[GTALUG] a simple security solution?

D. Hugh Redelmeier hugh at mimosa.com
Tue Apr 16 10:00:32 EDT 2019


I think many of the respondents are giving answers without an important 
piece of the puzzle.

Karen: I seem to remember that your vision is quite limited.  Would you be 
able to get anything out of the playback of a video?

If you cannot view a video directly, I don't think that any of the 
proposals would be of much use.

Some thoughts (but note that I have no particular expertise in this area):

- recordings *inside* you front door would reduce the false positives.  
  In other words, recordings would be limited to someone who actually came 
  inside.

  Bonus points: an outside recording that could be retroactively examined, 
  (by someone else) once "something of interest" were detected inside.

- recordings should include sound since that would be something you could 
  examine yourself.  You could develop the habit of saying "I'm home" to
  mark recordings of yourself.

- AI is getting better.  Some day it will recognize people sufficiently 
  well to do some of this for you.  But I don't know that we're there yet.

- There is an AI gold rush now.  Perhaps some startup could be
  interested in your problem as an untapped market.  But that's a long
  way from a simple off-the-shelf solution.

- I imagine that most systems ignore the accessibility features you need.  
  An Open Source solution might make it easier to hack in code to make 
  at least the logs accessible.

- Some home alarm companies might have solutions for you where they do the
  monitoring remotely.  I've not asked.  But that would compromise your 
  privacy and might be expensive.

I don't imagine that there is a simple affordable security solution.
Other than getting a dog (they have native intelligence).



More information about the talk mailing list