This means links get a semitransparent underline normally, and full-opacity on hover. I reckon it’s an excellent balance. Occasionally you’ll get double underlining due to people using border-bottom instead of text-decoration for some unfathomable reason, and occasionally there’ll be link-styled buttons that won’t get this underline, but all up I’ve found it pretty good as an intervention.
(I’ve been using this technique on websites I make since 2019, though I haven’t yet had the opportunity to use color-mix() on a public site, which only stabilised in browsers 6–11 months ago. My preferred technique there will be `:any-link:where(:not(:hover, :active, :focus)) { text-decoration-color: color-mix(in srgb, currentcolor 30%, transparent); }`.)
The user stylesheet version is deliberately blunt, done in two rules to forcibly override whatever the author may have done, to both the normal and the interacting experience.
If you control the site’s styles, it’s better to keep specificity as low as reasonable (which I decided meant one pseudoclass—you could reduce it to zero by shifting the `:where(` to the start, but I decided not to for some reason), and only change what you need to change (which is the text-decoration-color—notably, not the entire text-decoration—on links that aren’t being interacted with.
Why? Because as well as being shorter and more conceptually elegant (and frankly, I place likely-disproportionate value on it being just one declaration, pride probably factoring in), it makes it easier to override when you want to. Suppose, for example, you want <a class="button"> to not get underlines because you’re making it look like a button (for better or for worse): you can write `.button { text-decoration: none; }`, just like you would have done previously, and it’ll work fine. Against the user stylesheet approach, even stripped of its !importants, you’d still get an underline on hover until you increased specificity (e.g. `.button.button`, and I’m assuming this is coming after the :any-link:is(…) rule, or else you’ll want `.button.button.button`) or reduced its specificity (which I grant :is → :where can do).
At least in my experience..if you can get a bunch of your own, often cynical (/tired of bs) engineers to say something you built is good, it's very different than the marketing or product people saying it's good..
Is it possible to teach llm to read issues and answer questions such as how do I get angular 17 to output to dist instead of dist/browser with you can't do this with angular. However, in your CI CD pipeline, you can do a mv or robocopy or bla bla bla depending on context?
It would be very useful because there are so many edge cases that aren't covered in the docs but are probably available in some kind of WON'T FIX issues.
Love how the AI hype went from "it's going to write code for you!" to "oh okay, maybe just boring boilerplate" to "it's basically a glorified search" (and it probably won't be great at that, either).
Yeah, it saves me time + effort, and lets me reach for larger changes than I otherwise would. It acts as a safety net / scaffolding as well as just answering Qs / writing code.
Copilot for docs is a horrible name for this demo. It feels more like a search engine concept, the knowledge base is up to date, and the answer is personalized.
But again, the naming is horrible, I thought is was a Copilot just for markdown.
I thought it was an autocompleter for writing documentation.
I am very disappointed that it isn't.
I had some success having ChatGPT document my code, it fact, it does a better job than I do by myself. Sometimes I have to fix some misunderstanding, but when it comes to writing and finding the right words, it is way better than I am. For me, it works better than for code generation, and I think that kind of "Copilot for Docs" would make a lot of sense.
Like many of those projects, access is controlled through a Discord server that is dedicated to more than just this prototype. I’m parsing and seeing things about getting CoPilot and VSCode to work together.
If you are eager to try something similar without dealing with a community, I’m assuming it’s possible to download the documentation from your favorite framework and start a “GPT” on OpenAI by giving it the role of a helpful coding assistant.
I definitely feel like there’s something to be done beyond the RAG aspect and figure out how to address the “other question” problem (you ask how you can do A, but that’s because A is your way to finding a solution for B, and there’s a much easier solution to B, but you don’t get the connection because you don’t understand the abstraction C). That would require a much more deliberate community to formalise properly.
I switched to Cursor (VSCode fork) a couple of months ago. It does this already and I haven’t looked back since. Especially as the copilot extension (along with all my others) work out of the box.
It has full knowledge of your entire codebase rather than being limited to currently open files (I assume embeddings with RAG) and will index documentation (or any other URL).
I find it especially useful for things like Swift/SwiftUI APIs given how quickly they
can evolve, not to mention copilot’s other limitations when it comes to languages beyond JS/TS, Python, etc
Who made this website? I feel like I am in a vegas casino and having trouble focusing.
It was unclear in the quick pitch but I am wondering if they essentially throwing RAG/search on top of a GPT model? I am guessing so because I cannot imagine you could train the model on such a limited source and get meaningful information. If thats the case perhaps this will be interesting but I think there are other interesting angles to this approach than focusing on a single library/codebase.
This is an excellent idea, but it feels like an obvious application of RAGs (or a trivial agent to build with OpenAI’s latest ‘GPTs’ tool). Not dismissing the necessary and valuable work—it’s “obvious” in the way sleeping after a long day is obvious.
What I’d love to see is such an agent *learn* from the interactions with people asking:
* What questions they ask more often than other framework, and how can the documentation clarify that (explicitly or not);
* Can you detect when questions betray a naive grasp of the problem and when the person asking would benefit from dedicated training, not just a quick answer;
* Can an LLM structure programming concepts and suggest ways to describe a framework that would help people make sense of what each framework actually is?
GitHub Copilot for Docs™ is an award winning, Gartner® Magic Quadrant™ leading, blazing fast AI solution that empowers business leaders to maximize their ROI by leveraging state of the art GPU enabled Hybrid Cloud infrastructure on the Edge to break knowledge silos and improve productivity at scale.
Don't think they know yet either. Still early days, they're basically just dangling the carrot out of reach. They're GitHub researchers based on their Twitter.
An alternative to reading the docs; you can ask questions about the problem you are facing, and an LLM (presumably trained on coding questions, and RAGged to the docs) will give you configurable answers.
I always assumed copilot and ai code assistants were replacing docs, for most use cases. I guess I'd rather just ask the question I have, when I have it, instead of having copilot preemptively create docs for questions it thinks I'll have later on.
Hate how they announce this product by shitting on the hard work of docs teams.
I get that docs are uneven in quality but there’s also been what I think is a renewed focus on quality & interactive examples. There’s some truly fantastic OSS documentation out there.
I don't think I've used it in ages. Maybe part of maturing as a developer, most answers I started to find there were expected and not covering the problems I faced, so usually original docs are the only reliable source, and maybe a GH issue to interact directly with the author. So I guess a tool to surface information straight from docs beats SO. Also won't complain about not having to deal with legitimate questions being closed due to "reasons".
Same for me. The issue tracker is usually the most interesting place to figure out why it's not working as expected while in the beginning of my career it was oftentimes SO.
If you're only using stackoverflow to have someone read the docs to you, then I feel you're missing the point. There is so much between the lines, and the LLM isn't 100% there yet.
