This is a fun side project. What I'm curious to know is: has any blind person actually tried any of Alexander Fakoó's scripts?
The reason I'm curious is that there's something funny that happens when sighted people create tactile technologies for the blind: they often don't consult with blind people at all. There's something appealing about the idea of theoretically assistive technology that leads to very impractical systems like Boston Line Type [1] or "braille displays" that have only one single character, that's 10x the normal size. It's easy to assume that if you can technically feel something then that's sufficient for blind people, but the history of blind writing systems shows us that's not enough.
This particular idea seems more promising than other efforts (e.g. Moon), because bumps seem to be easier to feel than shaped figures. It can also be written using a regular slate and stylus. However, modern Braille [2] is full of contractions to reduce the number of characters, and even then braille books are massive and heavy compared to their print counterparts. Doubling the width of individual letters and forgoing contractions really limits the utility to very small snippets of text, and learning an entirely new alphabet just for that doesn't seem very practical.
The author's website [3] is full of promises that "anyone" can read the writing system, but it also says "Developer is Alexander Fakoó, who has learned to read the Braille Writing optically". That's great, but sight-reading Braille is a whole different medium from tactile reading. Personally, I used to be able to read grade 1 braille by sight, but could never read by touch.
So the visually impaired would need to learn a new alphabet/representation all to make things readable to the sighted.
I don’t think that suggestion would go over well.
Also it looks like it would be less informative dense for a given area, which seems like a downside as well.
It’s a neat exercise and could make a cool font for an ultra-low resolution display. But this does not seem like a better solution than just printing normal text above/below/on top of the Braille.
Maybe I'm wrong, but it seems like the advantage here is that as people loses their sight (slowly or quickly) they would quickly be able to pick up a tactile font with little to no additional training. Normal sighted people would also be able to read things in the dark (safety factor for things like power outages). Percent of people with blindness and vision impairment drastically increase by age so I think it is safe to say that most blind people aren't born blind.
They generally learn Latin letterforms already for reading embossed signage. These are designed to be composed from Braille pairs so that you get dual purpose text with existing production methods.
It's fine if the tactile letters intended for the blind don't look the same as the visual ones intended for the sighted; you can associate the whole set in a single day, provided that you're genuinely interested in doing so. (People don't do it because they don't see the point [pun not intended] in doing so.)
The actual issue with Braille, is that letters are often tactically similar. Even tactically, people identify shapes better than absolute positions over a grid; in this sense Braille is flawed by design (why just dots? Why not a mix of dots, curves and straight lines?), and Fakoo's alphabet doesn't solve this.
It was, but is it now? I don't think that we should restrict ourselves to what we could do 200y ago.
Specially when you could easily build a system already using the readers' knowledge of current Braille, something like this: https://i.imgur.com/faOWkxb.png
They're visually more distinctive, but more importantly: I think that they'd be easier to tell apart by touch.
If you do that you throw away everything that’s been written in Braille, or force the visually impaired to learn two alphabets. All in the name of an optimization that has an unproven benefit.
Let’s say it’s 10% better. Would that be worth it?
> throw away everything that’s been written in Braille, or force the visually impaired to learn two alphabets.
Neither.
Check the picture linked in the very comment that you're replying to. Then you'll notice my proposal is to modify the current system, by either a) tweaking the shapes of the individual dots or b) replacing certain sequences of dots with lines. The underlying associations are still the same.
If it helps, think on it as a different typeface for Braille instead of a new alphabet. The actual con of the idea is printing, not reading.
Fakoo's idea on the other hand is, effectively, a new alphabet. It might be more advantageous than Braille, I don't know; however keep in mind that learning a new alphabet is by no means such a Herculean cognitive burden as you're pretending it to be, provided that you're already used to the relevant medium.
>Let’s say it’s 10% better. Would that be worth it?
Nobody needs to assume anything in this case. _If_ the idea is worth trying, this can be measured by testing it.
Mind that visual reading – I actually do not know about tactile, but I'd guess it may be similar – isn't about letter by letter, but about the impression of entire words. Here, similarity helps a lot. (Otherwise, you have not only learn the new alphabet, but the gestalt of all the words anew.)
Regarding Braille, my grandmother became mostly blind at old age. Having maintained an interest in books, while finding herself increasingly restricted with regard to her radius of action, she tried to learn Braille, but eventually gave up. At her age, she couldn't feel the differences anymore. Tactile resolution and significance do matter. (I'd imagine, in her case a similarity to what she had been used to may have also helped.)
Generally – while not knowing much about the viability of the specific implementation –, I think, Fakoo is a great idea.
The similarity that I'm talking about (letters with excessively similar shapes) is a hindrance, not a help. Specially because, as you said, we don't read words letter-by-letter; if the shapes are too similar you need to pay more attention to each one of them, otherwise everything becomes a jumbled mess.
I'm not blind, mind you, although I can read Braille by touch. And sometimes I feel like Braille is the tactile version of this: https://www.boredpanda.com/blog/wp-content/uploads/2020/07/r... - sure, you can decode the letters, but you waste seconds doing so.
Fakoo doesn't address that, and I feel that it's a more pressing concern than making the letters look like Latin letters.
> Mind that visual reading – I actually do not know about tactile, but I'd guess it may be similar – isn't about letter by letter, but about the impression of entire words.
I'm under the impression that I'm reading words from both ends with a rather diffuse emphasis on the middle, progressing from both ends towards the middle (maybe at times with a slightly stronger emphasis on the ending) and stopping, when sufficiently ensured, both by the gestalt and and from context. It's more like grabbing the words than reading them sequentially. This is also true for entire groups of words. Generally speaking, reading fluently is a different processing from spelling words out.
Many years ago, I had tried developing a font style with each glyph resembling a blade of grass. The constraints were strongly enforced, so the optimized glyphs no longer resemble the original shapes.
https://imgur.com/a/8jTf7yn
It was 7 years ago. I have lost the optimization code I wrote. The error was defined as IOU metric and penalty term was deviation from three points (top, middle, bottom) on vertical straight line. The idea was to write as much of the glyph vertically as possible that I believed would lead to faster pen strokes.
I haven't looked it up, but I can't imagine this is acceptable under the ADA. For the vast majority of blind people I guess this would just be an incomprehensible bunch of bumps, failing in the accessibility mission.
Only because they haven't learned it. If you gave Braille to a blind person who has never experienced it before, they would also find it to an incomprehensible bunch of bumps failing in the accessibility mission.
At least with Fakoo, someone who used to be able to see letters before going blind, could probably figure out Fakoo without instruction.
Most people who are blind lost their sight as adults[0], and less than 10% of people who are blind can read Braille[1], so this font is already probably readable by more blind people than Braille is.
I'm not so sure. There are enough tactile alphabets that can be possibly decoded by who became blind recently, but they are sufficiently similar to the original alphabet when touched, unlike Fakoo's artificial 3x3 grid. And those tactile alphabets were not well received by blinds anyway; Braille is literally the most successful tactile alphabet (yes, even at the current <10% usages) and its success seems to prefer long-term efficiency over familiarity.
Dot-based tactile alphabets are far more readable than other types, because dots maximize the number of broken paper fibers sticking up vertically out of the tops of the paper to tickle the fingertips, which is what Braille readers actually feel when reading - they don't actually try to feel the bumps, they instead feel the much higher frequency stimulation from the raised fibers that reach above the bumps (I apologize for not having a citation on this, I spent a great deal of time 20 years ago researching tactile technology for the blind, and this was a highly surprising and broadly known and reproduced result at the time).
This is a fun side project. What I'm curious to know is: has any blind person actually tried any of Alexander Fakoó's scripts?
The reason I'm curious is that there's something funny that happens when sighted people create tactile technologies for the blind: they often don't consult with blind people at all. There's something appealing about the idea of theoretically assistive technology that leads to very impractical systems like Boston Line Type [1] or "braille displays" that have only one single character, that's 10x the normal size. It's easy to assume that if you can technically feel something then that's sufficient for blind people, but the history of blind writing systems shows us that's not enough.
This particular idea seems more promising than other efforts (e.g. Moon), because bumps seem to be easier to feel than shaped figures. It can also be written using a regular slate and stylus. However, modern Braille [2] is full of contractions to reduce the number of characters, and even then braille books are massive and heavy compared to their print counterparts. Doubling the width of individual letters and forgoing contractions really limits the utility to very small snippets of text, and learning an entirely new alphabet just for that doesn't seem very practical.
The author's website [3] is full of promises that "anyone" can read the writing system, but it also says "Developer is Alexander Fakoó, who has learned to read the Braille Writing optically". That's great, but sight-reading Braille is a whole different medium from tactile reading. Personally, I used to be able to read grade 1 braille by sight, but could never read by touch.
[1]: https://en.wikipedia.org/wiki/Boston_line_letter
[2]: https://en.wikipedia.org/wiki/English_Braille
[3]: https://fakoo.de/en/fakoo.html
A similarly low-res, for different reasons, font: https://simplifier.neocities.org/4x4.html
now also on the front page:
https://news.ycombinator.com/item?id=33127419
So the visually impaired would need to learn a new alphabet/representation all to make things readable to the sighted.
I don’t think that suggestion would go over well.
Also it looks like it would be less informative dense for a given area, which seems like a downside as well.
It’s a neat exercise and could make a cool font for an ultra-low resolution display. But this does not seem like a better solution than just printing normal text above/below/on top of the Braille.
Maybe I'm wrong, but it seems like the advantage here is that as people loses their sight (slowly or quickly) they would quickly be able to pick up a tactile font with little to no additional training. Normal sighted people would also be able to read things in the dark (safety factor for things like power outages). Percent of people with blindness and vision impairment drastically increase by age so I think it is safe to say that most blind people aren't born blind.
They generally learn Latin letterforms already for reading embossed signage. These are designed to be composed from Braille pairs so that you get dual purpose text with existing production methods.
The 7 is a glider.
https://en.wikipedia.org/wiki/Glider_%28Conway%27s_Life%29
q is a boat
Does anyone know why the "I" (pronounced 'eye') is an upside-down "T"?
I see that "Z" must take what would naturally be an "I" shape on a 3x3 grid, but why not use just a simple vertical line for "I"?
Is it possibly because the font should be monospaced and the reader might then mistake the distance before and after the "I" for a space?
I don’t get that either. Visually I should be sideways H. Z should be mirror of S.
Notice all characters use full 3px width
As in GP's proposal.
Some punctuation characters don't.
Similarly, in siekoo[1] the “I” is also not a straight line.
Perhaps predictable character width (monospace) makes reading easier.
[1]: https://omniglot.com/conscripts/siekoo.htm
It's fine if the tactile letters intended for the blind don't look the same as the visual ones intended for the sighted; you can associate the whole set in a single day, provided that you're genuinely interested in doing so. (People don't do it because they don't see the point [pun not intended] in doing so.)
The actual issue with Braille, is that letters are often tactically similar. Even tactically, people identify shapes better than absolute positions over a grid; in this sense Braille is flawed by design (why just dots? Why not a mix of dots, curves and straight lines?), and Fakoo's alphabet doesn't solve this.
> why just dots? Why not a mix of dots, curves and straight lines?
Because it was easier for the technology of the day (invented in 1824) to reliably emboss dots.
It was, but is it now? I don't think that we should restrict ourselves to what we could do 200y ago.
Specially when you could easily build a system already using the readers' knowledge of current Braille, something like this: https://i.imgur.com/faOWkxb.png
They're visually more distinctive, but more importantly: I think that they'd be easier to tell apart by touch.
If you do that you throw away everything that’s been written in Braille, or force the visually impaired to learn two alphabets. All in the name of an optimization that has an unproven benefit.
Let’s say it’s 10% better. Would that be worth it?
> throw away everything that’s been written in Braille, or force the visually impaired to learn two alphabets.
Neither.
Check the picture linked in the very comment that you're replying to. Then you'll notice my proposal is to modify the current system, by either a) tweaking the shapes of the individual dots or b) replacing certain sequences of dots with lines. The underlying associations are still the same.
If it helps, think on it as a different typeface for Braille instead of a new alphabet. The actual con of the idea is printing, not reading.
Fakoo's idea on the other hand is, effectively, a new alphabet. It might be more advantageous than Braille, I don't know; however keep in mind that learning a new alphabet is by no means such a Herculean cognitive burden as you're pretending it to be, provided that you're already used to the relevant medium.
>Let’s say it’s 10% better. Would that be worth it?
Nobody needs to assume anything in this case. _If_ the idea is worth trying, this can be measured by testing it.
Mind that visual reading – I actually do not know about tactile, but I'd guess it may be similar – isn't about letter by letter, but about the impression of entire words. Here, similarity helps a lot. (Otherwise, you have not only learn the new alphabet, but the gestalt of all the words anew.)
Regarding Braille, my grandmother became mostly blind at old age. Having maintained an interest in books, while finding herself increasingly restricted with regard to her radius of action, she tried to learn Braille, but eventually gave up. At her age, she couldn't feel the differences anymore. Tactile resolution and significance do matter. (I'd imagine, in her case a similarity to what she had been used to may have also helped.)
Generally – while not knowing much about the viability of the specific implementation –, I think, Fakoo is a great idea.
The similarity that I'm talking about (letters with excessively similar shapes) is a hindrance, not a help. Specially because, as you said, we don't read words letter-by-letter; if the shapes are too similar you need to pay more attention to each one of them, otherwise everything becomes a jumbled mess.
I'm not blind, mind you, although I can read Braille by touch. And sometimes I feel like Braille is the tactile version of this: https://www.boredpanda.com/blog/wp-content/uploads/2020/07/r... - sure, you can decode the letters, but you waste seconds doing so.
Fakoo doesn't address that, and I feel that it's a more pressing concern than making the letters look like Latin letters.
> Mind that visual reading – I actually do not know about tactile, but I'd guess it may be similar – isn't about letter by letter, but about the impression of entire words.
This is a strangely popular myth.
>> reading ... isn't about letter by letter, but about the impression of entire words
> This is a strangely popular myth
Praheps that starengly pouplar mtyh about rediang is gronuded in the nautre of cogitnive prcoesess.
I'm under the impression that I'm reading words from both ends with a rather diffuse emphasis on the middle, progressing from both ends towards the middle (maybe at times with a slightly stronger emphasis on the ending) and stopping, when sufficiently ensured, both by the gestalt and and from context. It's more like grabbing the words than reading them sequentially. This is also true for entire groups of words. Generally speaking, reading fluently is a different processing from spelling words out.
It's simplified but not a myth. This article from Science Alert explains it better than I do: https://www.sciencealert.com/word-jumble-meme-first-last-let...
ThE OvErAlL SHapE MatTeRs ToO, and that's what makes case-mixing so annoying.
Your link does not even purport to address the mechanics of reading.
https://learn.microsoft.com/en-us/typography/develop/word-re...
> ThE OvErAlL SHapE MatTeRs ToO, and that's what makes case-mixing so annoying.
This is false.
Interesting, but what about accents or non-English letters such as ø, þ pr ð?
Also, considering that so much is written in Braille, how many people use this?
Many years ago, I had tried developing a font style with each glyph resembling a blade of grass. The constraints were strongly enforced, so the optimized glyphs no longer resemble the original shapes. https://imgur.com/a/8jTf7yn
I like it. Do you have more information?
It was 7 years ago. I have lost the optimization code I wrote. The error was defined as IOU metric and penalty term was deviation from three points (top, middle, bottom) on vertical straight line. The idea was to write as much of the glyph vertically as possible that I believed would lead to faster pen strokes.
Does ADA legislation in the US allow use of a font like this, or does the legislation explicitly require Braille text?
I haven't looked it up, but I can't imagine this is acceptable under the ADA. For the vast majority of blind people I guess this would just be an incomprehensible bunch of bumps, failing in the accessibility mission.
Only because they haven't learned it. If you gave Braille to a blind person who has never experienced it before, they would also find it to an incomprehensible bunch of bumps failing in the accessibility mission.
At least with Fakoo, someone who used to be able to see letters before going blind, could probably figure out Fakoo without instruction.
Most people who are blind lost their sight as adults[0], and less than 10% of people who are blind can read Braille[1], so this font is already probably readable by more blind people than Braille is.
[0]https://www.who.int/news-room/fact-sheets/detail/blindness-a...
[1]https://nfb.org/images/nfb/documents/pdf/braille_literacy_re...
I'm not so sure. There are enough tactile alphabets that can be possibly decoded by who became blind recently, but they are sufficiently similar to the original alphabet when touched, unlike Fakoo's artificial 3x3 grid. And those tactile alphabets were not well received by blinds anyway; Braille is literally the most successful tactile alphabet (yes, even at the current <10% usages) and its success seems to prefer long-term efficiency over familiarity.
Dot-based tactile alphabets are far more readable than other types, because dots maximize the number of broken paper fibers sticking up vertically out of the tops of the paper to tickle the fingertips, which is what Braille readers actually feel when reading - they don't actually try to feel the bumps, they instead feel the much higher frequency stimulation from the raised fibers that reach above the bumps (I apologize for not having a citation on this, I spent a great deal of time 20 years ago researching tactile technology for the blind, and this was a highly surprising and broadly known and reproduced result at the time).
Proto-Marain! https://theculture.fandom.com/wiki/Marain