Message-ID: <41012684.5030501@g3ysx.org.uk>
Date: Fri, 23 Jul 2004 15:53:56 +0100
From: Stewart Bryant <stewart@g3ysx.org.uk>
User-Agent: Mozilla/5.0 (Windows; U; Windows NT 5.0; en-US; rv:1.4) Gecko/20030624 Netscape/7.1 (ax)
MIME-Version: 1.0
To: rsgb_lf_group@blacksheep.org
References: <01C470B0.228BAA70.actalbot@southsurf.com>
In-Reply-To: <01C470B0.228BAA70.actalbot@southsurf.com>
Subject: Re: LF: Re: Loops v Verticals
Content-Type: text/plain; charset=us-ascii; format=flowed
Content-Transfer-Encoding: 8bit
Sender: owner-rsgb_lf_group@blacksheep.org
Precedence: bulk
Reply-To: rsgb_lf_group@blacksheep.org


Andy wrote:

>Well...
>That may not always be the case.
>
>If experimentation disagrees with theory you need to check that everything is 
>accounted for, before assuming the theory is wrong and reworking it.
>
>It is too easy to overlook some seemingly trivial little thing or ignore the 
>seemingly obvious.
>
>Like baluns on antenna measurements - just as an example.
>
Andy

You are correct to point out that the experiment must be conducted 
competently and be
reproducable before modifying the theory.

However anomolous results can also be of interest. For example, it is 
not uncommon for
researchers to sweep minor, irritating, results under the table. I 
recall as an
undergraduate being taught quantum mechanics by Jocelyn Bell, and I 
remember it
being impressed on us how easily she might have dismissed the anomalous
LGM traces as being EMC from a passing police car - certainly that would 
have
been a view readily accepted by the rest of here department. As you 
probably
know she noticed that the "police car" got 4 mins earler each day* and 
persued
the results to discover Pulsars.

* A cynic might say that since the probability of police being early is 
absurd, one
could conclude that it was Little Green Men without further investigation :)

>I've often found that theory based on back of envelope calculations, made from 
>first principles, has this annoying habit of often being surprisingly accurate. 
>  
>
None the less it is always wise to challenge and probe conventional 
wisdom, and
although the instances of it being wrong are rare, I have personal 
experience (in
computer networking) where ignoring such wisdom yeilded important results.

73

Stewart G3YSX


>Andy  'JNT
>
>
>-----Original Message-----
>From:	Stewart Bryant [SMTP:stewart@g3ysx.org.uk]
>Sent:	2004/07/23 09:12
>To:	rsgb_lf_group@blacksheep.org
>Subject:	Re: LF: Re: Loops v Verticals
>
>
>
>Quite right. Experimental evidence is the only way to resolve the issue.
>
>If the experiment agrees with theory, then that is good. If it does not,
>then it is the theory that needs to be extended or abandoned.
>
>- Stewart G3YSX
>
>
>
>
>  
>