Mosaik purchases Sensorly

[lilotimz]

See: http://www.mosaik.com/news/mosaik-acquires-sensorly-to-improve-global-wireless-network-experiences

 

Thoughts? Comments?

[S4GRU]

See: http://www.mosaik.com/news/mosaik-acquires-sensorly-to-improve-global-wireless-network-experiences

 

Thoughts? Comments?

I know very little about Mosaik. But it will be nice that Sensorly will now be run stateside.

 

Sent from my Nexus 5 using Tapatalk

[WiWavelength]

I can remember back when Mosaik Solutions was formerly American Roamer.

 

AJ

[dkyeager]

Hopefully the become more responsive and further separate out their LTE to make it more accurate.

[JossMan]

No word on how much Mosaik gave for Sensorly? 

[red_dog007]

Combining the two seems very cool.  

 

Maybe we will get a treasure trove of data pulled from the crowed sourced data as well instead of just connection type and random speedtests. 

[Thomas L.]

Hopefully the become more responsive and further separate out their LTE to make it more accurate.

Separate their LTE to make it more accurate it what way?

[dkyeager]

Separate their LTE to make it more accurate it what way?

 

Since it is based on RSRP, the Sprint results can be rigged by only using B26.  By the same token B41 will be measured as much weaker.  The release notes in indicate attempts to address CA with certain phones.  Using my LG G4 I have yet to see any difference.

[SWMich4G]

It would also be nice to see them weight by how old the data is, or even put a filter that allows you to only show data collected in the last (whatever time interval).  Also giving stronger signals more weight, so it's not diluted by people mapping while indoors or with phones without every band.  Sprint results got diluted because so many people mapped so much when it was just B25, which makes the all-time aggregate signal strength appear weak almost everywhere.  I also don't like that it basically shows you "here's where at least one person ever got a signal, even if 99% of the time you can't get a signal here."  When multiple people map and fail to get LTE on a previously mapped area, it needs to count that against the fluke points and remove coverage from the map.  I think that's a big one: I don't like seeing so much purple on the map where I know no phone on any band can get a usable signal - you might as well use the carrier's own coverage maps at that point. 

 

Basically it's a great service for a casual viewer who wants to get a ballpark idea of coverage for carriers, but the way it was originally coded in many way makes it less accurate over time.  Under new ownership, and depending on how the data was originally collected, I'd love to see them implement a few changes that would make their maps much more accurate/valuable to consumers.