Ian Jackson wrote in
:

....
As a Windom is only likely to be used on the HF bands (and probably
mainly on the lower HF bands), I would have thought that a few dB of
attenuation would have negligible effect on the SNR.

Ian,

Ah, the statements in support of mediocrity about.

Yes, there is an element of truth in what you say... but looking at it a
little further....

On any bands where the external noise is very much greater than the
receiver internal noise, small extra antenna loss will not markedly
degrade S/N ratio on receive.

One of the issues when the feedline participates as an antenna condcutor
is that its proximity to noise sources (eg the power wiring of a
building) may lead to higher receive noise without increasing signal
strength, so feedline participation can degrade S/N by that mechanism.
You won't see that called out in Carolina Windom adds which claim
feedline participation as an advantage.

On the tx side, loss in the antenna system reduces EIRP, and so directly
degrades S/N at the other end. So, even on those bands where external
noise is very much greater than receiver internal noise, pay attention to
antenna efficiency to achieve optimal radiated power to be heard at best
S/N.

One of the bogus arguments often trotted out is that efficiency is less
important with QRP since there is less power to damage a lossy component.
Another view is that if you start out with a transmitter that is 10dB or
so behind the average transmitter power being used, then why exacerbate
the situation when an even lossier antenna system.

For a once technically based hobby, we do think up some phony rationale.

It is a personal judgement about whether 3dB (mentioned by one poster) or
any other number is acceptable in the compromise that is made with all
practical antenna systems.

Owen