cost increases IN BIG VOLUME) is far from "perfection"...
regardless of how well it fits the spec and cost targets!

(think: externalities)

[toc] | [prev] | [next] | [standalone]

#446169

"Don Y" <blockedofcourse@foo.invalid> wrote in message 
news:o75ju9$jc7$1@dont-email.me...
> On 2/4/2017 3:02 PM, John Larkin wrote:
>>> You spec a 2000hr part in a power supply.  It meets the design 
>>> specifications.
>>> It meets the expected lifespan of the product (e.g., clears warranty). 
>>> But,
>>> realistically, you KNOW that this will result in the $200 product being
>>> shit-canned in 2-3 years (of normal use).
>>
>> A typical aluminum electrolytic cap will have a rated lifetime of 2000
>> hours, and a shelf life the same. That's 80 days, totally insane. Does
>> Digikey stock them for hours, like dougnnuts? They actually work for
>> decades.
>
> The hour rating on an ecap indicates lifetime to a particular loss of
> capacitance AT WVDC AND OPERATING TEMPERATURE.
>
> So, spec a 105C cap in a device that is operating at an ambient of 25C
> with virtually no self-heating from ripple current and it will degrade
> at a rate ~250 times slower than that exact same component operating
> at a combined ambient + self-heating temperature of 100C.
>
> E.g., in a power supply where ripple currents tend to be high AND
> ambient temperature elevated (because other nearby components are
> dissipating power) and you WILL notice a marked difference in
> lifetimes between 2000hr components and 5000hr devices.
>
> Given that the cost (labor) of repairing damn near anything makes
> a $100-200 item effectively disposable, then specing a 2000hr part
> when you could just as easily spec a 5000hr part (without incurring
> HUGE cost increases IN BIG VOLUME) is far from "perfection"...
> regardless of how well it fits the spec and cost targets!
>
> (think: externalities)

Exactly. Think warranty timer.

[toc] | [prev] | [next] | [standalone]

#446173

On 2/4/2017 3:36 PM, tom wrote:
> "Don Y" <blockedofcourse@foo.invalid> wrote in message
> news:o75ju9$jc7$1@dont-email.me...
>> On 2/4/2017 3:02 PM, John Larkin wrote:
>>>> You spec a 2000hr part in a power supply.  It meets the design
>>>> specifications.
>>>> It meets the expected lifespan of the product (e.g., clears warranty).
>>>> But,
>>>> realistically, you KNOW that this will result in the $200 product being
>>>> shit-canned in 2-3 years (of normal use).
>>>
>>> A typical aluminum electrolytic cap will have a rated lifetime of 2000
>>> hours, and a shelf life the same. That's 80 days, totally insane. Does
>>> Digikey stock them for hours, like dougnnuts? They actually work for
>>> decades.
>>
>> The hour rating on an ecap indicates lifetime to a particular loss of
>> capacitance AT WVDC AND OPERATING TEMPERATURE.
>>
>> So, spec a 105C cap in a device that is operating at an ambient of 25C
>> with virtually no self-heating from ripple current and it will degrade
>> at a rate ~250 times slower than that exact same component operating
>> at a combined ambient + self-heating temperature of 100C.
>>
>> E.g., in a power supply where ripple currents tend to be high AND
>> ambient temperature elevated (because other nearby components are
>> dissipating power) and you WILL notice a marked difference in
>> lifetimes between 2000hr components and 5000hr devices.
>>
>> Given that the cost (labor) of repairing damn near anything makes
>> a $100-200 item effectively disposable, then specing a 2000hr part
>> when you could just as easily spec a 5000hr part (without incurring
>> HUGE cost increases IN BIG VOLUME) is far from "perfection"...
>> regardless of how well it fits the spec and cost targets!
>>
>> (think: externalities)
>
> Exactly. Think warranty timer.

Conversely, for a few pennies more, you could produce a QUALITY
product (though often depriving yourself of a REPLACEMENT SALE
a couple of years down the road).

[But, heck, you may not even be in business, then!]

[toc] | [prev] | [next] | [standalone]

#446174

Den lørdag den 4. februar 2017 kl. 23.46.28 UTC+1 skrev Don Y:
> On 2/4/2017 3:36 PM, tom wrote:
> > "Don Y" <blockedofcourse@foo.invalid> wrote in message
> > news:o75ju9$jc7$1@dont-email.me...
> >> On 2/4/2017 3:02 PM, John Larkin wrote:
> >>>> You spec a 2000hr part in a power supply.  It meets the design
> >>>> specifications.
> >>>> It meets the expected lifespan of the product (e.g., clears warranty).
> >>>> But,
> >>>> realistically, you KNOW that this will result in the $200 product being
> >>>> shit-canned in 2-3 years (of normal use).
> >>>
> >>> A typical aluminum electrolytic cap will have a rated lifetime of 2000
> >>> hours, and a shelf life the same. That's 80 days, totally insane. Does
> >>> Digikey stock them for hours, like dougnnuts? They actually work for
> >>> decades.
> >>
> >> The hour rating on an ecap indicates lifetime to a particular loss of
> >> capacitance AT WVDC AND OPERATING TEMPERATURE.
> >>
> >> So, spec a 105C cap in a device that is operating at an ambient of 25C
> >> with virtually no self-heating from ripple current and it will degrade
> >> at a rate ~250 times slower than that exact same component operating
> >> at a combined ambient + self-heating temperature of 100C.
> >>
> >> E.g., in a power supply where ripple currents tend to be high AND
> >> ambient temperature elevated (because other nearby components are
> >> dissipating power) and you WILL notice a marked difference in
> >> lifetimes between 2000hr components and 5000hr devices.
> >>
> >> Given that the cost (labor) of repairing damn near anything makes
> >> a $100-200 item effectively disposable, then specing a 2000hr part
> >> when you could just as easily spec a 5000hr part (without incurring
> >> HUGE cost increases IN BIG VOLUME) is far from "perfection"...
> >> regardless of how well it fits the spec and cost targets!
> >>
> >> (think: externalities)
> >
> > Exactly. Think warranty timer.
> 
> Conversely, for a few pennies more, you could produce a QUALITY
> product (though often depriving yourself of a REPLACEMENT SALE
> a couple of years down the road).
> 
> [But, heck, you may not even be in business, then!]

or the thing have been scrapped anyway because it was replaced with 
something better/shinier ...

[toc] | [prev] | [next] | [standalone]

#446176

On 2/4/2017 3:56 PM, Lasse Langwadt Christensen wrote:
> Den lørdag den 4. februar 2017 kl. 23.46.28 UTC+1 skrev Don Y:
>> On 2/4/2017 3:36 PM, tom wrote:
>>> "Don Y" <blockedofcourse@foo.invalid> wrote in message
>>> news:o75ju9$jc7$1@dont-email.me...
>>>> On 2/4/2017 3:02 PM, John Larkin wrote:
>>>>>> You spec a 2000hr part in a power supply.  It meets the design
>>>>>> specifications.
>>>>>> It meets the expected lifespan of the product (e.g., clears warranty).
>>>>>> But,
>>>>>> realistically, you KNOW that this will result in the $200 product being
>>>>>> shit-canned in 2-3 years (of normal use).
>>>>>
>>>>> A typical aluminum electrolytic cap will have a rated lifetime of 2000
>>>>> hours, and a shelf life the same. That's 80 days, totally insane. Does
>>>>> Digikey stock them for hours, like dougnnuts? They actually work for
>>>>> decades.
>>>>
>>>> The hour rating on an ecap indicates lifetime to a particular loss of
>>>> capacitance AT WVDC AND OPERATING TEMPERATURE.
>>>>
>>>> So, spec a 105C cap in a device that is operating at an ambient of 25C
>>>> with virtually no self-heating from ripple current and it will degrade
>>>> at a rate ~250 times slower than that exact same component operating
>>>> at a combined ambient + self-heating temperature of 100C.
>>>>
>>>> E.g., in a power supply where ripple currents tend to be high AND
>>>> ambient temperature elevated (because other nearby components are
>>>> dissipating power) and you WILL notice a marked difference in
>>>> lifetimes between 2000hr components and 5000hr devices.
>>>>
>>>> Given that the cost (labor) of repairing damn near anything makes
>>>> a $100-200 item effectively disposable, then specing a 2000hr part
>>>> when you could just as easily spec a 5000hr part (without incurring
>>>> HUGE cost increases IN BIG VOLUME) is far from "perfection"...
>>>> regardless of how well it fits the spec and cost targets!
>>>>
>>>> (think: externalities)
>>>
>>> Exactly. Think warranty timer.
>>
>> Conversely, for a few pennies more, you could produce a QUALITY
>> product (though often depriving yourself of a REPLACEMENT SALE
>> a couple of years down the road).
>>
>> [But, heck, you may not even be in business, then!]
>
> or the thing have been scrapped anyway because it was replaced with
> something better/shinier ...

That's typically true of University settings -- a grant runs out and
all the kit purchased for the grant is shit-canned.  Or, a donor
opts to upgrade a lab, etc.

But, business seems to be slower to discard working kit.  Esp
things like LCD monitors (the "tower/cpu" might get replaced as
part of an upgrade cycle but the monitor lingers -- "it still works")
They might opt to buy some new ones -- but the working old ones
find other uses ("Bob wants a second monitor on his workstation...")

[toc] | [prev] | [next] | [standalone]

#446177

"Don Y" <blockedofcourse@foo.invalid> wrote in message 
news:o75lgn$oa3$2@dont-email.me...
> On 2/4/2017 3:36 PM, tom wrote:
>> "Don Y" <blockedofcourse@foo.invalid> wrote in message
>> news:o75ju9$jc7$1@dont-email.me...
>>> On 2/4/2017 3:02 PM, John Larkin wrote:
>>>>> You spec a 2000hr part in a power supply.  It meets the design
>>>>> specifications.
>>>>> It meets the expected lifespan of the product (e.g., clears warranty).
>>>>> But,
>>>>> realistically, you KNOW that this will result in the $200 product 
>>>>> being
>>>>> shit-canned in 2-3 years (of normal use).
>>>>
>>>> A typical aluminum electrolytic cap will have a rated lifetime of 2000
>>>> hours, and a shelf life the same. That's 80 days, totally insane. Does
>>>> Digikey stock them for hours, like dougnnuts? They actually work for
>>>> decades.
>>>
>>> The hour rating on an ecap indicates lifetime to a particular loss of
>>> capacitance AT WVDC AND OPERATING TEMPERATURE.
>>>
>>> So, spec a 105C cap in a device that is operating at an ambient of 25C
>>> with virtually no self-heating from ripple current and it will degrade
>>> at a rate ~250 times slower than that exact same component operating
>>> at a combined ambient + self-heating temperature of 100C.
>>>
>>> E.g., in a power supply where ripple currents tend to be high AND
>>> ambient temperature elevated (because other nearby components are
>>> dissipating power) and you WILL notice a marked difference in
>>> lifetimes between 2000hr components and 5000hr devices.
>>>
>>> Given that the cost (labor) of repairing damn near anything makes
>>> a $100-200 item effectively disposable, then specing a 2000hr part
>>> when you could just as easily spec a 5000hr part (without incurring
>>> HUGE cost increases IN BIG VOLUME) is far from "perfection"...
>>> regardless of how well it fits the spec and cost targets!
>>>
>>> (think: externalities)
>>
>> Exactly. Think warranty timer.
>
> Conversely, for a few pennies more, you could produce a QUALITY
> product (though often depriving yourself of a REPLACEMENT SALE
> a couple of years down the road).
>
> [But, heck, you may not even be in business, then!]

I look at it as a blessing for us old engineers that could never afford to 
personally own some very high quality test equipment. As an example, a HP 
8566B 22 GHz spectrum analyzer for a few hundred dollars because the scrap 
yard can't get it to run. Replace a few power supply caps and you have a 
wonderful instrument.

That and NVRAM chips with internal Li cells that have finally after 20+ 
years failed.

It's a strange new world.

Regards 

[toc] | [prev] | [next] | [standalone]

#446183

On 2/4/2017 4:22 PM, tom wrote:
> I look at it as a blessing for us old engineers that could never afford to
> personally own some very high quality test equipment. As an example, a HP
> 8566B 22 GHz spectrum analyzer for a few hundred dollars because the scrap
> yard can't get it to run. Replace a few power supply caps and you have a
> wonderful instrument.
>
> That and NVRAM chips with internal Li cells that have finally after 20+
> years failed.
>
> It's a strange new world.

(sigh)  Disturbing.

I used to volunteer at a place that recycled donated kit -- corporate
donors, private citizens, etc.

Despite being located in an off-the-beaten-track part of town,
we'd process ~21,000 pounds of "stuff" every week, 50 weeks/year
(closed for 2 weeks per year).

Extrapolate that to other municipalities around the country and
the amount of cruft going into landfills is STAGGERING!

If you make a conscious effort to try to salvage some of that kit and
redirect it back into the community (schools, clinics, hospitals, etc.),
you also find that many folks don't WANT "last year's model" -- regardless
of how it compares to "this year's model"!

Personally, I have little hesitation to inheriting discards.  E.g., I
know that *I* am the limiting factor in any computer acquisition
(if the applications run 10 times faster, the total time spent on
a task decreases by 5% -- cuz so much of it is meatware limited!).

Plus, there are costs (time) to "upgrade" so the net gain is even less
(I'm seriously considering "passing" on an 8-core, 2GHz Xeon box
just because I don't want to spend the time to install the software
on it)

So, how many logic analyzers, DSO's, freq counters, ARB's, etc. do you
really need?  :<

And, if you want to funnel those items to folks/organizations that can
make use of them, you end up burning lots of time coordinating pickups,
etc.

[OTOH, its fun (and inexpensive!) to see what sorts of kit exists
"in the wild".  I've encountered all sorts of neat toys that, if
nothing else, proved challenging to try to reason out their uses!
And, a great way to see the sorts of things that fail in other
people's designs -- as well as fabrication/assembly techniques,
serviceability issues, etc.]

[toc] | [prev] | [next] | [standalone]

#446210

On 2017/02/04 3:22 PM, tom wrote:
> "Don Y" <blockedofcourse@foo.invalid> wrote in message
> news:o75lgn$oa3$2@dont-email.me...
>> On 2/4/2017 3:36 PM, tom wrote:
>>> "Don Y" <blockedofcourse@foo.invalid> wrote in message
>>> news:o75ju9$jc7$1@dont-email.me...
>>>> On 2/4/2017 3:02 PM, John Larkin wrote:
>>>>>> You spec a 2000hr part in a power supply.  It meets the design
>>>>>> specifications.
>>>>>> It meets the expected lifespan of the product (e.g., clears warranty).
>>>>>> But,
>>>>>> realistically, you KNOW that this will result in the $200 product
>>>>>> being
>>>>>> shit-canned in 2-3 years (of normal use).
>>>>>
>>>>> A typical aluminum electrolytic cap will have a rated lifetime of 2000
>>>>> hours, and a shelf life the same. That's 80 days, totally insane. Does
>>>>> Digikey stock them for hours, like dougnnuts? They actually work for
>>>>> decades.
>>>>
>>>> The hour rating on an ecap indicates lifetime to a particular loss of
>>>> capacitance AT WVDC AND OPERATING TEMPERATURE.
>>>>
>>>> So, spec a 105C cap in a device that is operating at an ambient of 25C
>>>> with virtually no self-heating from ripple current and it will degrade
>>>> at a rate ~250 times slower than that exact same component operating
>>>> at a combined ambient + self-heating temperature of 100C.
>>>>
>>>> E.g., in a power supply where ripple currents tend to be high AND
>>>> ambient temperature elevated (because other nearby components are
>>>> dissipating power) and you WILL notice a marked difference in
>>>> lifetimes between 2000hr components and 5000hr devices.
>>>>
>>>> Given that the cost (labor) of repairing damn near anything makes
>>>> a $100-200 item effectively disposable, then specing a 2000hr part
>>>> when you could just as easily spec a 5000hr part (without incurring
>>>> HUGE cost increases IN BIG VOLUME) is far from "perfection"...
>>>> regardless of how well it fits the spec and cost targets!
>>>>
>>>> (think: externalities)
>>>
>>> Exactly. Think warranty timer.
>>
>> Conversely, for a few pennies more, you could produce a QUALITY
>> product (though often depriving yourself of a REPLACEMENT SALE
>> a couple of years down the road).
>>
>> [But, heck, you may not even be in business, then!]

Job security if the item in question is collectable.

>
...
>
> That and NVRAM chips with internal Li cells that have finally after 20+
> years failed.

We call those Suicide Memory devices. In arcade games when the sealed 
battery dies the game also dies and you need some specialized knowledge 
to revive it. Lookup Capcom Suicide Battery for examples...

>
> It's a strange new world.

The suicide battery protected products from being copied - for about two 
weeks. However collectors down the time line are all cursing the companies.

> Regards
>
>

John

[toc] | [prev] | [next] | [standalone]

#446227

On 2/4/2017 7:22 PM, John Robertson wrote:
>> That and NVRAM chips with internal Li cells that have finally after 20+
>> years failed.
>
> We call those Suicide Memory devices. In arcade games when the sealed battery
> dies the game also dies and you need some specialized knowledge to revive it.
> Lookup Capcom Suicide Battery for examples...

Different sort of beast.  I think Tom is just talking about things like
packaged BBSRAM devices -- a low power CMOS RAM with a battery and some
signal gating logic bolted on.  There have been many produced over the years
and they tend to go out of favor -- except for the products that already have
them installed!

Sun was fond of them in their workstations.  When they'd inevitably fail
("Sun" hasn't produced a workstation in ages!), you'd be left trying to
dig the power connections for the battery out of the encasing plastic and
piggy-backing a battery atop the device.

The components you are referencing are intended to contain "details" that
will leak out when/if the device is copied/reverse engineered thereby
confounding the reverse engineering effort (the "information" is a
key component in the design).

Similar approaches, over the years, included removing device markings
from components (but, a counterfeiter could inspect the die to determine
the identity of the component), potting subassemblies as a step beyond
simply removing markings (but, they can be de-encapsulated, Xrayed, etc.),
embedding metal fibers in the potted assembly (to confound Xray), "full
custom" components (e.g., the "DMA" device used in Williams games, the
customs used in Atari games, etc).

>> It's a strange new world.
>
> The suicide battery protected products from being copied - for about two weeks.
> However collectors down the time line are all cursing the companies.

Blame the counterfeiters -- and the operators who would patronize them
and their "conversion kits".

[flippers.com?  I had a pin-table called Flipper -- first add-a-ball, ~1960.
Fun machine... 3 bells, all of the rubber was "live", etc.]

[toc] | [prev] | [next] | [standalone]

#446168

"John Larkin" <jjlarkin@highlandtechnology.com> wrote in message 
news:tqjc9ct91aa4dopd9rvu2d96j16ss8loug@4ax.com...
> On Sat, 4 Feb 2017 14:15:33 -0700, Don Y <blockedofcourse@foo.invalid>
> wrote:
>
>>On 2/4/2017 1:07 PM, Joerg wrote:
>>> On 2017-02-04 11:34, Don Y wrote:
>>>> On 2/4/2017 8:24 AM, Joerg wrote:
>>>>> On 2017-02-03 08:35, Don Y wrote:
>>>>>> On 2/3/2017 5:21 AM, Winfield Hill wrote:
>>>>>>> bill.sloman@ieee.org wrote...
>>>>>>>>
>>>>>>>> Atul Gawande has written another impressive piece for the New 
>>>>>>>> Yorker
>>>>>>>> http://www.newyorker.com/magazine/2017/01/23/the-heroism-of-incremental-care
>>>>>>>>
>>>>>>>> It is mainly a plea for better primary-care medicine - what your
>>>>>>>> own doctor does for you at your local practice - but it includes
>>>>>>>> a reference to the 1967 collapse of the Silver Bridge across the
>>>>>>>> Ohio River, which was due to a failure in inspection and 
>>>>>>>> maintenance,
>>>>>>>> making the point that while doing heroic operations and building
>>>>>>>> new bridges are both extremely glamorous, the cost-benefit analysis
>>>>>>>> says that it's better to spend time on making sure that what's
>>>>>>>> working stays working, and perhaps improve it a bit from time to
>>>>>>>> time.  Development rather than a new design from scratch.
>>>>>>>
>>>>>>>  Would that we could snap our fingers and get a doctor like that.
>>>>>>
>>>>>> Why do we EXPECT medical professionals to be more "dedicated"
>>>>>> than any other profession?
>>>>>
>>>>> Because we as engineers are dedicated?
>>>>
>>>> You've never met a "9-to-5" engineer?
>>>
>>> Yes, but usually in government positions. Very occasionally in private 
>>> industry
>>> but then those were very bright folks.
>>
>>I've experience/exposure to gummit labs (i.e., all of their funding
>>comes from the taxpayer), large multinationals (i.e., tens of thousands
>>of employees worldwide), large corporations (thousands of employees),
>>startups (one to four souls) and struggling wannabes (~100 bodies).
>>
>>In virtually all of those scenarios, it was rare to find folks
>>EAGER to spend THEIR TIME pursuing the corporate goal.  The few
>>cases I was able to single out were typically very bright/inquisitive
>>folks who were completely engaged in the CHALLENGE before them.
>>
>>>> I had a (regular) job, many years ago.  Bringing up the prototype
>>>> I naturally was excited and "didn't want the day to end" (until
>>>> seeing it run).
>>>>
>>>> Next day, boss called me in his office and casually mentioned:
>>>>     "I see you kept the guys here until 8PM last night..."
>>>> (keep in mind, *I* am not the project manager.  nor have any
>>>> other "special responsibilities" as far as the project was
>>>> concerned -- Just Another Engineer with my little portion of
>>>> the pie)
>>>>
>>>> I was startled by the comment -- as if it was an accusation.
>>>
>>> If it was a union shop it might have been.
>>
>>Nope.  Just a big Fortune 500.  Staff wasn't used to "working late"
>>("The problem will still be here, waiting for us, Monday morning.
>>If the company can afford the time for these meetings and other
>>cruft, they surely can afford to wait until then!") and management
>>didn't expect them to!
>>
>>>> He went on to say that he couldn't recall the last time ANY of
>>>> "the guys" had stayed past 5PM.
>>>
>>> That doesn't sound like a place I'd want to work at. Don't get me wrong, 
>>> I
>>> strongly believe in balancing one's private life with work and pulling
>>> all-nighters year round isn't good. Neither is living off Big Macs and 
>>> 128oz
>>> caffeine-laced sodas with tons of sugar. People make too many mistakes 
>>> then
>>> anyhow and they should have a life. Even engineers. However, dropping 
>>> the
>>> pencil at 5:00:00 no matter what is not cool. There has to be some 
>>> flexibility.
>>
>>Why?  Do you expect your dentist to be willing to work late as often as
>>folks show up WANTING him to work late (on their problems)?  Or, your
>>plumber?  Or, the landscaper?  Or...
>>
>>Just because you have *one* project (or ten!), doesn't mean you should
>>be slave to it.  What cost should your wife/kids bear for your
>>"dedication"?
>>
>>> In my consulting agreement I have a clause that if I have to work after 
>>> hours
>>> or on holidays because the situation demands it there will never be an 
>>> upcharge
>>> (which some other professions tend to do).
>>
>>IME, that's usually a result of someone's unrealistic expectations/demands
>>or poor planning.
>>
>>When I was young, bosses could brow beat me into agreeing to unrealistic
>>schedules.  I've literally run out of the shop with THE prototype in my 
>>hand
>>as I tried to finish wiring it before field trials.
>>
>>With time and experience, I learned that folks will ALWAYS want more.  Why
>>not?  *I* would want more!  Its up to me to discipline the use of my time
>>in a manner with which I'm comfortable -- NOT my employer/client.
>>
>>I recall a 5PM progress meeting with a client on a Friday.  After 
>>discussing
>>some changes to the design, the question was asked:  "So, do you think we 
>>can
>>see something MONDAY?"  I worked hard at keeping my expression "cool" and
>>replied:
>>    "Well, as it's now 6:30PM on /FRIDAY/, I don't see how that is 
>> possible..."
>>I.e., "No, I'm not keen on giving up MY weekend any more than I see YOU
>>willing to give up YOURS!"
>>
>>>> I did a gig at an IBM facility where the shifts were staggered
>>>> in 6 minute intervals.  I'd watch the engineers that I was
>>>> working with diligently gather up their belongings to ensure
>>>> they'd be leaving in their assigned 6 minute interval -- while
>>>> I'd typically spend much of the night continuing my tasks.
>>>
>>> That would drive me crazy.
>>
>>Apparently, they found the benefits and demands (lack thereof)
>>more than compensatory!
>>
>>I'm lucky in that I can work when I want.  And, largely, under the
>>circumstances of my own choosing.  So, if I'm engrossed in a
>>particular aspect of a problem, I can continue to work on it
>>regardless of the time of day or other potential commitments
>>("Why don't you get yourself something to eat as I'm going to be
>>busy with this for a while...")
>>
>>But, I *love* my work.  Few other folks that I've met can say
>>that -- even if they revel in  the challenges presented.
>>
>>>>>> Next time you "cut a corner", think about the likelihood that
>>>>>> your healthcare providers are behaving similarly -- do you think
>>>>>> YOU cut that corner out of malice, laziness, ignorance, etc.?
>>>>>
>>>>> Some of us don't cut corners.
>>>>
>>>> Every engineering decision cuts a corner -- everything is a tradeoff.
>>>> *You* decided which way to cut the corner (in favor of X over Y).
>>>>
>>>> Do you strive for -- and achieve -- "perfection" with every design?
>>>
>>> Generally yes. Perfection to me and in the eyes of my clients is when 
>>> the
>>> design is reliable, easily produceable and below cost target.
>>
>>To the healthcare provider, we KNOW the patient WILL DIE.
>>So, as long as he gets to live another day, is that "perfection"?
>>If you could have eeked out a few more cents on the product cost...
>>or, increased the MTBF or decreased the MTTR by a bit more, isn't
>>that MORE perfect?  In the health care analogy, if you could
>>ensure the patients chances of surviving for N years were increased
>>by X%, wouldn't that be a more dedicated professional?
>>
>>>> If you leave the MD's office in better condition than when you
>>>> arrived, has he *not* cut a corner?  Could you have left in
>>>> EVEN BETTER condition?  Should he hound you AT HOME to follow
>>>> his prescribed treatment plan?  How *dedicated* is he -- not?
>>>
>>> What is usually meant by cutting corners is not small compromises. It's 
>>> the
>>> bigger stuff, things that can have dire consequences. Example: I had 
>>> major back
>>> issues in the past. What did most of the medical pros do? The usual, 
>>> inject
>>> cortizone, prescribe pain pills, prescribe muscle relaxant. _That_ is 
>>> cutting a
>>> corner. Finally another professional (not an MD but a CPA!) told me I 
>>> might
>>> have a magnesium deficiency and it couldn't hurt to take supplements for 
>>> a
>>> while. Half a year later I felt a lot better and have never had a 
>>> debilitating
>>> back pain attack ever since. It has been over a decade now so I can be 
>>> quite
>>> sure that the medical diagnosis of the CPA was better than that of 
>>> several MDs.
>>
>>You spec a 2000hr part in a power supply.  It meets the design 
>>specifications.
>>It meets the expected lifespan of the product (e.g., clears warranty). 
>>But,
>>realistically, you KNOW that this will result in the $200 product being
>>shit-canned in 2-3 years (of normal use).
>>
>
> A typical aluminum electrolytic cap will have a rated lifetime of 2000
> hours, and a shelf life the same. That's 80 days, totally insane. Does
> Digikey stock them for hours, like dougnnuts? They actually work for
> decades.
>
>
> -- 

That is at the rated temperature. 85 or 105 °C

[toc] | [prev] | [next] | [standalone]

#446171

On 2/4/2017 3:35 PM, tom wrote:

>>> You spec a 2000hr part in a power supply.  It meets the design
>>> specifications.
>>> It meets the expected lifespan of the product (e.g., clears warranty).
>>> But, realistically, you KNOW that this will result in the $200 product being
>>> shit-canned in 2-3 years (of normal use).
>>
>> A typical aluminum electrolytic cap will have a rated lifetime of 2000
>> hours, and a shelf life the same. That's 80 days, totally insane. Does
>> Digikey stock them for hours, like dougnnuts? They actually work for
>> decades.
>
> That is at the rated temperature. 85 or 105 °C

And operating voltage (though you usually can't get more than a 2X increase in
life by operating at a reduced voltage).

So, caps sitting on a shelf in a warehouse at "STP" have life expectancies
~500 times greater than when operating at ~100C *internal* temperatures.
2000 hours becomes 100 years in storage.

(realistically, seals degrade after 10-20 years)

[toc] | [prev] | [next] | [standalone]

#446203

On Saturday, February 4, 2017 at 5:44:50 PM UTC-5, Don Y wrote:

> > That is at the rated temperature. 85 or 105 °C
> 
> And operating voltage (though you usually can't get more than a 2X increase in
> life by operating at a reduced voltage).
> 
> So, caps sitting on a shelf in a warehouse at "STP" have life expectancies
> ~500 times greater than when operating at ~100C *internal* temperatures.
> 2000 hours becomes 100 years in storage.
> 
> (realistically, seals degrade after 10-20 years)

I thought that aluminum electrolytic caps had a longer life if they had voltage applied.  

                                     Dan

[toc] | [prev] | [next] | [standalone]

#446206

On Sat, 4 Feb 2017 15:44:39 -0700, Don Y <blockedofcourse@foo.invalid>
wrote:

>On 2/4/2017 3:35 PM, tom wrote:
>
>>>> You spec a 2000hr part in a power supply.  It meets the design
>>>> specifications.
>>>> It meets the expected lifespan of the product (e.g., clears warranty).
>>>> But, realistically, you KNOW that this will result in the $200 product being
>>>> shit-canned in 2-3 years (of normal use).
>>>
>>> A typical aluminum electrolytic cap will have a rated lifetime of 2000
>>> hours, and a shelf life the same. That's 80 days, totally insane. Does
>>> Digikey stock them for hours, like dougnnuts? They actually work for
>>> decades.
>>
>> That is at the rated temperature. 85 or 105 °C
>
>And operating voltage (though you usually can't get more than a 2X increase in
>life by operating at a reduced voltage).

But you get more than a 2x increase for every 10C below the rated
temperature.  Reduce the ripple current and you can do a lot better. I
don't rely on elecrolytics for ripple, anyway.
>
>So, caps sitting on a shelf in a warehouse at "STP" have life expectancies
>~500 times greater than when operating at ~100C *internal* temperatures.
>2000 hours becomes 100 years in storage.
>
>(realistically, seals degrade after 10-20 years)

[toc] | [prev] | [next] | [standalone]

#446293

On 2017-02-04 13:15, Don Y wrote:
> On 2/4/2017 1:07 PM, Joerg wrote:
>> On 2017-02-04 11:34, Don Y wrote:
>>> On 2/4/2017 8:24 AM, Joerg wrote:
>>>> On 2017-02-03 08:35, Don Y wrote:
>>>>> On 2/3/2017 5:21 AM, Winfield Hill wrote:
>>>>>> bill.sloman@ieee.org wrote...
>>>>>>>
>>>>>>> Atul Gawande has written another impressive piece for the New Yorker
>>>>>>> http://www.newyorker.com/magazine/2017/01/23/the-heroism-of-incremental-care
>>>>>>>
>>>>>>>
>>>>>>> It is mainly a plea for better primary-care medicine - what your
>>>>>>> own doctor does for you at your local practice - but it includes
>>>>>>> a reference to the 1967 collapse of the Silver Bridge across the
>>>>>>> Ohio River, which was due to a failure in inspection and
>>>>>>> maintenance,
>>>>>>> making the point that while doing heroic operations and building
>>>>>>> new bridges are both extremely glamorous, the cost-benefit analysis
>>>>>>> says that it's better to spend time on making sure that what's
>>>>>>> working stays working, and perhaps improve it a bit from time to
>>>>>>> time.  Development rather than a new design from scratch.
>>>>>>
>>>>>>  Would that we could snap our fingers and get a doctor like that.
>>>>>
>>>>> Why do we EXPECT medical professionals to be more "dedicated"
>>>>> than any other profession?
>>>>
>>>> Because we as engineers are dedicated?
>>>
>>> You've never met a "9-to-5" engineer?
>>
>> Yes, but usually in government positions. Very occasionally in private
>> industry
>> but then those were very bright folks.
>
> I've experience/exposure to gummit labs (i.e., all of their funding
> comes from the taxpayer), large multinationals (i.e., tens of thousands
> of employees worldwide), large corporations (thousands of employees),
> startups (one to four souls) and struggling wannabes (~100 bodies).
>

There's part of the problem. You should have looked at companies much 
smaller than that. That is also, and not by chance, where most 
innovation happens.


> In virtually all of those scenarios, it was rare to find folks
> EAGER to spend THEIR TIME pursuing the corporate goal.  The few
> cases I was able to single out were typically very bright/inquisitive
> folks who were completely engaged in the CHALLENGE before them.
>

A very nice countermeasure to such passive behavior is also a good old 
American one: Stock options. I never had any job without those. At my 
last employer before starting self-employed every employee had them, 
plus a stock purchase program. IOW we were de facto partly employee-owned.


>>> I had a (regular) job, many years ago.  Bringing up the prototype
>>> I naturally was excited and "didn't want the day to end" (until
>>> seeing it run).
>>>
>>> Next day, boss called me in his office and casually mentioned:
>>>     "I see you kept the guys here until 8PM last night..."
>>> (keep in mind, *I* am not the project manager.  nor have any
>>> other "special responsibilities" as far as the project was
>>> concerned -- Just Another Engineer with my little portion of
>>> the pie)
>>>
>>> I was startled by the comment -- as if it was an accusation.
>>
>> If it was a union shop it might have been.
>
> Nope.  Just a big Fortune 500.  Staff wasn't used to "working late"
> ("The problem will still be here, waiting for us, Monday morning.
> If the company can afford the time for these meetings and other
> cruft, they surely can afford to wait until then!") and management
> didn't expect them to!
>

Not a place I'd like to work at as an employee. As a consultant, no 
problem, but not as an employee.


>>> He went on to say that he couldn't recall the last time ANY of
>>> "the guys" had stayed past 5PM.
>>
>> That doesn't sound like a place I'd want to work at. Don't get me
>> wrong, I
>> strongly believe in balancing one's private life with work and pulling
>> all-nighters year round isn't good. Neither is living off Big Macs and
>> 128oz
>> caffeine-laced sodas with tons of sugar. People make too many mistakes
>> then
>> anyhow and they should have a life. Even engineers. However, dropping the
>> pencil at 5:00:00 no matter what is not cool. There has to be some
>> flexibility.
>
> Why?  Do you expect your dentist to be willing to work late as often as
> folks show up WANTING him to work late (on their problems)?  Or, your
> plumber?  Or, the landscaper?  Or...
>

If there is an emergency I prefer to have a dentist who won't simply say 
"Oh, I'll give you a prescription for the mother of all pain pills and 
then you come in some times next week when we have an opening, ok?"

I know dentists who'd do that but I always made sure we had ones who didn't.


> Just because you have *one* project (or ten!), doesn't mean you should
> be slave to it.  What cost should your wife/kids bear for your
> "dedication"?
>

That's why I wrote that I believe in work/life balance. You can't make 
overtime routine and it wouldn't be healthy for anyone.


>> In my consulting agreement I have a clause that if I have to work
>> after hours
>> or on holidays because the situation demands it there will never be an
>> upcharge
>> (which some other professions tend to do).
>
> IME, that's usually a result of someone's unrealistic expectations/demands
> or poor planning.
>

No, it is usually the result of something that unexpectedly went wrong, 
typically very wrong. My first consulting job ever happened on the 
morning of my first day in the office. I had clients signed up before 
starting. Phone rang around 6am, over and over again. So I went down the 
hallway in my pajamas, picked it up. "We are at a trade show, our 
flagship system went PHUT *POOF* and the service guys can't get it going 
again. Could you come here, and please floor it?"


> When I was young, bosses could brow beat me into agreeing to unrealistic
> schedules.  I've literally run out of the shop with THE prototype in my
> hand
> as I tried to finish wiring it before field trials.
>

Yeah, unrealistic schedules are especially typical in med tech. 
Unfortunately.


> With time and experience, I learned that folks will ALWAYS want more.  Why
> not?  *I* would want more!  Its up to me to discipline the use of my time
> in a manner with which I'm comfortable -- NOT my employer/client.
>

With me, if they want more they also pay more :-)

I strictly work by time and materials, never fixed bid. Ever.


> I recall a 5PM progress meeting with a client on a Friday.  After
> discussing
> some changes to the design, the question was asked:  "So, do you think
> we can
> see something MONDAY?"  I worked hard at keeping my expression "cool" and
> replied:
>     "Well, as it's now 6:30PM on /FRIDAY/, I don't see how that is
> possible..."
> I.e., "No, I'm not keen on giving up MY weekend any more than I see YOU
> willing to give up YOURS!"
>

Sure, got to be assertive. Unless it is an unexpected issue and the 
trade show starts Monday. And then I'd expect the employer to rewards 
these employees nicely, in a way that their family is also rewarded. 
Fully paid trip to Disneyland with the kids, things like that.


>>> I did a gig at an IBM facility where the shifts were staggered
>>> in 6 minute intervals.  I'd watch the engineers that I was
>>> working with diligently gather up their belongings to ensure
>>> they'd be leaving in their assigned 6 minute interval -- while
>>> I'd typically spend much of the night continuing my tasks.
>>
>> That would drive me crazy.
>
> Apparently, they found the benefits and demands (lack thereof)
> more than compensatory!
>
> I'm lucky in that I can work when I want.  And, largely, under the
> circumstances of my own choosing.  So, if I'm engrossed in a
> particular aspect of a problem, I can continue to work on it
> regardless of the time of day or other potential commitments
> ("Why don't you get yourself something to eat as I'm going to be
> busy with this for a while...")
>
> But, I *love* my work.  Few other folks that I've met can say
> that -- even if they revel in  the challenges presented.
>

Same here. I really like my work and I am self-employed. Of course, this 
also means that clients can have an emergency and then I am there for 
them. They know that and they do not take advantage of me just like I 
would not call a dentist on a weekend unless the pain was truly unbearable.


>>>>> Next time you "cut a corner", think about the likelihood that
>>>>> your healthcare providers are behaving similarly -- do you think
>>>>> YOU cut that corner out of malice, laziness, ignorance, etc.?
>>>>
>>>> Some of us don't cut corners.
>>>
>>> Every engineering decision cuts a corner -- everything is a tradeoff.
>>> *You* decided which way to cut the corner (in favor of X over Y).
>>>
>>> Do you strive for -- and achieve -- "perfection" with every design?
>>
>> Generally yes. Perfection to me and in the eyes of my clients is when the
>> design is reliable, easily produceable and below cost target.
>
> To the healthcare provider, we KNOW the patient WILL DIE.
> So, as long as he gets to live another day, is that "perfection"?
> If you could have eeked out a few more cents on the product cost...
> or, increased the MTBF or decreased the MTTR by a bit more, isn't
> that MORE perfect?  In the health care analogy, if you could
> ensure the patients chances of surviving for N years were increased
> by X%, wouldn't that be a more dedicated professional?
>

Perfection is more multi-faceted than that. The client needs to feel all 
around happy about a design. Or a problem presented to me to be remedied 
must completely vanish. That to me and to my clients is perfection.


>>> If you leave the MD's office in better condition than when you
>>> arrived, has he *not* cut a corner?  Could you have left in
>>> EVEN BETTER condition?  Should he hound you AT HOME to follow
>>> his prescribed treatment plan?  How *dedicated* is he -- not?
>>
>> What is usually meant by cutting corners is not small compromises.
>> It's the
>> bigger stuff, things that can have dire consequences. Example: I had
>> major back
>> issues in the past. What did most of the medical pros do? The usual,
>> inject
>> cortizone, prescribe pain pills, prescribe muscle relaxant. _That_ is
>> cutting a
>> corner. Finally another professional (not an MD but a CPA!) told me I
>> might
>> have a magnesium deficiency and it couldn't hurt to take supplements
>> for a
>> while. Half a year later I felt a lot better and have never had a
>> debilitating
>> back pain attack ever since. It has been over a decade now so I can be
>> quite
>> sure that the medical diagnosis of the CPA was better than that of
>> several MDs.
>
> You spec a 2000hr part in a power supply.  It meets the design
> specifications.


I don't do that. First, electrolytics have a measured loss of 
capacitance over xxxx hours if constantly at the temperature in that 
rating. Even then they do not die after 2000h. Secondly, a 2000h/85C cap 
will never be granted entry into any design of mine. Third, I'd never 
spec a bulk capacitance so close that a 50% drop would cause the circuit 
to become iffy. In fact, I avoid electrolytics altogether whenever possible.

BTW, some electrolytics are stunning. On Hammond organs they place the 
big electrolytc right next to the rectifier tube which gets really hot. 
Beats me why, I'd never do that. The organ we have sometimes ran to days 
on end at the previous owner, mostly because they forgot to turn it off. 
This electrolytic entered service in 1961 and is still going strong.


> It meets the expected lifespan of the product (e.g., clears warranty).


Barely clearing a warranty phase is not a solid design in my eyes and 
those of my clients. Should it ever happen that a client wanted me to do 
that (never has so far) I'd likely turn down the assignment.


>                                                             ... But,
> realistically, you KNOW that this will result in the $200 product being
> shit-canned in 2-3 years (of normal use).
>

Which is why I would not design it. They'd have to find someone else.


> Have you cut a corner?  Have you achieved perfection?  Or, is it someone
> else's "problem"/concern cuz that's what they asked for?
>

It won't happen at this here desk.

-- 
Regards, Joerg

http://www.analogconsultants.com/

[toc] | [prev] | [next] | [standalone]

#446314

On 2/5/2017 9:14 AM, Joerg wrote:
>>>>>>>  Would that we could snap our fingers and get a doctor like that.
>>>>>>
>>>>>> Why do we EXPECT medical professionals to be more "dedicated"
>>>>>> than any other profession?
>>>>>
>>>>> Because we as engineers are dedicated?
>>>>
>>>> You've never met a "9-to-5" engineer?
>>>
>>> Yes, but usually in government positions. Very occasionally in private
>>> industry
>>> but then those were very bright folks.
>>
>> I've experience/exposure to gummit labs (i.e., all of their funding
>> comes from the taxpayer), large multinationals (i.e., tens of thousands
>> of employees worldwide), large corporations (thousands of employees),
>> startups (one to four souls) and struggling wannabes (~100 bodies).
>
> There's part of the problem. You should have looked at companies much smaller
> than that. That is also, and not by chance, where most innovation happens.

I've *been* the "Engineering Department" at firms.  I've also been employed
at places where there were 6 or 8 other souls in the building, each of us with
a key to the front door and alarm code -- and, responsible to fix the toilet
if it ever backed up!

IME, larger tends to "tolerate" more of a 9-to-5 attitude.  But, conversely,
smaller doesn't imply all-hours-of-the-day-and-night heroics.

I've met many engineers who simply treat it as a job.  And, others who
consider it a *hobby* (in addition to a job) -- with an extra findness
for the challenges that it presents.

>> In virtually all of those scenarios, it was rare to find folks
>> EAGER to spend THEIR TIME pursuing the corporate goal.  The few
>> cases I was able to single out were typically very bright/inquisitive
>> folks who were completely engaged in the CHALLENGE before them.
>
> A very nice countermeasure to such passive behavior is also a good old American
> one: Stock options. I never had any job without those. At my last employer
> before starting self-employed every employee had them, plus a stock purchase
> program. IOW we were de facto partly employee-owned.

I've worked at places where annual bonuses exceeded your base pay.
Money is not a motivator.  Lack of money can demotivate but dangling
a carrot doesn't tend to produce results.

The colleagues with shich I've spent many "late nights" were invariably NOT
motivated by financial issues -- rewards OR penalties.  But, rather, a
genuine interest in the problem they had before them.  The knowledge that
"leaving work" at this time would still have their minds fully engaged with
the problem -- so, why "leave work"?

>>>> Next day, boss called me in his office and casually mentioned:
>>>>     "I see you kept the guys here until 8PM last night..."
>>>> (keep in mind, *I* am not the project manager.  nor have any
>>>> other "special responsibilities" as far as the project was
>>>> concerned -- Just Another Engineer with my little portion of
>>>> the pie)
>>>>
>>>> I was startled by the comment -- as if it was an accusation.
>>>
>>> If it was a union shop it might have been.
>>
>> Nope.  Just a big Fortune 500.  Staff wasn't used to "working late"
>> ("The problem will still be here, waiting for us, Monday morning.
>> If the company can afford the time for these meetings and other
>> cruft, they surely can afford to wait until then!") and management
>> didn't expect them to!
>
> Not a place I'd like to work at as an employee. As a consultant, no problem,
> but not as an employee.

Having lived with the turmoils of small companies not being able to
meet payroll, buy needed equipment, get VC funding, etc. I opted for a
"large, well established firm".  As I'd not, prior to that, encountered
the 9-to-5 mentality, I wasn't prepared for it.  I naively thought
ALL engineers were driven by the challenges they were tackling and
it didn't occur to me that some/many would just see it as a job:
     "The prototype will STILL be here Monday morning.  No need to
     keep working on it, tonight..."

>> Just because you have *one* project (or ten!), doesn't mean you should
>> be slave to it.  What cost should your wife/kids bear for your
>> "dedication"?
>
> That's why I wrote that I believe in work/life balance. You can't make overtime
> routine and it wouldn't be healthy for anyone.

There's a gulf between "routine" and "exception".  I've frequently worked
with crews where individuals will get up and "excuse" themselves:
     "Sorry, guys, my wife is expecting me for dinner..."
No one grumbling that he's "bailing out" on them, etc.

Likewise, been the sole person in a facility in the wee hours of
the morning and find a colleague wandering in to ALSO work on the
project:
     "Gee, I didn't expect to find you here, Don..."

But, there are others that I would never expect to be apologetic (in the first
case) or surprised (in the second) as the idea of working on "work" any
time "after hours" is anathema to them.

>> When I was young, bosses could brow beat me into agreeing to unrealistic
>> schedules.  I've literally run out of the shop with THE prototype in my
>> hand as I tried to finish wiring it before field trials.
>
> Yeah, unrealistic schedules are especially typical in med tech. Unfortunately.

When you have a manufacturing facility tied to your engineering efforts,
you see the faces of the people that will be "idled" (laid off) when you
miss a deadline.  Management thinks this is a means of coercing folks
to meet those deadlines; responsible engineers see it as a means of
ensuring they only *commit* to realistic deadlines.

The Venn diagram isn't guaranteed to show an intersection of those
event spaces!  :>

>> With time and experience, I learned that folks will ALWAYS want more.  Why
>> not?  *I* would want more!  Its up to me to discipline the use of my time
>> in a manner with which I'm comfortable -- NOT my employer/client.
>
> With me, if they want more they also pay more :-)
>
> I strictly work by time and materials, never fixed bid. Ever.

I started my business with that approach -- on the advice of others.
Then, realized it let clients drag projects out far longer than *I*
was interested (I'm not eager to dig ditches regardless of the pay!)

So, I take the exact opposite approach:  fixed cost for everything.
*But*, that means the project's extents are completely defined
before hand!  If *you* don't know what you want (and expect me to
design N different concepts so you can see WHY you don't like
them), then I surely don't want to be your entertainment.

I want a known exit from a project -- so I can move on to something else.
Diversity being far more important than a "continuous paycheck" from
an endless project!

This criteria also tends to filter out the clients that are "insecure";
that want someone to hold their hands while they try to make up their
minds.

And, lets clients nibble at a problem instead of signing a blank check
and wondering where it will lead:
     "We've spent $X and have nothing to show for it!"
     "Sure you do!  All of those approaches that you spent money reifying
     and then ruled out!  Had you done that on paper -- or in your own
     mind -- then you wouldn't have spent the $X and wasted all of those
     hours of MY LIFE!"

>> I recall a 5PM progress meeting with a client on a Friday.  After
>> discussing
>> some changes to the design, the question was asked:  "So, do you think
>> we can
>> see something MONDAY?"  I worked hard at keeping my expression "cool" and
>> replied:
>>     "Well, as it's now 6:30PM on /FRIDAY/, I don't see how that is
>> possible..."
>> I.e., "No, I'm not keen on giving up MY weekend any more than I see YOU
>> willing to give up YOURS!"
>
> Sure, got to be assertive. Unless it is an unexpected issue and the trade show
> starts Monday. And then I'd expect the employer to rewards these employees
> nicely, in a way that their family is also rewarded. Fully paid trip to
> Disneyland with the kids, things like that.

"If you don't like standing in shit, then you shouldn't be a plumber"

>> I'm lucky in that I can work when I want.  And, largely, under the
>> circumstances of my own choosing.  So, if I'm engrossed in a
>> particular aspect of a problem, I can continue to work on it
>> regardless of the time of day or other potential commitments
>> ("Why don't you get yourself something to eat as I'm going to be
>> busy with this for a while...")
>>
>> But, I *love* my work.  Few other folks that I've met can say
>> that -- even if they revel in  the challenges presented.
>
> Same here. I really like my work and I am self-employed. Of course, this also
> means that clients can have an emergency and then I am there for them. They

There's a difference between "having an emergency" and "poor planning".
We accept the term "emergency" far too willingly.  Like "having an
accident" and relying on the interpretation that it was "unexpected"
vs. "by chance".

Failing to plan for Y2K isn't an emergency.  Its just a defect in your
planning process!

> know that and they do not take advantage of me just like I would not call a
> dentist on a weekend unless the pain was truly unbearable.
>
>>>>>> Next time you "cut a corner", think about the likelihood that
>>>>>> your healthcare providers are behaving similarly -- do you think
>>>>>> YOU cut that corner out of malice, laziness, ignorance, etc.?
>>>>>
>>>>> Some of us don't cut corners.
>>>>
>>>> Every engineering decision cuts a corner -- everything is a tradeoff.
>>>> *You* decided which way to cut the corner (in favor of X over Y).
>>>>
>>>> Do you strive for -- and achieve -- "perfection" with every design?
>>>
>>> Generally yes. Perfection to me and in the eyes of my clients is when the
>>> design is reliable, easily produceable and below cost target.
>>
>> To the healthcare provider, we KNOW the patient WILL DIE.
>> So, as long as he gets to live another day, is that "perfection"?
>> If you could have eeked out a few more cents on the product cost...
>> or, increased the MTBF or decreased the MTTR by a bit more, isn't
>> that MORE perfect?  In the health care analogy, if you could
>> ensure the patients chances of surviving for N years were increased
>> by X%, wouldn't that be a more dedicated professional?
>
> Perfection is more multi-faceted than that. The client needs to feel all around
> happy about a design. Or a problem presented to me to be remedied must
> completely vanish. That to me and to my clients is perfection.

So, if the MD sends the patient home with a narcotic and the pain vanishes,
has he achieved perfection?  Despite the fact that the underlying cause
hasn't been addressed?

If the docotr makes a best-guess assessment that the patient will likely
die from X when he is 78 years old and his nominal life expectancy is 75,
then is it acceptable to NOT treat X?

>> It meets the expected lifespan of the product (e.g., clears warranty).
>
> Barely clearing a warranty phase is not a solid design in my eyes and those of
> my clients. Should it ever happen that a client wanted me to do that (never has
> so far) I'd likely turn down the assignment.

Why?  A client can opt to maximize their profit at the expense of their
reputation (People's Electronic Device Manufacturer #24566).  To do so,
model "in warranty" replacements (and their costs) vs. the cost to
provide a QUALITY product that will continue to operate far beyond the
warranty period.

Is the purchaser likely to buy a SECOND unit -- even if the first NEVER FAILS?
If not, then what value to increasing the consumer's impression of the product
or its manufacturer -- as offset by the added costs (lost profits) of making
that better device?

How many people bought ANOTHER DTV decoder because they were so satisfied
with the first one they purchased?  How many people had BOTH of the DTV
converters that they purchased (subsidized) *fail* in short order?
How many of the DTV manufacturers are still making DTV converters and
expecting to tap into that market?

There are lots of markets where reputation means little and cost is king.
When there are no regulatory agencies involved, it is stunning how many
corners will/can be cut in crafting products!

>>                                                             ... But,
>> realistically, you KNOW that this will result in the $200 product being
>> shit-canned in 2-3 years (of normal use).
>
> Which is why I would not design it. They'd have to find someone else.

A 9-to-5 engineer?  Or, someone who is REALLY CLEVER at figuring out how to
design RIGHT to the margins??  :>

>> Have you cut a corner?  Have you achieved perfection?  Or, is it someone
>> else's "problem"/concern cuz that's what they asked for?
>
> It won't happen at this here desk.

[toc] | [prev] | [next] | [standalone]

Page 2 of 3 — ← Prev page 1 [2] 3  Next page →

Back to top | Article view | sci.electronics.design

csiph-web