offtopic question about high density wifi

[david at lang.hm]

On Tue, 26 Jan 2010, Martin Langhoff wrote:

> On Tue, Jan 26, 2010 at 4:26 PM,  <david at lang.hm> wrote:
>> Ok, having gone back and re-reading this, I'm confused. This seems to be
>> exactly the RF contention issues that I understand and am prepared
>
> Yes, RF contention, but what Scott is pointing out is that due to
> various implementation glitches, it's a lot worse than any theoretical
> anything.
>
>> you seem to be telling me that I don't need to worry about RF contention,
>
> I think he's saying "be pessimistic", dodgy wireless drivers, stupidly
> implemented programs will spew crap in the RF and mess things up :-)
>
> See the mention of a bogus ad-hoc network in the 2007 report. I've
> seen quite a few other mishaps where a single misbehaving node messes
> it all up.

yeah, two years ago when I attened this event I found that just about 
every other vendor booth was running their own AP as well. I mentioned it 
at the time and got a 'what can we do' response. This year I plan to do 
some patrolling to spot such things and try and convince them to turn them 
off, but I have to have a usable system to do so.

>> As these articles say (and I was already planning), I want to have the
>> access points with power turned down so that I can fit more in a given
>> area without them overlapping on the same frequency.
>
> Yep, that's the other good point.
>
> I am with the Pycon guys that the all the multi-antenna enterprisey
> APs that I've seen don't deliver. Lots of carefully configured
> mid-range APs do better.

that's my impression as well. The enterprisey things may work well for a 
hotel or office where you have a lot of area to cover and only a few 
people really using it, but they just don't work when you have a lot of 
geeks in a tight area.

David Lang