Skip to Content

A Request for Mr. S. Claus - Some Sensor Tools!

Printer-friendly versionPrinter-friendly version
Issue: 
Fall 2012

THE FOLLOWING LETTER WAS FOUND IN A SNOW DRIFT OUTSIDE THE NSF HQ EARLY LAST WINTER NEAR WHAT APPEARS TO BE A REINDEER HOOF TRACK.

Dear Santa,

I know that sensors pose a challenge for you in your work (especially “roof-cams”). For Christmas, some people might ask for new sensors, but I’d like some software tools for managing them.  In particular, I’d like:

  1. Improved capabilities for tracking sensor events, including deployment, calibration and failure.  When I started with the basic sensor network you bought me 3 Christmases ago, it wasn’t hard. First, there weren’t so many different sensors. Also, they were all new, so they worked great. However, now with the advanced sensors you’ve brought me on subsequent Christmases, I’m getting overloaded! I’d like the data about the sensors and their deployments to be in a database so that I could:
    1. Use queries to identify sensors that may be reaching the end of their useful life so that I can replace them before they fail.
    2. Use queries to identify time periods where sensors were inactive or providing less-than premium data so that I can put the appropriate “flags” in my datasets.
    3. Provide a basis for statistical analyses focused on identifying causes of sensor failure, so that I can anticipate future problems.

    Please make it so that populating the database doesn’t require too much labor in and of itself. When I or my technician is in the field, it’s not always the best time to record copious notes, especially when we are rushing home to wrap gifts or bake cookies.

  2. Better tool sets for identifying sensor problems.  As you know - especially following the unfortunate episode of the “defective Rooty-Toot-Toots” - the key to avoiding big problems is early detection of small ones.  I can generate some basic QC tools, but it would be great if you could integrate a bunch of standard tools into a set where I could easily apply them to new datasets.  John C. and Don H. have been talking about developing some standards, and it would be great if you could integrate those into some tools that I could easily use with minor configuration for new datasets.

  3. More sophisticated image and soundscape analysis tools.  As a world traveler of some renown, you know how valuable sight and sound are. After all, if it weren’t for your “sees you when you’re sleeping” device, you’d have been “busted” years ago for burglary.  However, unlike you, we don’t have an unlimited supply of elves to watch and listen.  It would be great to have some programs that would automatically extract information from image and sound streams.

I know that a lot of other information managers will want these kinds of things, too, because they all have sensors, sometimes the same kind, sometimes different, some old, some new. So Santa – this line of toys will make lots of people happy. I have been a good information manager all year, and have faithfully worked on improving my metadata, so I’m hoping you and your elves will come through with the goods!

Sincerely,

NAME TOO SMUDGED TO READ