This is a reply to the following bbs thread:
gemini://bbs.geminispace.org/s/Gemini/16233
My original article that triggered this discussion.
So as it turns out, I was late to the party!
gemini://bbs.geminispace.org/s/Geminispace/3228
Those things have been discussed before, and my accidental discovery of a keyboard app that decorates text, was not the first time someone – perhaps more knowledgeable than me – tried to "abuse" utf-8 characters for their visual appearance.
It turns out that one of my most important concerns, is true:
This will mess up screen readers, and be a nightmare for accessibility.
Somehow I was hopping that a bold, or italic "a" character, would still carry the semantic value of the letter "a". After all it is an "a"—right?
Well it turns out that it is not true; as I have been informed by feedback. (thanks Alexis)
Whether or not these characters render properly will depend upon the context. My system does not render these characters properly.
But in the terminal, it does not. Even Vimini, a graphical browser, does not render them correctly.
There lays the problem that I was not expecting!
I tested the appearance of my article, on Lagrange, on Buran, and on Deedum. (Those are all mobile clients) And everything looked fine.
I don't have any terminal client installed, so just for good measure, I copy-pasted the raw text on termux, using nano, vim, and cat. No problems there either!
I even tested it on mozz.us portal.
I did all that, even though I was under the impression, that rendering those symbols, is a system thing, and not a client thing.
I thought that if a client supports utf-8, it should be able to display all symbols.
Well that was also not correct as it appears.
Although I must point out, that while we are right to be concerned about accessibility, internationalization Is equally important.
Why would a client not display mathematical symbols? Would it display Mandarin Chinese? Cherokee? Can I write ancient Greek, with paragraphoi "⸐", and coronis "⸎"
I don't know how those things actually work. Is it just a matter of installing some font-set?
Does support for "fancy" symbols increase the complexity of building a client? If it does, then using such "techniques" Is also in contradiction with the Gemini philosophy, that kept emphasis out of gemtext in the first place.
And also the opposite of what I was trying to achieve!
I have to keep reminding myself that a lot of this formatting is a crutch for good writing, and only really exists in an ad hoc sense due to the digital publishing revolution.
I would like to point out, that writen text has evolved a lot historically, and major role in that evolution, has been the technology and the medium.
Writing began in Scriptio continua, boustrophedon, without punctuation. (In some places)
Centuries of use, pointed out the needs for those things to change and be improved, mostly with additions.
Emphasis appeared because it was needed, and because it was possible by the technology of the time.
Losing that, is limiting our options for expression.
We should not disregard it as lesser, and most importantly, gemtext should not be just for good writers.
It could be helpful for non native speakers, or writers of suboptimal linguistic capabilities, to express themselves more effectively.
Also in my honest opinion, it is really helpful for the reader, when key words are emphasized.
It helps focusing the eye, and I am wondering weather it is helpful for people with reading difficulties, like dyslexia.
As a matter of principle, Unicode characters should be used according to their semantic meaning and not their visual appearance.
For me the deal breaker, is the screen reader issue, as well as the auto translation, and search issue.
But speaking of semantic meaning, what is the semantic meaning of an asterisk, or an underscore?
They are fine as markup symbols, but when they are not rendered, they are out of place.
Also how well would the word **emphasis ** sound in a screen reader?
Will the reader say: "asterisk asterisk emphasis asterisk asterisk", or can it be configured to ignore asterisks?
In that case perhaps a screen reader could be configured to ignore non-breaking-space between single characters, and give everyone the "ability" to use w i d e words as substitute for italics.
But we need feedback from someone actually familiar with screen readers, as to how possible or feasible this could be.
If it turns out to be feasible, perhaps some instruction for screen-reader users, can be posted somewhere on the gemtext instructions page, and behold! We have an extra way of expressing emphasis. (Unless this breaks something else, and I am just talking nonsense!)
For me in the meantime, CAPITALIZATION is a good approximation for bold, because bold is kinda like shouting!
And w i d e, is a great substitute for italics, but I will need to include an unemphasized version of every article, for accessibility.
The conclusion of this whole discussion for me is that unicode "decorated" text should be avoided, and I – the author of the original post – do not endorse it.
Sorry to everyone for beating up a dead horse!
gemini://bbs.geminispace.org/s/Gemini/16233
My original article that triggered this discussion.
gemini://bbs.geminispace.org/s/Geminispace/3228
2024-04-21 corrected a typo
❇️ ❇️ ❇️