Notes from the Bell Mobility/ESRI Location-Based Services Seminar, Part 2

Here's part 2 of my notes from the Bell Mobility/ESRI Location-Based Services Seminar. If you missed part 1, here it is.
The LBS Partner Programs"I need your help," began Bell Mobility's David Cheung during the "LBS Partner Programs" segment of the presentation. "I'm not here to be a hindrance." He pointed out that you don't have to be a Bell Mobility partner to get started. There are white papers and an SDK that anyone can download, and they can provide access to a sandbox with dummy phone numbers that your location-aware applications can locate.He then covered the three levels of partnership: Bronze, Gold and Business. I have no idea why Bell didn't go with the perfectly well-established convention of "bronze, silver, gold".
Bronze
Essentially a "sanity check"
The criteria for qualification is simple: can you build a location-aware application or not?
Your application gets featured at businessonthego.ca
Gold
At this level, you have an application that's ready for the market
You're trying to prove that you can launch, using 5 to 50 mobile devices as your testing ground
Qualification includes passing a tech assessment questionnaire
Business
At this level, there's a contractual obligation
Bell does a business assessment of your company
In return, you have access to Bell's national sales team
The types of partnership can vary -- it could be a marketing alliance or a revenue-sharing program. This seems to be determined on a case-by-case basis.
Location ChargesThere is a charge associated with every "locate" -- that is, every time a phone is pinged for its location. The charges can be applied in one of two ways:
A "one-shot" locate costs 5 cents
Businesses can buy locates in packages, as shown in the table below
"Locates" per month
Approximate frequency
Package cost
20
Once every 1.5 days(in the presentation, they say "once every day")
$5
200
Once every 3.5 hours(in the presentation, they say "once every hour")
$10
500
Once every 1.5 hours(in the presentation, they say "once every half-hour")
$15
1000
Once every 45 minutes(in the presentation, they say "once every 15 minutes")
$20
2500
Once every 15 minutes(in the presentation, they say "once every 5 minutes")
$30
Note that these are "business" rates. They didn't show any rates for "consumers". When asked how the system differentiates between a "business" and "consumer" customer, the presenter replied that every application has a "feature code" attached to it which makes the differentiation possible.
Locates are billed directly to the person being located. Someone asked if there would be a way for the application developer to pay for locates, and the response was that it would probably have to be worked out on a case-by-case basis with Bell.There are volume discounts for locates -- they range from 2% to 30%.
Technical Considerations (and Annoyed Developers)
Ping Throttling and Caching
It's possible to ping a phone once every 35 seconds, but currently, pinging a phone more than once every 5 minutes tend to increase the number of location errors. Apparently, this is due to the limitations of the current hardware. Hence, pings are throttled.
Someone asked: "What if say 40 different apps are trying to locate the same device? Do you send 40 pings?" A Bell systems guy in the audience said that some caching was used to avoid this sort of problem, but didn't get into specifics.
Per-Locate Billing Directly to the User: Not Popular
Another guy in the audience balked at the 5 cents-a-locate rate, saying "even at 1 cent a locate, it's too pricey" and that he'd never get customers to use his service at that rate. A per-use charge billed directly to the user, he argued, would make it impossible for him to create a "consumer model". He pointed out that his phone from a rival telco, Telus, did locates for free, and the presenter pointed out that it could locate itself for free -- there would be a charge involved if an application not based within the phone had to get the location.
A handful of developers in the room echoed his concerns about the rates, also thinking that they'd discourage customers from using their applications. From the sound of it, they seemed to be writing applications targeting "consumers" rather than "businesses", and their annoyance was apparent.
"Customers are paying for location-based services right now," said one of the presenters, in an attempt to assure us that customers wouldn't balk at being billed for every locate.
Too Big a Window?
Someone else pointed out that a maximum locate rate of once every 35 seconds was too big a "window"; he was hoping to do locates once every 5 seconds. The presenter pointed out that this was beyong the current network's capabilities for reasons that would be explained in the next session, and that it would probably be a very big drain on a mobile phone's batteries. "Locates consume power," he reminded the audience, "and you have to take power consumption into consideration."
The presenter said that LBS was for making applications that were location-aware and not primarily for monitoring, navigation or geocoding -- in spite of the fact that the examples of LBS in action mentioned less than an hour prior were just that. "You could use it to track a missing child," he said; that sounds like a "monitoring" app to me. Perhaps what he should've said was that LBS wasn't for mission-critical monitoring, navigation or geocoding, but I'd be hard-pressed to convince a parent that locating a missing child wasn't mission-critical.
The PATRIOT Act Rears its Ugly Head
Someone pointed out that although developers working with ESRI would be signing agreements with a Canadian branch of a company, ESRI's servers are based in the U.S., meaning that ESRI-hosted information would be accessible to the U.S. government if they invoked search-and-seizure under the PATRIOT act. That's something to consider when dealing with the privacy issues of your application. Print Article
-->
Posted to:
Mobile
Comments
Post a comment