On Thu, 22 Sep 2005 14:09:27 GMT, "Steve Berry" <reachnet@hotmail.com>
wrote:
>Yes - but why is the "Bull in a China Shop" approach - bombarding users with
>tech information you're not sure they'll even understand beter
>than taking it slower and trying to ascertain what level of knowledge is
>actually going to be valuable to them ?
Because that's the way they learn. Or rather, that's the way I learn
and expect others to do the same. It is impossible to determine the
experience level of the person asking a question on usenet or in a
mailing list. I therefore tend to supply answers and information well
over most peoples experience level. Most beginners will pickup about
10% of my ranting. That's sufficient for them to do some reading,
thinking, and research. I try to supply complete, stand alone,
detailed answers that if Googled, will supply enough information to
solve whatever problem is at hand. Very few people supply enough
information to actually solve their problem, so the research aspect is
almost a requirement. How "valuable" is my information largely
depends on the subsequent research.
What would you have preferred I posted? A one line opinion of she
should do consisting of "I think WDS sucks. Get a better antenna"?
Is that what you're suggesting?
>If you notice, the poster was a single mum - I doubt she's ever heard of
>WDS, so what would be the point in hitting that user with info
>about Omin-directional antennas if they don't understand what the **** that
>is ?
Single Mom's are capable of learning. I don't insult anyone's
intelligence by suggesting otherwise. Incidentally, I will confess
that I often write as if I'm lecturing to an audience instead of
supplying a personalized answer. If you've been in a position of
answering questions from an audience, you'll find that directing the
answer to the entire audience, instead of asking them to listen to a
personal conversation, is far more effective.
>The "tone" of your email was "here's reasons not to use WDS" - negative but
>factual - not here's what you could do with WDS, which is slightly more
>optimistic.
You should read some of my other postings. The "tone" is more
insulting, humiliating, brutal, arrogant, and obnoxious than my
previous answer. Want tech answers without supplying enough info?
Fine, be prepared for my wrath.
In this case, no "tone" is required to understand that I don't like
WDS and mesh networks and that methinks a better antenna or location
would be better. I supply the information. The person asking the
question can supply the conclusions. If she wants to use WDS despite
the itemized limitations, then at least she has an understanding of
what might be involved, and what to look for when things go wrong.
Note that I did not address the answer to you, but to the original
question. If I had addressed it directly to your one line comment, I
would have started with something like "How do you know that she can
use WDS when you don't have a clue as to what equipment she currently
owns".
>> That's true. I didn't list the advantages of WDS because I thought
>> you would supply the necessary detail. After all, it is your
>> suggestion.
>
>I didn't specifically for the reason staed above.
I'm disappointed. Confuse them with facts? Want me to list a few
advantages to WDS? I don't mind.
>Ok - that's valuable info thanks - that much I didn't know.
Would you have preferred a non-confusing, non-detailed, one line
pontification instead? I have no clue as to your experience level,
therefore I should aim low and presume minimal competence? Don't you
think that the detail, despite some of it going over your head, might
be a bit more useful than a one line conclusion?
>I bow to your superior knowledge.
In the colonies, we don't have royalty and don't do that.
>> Any benchmarks?
>No.
I think you'll be surprised. A single WDS repeater actually works
fairly well if the connecting laptops can only see the repeater and
not the main access point. However, if they are setup in a way where
everyone sees everyone else, collisions will become epidemic. Even if
you turn on CTS/RTS flow control on the access points, there will
still be collisions because the access points and clients transmit
asynchronously.
A really simple clue can be obtained with just ping. I suggest using
"fping" instead of Microsoft ping. Just ping the DSL gateway IP from
the client machine going through a WDS repeater. What you'll see is
that the normal ping times vary radically depending on how many
retransmissions are required to get the packet through. With WDS, a
lost packet requires two retransmissions so the effect will be very
obvious.
Most cheapo IP access points hide these resends. There's no internal
diagnostic commands that will show multiple transmissions. However,
the delays will be evident. If there were no collisions or resends,
then the ping times would all be constant and fairly low (about 30msec
for 1.5Mbits/sec). Also try the same test without the WDS repeater.
It might still vary, but probably not as radically as with a WDS
repeater.
>> Flakey wireless connectivity is the
>> surest way to screw up reliability.
>True.
You might want to remember that when offering advice. In general, an
RF solution or improvement will offer a more stable and reliable
connection than adding a repeater. I also commonly suggest adding a
2nd access point via a CAT5 cable to areas where coverage is lacking.
Separating the modem/router from the wireless access point is also a
good idea because it allows the wireless access point to be elevated
from its traditional location under a desk to a much less RF
obstructed location on perhaps a bookshelf. Can't do that will a
tangle of CAT5 wires going to the switch ports. Lots of other options
are available before going to a WDS repeater.
>> Any more questions or comments?
>Nope.
Fine. For homework, chew on this for a while and write a few
paragraphs on the advantages of WDS. Quiz in about a week.
--
Jeff Liebermann
jeffl@comix.santa-cruz.ca.us
150 Felker St #D
http://www.LearnByDestroying.com
Santa Cruz CA 95060
http://802.11junk.com
Skype: JeffLiebermann AE6KS 831-336-2558
Extending a wireless network 
Linear Mode
