mark handler
SAWHORSE
Off again, on again --- Contrast standards are now "On Again!"
Sharon Toji
President/ADA consultant at ADA Sign Products
ANSI 117.1 Standards commission member
If anyone has been following the ups and downs of my attempt to get some kind of rational measurable standards for dark/light contrast established, they know that on my third try with ANSI, I had finally reached what I thought was a solid majority "yes" earlier in the 2015 cycle. Then, some of my friends in the design field came up with a counter-argument and convinced enough people, that the vote was changed.
I'm not one to give up easily when I believe something is the right thing to do, so I was able to return to the next meeting with another expert, and the vote changed again. But it still had to go through another hoop, the written ballot.
Yesterday, I got the result, and the standard, as it was last approved, survived, and quite well.
The only sad part was that, due to a "typo-type" error in the material I sent in, we lost a measurable standard for stair striping. I know you may think that's a strange issue for the ADA Sign Lady to concern herself with, but I take all the standards seriously, and particularly anything to do with how the environment is designed to communicate. And frankly, I'm not just a "Lady," but an "Old Lady," so stair striping is very important to me and to friends of my own age -- 'cause guess what, I just turned 79. Yes, next year, I'll it the big 80!
But enough about me. Let's talk instead about the fact that we need to broaden our outlook on just who does benefit from contrast in the built environment, and eventually, broaden our scope of what matters.
Of course the major target group for the benefit of the Americans with Disabilities Act is those who can be diagnosed with actual vision, hearing, cognitive, and mobility impairments. But even the federal government is slowly coming around to the fact that we have a rapidly growing aging population, and that with aging, inevitably come reductions in all those abilities. And for vision, we need to add the rather large group of people, particularly men, who have color deficiencies in their vision. Of course the very reason that we talk about signs not in terms of color, but of dark and light contrast, is to benefit all those whose color vision is deficient. Don't think that color deficient vision happens only in isolation. Various types of vision problems that cause legal blindness reduce the ability to recognize colors, as does aging vision.
There is only one fairly lengthy written reason for the few negative ballots that were received. I think it was based on a misunderstanding of the basic crux of the standard that may actually, this time, make it all the way to the end of the process. Because it came from the ISA, it will be taken seriously, however.
The standard is actually our own, "Made in the USA." It is not the British Standard, not the Australian Standard, not the European Standard. It is very similar in many respects to all of those, and probably the differences among them are really of little consequence, anyway.
Our standard is unique because it seemed best to start with what we already had, and not try to reinvent the wheel. Years ago, a standard that is used world-wide for minimum contrast in the built environment, and that was supported by a mathematical formula so it could be determined, entered our code system.
That is the well-known "70 percent minimum" for dark/light contrast. Even today, it survives intact in one section of the California Building Code, and undoubtedly could be found in other codes, nationwide, as well. The minimum is often backed up by a formula that uses LRV, or Light Reflectance Values, to determine light/dark contrast.
Scientists pretty much agree that Light Reflectance Values are the proper way to measure this, and architects routinely refer to them when they specify all kinds of architectural surface colors, inside and out. Most manufacturers of colored materials, such as paints and plastics, measure them routinely, because its a way to determine that their colors remain constant, one batch to the next.
There is one problem when it comes to using the formula that determines the 70 percent minimum contrast: It creates a strange curve (sorry -- I've had it drawn out and demonstrated, but my advanced mathematics skills are non-existent, so I don't pretend to completely understand it) -- so that if you compare two dark colors, you get a much higher percentage than if you compare two light colors that vary by the same arithmetical amount. In other words, 30 points of difference between 50 and 100 get a different percentage than 30 points of difference between 1 and 50.
You can try it yourself. Compare 80 and 50 and see what percentage you get, and 50 and 20. It will be a large difference. The basic reason is, because the points on the LRV measurement scale are not equal. They get further and further apart as you move along the scale.
That means, that the only way you can use the formula, and can use the 70 percent with reliability so that it will determine contrast in the "real world" of someone trying to see the letters on a sign, is to force the choice into specific areas of the scale. In other words, you have to come to a decision of where on the scale you put the dividing line between light and dark.
We have read the reports of several scientists who have evaluated contrast over the years, and this has been their conclusion: In order to use the 70 percent minimum, and use the formula, choose a point on the scale and say the lighter of the two colors must have that LRV or higher. Then apply the formula, and you will get reasonable results.
So, that is what our ANSI Contrast Committee did. We checked it out with samples, compared it to other national standards, and even had a small field test with people at the national convention of the American Council of the Blind one year. The number we determined would be a reasonable and sensible number is 45.
Since the scale is not even, 50 is not the "perfect" number. Because of the scale, we decided to put the point lower than halfway. We could have placed it lower still, but we were trying to provide designers with the greatest range of choices possible, while providing reasonable contrast. Of course the hope always is, that designers will not design to the minimum. Choosing the number is somewhat arbitrary. Most of the numbers we choose are arbitrary. The last scientist to testify on our behalf asked, "why is 20/40 considered the point at which you can drive safely with corrective lenses?" It was determined because it was sensible, not because 1000 people were tested and people with that exact vision or higher were safer drivers. The fact is, there is no "perfect number," and not for most things we regulate.
So why does ISA argue that the British didn't test the British Standard of Test on small signs, so our proposed standard is not supportable? They are confusing the Standard of Test with the Standard. We added the Standard of Test only to have a specific scientific standard for determining, by the same method, what the LRV is. It really doesn't matter what you are using the material for, the LRV will be the same, swatch to swatch, if you test it according to the Standard of Test.
And yes, it is also true that the British did not base their initial studies on their own formula using signs. They express their minimums differently than we do. The Australian Standard is much closer to our proposed standard. The British merely require a certain number of LRV points between the numbers obtained from the two elements. If it is a floor and wall, it must be 30 points apart on the scale. Their method actually has the same flaw ours does.
So let's be very clear. Our "Standard" is the same one you can find in the original ADAAG Appendix. Even SEGD has long cited this standard and this formula in their "White Paper." Several nationally prominent sign companies like ASI and APCO have also used this standard in their material. However, they have never acknowledged the flaw. We have made one very tiny and important change to that original standard. We will now require that the lighter of the two colors have an LRV of 45 or higher. That's it. That's the entire change we are suggesting.
It gives designers huge leeway in the colors they use for signs. It is modest, and it is conservative, in that it does not require extensive change on anyone's part, if they have been using the formula all along. We now have available a tiny instrument that costs about $250 that will even test the LRV out in the field.
So where does the British Standard of Test come in? It is merely a way, if a case goes to court, that you could determine whether the LRV of the two colors was within the formula. If the designer is going to rely on using minimums, and always chooses a light color of 45, and always chooses a dark color that will achieve no higher than 70 percent contrast, then he or she is treading dangerously if the case goes to court. So, our advice is, don't shoot for minimums. Use 47 as your lighter color, or higher. Try for 75 percent minimum. Then, you can safely use the LRV that you determine with your own inexpensive equipment, which I have found to be accurate to one or two points, or you can rely on the manufacturer's LRV.
Sharon Toji
President/ADA consultant at ADA Sign Products
ANSI 117.1 Standards commission member
If anyone has been following the ups and downs of my attempt to get some kind of rational measurable standards for dark/light contrast established, they know that on my third try with ANSI, I had finally reached what I thought was a solid majority "yes" earlier in the 2015 cycle. Then, some of my friends in the design field came up with a counter-argument and convinced enough people, that the vote was changed.
I'm not one to give up easily when I believe something is the right thing to do, so I was able to return to the next meeting with another expert, and the vote changed again. But it still had to go through another hoop, the written ballot.
Yesterday, I got the result, and the standard, as it was last approved, survived, and quite well.
The only sad part was that, due to a "typo-type" error in the material I sent in, we lost a measurable standard for stair striping. I know you may think that's a strange issue for the ADA Sign Lady to concern herself with, but I take all the standards seriously, and particularly anything to do with how the environment is designed to communicate. And frankly, I'm not just a "Lady," but an "Old Lady," so stair striping is very important to me and to friends of my own age -- 'cause guess what, I just turned 79. Yes, next year, I'll it the big 80!
But enough about me. Let's talk instead about the fact that we need to broaden our outlook on just who does benefit from contrast in the built environment, and eventually, broaden our scope of what matters.
Of course the major target group for the benefit of the Americans with Disabilities Act is those who can be diagnosed with actual vision, hearing, cognitive, and mobility impairments. But even the federal government is slowly coming around to the fact that we have a rapidly growing aging population, and that with aging, inevitably come reductions in all those abilities. And for vision, we need to add the rather large group of people, particularly men, who have color deficiencies in their vision. Of course the very reason that we talk about signs not in terms of color, but of dark and light contrast, is to benefit all those whose color vision is deficient. Don't think that color deficient vision happens only in isolation. Various types of vision problems that cause legal blindness reduce the ability to recognize colors, as does aging vision.
There is only one fairly lengthy written reason for the few negative ballots that were received. I think it was based on a misunderstanding of the basic crux of the standard that may actually, this time, make it all the way to the end of the process. Because it came from the ISA, it will be taken seriously, however.
The standard is actually our own, "Made in the USA." It is not the British Standard, not the Australian Standard, not the European Standard. It is very similar in many respects to all of those, and probably the differences among them are really of little consequence, anyway.
Our standard is unique because it seemed best to start with what we already had, and not try to reinvent the wheel. Years ago, a standard that is used world-wide for minimum contrast in the built environment, and that was supported by a mathematical formula so it could be determined, entered our code system.
That is the well-known "70 percent minimum" for dark/light contrast. Even today, it survives intact in one section of the California Building Code, and undoubtedly could be found in other codes, nationwide, as well. The minimum is often backed up by a formula that uses LRV, or Light Reflectance Values, to determine light/dark contrast.
Scientists pretty much agree that Light Reflectance Values are the proper way to measure this, and architects routinely refer to them when they specify all kinds of architectural surface colors, inside and out. Most manufacturers of colored materials, such as paints and plastics, measure them routinely, because its a way to determine that their colors remain constant, one batch to the next.
There is one problem when it comes to using the formula that determines the 70 percent minimum contrast: It creates a strange curve (sorry -- I've had it drawn out and demonstrated, but my advanced mathematics skills are non-existent, so I don't pretend to completely understand it) -- so that if you compare two dark colors, you get a much higher percentage than if you compare two light colors that vary by the same arithmetical amount. In other words, 30 points of difference between 50 and 100 get a different percentage than 30 points of difference between 1 and 50.
You can try it yourself. Compare 80 and 50 and see what percentage you get, and 50 and 20. It will be a large difference. The basic reason is, because the points on the LRV measurement scale are not equal. They get further and further apart as you move along the scale.
That means, that the only way you can use the formula, and can use the 70 percent with reliability so that it will determine contrast in the "real world" of someone trying to see the letters on a sign, is to force the choice into specific areas of the scale. In other words, you have to come to a decision of where on the scale you put the dividing line between light and dark.
We have read the reports of several scientists who have evaluated contrast over the years, and this has been their conclusion: In order to use the 70 percent minimum, and use the formula, choose a point on the scale and say the lighter of the two colors must have that LRV or higher. Then apply the formula, and you will get reasonable results.
So, that is what our ANSI Contrast Committee did. We checked it out with samples, compared it to other national standards, and even had a small field test with people at the national convention of the American Council of the Blind one year. The number we determined would be a reasonable and sensible number is 45.
Since the scale is not even, 50 is not the "perfect" number. Because of the scale, we decided to put the point lower than halfway. We could have placed it lower still, but we were trying to provide designers with the greatest range of choices possible, while providing reasonable contrast. Of course the hope always is, that designers will not design to the minimum. Choosing the number is somewhat arbitrary. Most of the numbers we choose are arbitrary. The last scientist to testify on our behalf asked, "why is 20/40 considered the point at which you can drive safely with corrective lenses?" It was determined because it was sensible, not because 1000 people were tested and people with that exact vision or higher were safer drivers. The fact is, there is no "perfect number," and not for most things we regulate.
So why does ISA argue that the British didn't test the British Standard of Test on small signs, so our proposed standard is not supportable? They are confusing the Standard of Test with the Standard. We added the Standard of Test only to have a specific scientific standard for determining, by the same method, what the LRV is. It really doesn't matter what you are using the material for, the LRV will be the same, swatch to swatch, if you test it according to the Standard of Test.
And yes, it is also true that the British did not base their initial studies on their own formula using signs. They express their minimums differently than we do. The Australian Standard is much closer to our proposed standard. The British merely require a certain number of LRV points between the numbers obtained from the two elements. If it is a floor and wall, it must be 30 points apart on the scale. Their method actually has the same flaw ours does.
So let's be very clear. Our "Standard" is the same one you can find in the original ADAAG Appendix. Even SEGD has long cited this standard and this formula in their "White Paper." Several nationally prominent sign companies like ASI and APCO have also used this standard in their material. However, they have never acknowledged the flaw. We have made one very tiny and important change to that original standard. We will now require that the lighter of the two colors have an LRV of 45 or higher. That's it. That's the entire change we are suggesting.
It gives designers huge leeway in the colors they use for signs. It is modest, and it is conservative, in that it does not require extensive change on anyone's part, if they have been using the formula all along. We now have available a tiny instrument that costs about $250 that will even test the LRV out in the field.
So where does the British Standard of Test come in? It is merely a way, if a case goes to court, that you could determine whether the LRV of the two colors was within the formula. If the designer is going to rely on using minimums, and always chooses a light color of 45, and always chooses a dark color that will achieve no higher than 70 percent contrast, then he or she is treading dangerously if the case goes to court. So, our advice is, don't shoot for minimums. Use 47 as your lighter color, or higher. Try for 75 percent minimum. Then, you can safely use the LRV that you determine with your own inexpensive equipment, which I have found to be accurate to one or two points, or you can rely on the manufacturer's LRV.