Searching \ for '[OT] Yet Another Piclist Suggestion: Marking Posts' in subject line. ()
Make payments with PayPal - it's fast, free and secure! Help us get a faster server
FAQ page: piclist.com/techref/microchip/devices.htm?key=pic
Search entire site for: 'Yet Another Piclist Suggestion: Marking Posts'.

Exact match. Not showing close matches.
PICList Thread
'[OT] Yet Another Piclist Suggestion: Marking Posts'
1999\03\27@154909 by Bob Drzyzgula

flavicon
face
On Sat, Mar 27, 1999 at 11:08:42AM -0800, Gerhard Fiedler wrote:
> At 09:39 03/27/99 -0600, Eric Oliver wrote:
> >Alas, often my questions went unanswered.
>
> it would be interesting to hear from ex-members who have been considered
> valuable, contributing members by a majority as to why they left. probably
> some members have still some kind of contact with these guys (and gals?),
> so maybe someone could ask such a person for a helpful statement?

Myke Predko is one person who was mentioned as having
a much-lowered profile on this list over the past few
months. But I know that Myke has been writing another book
over the past few months and has generally been swamped,
so perhaps we should be careful with what we read into
changes in the posting patterns of others; there are myriad
reasons for a person to drift away.

As others have suggested, there are two ways to raise
a raw signal/noise ratio: lower the noise and raise the
signal. If you are designing a system that will operate
in an enclosed, controlled environment, it is generally
practical to work on the denominator. One can shield an
enclosure, shorten leads, use higher-quality parts and
generally tighten up the specs in various ways.

But if you are trying to ship a signal through an
environment over which you have no control, then there
really isn't all that much you can do about that pesky
denominator. In such a case, you pretty much have one
choice: Use a signal that is markedly different than the
noise in some detectable way. One way is of course to give
the signal a substantially higher intensity.  But another
means toward the same end is to give a signal a clearly
identifiable pattern that will carry though the the noisy
environment and still be detectable at the other end;
basically the idea being to pre-condition the signal
so as to target the sweet spot in a realizable filter.

So in thinking about what to do with the piclist (if
anything indeed "needs" to be done), one probably needs
to consider the following questions:

1. Is the noise level in the transmission medium
   controllable?

2. Is there any way to reliably modify the character
   of the noise so that it will be distinguishable from
   the signal?

3. Is there any way to reliably modify the character
   of the signal so that it will be distinguishable from
   the noise?

On question 1, my personal opinion is that it is not, at
least not without great and possibly unacceptable cost.
Sadly, it seems that this is where people have placed
the greatest effort. There have been no end of messages
on this list attempting to shame people into lowering the
noise level, but this does not appeared to have worked.
I realize that this is somewhat different than what I've
said before, but I plead guilty to thinking about it
a little more.

On question 2, I think that some of this can be done,
and the [OT] designation in the subject line is a step
in this direction that is genuinely useful. However, there
is still a great deal of traffic that (a) is interesting
to many if not most subscribers but marked [OT], and (b)
uninteresting to most subscribers but not marked [OT].
In the end, attempting to insist that all noise be so
marked is probably a losing approach.

On question 3, I think that this is an area of great
promise that has, as far as I can tell, largely been
left unexplored. Some have suggested creating a new
list for this purpose, but the inertial forces against
this are overwhelming, and in the end I believe that
either the elite list will wither away from disuse
or the entire body of piclist discussion will migrate
there.

So here's another suggestion: If you post a clearly
on-topic message to piclist, mark it [PIC], or possibly
[ON TOPIC] (or maybe [ON PIC]?). You have to write a
subject line anyway, it isn't that hard. Certainly there
will be a much greater resistance and stigma attached to
marking noise as signal than there is to marking signal
noise. People mark fascinating topics [OT] only because
they aren't sure of the relevance. But I find it much
less likely that someone would have the gall to mark as
on-topic a question on "k vs K" or "what means american".

It is fairly easy to argue that this shouldn't have to
be -- that the people posting the list's intended content
shouldn't have to go to this trouble. But I would say
that it really is not that much trouble, really, and
it *has* to be better than reading anti-OT whining all
the time.

FWIW.

--Bob

--
============================================================
Bob Drzyzgula                             It's not a problem
spam_OUTbobTakeThisOuTspamdrzyzgula.org                until something bad happens
============================================================

1999\03\27@162040 by Gerhard Fiedler

picon face
At 15:46 03/27/99 -0500, Bob Drzyzgula wrote:
>So here's another suggestion: If you post a clearly
>on-topic message to piclist, mark it [PIC], or possibly

this in effect could achieve everything the suggesters of a second list
wanted to have, without the drawbacks. whoever wants a really restricted
amount of messages, sets a filter to delete everything that has a subject
not containing [PIC]... which probably would include a number of pic
related questions: the ones which would've been posted to the wrong list if
there were two lists :)

this is a pretty democratic issue: if people start to use it, one can start
to use a filter. i'd suggest [PIC] (among others because it's short and
obvious :).

ge

More... (looser matching)
- Last day of these posts
- In 1999 , 2000 only
- Today
- New search...