I built this proof of concept of a tool called https://text.makeup. It is meant to be a friendly Unicode explainer – meant not just for Unicode nerds, but nerds of any kind. Useful for debugging, but also learning.
You can go there now to play (much more fun on desktop!), but I also recorded a 5-minute video that explains it further.
I am curious: Does this feel like fun? Is it worth building out for real? What would you like to see in it if so?
I am a bit overwhelmed with the idea of finishing it and maintaining it since Unicode is so vast and complex. But maybe this could be open sourced, for example?
I have an About page (https://text.makeup/about) that talks more about what’s in there and what *could* be in there. (I have already learned *a lot* from building it, so there’s no skin off my back.)
Thanks for any thoughts you might have!
@mwichary it looks like it's already better for the purpose of exploring text than my previous go-to, Babelstone's venerable "what unicode character is this". https://babelstone.co.uk/Unicode/whatisit.html
(and it has a more memorable URL)
@mwichary This is very cool, and the kind of thing I am very interested in - so much so that I snagged the unicode.exposed domain a while back (a la https://float.exposed/) and then didn't get far enough in that project to be worth putting up there 
If you'd like to take over that domain for this project, feel free to message me! I'd also be interested in collaborating a bit, but as indicated by my half-finished attempt, that may be overambitious. Here's where I left off on my own explorations, you made much more progress than I did, and obviously took a different tack: https://mahogany-waiting-treatment.glitch.me/
My initial interest is in breaking down the mystery of "how do computers represent things", particularly around text encoding, but I also had ambitions parallel to yours - particularly the Observations panel is a brilliant little way to highlight some of the info I wanted to capture. Thanks for this!
taichimaster
Marcin Wichary
80columns
a wandering happenstance
Frédéric Grosshans
Grant Hutchinson
Ell 🏳️⚧️
shork:) and certainly not about \nThe "Information" and "Observation" boxes do not change when I type something in the "Text" box.
The Javascript console starts out with:
Adam SjøgrenTypeError: s is undefined
and as I type I get a bunch of:
Uncaught TypeError: Intl.Segmenter is not a constructor
stringSplit https://text.makeup/main.js?v8:965
senseQuotedPrintableEscape https://text.makeup/main.js?v8:479
hasQuotedPrintableEscapes https://text.makeup/main.js?v8:912
detectStringType https://text.makeup/main.js?v8:931
processInputIntoOutput https://text.makeup/main.js?v8:2201
onInput https://text.makeup/main.js?v8:4825
main https://text.makeup/main.js?v8:6473
main https://text.makeup/main.js?v8:6473
<anonymous> https://text.makeup/:123
main.js:965:28
stringSplit https://text.makeup/main.js?v8:965
senseQuotedPrintableEscape https://text.makeup/main.js?v8:479
hasQuotedPrintableEscapes https://text.makeup/main.js?v8:912
detectStringType https://text.makeup/main.js?v8:931
processInputIntoOutput https://text.makeup/main.js?v8:2201
onInput https://text.makeup/main.js?v8:4825
main https://text.makeup/main.js?v8:6473
(Async: EventListener.handleEvent)
main https://text.makeup/main.js?v8:6473
<anonymous> https://text.makeup/:123
errors.
This is in Firefox 114.0 on Debian GNU/Linux 12.
@asjo Thanks for the report! It seems like the Intl.Segmenter has only been added in Firefox 125 released in April of this year.
I should throw a more gentle, user-facing error on this…
@mwichary @nelson For new additions in the upcoming version of Unicode, the Pipeline page is great, as it links to sources: https://www.unicode.org/alloc/Pipeline.html
The versioned charts page for each release version is nice as it highlights additions/changes, but doesn't link sources: https://www.unicode.org/charts/PDF/Unicode-16.0/
@mwichary @tedmielczarek I asked an expert about this back in 2015 and got this reply:
I have long wanted to create an index of all UTC/WG2 documents, which would go part way to achieving what you want, but I have not had the time to do so unfortunately. For most characters the paper trail is quite straight forward, but for some characters the history is convoluted, and you have to delve into the minutes and resolutions of UTC and WG2 meetings to determine what happened.
@nelson @tedmielczarek I’m interested in a) good stories, and b) stuff that helps understand other things (“how we got there”).
For example, very curious what made “symbols for legacy computing” made it in this year particularly!
@mwichary Maybe it's just me, but when I see "This sequence can be encoded in a more complex way" my first thought is "I must find the _most_ complex way" and then comes the supervillain laugh. Maybe replace "Just so you know" with "<abbr title='Just so you know. This is not a dare. This is not a challenge. OMG do not do this'>Just so you know</abbr>"
Or maybe add that meme "Your scientists were so preoccupied with whether or not they could, they didn’t stop to think if they should"
@mwichary @kayserifserif This looks *great*! Fun for sure, and i’m betting also very useful.
Can you (or have you already somewhere?) talk about how much of this is manual versus automated? Like, I know things like the composition of the flags and handshake can be deconstructed programmatically, but are things like the “fake font“ manual? Are you creating those all yourself?
@a @kayserifserif It’s a huge spectrum. Some things are almost all automated (like flags), but other things (like indeed fake fonts) very manual. There is also a secondary spectrum on “manual based on manually compiled data” or “automatic based on manually compiled data” and so on. The About page has some details on data sources, if you are interested.
Some fully automatic things like confusables and compiled stuff will require slight manual assists/additions, too.
@mwichary this is super cool!
I think many people would benefit from something like a trail map for exploring the data
@DunwichType Thanks! Is it even remotely useful? I haven’t tested at all.
If you have any example URLs or strings handy that you tested, that’d be useful for my orientation!
@compfu Thanks! I figured out the original issue – turns out I wasn’t doing exactly the steps you were doing, and it’s not Vivaldi-specific after all.
It would be fun to support and explain Punycode, but it will a bit more work. But in the meantime I fixed the international domains from breaking at least, if you reload the tool. Gave you credit at https://text.makeup/about/ – thanks again!
Nelson Minar
Simon Cozens
Ted Mielczarek
LAHosken 🇺🇸👀
Jorge Toledo 
Szescstopni
sam henri gold
Henner Zeller
Anthony Sorace
Joe Germuska
Dunwich Type Founders
Stefan Ihringer
Alice Averlong🏳️⚧️
Josh