As far as I know the DION Reloaded project is just a visualization.
The DION Systems editor 4coder was bought in 2020, turned into the coolest AST demo I have ever seen (already posted), then abandoned in 2022.
There is also Ki Editor which claims to be an AST modal editor, but seems to be more a prototype than a product at the moment.
Finally you have Kiro which claims to reduce cost for AI agents since one single modification to a node in an AST would equate to multiple modifications in plain text (saving tokens). Sadly, as a the hooman cowboy that I am, I donāt care about about this use case.
Every AST editor I have tried had too many bugs, needed hand written plugins, is extremely ugly or refuses to launch. That is not to say that the tech is not know, understood or practiced. Itās a product about to be born, waiting for some sweet sweet VC money and maybe a rewrite.
My statement about it being unprecedented was prefaced with the context of my viewpoint, I even used a semi-colon to explicitly state that theyāre linked and āsoā which also links. I feel like I should be clear enough but it seems like A11y is a sensitive topic with a lot to hate about it.
After looking at the options, they donāt do what Iād find useful when customising the formatter does do what Iād want them to do. Dion Reloaded is neat but is not what Iām asking for, I just want to program but with some control to aid readability.
I keep being precise with how I explain things and itās constantly misinterpreted and then misrepresentations get a load of likes - I feel like Iām making things worse by existing and defending my access.
Iām getting really tired of this. I felt like I already said everything I had to say in my first message, so Iām going to leave. Hopefully the Zig team will care about A11y eventually, even if @andrewrk has already shown a stance that I read as being ableist and hostile, so I shouldnāt hope for any considerations.
People with disabilities deserve considerations, we feel excluded enough already.
That is not what tab stops were invented for. They were invented for manual typewriters to assist with generating a variety of fixed-format documents on paper. For typing up invoices, for example. True, the user upon entry can define them, but they are unchangeable once the document has been typed. Everything else is an evolution from typewriters to terminals to GUI editors where new functionality and interpretation of ātabā evolved.
I agree with Andrew that itās code editors and display systems that need evolution. I find it amusing that people hell bent on bringing back tabs grasp at any new argument, accessibility in this case, to relitigate a decision long since made.
These are my own impressions, should not be attributed to anyone else, but beyond the simple explanation that more people may just disagree with your opinion on zig fmt, it may have to do with your communication style. People disagreeing with your positions on the topic seem to be interpreted by you as āattacksā. You would rather argue over the use of a semicolon or the third most popular formal definition of a word to āproveā that you technically didnāt say what someone gleaned from your reply.
This is an informal place, not a courtroom. Your precision in language, grammar, and punctuation is not paramount here. If someone interprets you incorrectly, just explain that wasnāt your intent, and restate it. You donāt need to explain in detail how they were wrong to interpret it that way. This is simply an abrasive communication style, and mentally exhausting to participate in.
You are not a victim, and nobody appreciates having attempts at their empathy being weaponized against them. We can discuss the pros/cons of the topic without it devolving into this nonsense. You simply have the minority opinion on what ought zig fmt do, nothing more. No reason to get all existential over it.
Yes, tab stops on typewriters were made to assist in writing up tabular data by helping with alignment, so that each column actually begin on the same level instead of needing to do this manually.
When automatic input was invented when the very first computers came around (which either outputted in punch cards, lights, the print part of a typewriter or a combination of these), that led to the ability of changing where the tab stops are and with that making tables more ādynamicā. So if you suddenly needed one column a bit bigger, you could easily do that.
Indentation is just a form of alignment.
But either way, using spaces when you want to have the ability for users to decide the indentation level is still stupid. Spaces are to convey a fixed amount.
There are even company coding standards who give as a reason for using spaces that they donāt want employees to be able to customise the rendered indentation level. They would consider an editor where you could modify the width of a space an anti-feature, even if that inability is a problem for people with disabilities.
No, really, thatās why people use spaces in Java (which was the first PL which seemed to actually have an opinion on this question from what I can tell).
I have heard from some people that some extreme companies donāt even let you chose font, syntax highlighting (colour) or even font size (the latter of which I am quite sure violates some anti-discrimination laws in some countries).
There are some reasons why one might want to use spaces over tabs besides this, but these always boil down to bad tooling (or using a mix of both for indentation, but that this is a bad idea is hopefully obvious).
So to sum it up: use tabs if you want the reader to be able to customise, use spaces if you want the author to be able to customise.
But since we use formatters these days, the author canāt actually customise these days anyway.
VC money