* The main page search bar is really choppy.
I typed in “lisp” and, when holding down backspace, the cursor skipped around.
* The website is unwelcoming.
If this is supposed to be a wiki, then maybe the landing would be better off if it just contained a logo, short description, search bar, and a random-article button.
What about "modern wiki" is interesting to you?
One thing that concerns me slightly is the presentation of opinion as fact, which to some extent is inevitable in a public wiki that has no citation requirement. There also seem to be articles with a wide variety of intents - there are informative articles like the keto diet instructions and the hierarchy of evidence, but also laundry lists like "no brainers" and "scientific study requests". I don't know if there is an intended purpose for the contents of the site, but having this mix creates some ambiguity in my eyes.
Overall, I'm keen to see where this goes - especially if it can help me to learn new useful ideas and practices (as opposed to edutainment-style wikipedia browsing).
Citations are currently encouraged, but not required. They might be required in the future.
If you were to pick a domain of knowledge to focus the site content on (rather than "everything"), what would you pick?
What new ideas and practices are you interested in?
The content I'm interested in is twofold - practical/actionable info, and high-level summaries. The former is somewhat covered by WikiHow but their content is notoriously vapid. There's nowhere currently that I'm aware of that I can quickly learn the general mental model for a new topic. If I was new to keto, for example, the article on your wiki would provide me with all the high-level knowledge I need to get started, and to roughly understand what's going on. There are numerous concepts that I want to learn about but often that requires buying 2-3 books or reading a lot of articles before I can start to understand the mental model. Self-help and business books for example are well known for stretching an easy-to-understand concept out into hundreds of pages.
I'd also suggest smaller default font. Sites like this benefit from high informational density (look at craigslist, reddit (old design) and wikipedia as examples). That's my vote anyway.
You may be right about the font size. For now, Grok's larger font size is deliberately lower density. It's meant to encourage conciseness, be comfortable to read, and avoid overwhelm.
One guideline I'm trying to respect is what UX people claim is the ideal text line width: 50-75 chars. Wikipedia, on other hand, is ~279 chars on my present screen (it scales with screen width), which feels awkward to me.
Search doesn't do anything when I type a term and hit enter.
The most impact will come from the community of contributors who are committed to these principles. I'll be supporting them.
It's a big challenge to maximize accuracy while allowing open contribution. We'll need to experiment and iterate on this to improve.
What are your suggestions?
Articles might need a warning before editing to warn users that certain edits or information will be reverted/removed.
For topics where the truth is a matter of opinion, the topic should probably just describe the different points of view, and qualify their use-cases and pros and cons, rather than arbitrarily picking one.
For the remainder, the side with the strongest references should win. I'll probably need to stratify the community a bit as Grok grows. Those with more influence should judge based on the site principles, similar to the US Supreme Court interpreting the law. I'm cautious to avoid fiefdoms, ulterior motives, and conflicts of interest.
I marked it as not spam.
By the way, I think the world can live with a wiki called grok and an unaffiliated app of the same name.