On Thursday 19 July 2007 10:18:32 pm Dave Bour wrote: > Today, the perception is "first to market wins" regardless of how > complete it is. And there's many an example to cite the business case > of doing so. The best solution doesn't always win and the challenge is > to find that balance. I'll agree, that the "bad guys" are the numbers > they expect, and willing to bet that half the "IP theft" is internal > based anyhow.
First to market doesn't mean it has to be closed file formats and no documentation. Poor documentation sure, but none? Honestly I think it's a little like the HitchHiker's Guide to the Galaxy editorial lawyers who cut out anything remotely useful and then send it out. :-) > True enough. For me, the simple algorithm though of max power wins..if > full, fall back to next base station...leaves something to be desired. > Using today's conference room example...a base station in each > corner...every device in that room would likely see it. The algorthm > could be something like... > 1. anything above a min threshold...ie..I think they said -90db or > something was their min power needed ...so take half that...Any base > station showing power of at least half...find the lowest loaded of > it...since they use the oom software anyhow..it makes the decision...not > the base stations...let it load them accordingly... > Then walk-ins, etc have less chance of switching etc...and nobody should > get a "overflow" and unavailable. Extend that situation to a warehouse > with 20-30% overlap...it becomes even more viable since workers tend to > concentrate in locations... The conference room though with 5 RFPs would still allocate more or less evenly. And there wouldn't be any overflows in the original situation, as any RFP at capacity would not be eligible, even if it had the strongest link quality of all of them. I guess there is a bit of an advantage for splitting the load, but with the handsets moving around the chances of it having to use a transfer channel are likely high anyway, so your careful attempt to balance load would be wasted. > I missed the sip reinvite ..guess that was session 1. To me, sip > reinvite is a moot point as it's only merit comes in offloading heavily > loaded servers...and in today's environment, that 512 phone capacity is > well short of a modern server anyhow. ... or heavily loaded RFPs who are doing useless transfers and wasting bandwidth. :-) > SNMP and the whole host of reporting...way short on that ... Haven't > even begun to think about what I'd like to see here but at the minimum, > dropped calls, call quality, handoffs, rfp loading... > Then items like visible phones from each rfp (with signal > strength)...tie that into some mapping program...get a 3d picture of > where a phone is located... It will be interesting to see how well you can pinpoint someone with two or three rfps over a number of floors. :-) > I guess I'd like to see something like the AGPS on most cell phones now > for emergency locating. I actually had the experience a year ago to use > it...guy near drowning up at cottage on one of those kite surf boards. > Did a 911 call, get put through to OPP marine unit but got disconnected > due to loss of signal. Wasn't 5 minutes and 5 OPP as well as a couple > of the local reserve police showed up parked behind my vehicle where the > phone was located. Marine unit did their job but was surprised that > they pinpointed the phone's location as well as they did. Wow, and so quickly too... did the guy make it? -A. --------------------------------------------------------------------- To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
