Seamful design and cell phone reception bars

Different approaches have been developed under the "seamful design" term. Chalmers, McColl and Bell indeed proposes to reveal seams and technology limites to empower users. In a paper from Eurowearable in 2005, they give an example: "By revealing such seams, users can better understand when and where to use digital resources such as network connectivity—and when not to—as they go about their work and use our systems in their ways". A common example is the one of cell phone reception bars that allows people to adjust their behavior (one bar = SMS, 3-4 bars = voice communication, 1-2 bars = assumptions that the communication quality would be bad).

Reception bars

But what does those reception bars actually mean? I cannot remember how I ran across this Metafilter discussion about "this topic. Some excerpts:

"They don't mean much of anything, it turns out.

I don't know what they're displaying for GSM, but probably what they're displaying is the signal strength. For CDMA (which is what I know about) that's what they display, but in CDMA the signal strength is highly deceptive because it doesn't inform you of what the noise floor is.

The technical term is "EC/I0" (pronounced "ee-see-over-eye-naught") and it refers to the amount of the signal which is usable. In CDMA you can have strong signal (4 bars) and lousy EC/I0 and not be able to carry a call, and you can have low signal (zero bars) and excellent EC/I0 and carry a call fine. (...) Even worse... there is no industry standard for what "one bar" or "two bars" means. None. Everyone just sort of sets some thresholds, and even from the same manufacturer it can change from phone model to phone model. (...) The GSM standard does not specify the meaning of the signal bars on your handset (correctly known as the "signal quality estimate"). Each manufacturer uses their own formula to work out how many bars you see. This varies not only between phone makers, but also between models, and between firmware versions of the same model. In short, you can't compare phones using signal bars. You *can* - to a limited extent - compare the signal strength in different locations using the same phone, but even that isn't reliable."

Why do I blog this? this is an interesting example of how seamful design is hard to put in place. However, it would be intriguing to have behavioral adjustments (such as the one we often see with reception bars) even with reception bars that do not mean anything. As if the design itself was more important that the meaning of the information represented.