Stress test schema about fossa

[walterj]

Quick question. We will very soon need to be able to filter (and exclude from visualization, or show in different color) points by month and year. This straightforward in this scheme?

[eightysteele]

Right now, for our immediate goal, we're looking to quickly support a specific use case: Given a scientific name, quickly map all points. So as it is above, we're not factoring in date information. But I think we can do that another way.

Andrew at Vizzuality created a pretty amazing visualization of the VertNet data in CartoDB. Basically it's an animation that shows specimens collected over time. Click the link below and watch. The dates are in the lower right corner:

http://cartodb.github.com/torque/examples/vertnet.html

[walterj]

Aaron, thanks. Do say which other way. I don’t think animations are going to cut it. I mean interactive ability to filter for key fields. We need that essentially as soon as we put the new points on the ap.

[eightysteele]

Totally. So in addition to month and year, what are the other key fields to filter on.

[walterj]

Acccuracy, Institution.

Rob?

Is this all useful, or am I complicating things too much?

[robgur]

Hi guys --- I can see a bunch of use cases we need to support re:
filtering and I think it is essential to have this conversation before we
get too far down the road. Walter hit the very most important ones right
up front: 1) Month in order to get seasonality; 2) Year in order to look
at temporal trends. This gives us sciname/date/location which are
crucial for us in terms of mass filtering of points. The other use cases
strike me as less critical and we could provide a way for users to generate
temporary tables from dynamic queries if needed for other kinds of point
data returns. I see the point of filtering by geospatial "uncertainty" as
also potentially important. One nice thing: these are all stored in
formats that are pretty easy to index/small.

I also want us to be able to flag records --- that is to write into
these tables for records where there is some assessment of a problem (zoo
record, etc). So we really need to plan for that use case too for points.
I think this covers it though. Aaron, thanks for keeping us in the loop
and this really exposes how that loop-keeping is super useful to nip issues
before we get too far downstream.

Best, Rob

On Wed, Oct 31, 2012 at 4:05 PM, Walter Jetz [email protected]:

Acccuracy, Institution.

Rob?

Is this all useful, or am I complicating things too much?

Walter

From: Aaron Steele [mailto:[email protected]]
Sent: Wednesday, October 31, 2012 5:49 PM
To: MapofLife/fossa
Cc: Jetz, Walter
Subject: Re: [fossa] Stress test schema (#1)

Totally. So in addition to month and year, what are the other key fields
to filter on.

—
Reply to this email directly or view it on GitHub<
https://github.com/MapofLife/fossa/issues/1#issuecomment-9963420>.

—
Reply to this email directly or view it on GitHubhttps://github.com//issues/1#issuecomment-9963869.

[eightysteele]

Yup, yup, good feedback guys. Let me fold this in and marinate. More thoughts soon.