pgtruspace's blog

about things that interest me.

An Engineers’ Tale

An Engineers tale

 Berkeley/ Livermore; Former Chief Engineer, Lawrence Radiation Laboratories. .

In the early 1980s I was visiting the home of a good friend, his recently retired father was there. We were sharing beers and swapping lies. When the subject of his fathers’ engineering work at Lawrence Radiation Labs. came up. His father said that the scientists would come to the Laboratories with Large grants to carry out a test of their pet theory. It was up to him to create the experiment that would yield the results that were required. When the needed results were created, they would then write up their paper and publish it. It would then become the foundation of the next step in the “science” of physics. As the cost of the experiment was so high, it was not likely that anyone would ever get the funding to repeat it.

So, he said, I should not believe everything that what was published is an actual fact in physics.

>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>.

Berkeley Physics Laboratory

On my 1963 visit to a symposium of physics papers and tour of the Berkeley science labs.

As a budding young scientist, the government funded my travel to California  Berkeley University to encourage my travel on that path. This allowed me to visit much of the Science Department. In particular the radiation lab run by Lawrence Laboratories. This entailed their Cyclotron, linear accelerator, Detector system, computers, cloud chambers, and photo labs. The biggest and best of that era. They were very proud to show off their toys to the budding, want to be,  science students. The Physics work was of some interest to me. The engineering needed! Magnets, shielding, detectors, computers, right down to the time measurements of signal travel in the wires and equipment in nano and pico seconds. That was fascinating! The hell with being a Physicist! I wanted to be an Engineer. Applied Science, real stuff, not lame theory based on Assumptions  and verbal higher mathematics  priest talk.

After the presentation of the grad students papers on their experiments, facts ( the real science) examined and conclusions that they had drawn. We questioned them. I was struck by the lack of exploration of other things that might have caused the results that they had detected. Generally the experiment was to prove their theory and their argument rather then explore all the possibilities. But then what did I know? I was just a 16 year old, Hick from the Sticks, kid.

When I was a science student we were told that Science was done by creating a theory, Develop an experiment that gathers the facts that you will use to prove your argument. The proven theory then becomes scientific fact.

That can be the path to BS ( Bad Science) if the “proved” facts are bs (bull scat) and the key parts of the argument are based on assumptions from “cited authorities” previous work. pg

Advertisements

18 responses to “An Engineers’ Tale

  1. p.g.sharrow February 23, 2014 at 9:46 pm

    While I was in discussion in comments at EMSmith’s: ChiefIO blog, I made mention of the my opinion of the quality of some scientific research in physics and a tale told to me by an old engineer. Engineers swap tales just like sailors or farmers of their life experiences. It is another way to learn from others mistakes. 😎

  2. Simon Derricutt February 24, 2014 at 3:35 am

    Even with the prestigious organisations, the people producing the data are still just people and neither omnipotent nor always right. I’ve never yet had a chip manual that I haven’t had to scribble corrections into. There’s thus always the possibility that something you know to be true from the books just isn’t when you come to have to make it work. Most of the time, the old engineers have a work-around that makes the things run. Sometimes, the gap between what is supposed to happen and what does happen can be very interesting.

  3. p.g.sharrow February 26, 2014 at 9:32 pm

    @Simon; That manual problem seems to be wide spread. It maybe that the person that writes the manual is a writer and not the engineer that created the device. I try to first read the manual and then work the device and keep in mind that the manual may be incorrect. Lucky for us there is now an internet with some group that works with that device and comments on their findings and work through. Although you often have to find them, read through hundreds of comments to find the few that pertains to your needs. It is fairly rare that you have a, one of problem. pg

  4. p.g.sharrow February 26, 2014 at 9:43 pm

    90% of the people are fascinated by the activities of others. 9% of the people concern themselves with the reasons for those actions. The last 1% examine the ramifications of those actions.
    1% figure things out. 9% organize and control actions that the 90% carry out.
    In the New Age that 1% can communicate and work together and directly communicate to the 90% without going through the 9% that pursue the tollgate position of gathering “Wealth Unearned” by their control of information and money flow. Engineers need sound information to create sound solutions to mankind’s problems. We really don’t need to waste our efforts on solutions for bureaucratic aggrandizement based on faulty science created in the “Dark” by political correct savants. pg

  5. Simon Derricutt February 27, 2014 at 3:49 am

    pg – I generalise the “manual” problem to the idea that all books (and teachings) have errors, and we don’t know what the errors are until we’ve tested them in all situations. Since we can’t test them in all possible situations, we can’t be totally sure. Even then, there are those Black Swan events where something that’s been done a long time with one outcome might occasionally have a low-probability alternate outcome. Those low-probability outcomes in computers are actually getting to be higher probabilities as the devices get smaller and we’re dealing with fewer electrons to give us a valid logic level. A nuclear decay in the encapsulation or a stray cosmic ray can changed the expected signal and give a wrong answer.
    Those dice are rattling everywhere. As I’ve gotten older I’ve become less and less sure that what I know is always true.
    Your 9% can also add value to a project by organising the productive people better – they aren’t all parasitic. It’s all down to the individual again. Even where they cock up from our point of view, it’s likely that they are only following the book that they believe is correct. I think it’s only ever been a small percentage of people that set out to do things that they know are wrong, but they do have a large effect.
    The way our political systems are set up, only the people who really want to tell other people what to do will have the stamina to get to the position where they can do so. Such people also tend to be sure that they are absolutely right. Makes for interesting times.

  6. p.g.sharrow February 27, 2014 at 1:23 pm

    Those 9% are necessary, they just have a very high opinion of themselves and are generally way too greedy. Oh yes! and someone else is responsible for their failures.
    Bureaucrats are the worst. They are mentally incompetent by legal definition.
    “Neither they or their organization can be held responsible for the outcome of their actions.”
    I have had them order actions that I knew would fail and told them so. They insisted and as well as required that I guarantee the outcome. They were not pleased when I told them that I would only guarantee that it would fail. 🙂 They got a different boy for their project. 😆 pg

  7. Jason Calley March 5, 2014 at 4:59 am

    Hey p.g.sharrow, thanks for the old engineer story. So much of physics (and other branches of science as well) has become not reproducible; sometimes not reproducible because of expense, sometimes not reproducible because of the statistical nature of the results. Either way, anything not reproducible becomes an article of faith, not science. Some of my younger friends are mystified when I refuse to jump on the newest bandwagons of “dark matter” or “Higgs bosons”, or some newest twist on “the history of the first 1/1000000 of a second of the Big Bang.” I tell them, “wait a decade or two and see what happens to that idea…”

  8. p.g.sharrow March 5, 2014 at 7:53 am

    @Jason; Yes, at times the line between science and Si-Fi is hard to determine, kind of sounds like startrek techno-babble. At times I just look a theory such as “Big Bang” with an engineers’ logic and say “no way” the present universe can fit into a point singularity. Examination of the observations that lead to that theory indicate a string of sloppy logic, one built on the next. A wonderful Science – Fiction story. I grew up reading Si-Fi for entertainment, wonderful stuff, just not possible in the real universe. On the other hand, some Si-Fi is a logical extension of invention. In the mid 1950s a Si-Fi story described creation of a multi layer solid state device, 10 years before the first integrated circuit was created by Fairchild Camera, and 15 years before the first microprocessor. Today they are beginning to create devices that are true multi layer solid state. It remains to be seen if the function of the device, described in that story, can be duplicated in the next 50 years, maybe. pg

  9. p.g.sharrow March 6, 2014 at 9:59 pm

    Ever wonder about the quality of peer reviewed published scientific papers.
    Do scientific papers ever seem like unreadable gibberish to you? Well, sometimes they really are. Some 120 papers published in established scientific journals over the last few years have been found to be frauds, created by nothing more than an automated word generator that puts random, fancy-sounding words together in plausible sentence structures. As a result they have been pulled from the journals that originally published them.
    http://www.foxnews.com/science/2014/03/01/over-100-published-science-journal-articles-just-gibberish/
    Some professors said that pay rules that base professor salaries on the number of papers they publish may lead to fakes.“Most schools have merit raise systems of some kind, and a professor’s merit score is affected by his or her success in publishing scholarly papers,”
    A fatter paycheck may be the driver, not the science. pg

  10. M Simon July 5, 2014 at 11:30 pm

    p.g. You might want to look into the Forth language. It is interactive (fig model) and very handy for exploring how chips actually work:

    http://spacetimepro.blogspot.com/2014/04/lpc812-devl-20-march-2014.html

  11. p.g.sharrow July 6, 2014 at 7:11 am

    @M Simon; actually 1988 I worked with a guy that was trying to create a Forth kernel for the OS firmware in a 8086 machine. He wanted to write it on a Commador 64 and execute it in a 8086 based cash-register! My contribution was a RS232 port driven by the Commador and it’s power supply. We needed a +12 / -12 volt DC supply to drive the RS232 from the always overtaxed Commador power supply that had no 12 volt DC. There was a single 9 volt AC wire that was unused from the transformer, so I created a set of regulated capacitor tanks that yielded a clean center tapped 24 volt DC power supply for the RS232 port, that was hooked up to that 1 wire. Worked well enough. Don’t know if he finished his project. Writing code is one field that I have avoided, but Forth seemed to be a clean way to go to the machine code. pg

  12. Steve C July 9, 2014 at 9:56 am

    I love Forth, it’s one of the most elegant things out there. Back in the day, I wrote a FIG system for my Nascom 1 (2MHz Z80, “15+1” lines by 48 character screen, bring your own memory if you need more than 896 bytes). It’s *really* satisfying to see your own system message come up first time! I also recall reading, not long after, of (IIRC) the Novix 4000 processor, which had single byte instructions for all the core Forth words and a solidly stack-based architecture, but never found one, though I later heard that many ATMs had used them.

    I did, however, have an 8K Forth ROM, written by David Husband, in my little Sinclair ZX81, which genuinely multi-tasked and was about 100x faster than the Sinclair Basic – most impressive. There’s a splendid emulator called “EightyOne” if you want to play with it – it includes the David Husband ROM among many others. (It can also emulate pretty much any Sinclair /Timex machine you’ve ever heard of, along with the (native Forth) Jupiter Ace.)

    What the world needs is a PIC with native Forth. That would be seriously interesting, and a natural sort of environment for a compact and powerful system. If the interest ever turns into an actual need, well, there are already PICs around which amount to pretty much a complete Z80 chipset, plus memory, in one package – I know I could program that.

    Agree with your main thrust, though. I think you nailed the problem when you said that modern scientists are aiming to “prove” their theory. They’re using the word as in propositional logic – it’s a “can I make this work?” process where the end justifies the means, where the aim is to reproduce the desired result by fair means or (not necessarily intentionally) foul. As Popper and Feynman would agree, though, they should be looking for “the exception that proves (i.e. tests) the rule”, as a “can I break it?” exercise where they systematically exclude everything they can think of that might be responsible for that result. Throw in too many “consensuses”, and all those papers that seem to have been generated by machine, and science just doesn’t seem to have that reassuring feel of solid knowledge any more. Oh, well. They also told us we should check everything for ourselves …

  13. p.g.sharrow July 9, 2014 at 8:03 pm

    @Steve C; I have often considered the Forth concept of language to computer code writing the best way to go. IIRC much of industrial computer control was done in Forth, as it was the best language for machine control. simple and direct with low overhead. But as I am 67 years old and have way too much on my plate as it is. This is one craft That doesn’t appeal to me. Actually it didn’t appeal to me 25 years ago ether. I am more of a hardware guy. Just a mental defect I guess. 😉 I was born a sceptic and required proof of everything. Would even get into arguments with the Catholic School sisters about dogma that they were teaching! Just a little 8 year old brat that would not accept the pronouncements of the “Holy Father” about GOD’s intent! Scientific “Authorities” have no special standing with me. As I was nearly untrainable I was pretty much left to learn on my own. I find it interesting that many of my view points are becoming main stream “NEW” science. Just logic to me. The older I get the more fun I am having. Past, present and future are just a river that we ride, Enjoy the scenery and paddle at least enough for steerage. pg

  14. M Simon July 9, 2014 at 8:30 pm

    p.g.

    I’m a hardware/software guy. I have written a Forth for the Super 8. A now defunct bit of silicon. I have found no better tool for bringing up a system. Where the manual is unclear I can quickly try things and find out how the silicon actually works.

    In any case if you are doing ARM stuff we have the tools for you. If you want us to do something out of sequence (we have a plan) we can port the kernel to your favorite ARM for a modest price. Contact us.

    Steve C.

    Can we help you?

  15. M Simon July 11, 2014 at 9:59 pm

    We have a free Forth for the LPC812 and LPC1114 available here:
    http://spacetimepro.blogspot.com/2014/04/new-fish-lpc812-support.html
    Just in case you want to roll your own.

    A Forth for the STM3F5 is coming.

  16. p.g.sharrow July 29, 2014 at 8:25 pm

    @M Simon;
    I have often considered Forth as possible foundation for a firmware/software OS. Modern systems have sort of grow up from the needs to think for the lack of early capacity and now bloat because we can just throw more code on top of the old.
    It just appears to me that a rethink is needed that would operate everything from the thermostat to the A I of an interactive system.
    I would put up a post on this but have too little personal knowledge for the effort. pg

  17. M Simon July 29, 2014 at 10:50 pm

    p.g.sharrow July 29, 2014 at 8:25 pm

    Just did a post on that.

    http://spacetimepro.blogspot.com/2014/07/the-software-has-to-get-better.html

    You might also like Moore’s Law is running out of gas: https://www.youtube.com/watch?v=JpgV6rCn5-g

please say something interesting

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

%d bloggers like this: