Hacker News Reader

Comment by shiado | Hacker News Reader

While they are at it please remove LinkedIn. You can't actually view anything on any page without logging in. Why is it listed at all? Seems pretty deceptive to me.

View on HN

Oh and w3schools.com needs to go below MDN because it's less informative.

Even today I wanted to read about transform-origin and landed on w3schools [0], where MDN article [1] is clearly superior.

[0] https://www.w3schools.com/cssref/css3_pr_transform-origin.as...

[1] https://developer.mozilla.org/en-US/docs/Web/CSS/transform-o...

Perhaps, but at least they don't require you to signup just to view a stupid image.

Does Google manually adjust weights? While I agree with you, I always supposed that Google weights websites according to their own algorithms, not because some moderator thinks that it's less informative. If w3schools is higher, probably there are technical reasons, for example more users prefer it (may be content is easier to understand).

They apply “manual actions” all the time.

Sure, so they need to improve the algorithms in this way so they produce the desired results. In other words, algorithmically driven manual actions.

> The Manual Actions report lists instances where a human reviewer has determined that pages on your site are not compliant with Google's webmaster quality guidelines. Google's algorithms can detect the vast majority of spam and demote it automatically; for the rest, we use human reviewers to manually review pages and flag them if they violate the guidelines. Flagged sites can be demoted or even removed entirely from Google search results.

https://support.google.com/webmasters/answer/2604824?hl=en

Specifically mentioned on that page is "Redirecting users away from the image on the site when the user clicks "view image" in Google search results" as an example of "image cloaking behavior".

When searching for things like that, I usually prefix my search with MDN, or even better, just use ddg for technical searches. These days, ddg is getting pretty good at showing results of typical coding/technical questions. If my search is for a very fringe/specific/new topic, I fall back to google (which you can from ddg with the `!g` prefix)

I use the personal blocklist extension (https://chrome.google.com/webstore/detail/personal-blocklist...) just to block w3schools.

Exactly the same here!

Back when I was writing a lot of C++ professionally I felt the same way about cplusplus.com vs. the (IMO much superior) cppreference.com. I'd love to be able to teach my search engine simple rules like "if a Wikipedia article matches my search, show it first" or "never show results from linkedin".

While we're in the same vein, Google should remove the dumbed-down content altogether for anyone flagged as a "developer" by their algos, and replace it with man pages and source docs. MDN is a step up, but I usually skip past it, because the actual specification papers are most always better (except for ATOM and OpenLayers!).

Generally I prefer specifications too, but ever since the WHATWG-W3C split, it's become a choice between stuff that isn't yet implemented (WHATWG), and stuff that never was implemented (W3C). At least MDN documents accurately what one major browser supports, and includes compatibility tables for others.

I know it’s not a popular opinion, but I actually like W3schools. Most of their infamous errors are gone and they are usually more “to the point” than Mozilla docs, which are a sometimes too prolific and not as clear.

It's "to the point" in the sense of being a more concise basic introduction. But MDN hits the right balance of being comprehensive yet accessible in a way nothing else does.

Plus W3Schools' very name is an SEO hack, so their Google rank is built upon fundamental dishonesty.

How is "w3schools" an SEO hack more than any other name?

Because it implies an association with the W3C

The biggest problem with that site is that it's sometimes good. Sometimes. I never know that before I click if the required info is there or not or it's good enough or still relevant. With MDN my expectations are high but they are always matched.

That's why I installed the W3Schools removal chrome extension :)

All my web platforms searches are now "mdn <web-thing>" cause w3schools is never informative enough for me.

I actually have a couple of bash scripts just for this when I want to look something up specifically at mdn, and thought about expanding it to stack overflow and reddit.

What's that? Why yes I have heard the phrase 'bikeshedding', why do you ask? ;)

DDG's `!` searches are your friend:

`!mdn whatever` will search MDN. Similarly !so, !reddit and a whole lot more.

https://duckduckgo.com/bang

I like W3schools too, you're correct it's better these days. The site has helped me a lot when I needed to quickly get help usually with CSS or HTML. I like the "try it yourself" feature, and the examples are usually good. Not sure how good it is for JS and Jquery, I usually use Jquery.com and stackoverflow for JS.

The examples w3schools have are useful, and MDN has less of them. But maybe that's my subjective experience. Still, I've made lots of use from both of them, and looking past the first hit should not be that difficult, especially for us programmers.

Edit: funny that people downvote this.

As a noob I actually prefer w3school's page to start ... and then I'd read MDN.

Maybe the noob clicking behavior is influencing the ranking?

Both are good and I use both. There's something about the layout of w3schools I find nicer and quicker to read. Sometimes I spend about 10 seconds on the site and I'm done. That's the mark of a good UI when you're in and out in 10 seconds with the info you need.

LinkedIn is one of the top sites for dark UX patterns. It is 100% purposeful, just like Pinterest, and designed to increase the all important "engagement" metric.

Does anyone actually ‘engage’ in way other than trying to block their spam? Every 6 months or so they seem to break through and the crap piles up. I was just fighting them off this morning.

The most important rule to use LinkedIn is to never ever log in. Or if you do, do it on a computer and shhh stop pressing those buttons. Just do the minimal and leave.

I think so. Not many, but I have seen friends make posts on Linkedin.

IIRC, they were about some technical topics, similar to what you'd expect to see in a blog. Why they close to use Linkedin I don't know, but apparently it does happen.

Yes, 100s of millions of people around the world.

Millions of new posts, job searches, private messages and other activities take place everyday. They do have plenty of spam and a terrible UX but that doesn't mean the site and community is anywhere near empty or useless.

People take spam for granted. Most people have more unread spam in their gmail mbox than _all_ of my mail archive.

As a non recruiter that isn't looking for a job I just find it useful for looking into people from 3rd parties I am currently working with.

Things such as who else they work with, their exact position in their company, where they have worked in the past.

All the social features bolted on are useless to me.

And how about Google finally do something about Quora's aggressive abuse of their quality guidelines:

https://support.google.com/webmasters/answer/35769?hl=en

Quora presents a very different site to Google than it does to the reader ultimately. Google intentionally, knowingly ignores this fact and has for years. They could manually penalize Quora for breaking Google's search quality guidelines in numerous ways and they choose not to.

A user can't freely browse Quora page to page, whereas Google can. Presenting a different site to Google than you do to the user, is about as direct of a violation of their guidelines as you can get.

AFAIK, you always get the same page that was shown when coming from Google results. That IMO is enough - what is shown in the search results is an accurate reflection of the page you reach. Requiring that subsequent navigations within the site are also the same seems excessive (as you don't have the context of a results page that would make it seem misleading) and would be an overreach on Google's part IMO.

The Google spider has an entirely different experience, when going from page to page on Quora, than what the user does (someone not signed in).

Google hammers any normal site that gets caught presenting a fundamentally different experience to the user than they do the spider. Google is not unaware that Quora, Pinterest and LinkedIn present different experiences to the bot vs the user, they knowingly allow it for giant services. That's the hypocrisy and double standard.

That’s because the policy isn’t derived from first principles, it’s a flexing of relative muscles. Google knows that small sites don’t have any choice but to comply with their guidelines. They also know that if they start filtering out LinkedIn results, if people are looking for something they expect to find on LinkedIn, they’ll go and search LinkedIn, possibly even before they search Google. Google will not allow that to happen.

Also, Google doesn’t also compete with those companies. I would bet if it were someone who they were competing with on another vertical , they’d be enforcing this policy.

Quora is shit anyway, StackOverflow&Co is the superior platform.

I agree generally speaking. Quora has raised $452 million in venture capital. It's incredible frankly. There's no actual sound business model there and never has been (which is why eight years later they still don't have a functioning business and are living on VC, while being subsidized by Google abusing its own search policies).

A lot of their funding is from Adam D’Angelo’s personal piggybank.

No sound business model? Quora has a sound business model. The same model as many VC backed SV companies.

No, it's not to simply sell ads. You build something slick and shiny, then you pump VC money into it and you keep hyping (otherwise known as "pumping") until you can get numbers that look good. You can just spend that sweet, easy VC money to buy users and spend a dollar to make 10 cents because of that "growth" you are buying.

Then, when those numbers you are buying (and also all those fake accounts you are ignoring) start looking really good, you go public, the VCs make out like bandits and all those institutional and private suckers buy those bloated shares, trade places with the VCs and slowly ride the shares back down to zero where they belong.

Do you have an example? You speak as if this is a very common pattern, but I can't think of any good examples. I can think of some that look like they might fit this pattern if they were to IPO, but that's a big if.

snap

Groupon IPO'd at $28 a share. Currently trading at around $4

Zynga IPO: $23. Now around $3

Snap $22 ... ~$16 (LA not SV)

Blue Apron $10 ... ~$2

Box plummeted more than 50% before coming back up closer to the IPO price but has almost never exceeded the IPO price since inception.

And there are a lot more that are newer and still riding the capital injection from the IPO (like twitter).

> Quora has raised $452 million in venture capital

Wow, really? I figured they're a bit fancier, more serious version of Yahoo Answers. I wouldn't have thought they'd have any venture capital at all.

if you 3rd party log in w/ your google account you can browse pages the same as the spider. quora knows the google spider is the google spider, so they've got a pretty good argument of equivalence there... plus you can still open links in a new incognito tab if you want to bypass registration like I did for years (kinda clunky though, & I was definitely feeling annoyed w/ them at times too)

Can we also ban Elsevier and get Google to replace their links with Sci-Hub links?

ScienceDirect articles shouldn't be indexed unless they have a readable abstract. I realise that isn't quite where you're going though ;)

On the version of Internet from the 90's?

Totally.

Could anything like that happen on the CommerceNet, post EME (Encrypted Media Extensions)? (Sept, 2017)

Hardly. (And your name is now on a list for suggesting something so subersive, citizen.)

Google didn't exist on the version of the Internet from most of the 90's. :)

Those few years of overlap were epic.

Fairly sure google isn't going to promote pirated content over above-board content, regardless of how a large part of the community feels about it.

It’s also a trap if you forget you’re logged in and the user immediately gets notified you viewed their profile.

There's an 'anonymous mode' setting which doesn't tell people when you view their profiles. I'm not sure how it works vs. the 'see who viewed your profile' perk that they offer in their premium accounts.

If you come with a google referrer you can still see the basic information of a profile.

I haven't found this to be true in the last few months

Even if that's true, that means LinkedIn is presenting a different site to the reader than they are to the search engine spiders, which breaks Google's guidelines and should be heavily penalized accordingly (as it would be for any mortals).

Google can spider every page freely, going from link to link without having to sign up. A regular user can't. Where's the penalty from Google for that? Why is so much LinkedIn content indexed when the user can't have the same experience as Google does? It's very blatant hypocrisy.

Agree. On my profile page I have option to make it ‘public’ but still requires login to be seen. It’s completely deceptive imho.

There is an extension for Chrome named "LinkedIn for Unregistered Users" that works on the basic profile page.

They didn't used to require this, unless I'm mistaken that started post Microsoft acquisition.

> You can't actually view anything on any page without logging in

I believe you can, actually, in some cases. It's just most of the time they try to lock you out with a login screen.

And while they're at it, any news sites that hit you with a paywall when coming from Google results.

Sometimes you can, and find out personal information or that used to be the case not long ago. I don't know under what conditions it takes but it's a bit like some news paywalls, I think. Incognito tabs won't work.

And Facebook.

Reply on news.ycombinator.com