[Battlemesh] Linksys promises not to block free firmware

[David Lang]

On Sat, 14 May 2016, Mitar wrote:

> Hi!
>
>> making an app that scans/reports on questionable use of APs would show
>> how small the problem is. If we can show that there are a lot of people
>> running the app and reporting that there are not problems near the
>> airports in question, and shows where people are using these
>> frequencies, then we can show that the problem they fear happening is
>> not actually happening.
>
> I like this idea. We have so many devices in the air already. We could
> just do regular site surveys and collect information into one common
> database and then draw it.
>
> In some way this is the same as HAM operators do. If they find somebody
> misusing licensing, they go and discover who they are (at least this is
> how I heard).

We had a lengthy discussion on the bufferbloat list and ended up with two 
incompatible viewpoints

one viewpoint said that every report needed to be authenticated to prevent false 
reporting.

the other viewpoint said that every report should be absolutly anonymous so that 
the database of reports could not be misused to track reporters.

I'm in that second camp.

I think the app should periodically report it's location and what it's heard or 
the fact that it doesn't hear anything. The database should record the 
timestamp, and the report of what's heard, nothing else, especially not the IP 
or anything else identifying the reporter.

My view is that negative reports (nothing heard here) are only useful in 
convincing people that the app is actually producing good coverage in what I 
expect is going to be the common case where nothing bad is detected.

false positives will happen, and are acceptable because this is not a database 
of "this is absolutly bad", but rather "this is odd/suspicious and somene should 
possibly look into it"

If someone floods the database with negative reports, they won't matter for 
someone who is looking at the positive reports.

If someone tries to flood the database with false positive reports to make their 
presense harder to find, they just call more attention to that area (unless they 
decide to flood the database with reports from everywhere, but that is a rather 
significant amount of work, and can be dealt with as an exception)

for that matter, the positive and negative reports are really separate 
databases, because you never take one into account when querying the other.

David Lang