Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

Wikipedia is 20

economist.com

Post reply on HN
Print view View original

281–290 of 391 posts

Re: Wikipedia is 20

#281

Post by 70rd »

nsajko wrote: post #143
disown wrote: post #129

Earlier quoted context omitted.

> Wikipedia is incredible. It's good for surface level stuff. The apolitical superficial knowledge is where wikipedia excels. > I find it frustrating when friends say it's unreliable/not to be trusted. In terms of depth and breadth of knowledge of topics, it is sorely lacking. And it is highly untrustworthy when it comes to anything remotely political/historical/economic/etc - which are ultimately all political. > Of…

Doesn't even have to be political. Moneyed interests and fanboys and cultists use Wikipedia for PR and similar all the time. See Falun Gong, for example. There are also groups of Wikipedia "editors" who coordinate off-wiki to influence its workings. (Sock-puppeting and meat-puppeting) Even if we disregarded all the malicious actors, harmful incompetence is rampant on Wikipedia. And I was just talking about the Englis…

What's the issue with the page on Falun Gong?

Re: Wikipedia is 20

#282

Post by aerosmile »

somedude895 wrote: post #177
creato wrote: post #72

Earlier quoted context omitted.

It is somehow sad that "caring about content quality" is considered SEO and not just making a good website. I think more than half the things you mentioned are only good SEO because search engines want to send people to websites they will like reading. I think when that is the case, we should be crediting people for making good websites, not good SEO.

Yeah, looking at it from a purely SEO perspective always makes decisions look kinda shady, as in they're trying to game some algorithm in order to gain more exposure. Wikipedia's biggest SEO factor has to be the massive amount of backlinks it gets. Those are purely organic and happen simply because it is generally the best authority on most any topic. It's more of a testament to how genuinely good Google's algorithm…

Once you have 150 million inbound links, the strategy choices are easy - focus on content quality! But you have to remember what Day 1 was looking like: a tiny community, no inbound links, and a fair amount of other encyclopedia competitors trying to attract authors. Now the strategy choices are quite interesting - at the beginning of a startup you have just enough energy/runway to "kill it" in one area. Which one do you focus on? Generating the world's best content alone without the heavy lifting they've done on deep interlinking and other SEO-friendly moves wouldn't have cut it.

Re: Wikipedia is 20

#283

Post by concordDance »

nsajko wrote: post #144
libraryofbabel wrote: post #63

Three aphorisms in honor of Wikipedia, greatest encyclopedia in world history, and its 20 years of free knowledge uncorrupted by advertising: * Wikipedia works in practice, but not in theory. * Wikipedia is the worst source of information, except for all the others. * Wikipedia: the Internet’s greatest reason to feel a little bit optimistic about human nature.

Here, I'll fix that for you: * Wikipedia works in theory, but not in practice (as soon as you scratch the surface). - The problem is that there are many good pages, but that just lulls one out of the necessary skepticism. * Try the Wikipedia sources that are hopefully on the bottom of each page instead. Also search on Stack Exchange and Reddit for book recommendations. * Wikipedia: the Internet’s greatest reason to f…

Providing a link to further information, particularly on the disinformation, would help.

Re: Wikipedia is 20

#284

Post by SilasX »

creato wrote: post #72
aerosmile wrote: post #40

I am surprised that the comments haven't mentioned the role of SEO in Wikipedia's growth and defensibility. Wikipedia's habit of deep interlinking helped it rank back in the early aughts when the SEO rules were rather simple. Add to that the subdomain-driven localization strategy and many other moves that were considered SEO best practices back in those years when the on-page factors used to matter. But that was just…

It is somehow sad that "caring about content quality" is considered SEO and not just making a good website. I think more than half the things you mentioned are only good SEO because search engines want to send people to websites they will like reading. I think when that is the case, we should be crediting people for making good websites, not good SEO.

Equivalently, the challenge in running a search engine is to decrease the divergence between “what makes a website good” and “what makes us rank you higher”.

Re: Wikipedia is 20

#285

Post by aerosmile »

pvorb wrote: post #169
creato wrote: post #72

Earlier quoted context omitted.

It is somehow sad that "caring about content quality" is considered SEO and not just making a good website. I think more than half the things you mentioned are only good SEO because search engines want to send people to websites they will like reading. I think when that is the case, we should be crediting people for making good websites, not good SEO.

Yes, I believe that Wikipedia and its authors never put much thought into SEO. They just think about how to best structure the information and make heavy use of links, which also happens to be a good strategy for SEO. The Google search rank algorithms changed a lot more than the overall structure of a good Wikipedia article in the last 20 years.

Have you ever encountered on Wikipedia a sentence like this:

"...because a blue whale did..."

rather than

"...because a blue whale did..."

Obviously, the latter version would have been more useful, and I find it difficult to believe that a human being would have made such a mistake. Don't get me wrong - such instances are rare, but they do happen and are an indicator that not all links are generated manually. I don't know what they are using today (if anything), but as someone else pointed out, in the early days they used UseModWiki to ensure a high level of deep interlinking. We can argue that this was done to improve the UX, but the level of ambition that went into it signals that they also saw it as a strategic move (and they would have been right to assume that - 20 years ago, a highly interlinked site was likely the best bang for the buck in SEO when it came to how to prioritize your time and resources).

Re: Wikipedia is 20

#286

Post by mvc »

ricardo81 wrote: post #210
tshaddox wrote: post #105

Earlier quoted context omitted.

Just an anecdote, but those are the top 2 and only 2 things I know him for. Didn’t know (or at least remember) that he created Brave or was involved with Mozilla. I’d be willing to bet that, indeed, those are the top 2 things he is known for.

Same for me. Definitely would be an interesting family fortunes/family feud question

It would have to be a special "software engineer" edition of family fortunes otherwise 99% of contestants wouldn't have a clue who he is.

Re: Wikipedia is 20

#287

Post by progre »

chrisbolt wrote: post #115
aerosmile wrote: post #40

I am surprised that the comments haven't mentioned the role of SEO in Wikipedia's growth and defensibility. Wikipedia's habit of deep interlinking helped it rank back in the early aughts when the SEO rules were rather simple. Add to that the subdomain-driven localization strategy and many other moves that were considered SEO best practices back in those years when the on-page factors used to matter. But that was just…

> So when Panda came around, Google correctly rewarded Wikipedia with #1 rankings for over 50% of its terms (!!), and Jason Calacanis had to shut down Mahalo which got destroyed by Panda. For context, this is the type of content that Mahalo was producing to try to game SEO: https://www.youtube.com/watch?v=vdNk1xmDpxo

Game how? Pretty girl says she will show how to mix a drink. Then proceeds to show how to mix a drink. Not ending world hunger or anything, but nothing deceitful here at least. Maybe I'm missing something

Re: Wikipedia is 20

#288

Post by concordDance »

ant6n wrote: post #219
Acrobatic_Road wrote: post #83

Earlier quoted context omitted.

It was even worse before. When I read it, the opening line suggested that all of the rioters collectively murdered that cop. I think that was a little too biased for wikipedia. Does anyone have the revision?

„(...) Summoned by Trump,[25] thousands of supporters gathered in Washington, D.C., on January 5 and 6 to demand that Vice President Mike Pence and Congress reject Biden's victory.[26][27][28] The rioters quickly became extremely violent, assaulting a police officer who later died, erecting a gallows on the Capitol grounds, assaulting the press, and desiring to take hostage and harm lawmakers such as House Speaker Na…

Yeah, a good example. Having seen 50 solid minutes of raw footage, describing it them as "extremely violent" is about as accurate as describing BLM protestors as criminals (e.g. it describes a very small minority).

Re: Wikipedia is 20

#289

Post by selestify »

aerosmile wrote: post #282
somedude895 wrote: post #177

Earlier quoted context omitted.

Yeah, looking at it from a purely SEO perspective always makes decisions look kinda shady, as in they're trying to game some algorithm in order to gain more exposure. Wikipedia's biggest SEO factor has to be the massive amount of backlinks it gets. Those are purely organic and happen simply because it is generally the best authority on most any topic. It's more of a testament to how genuinely good Google's algorithm…

Once you have 150 million inbound links, the strategy choices are easy - focus on content quality! But you have to remember what Day 1 was looking like: a tiny community, no inbound links, and a fair amount of other encyclopedia competitors trying to attract authors. Now the strategy choices are quite interesting - at the beginning of a startup you have just enough energy/runway to "kill it" in one area. Which one do…

What were the competitors in the early days?

Re: Wikipedia is 20

#290

Post by concordDance »

Merman_Mike wrote: post #59
admiralspoo wrote: post #23

Its reputation is severely declining, especially on topics relating to current events, figures, or politics. It, however, remains excellent for technical knowledge.

This is a current event that is extremely political: https://en.wikipedia.org/wiki/2021_United_States_Capitol_pro... It cites 462 sources and the parts I read are at least as objective as any news outlet. I know that Wikipedia editors lean left. I'd be interested to see examples of this bias in Wiki entries though.

Yikes, yeah, that's a good example of bias. The main bias being describing a group doing something that only a tiny minority did. It's not as bad as saying things like "Muslims then bombed the world trade center", but it's still pretty bad.

Based on the 50 minutes of raw footage I've seen, a description of "mostly peaceful" wouldn't be too far off.

Post reply on HN