Re: Epistemology 201: The Science of Science

From: Albert (albertwagner_at_cox.net)
Date: 02/21/05 

Date: Mon, 21 Feb 2005 09:39:58 -0600

Richard Herring wrote:
> In message <2_oRd.8212$zs.7245@okepread04>, Albert
> <albertwagner@cox.net> writes
>
>> Richard Herring wrote:
>>
>>> In message <lDnRd.8204$zs.5379@okepread04>, Albert
>>> <albertwagner@cox.net> writes
>>>
>>>> Richard Herring wrote:
>>>>
>>>>> In message <1F4Rd.6885$zs.221@okepread04>, Albert
>>>>> <albertwagner@cox.net> writes
>>>>> I didn't have time to address this yesterday.
>>>>>
>>>>>>
>>>>>> He also knew of the prior art, which was conveniently omitted from
>>>>>> the patent application.
>>>>>>
>>>>>> I know you have read this:
>>>>>>
>>>>>> http://www.bellsystemmemorial.com/belllabs_transistor1.html
>>>>>
>>>>>
>>>>> <snip>
>>>>> Yes, people were making crystal oscillators and even amplifiers
>>>>> long before the Bell Labs transistor. But a point you may have
>>>>> missed is that they were all diodes: *two*-terminal devices.
>>>>
>>>>
>>>> No. You apparently *you* missed the two whisker devices with one
>>>> whisker attached to a battery:
>>>>
>>> No, I saw that, too.
>>> It was common to have two (or even three) whiskers available,
>>> because detectors made from natural crystals were unreliable.
>>
>>
>> The second whisker, the one attached to a battery, served a different
>> function than simple redundancy.
>
>
> Not in the way you might be thinking. The whisker with bias battery was
> a sensitive detector for long ranges; the one without bias was used in
> harbour as a deliberately desensitised detector to avoid burnout by
> strong signals.

Yes. That statement is from the quoted text. Of course I read
it. It makes very clear that the bias was used to enhance
amplification.

>
>>
>>> Two diodes don't make a triode, as many later hobbyists who tried to
>>> make their own point-contact transistors with cat's-whisker
>>> technology have found..
>>
>>
>> Irrelevant.
>
>
> No, totally relevant. If you don't understand the difference between
> diode and triode there's no point in continuing this. It also shows how
> difficult it would be to discover transistor action by accident.

The Navy trained me in electronics, so yes I understand the
difference. Your attempts at 'appeal to authority' fail.
>
>>
>>> And carborundum crystals had to be biased by a direct current before
>>> they would work at all.

Imagine that. Of course, that too was in the quoted text.
>>
>>
>> Just like a transistor.
>
>
> Not in the least "just like", except in the trivial sense that both use
> a battery.

And both had three leads. And both amplified a weak signal. And
both depended on known properties of semiconductors.

>
>>> The presence of a battery doesn't imply amplification.
>>
>>
>> But there was, in fact, amplification. A ten-fold increase.
>
>
> But you've presented no evidence that that amplification was by one
> current modulating another,

LOL. How else might amplification take place?

> rather than the "negative resistance" diode
> amplification which was demonstrated by Eccles and others around that
> time. And negative resistance wasn't a new feature, nor one confined to
> semiconductors. Early spark transmitters relied on the nonlinearity of a
> carbon arc to produce oscillation.

Strawman.
>
> In fact, there's not a lot of hard *evidence* even for that alleged
> ten-fold gain. An oral reference to an unnamed "he" who (60-odd years
> earlier) had an unnamed magazine which had a translation of a Russian
> paper, author unknown. I'd be happier if we could see what the Russian
> paper really claimed.

"Hard evidence" should be easily obtainable by repeating what he
did. The rest of your statement is another strawman.
>
>>
>>>>> But it's very hard to get controlled amplification with such devices.
>>>>
>>>>
>>>> So? That would still leave the Bell transistor simply a refinement,
>>>> and not a new device.
>>>
>>> A *very very major* refinement, definitely novel in the patentese
>>> sense.
>>
>>
>> Legal matters are irrelevant.
>
>
> Also a very very major refinement in the physics sense: it allowed one
> current to control the flow of another, without itself being affected by
> the controlled current.
>
> Also a very very major refinement in the engineering sense: it allowed
> the construction of practical amplifiers which could be cascaded to give
> practically unlimited gain, and switches which could process digital
> information..

Your opinion. I disagree.

>
>>
>>>>> The transistor's third terminal, controlling the current between
>>>>> the other two, separating input from output, is what allows the
>>>>> easy construction of practical amplifiers, and none of your "prior
>>>>> art" did
>>>>
>>>>
>>>> Please re-read the article. Even assuming you were correct you are
>>>> still left with the transistor being only a refinement of earlier
>>>> devices and *not* a new 'invention.'
>>>
>>> It's a matter of degree. I say that being able to separate the input
>>> and output of an amplifying device is not just a refinement but a
>>> major and novel step change. It facilitates the design of entire
>>> classes of novel amplifying and oscillating circuits that could not
>>> be made with a two-terminal device. Three legs good, two legs bad.
>>> You evidently disagree.

No. You can't count.

>>>
>>>> The article also places the use of semiconductors and solid state
>>>> electronics well before the phenomena could be explained by QM.
>>>>
>>> *Some* very limited use, yes. But there's no evidence that anybody
>>> made an amplifying triode, and nobody managed with any of these
>>> earlier inventions to made systematic improvements to the point
>>> where something could be mass-produced (were any of these earlier
>>> devices patented?)
>>> Note that I'm not saying that it's *impossible* for someone to have
>>> developed these prior discoveries into a working transistor with no
>>> knowledge of QM. Maybe if you took someone with the creative genius
>>> of Tesla to think of the possibility that a third electrode would be
>>> useful (and there's no reason to think it would be without some QM
>>> input), allied with Edison's dogged approach to testing every
>>> possible combination to find the one that worked, it might have been
>>> done.
>>> The fact remains that it was not.
>>
>>
>> And the fact also remains that I was correct in stating that
>> semiconductors were discovered and used without any need for
>> explanatory input from QM;
>
>
> And if that's all you'd stated I wouldn't argue.
>
> But in <vNnQd.6740$zs.2469@okepread04> you stated that the invention of
> *transistors* was not dependent on QM.
>
>> And that Bell, essentially, re-invented the transistor
>
>
> But you haven't shown any evidence of the prior invention of the
> *transistor*, specifically a *three*-terminal semiconductor device in
> which one current (or voltage) controls the flow of another without
> itself being affected by the controlled current.

You can't count.
>
>> without any need for explanatory input from QM.
>
>
> That can't be ascertained, since the Bell team undoubtedly had the
> knowledge of QM.

They also had knowledge of prior art, which I already pointed out.
>
>> The transistor would have been invented with or without any input from
>> QM, using only prior experimental evidence.
>
>
> whose existence you haven't established.

Yes, I have. You just ignore it.
>
>>
>> I would remind you that this subthread grew out of Bob Kolker's
>> categorical statement that the computer on my desk would not exist
>> were it not for QM.
>
>
> I would remind you that I am not Bob Kolker.

Then why did you *** in?
>
>> That is proved false by historical evidence.
>>
> Your opinion.
>

Which is as good, or better, than your opinion, in that, I have
produced considerably more historical evidence than have you. 

-- 
"Mercifully free of the ravages of intelligence"
	-- Time Bandits
Quantcast