It not just poor client design, however (and I can't really always call it poor design, because who here doesn't get peeved with a short battery life device?? Which is what low transmit power helps). We are really switching from a coverage based design to capacity based design. If we want people to be able to do more with wireless, it can't come from just a change in wireless PHY. It also has to come from increased density, which will in turn lead to less shared bandwidth and higher through-put.
I think the days of getting by with a coverage based deployment are coming to an end, and the days of planning for capacity is already here. The big question is how many of us have adjusted for this model? We still haven't. Ryan From: The EDUCAUSE Wireless Issues Constituent Group Listserv [mailto:WIRELESS-LAN@LISTSERV.EDUCAUSE.EDU] On Behalf Of Jeff Kell Sent: Thursday, December 12, 2013 6:30 PM To: WIRELESS-LAN@LISTSERV.EDUCAUSE.EDU Subject: Re: [WIRELESS-LAN] WiFi planning On 12/12/2013 5:11 PM, Ian McDonald wrote: It seems to me to be completely impractical from a planning and budgetary perspective to be increasing the density of AP's on an annual basis due to poor client design, whether low transmit power, antenna deficiency, or insufficiently well designed front-ends. If a device can't connect to the same wireless network, side by side with last year's device, then from my perspective, that's an issue with the device, not an infrastructure issue. Well, when most of us started wireless deployment, it was pretty optimistic to plan for a laptop per student / class seat / dorm bed, this was the same time we were doing ResNet plans with a "port per pillow" -- a plan which game consoles initially wrecked, now followed by BluRays and Smart TVs and femtocells and who-knows-what-else. And now for wireless, it's certainly not just laptops (we have more registered/identified BYODs than computers now). Wireless devices continue to explode... its not last year's device that can't necessarily communicate, it's the 3-4 extras today over the original device that cause the issues. If you designed for 2.4G power/distance back when 2.4G was in vogue, and 5G was either ancient (11a) or new again (11n), it wasn't necessarily a design goal, and 5G doesn't tolerate walls, etc as well. Not sure about 11ac, but 11ad at even higher frequencies will penetrate even less. So yeah, if we had to do it over again AND knew what we know today... sure. How many deployed 11a/b/g over 100Mb ports? And out of those, how many were Cat6/6A? Regretting any of those decisions yet? Just give it time :) Things evolve. I'd agree they should last longer than "last year" but things change *fast* in this business :) Jeff ********** Participation and subscription information for this EDUCAUSE Constituent Group discussion list can be found at http://www.educause.edu/groups/. ********** Participation and subscription information for this EDUCAUSE Constituent Group discussion list can be found at http://www.educause.edu/groups/.
