Map viz doesn't work well with data sets having multiple rows that share the same GPS coordinates.

[ivanrudnicki]

In cases where data sets have multiple rows that share the same GPS coordinates, the performance of the map viz is sub-optimal. Clicking on the map pointer opens a window that shows the data from only one of the rows associated with those coordinates, without giving any indication that there are other rows. See, e.g., http://isenseproject.org/highvis.php?sessions=6758

I have a couple questions/thoughts about this:

1. Is it true that, in these cases, the map is loading a pointer for every row in the table, even though most are just duplicates of one another (and therefore invisible)? If so, would it be possible to do a data-reduction operation to enhance the map performance? (For the set shown above, this would reduce the number of pointers from 1200 to 15.)

2. Could we build tools to give users the ability to see multiple data rows associated with the same map pointer? Ideally, the pop-up window would allow the user to page back and forth through data from each of the rows associated with the corresponding point on the map. (I have no idea whether this would be possible!)

I'm hoping to use the data set shown above at the boathouse this spring.

[mmcguinn]

We used to have this sort of de-duping, and it was a bit concerning to me in a couple of respects. The old way simply only showed some %'age of datapoints, which is pretty disingenuous. The other obvious solution is to average the collapsed points, but that could be equally incorrect, and you can't really average time or text (or some numeric data for that matter).

[ivanrudnicki]

So, when there are multiple data points with exactly the same GPS
coordinates, is the map trying to load all of the pointers on top of one
another? If so, that must surely slow it down. We should at least keep it
from doing that, and put something in the pop-up saying that there are
multiple data points for this location?

On Thu, Apr 25, 2013 at 2:39 PM, mmcguinn [email protected] wrote:

We used to have this sort of de-duping, and it was a bit concerning to me
in a couple of respects. The old way simply only showed some %'age of
datapoints, which is pretty disingenuous. The other obvious solution is to
average the collapsed points, but that could be equally incorrect, and you
can't really average time or text (or some numeric data for that matter).

—
Reply to this email directly or view it on GitHubhttps://github.com//issues/607#issuecomment-17032727
.

[mmcguinn]

I'm not sure, a popup doesn't really work because there might be many such points, and then you have to consider what makes two points 'the same'. We should probably have a more in-depth discussion about this.