IETF attendees reengineer their hotel's Wi-Fi network
networkworld.com
networkworld.com
If you are the organizer of a tech conference, please, communicate that you need special network considerations and get in contact with the hotel's provider, or arrange to have your own provider brought in as soon as possible. 3 months lead time is cutting it close.
http://www.networkworld.com/cgi-bin/mailto/x.cgi?pagetosend=...
https://addons.mozilla.org/en-US/firefox/addon/autopager/ https://chrome.google.com/webstore/detail/mmgagnmbebdebebbcl...
Less is more in some circumstances.
And later they increased interference, in order to add a 4th channel.
It's all about tradeoffs...that's why they pay engineers the (market-competitive) bucks.
(pretty sure Cisco has an equivalent but I forget its name...)
Postscript: They talk about HP buying Colubris, but don't think the Colubris-legacy equipment is current-HP stuff; they went and bought 3Com in 2010, and I think that's where most of their current offerings come from.
But you will never be able to completely remove the human from the equation. A proper wireless site survey with someone walking around and taking readings pre/post AP installation is really the best way to do things. I'm not really a WLAN guy but I work with some really bright people who are. There is a lot of depth to this problem that we should not expect to be solved anytime soon by AI.
If you're interested in some of the theory about applying AI to RF determination check out the work of Tim Brown http://ecee.colorado.edu/~timxb/
1. Reset Router.
Hardly rocket science...
The shitty crap $20 routers I've seen where plugging it out and in again makes them work again for about 40 seconds make me wish there was a button on them that'd give everyone in the company that pretends to have made them an electric shock.
Dear everyone - you know not to buy anything that's sold through spam emails by now. How about not buying a router that, when you plug it in, makes a faint fizzing sound, costs less than what you think it should and isn't a good brand? No, Belkin is a level of shit yet to be reached by the crap factories of Turdtown. If I had a dollar for every F5D7230 that I wanted to fold in half... just don't trust the reviews. Don't get that Netgear, their firmware is utter crap. Oh, you want bridging mode? You'll need to buy the v8, we've enabled it in the firmware for that. The v7 won't cut it, oh no. WGR614 was the name for that abortion of a craptardation. Just don't buy that crap. If, for a fucking router, they need to do ten revisions of the hardware WGR614v1 to WGR614v10, to get their supplies they need and their shit together, don't buy. Cisco, Linksys, Billion, TP-LINK, Fritz. That's not a whitelist though. Even they have produced utterly worthless things that manage to give LEDs their 5 Volts and whatnot milli-Ampere and give you a nice UI to manage a Wi-Fi connection that just doesn't work.
You want a router for grown-ups? Get the WRT54GL. Ten year and then some track record. Put your own damn firmware on it. How do you know it's for adults? The Amazon page says WARNING CHOKING HAZARD -- Small parts. Not for children under 3 yrs. That warning ain't on D-Link.
The time I have wasted on these pieces of crap you wouldn't believe. Thank you Hacker News, for giving me a place to vent.
You want a simple page you can open in your browser? Noooooo way, that's way too industry standard. Rather have something where you need to keep your configuration software up to date too! It even doesn't work on Linux! What a feature!
It's not as bad as the ICANN meetings, which were deliberately held in obscure/expensive places around the world just to keep people from attending, in the guise of "being globally representative". The funny/sad thing is that this isn't one of the worst decisions ICANN has made.
I'm not especially defending the choice of Paris. Brussels or London or Berlin would have been at least as good. Sometimes it's nice to hold engineering meetings elsewhere than in the Valley.
So, more like lumps of coal.
In wifi networks, you can significantly reduce on-air collisions with other packets if you can send at a higher rate. A packet being sent at 2Mbps spends 1/2 the time on the wireless channel as compared to a one that is sent at 1Mbps. Therefore halving the probability that this packet will collide with some other transmission.