[Date Prev][Date Next] [Chronological] [Thread] [Top]

Re: Red Ovals - the technical response

If the red on the famous names ballot was created for infra-red then why do all the visible light lottery readers located in every little corner store use red 032 Warm Red ovals (rectangles) on the lottery tickets?
  The lottery ticket readers may be using a different frequency of red light.

  I remember the original demo ballots for the infrared reader having ovals in a lovely sky blue.  There are only a few substances that effectively absorb the particular infrared light used in the old readers so most any non-black ink was invisible to them (as well as many of the black ones).  The point being that for those readers, we could use most anything for ovals.

  The new visible light readers use a particular frequency of visible red light and any ink substance that absorbs that light will appear grey or black.  White, by definition, reflects all visible light frequencies.  White paper is never perfectly white and this is one reason why our readers need to calibrate for the background paper color for every ballot.  What we're interested in are ovals that do not interfere with detecting voting marks.  Since the readers measure the reflectance of our red light in a window measuring about 1/16th inch high by 3/16th inch wide, we always pick up a bit of the oval.  It's easiest to think of it as completely blurring everything in the viewing window.  As long as our blurred bit of oval doesn't cause our window to noticably darken, then we're okay.

  Our standard ovals are designed so that the bit of oval that we pick up is a small segment of a very thin line.  By making the ovals thicker, we have to be careful that we don't absorb any more of the light in our window than the standard ovals.  For the Riso duplicator ballots we do this by using a dotted line.  With color we can use inks that effectively reflect more of our red light than do our standard black inks.  However, since an ink is composed of components that work by a combination of reflecting, absorbing or transmitting each frequency across the spectrum, an ink that will reflect our red light doesn't have to look red.  Under common white light, they could be brown, orange, or maybe even purple.  The less that the ink substances absorb our red light, the less of a problem that they are.  It may be possible to find an ink substance that is totally invisible under our red light in which case we could totally fill the voting position.

  Thus the difficulty that we have in "specifying" an ink color.  Different ink manufacturers could use different coloring substances in combination to get the same specified color.  These different inks may look the same to us under white light but very different under our red light (ever color match at the store and then find that it's different at home?)  What we would need to spec would be the absorption characteristics of the hundreds of ink coloring chemicals in use.  Then, we would need to know the formula for any particular ink and we would add up the absorption levels of each of the components.  Believe me that this would be a hurculean task.

  In practice, the best that we can do is to test the particular ink that a printer wants to use.  It sounds like Ian has the equipment to measure the absorption characteristics of a particular ink.  However, unless the ink is invisible to our red light, it doesn't stop there.  Once we know the absorption characteristics, we have to estimate the effect of using that ink for particular sizes and shapes of ovals.  Our current oval specs evolved over dozens of modifications made to deal with situations encountered in real use.  And there are dozens of other factors that come in to play (i.e. since the ovals are printed from a separate plate, how good is the alignment with the timing marks?)

  We went through these kinds of issues trying to spec marking instruments for the infrared readers.  It was a constant juggle of tradeoffs (i.e. the carbon black settles out in the Berol markers!)  If you want to use something other than black, my advice would be to get your printer to select a small assortment of standard inks and print some stripes of each on ballot stock.  Ask (politely) Ian to measure the absorption levels of each sample under our red light.  If you find something that is sufficiently light to allow thicker oval lines (say a less than a quarter of the absorption of our black ink), then we could consider adopting that ink as a standard for thick ovals.  Make sure that you specify the ink manufacturer, formula name, dilution factor, and any other relevant variables and then hope that they're consistent in each ink batch.

  Yikes!  And I'll bet that you thought that color was for artists.  ;-)  (wink, wink)