Hacker News Reader

Comment by andirk | Hacker News Reader

It often outranks MDN official docs which these days are very readable, in-depth, with examples, and a simple UI and the only time I click to inappropriately named W3Schools is by accident because it's the top result and I'm expecting MDN and then I click back and that's why I don't like it.

View on HN

Nothing about that makes w3 a poor source. I get why you block it - so MDN is shown first. That’s fair. Im just saying its not a fault of W3. Because I tend to agree with that guy. W3 really isnt bad. A bit simplistic, and MDN is better, but its not bad.

W3 has a bad reputation because, at least a long while ago, it was rife with blatantly false or misleading information.

There's literally working code demos for everything on the site that's front-end or client-side.

Yes, and historically those code demos were often wrong in many important ways (like not working in all browsers, or insecure). They are much better now but reputation is hard to recover.

That must have been like 12 years ago or more at least because in my career I don't have much memory of feeling betrayed by the information on their site.

Yes some people have been doing web development for more than 12 years (:

https://web.archive.org/web/20110412103745/http://w3fools.co...

It's exactly 12 years ago. I started learning JavaScript in summer 2011 and I remember w3school was often misleading.

Wouldn't that have been the case everywhere before JS browser support standardized to ES 2015 and above?

It's a question of presentation. Back then (and to be clear, I'm talking about 00s here) the correct way was to point out that something might not even be doable in cross-browser way at all, and listing the proprietary features that'd let you do it in specific browsers while clearly labeling them as proprietary. Quality websites would do that, while w3schools and the likes would just give you the snippet for (usually) IE. Worse yet, in many cases they would keep suggesting the old proprietary approach even after a new standardized way of implementing something showed up.

Cross-platform demos?

That's why Kagi's personalization is so cool here. Kagi doesn't have to block it, and neither does the user. The user can just lower it, and raise MDN, because that's what they want in their search, and it helps them more.

Yeah the simplicity is what I prefer about W3schools. A lot of times I want a quick code example and don’t really want to go in depth about how something works.

> Nothing about that makes w3 a poor source.

I'm sure you meant w3schools but w3.org is where a lot of the official documentation for web technologies is and not affiliated with w3schools.

I have similar feelings about baeldung.com. It isn't that the content is inherently bad, but it is super annoying that when I search for a java class the first few results are from baeldung instead of the official javadocs.

I think the java official docs are rarely explanatory, they are just plain description with a lot of fluff and useless information

Yeah. Search for class, open official docs, read very short description. And still have absolutely zero idea is where this class is used, and how it should be created, and what alternatives.

How wonderful to differ - I actually love bite sized, "tldr" or baeldung and almost always get what I need. (not to mention that quite often javadocs of many project are just empty or severely lacking... )

Reply on news.ycombinator.com