Maybe, but to tell a story, these projects ended up making flights to a non-existent place.

I'm from Bermuda, a small group of islands in the Atlantic ocean.

What's probably now approaching a decade ago, we were alarmed to see airlines starting to sell tickets to "Greater Bermuda". This isn't a thing that exists - and initially it was assumed that someone at the airline had made some kind of mistake.

Over the next six months since we first saw this, the concept of "Greater Bermuda" and "Great Bermuda" started cropping up all over the place, eventually making it into Google's knowledge graph as well.

At this point, and encouraged by some fellow Bermudians, I dug in. It turns out that this all started as a fiction Wikipedia article written on the de subdomain. At some point someone translated all these orphaned articles to English bringing them over to the main wikipedia.org. Sometime later, an editor came along and merged several Bermuda articles into the Greater Bermuda article, and then various knowledge bases that absorb data from Wikipedia started aligning to this new "reality". Some time after that, various airline databases also started being reseeded with metadata from these data sets.

I happened to work for Google (still do) and was able to start filing bugs. At first, most were ignored/discarded, but after some persistence, both on Wikipedia and inside Google, I was able to start removing some of these invalid sources.

As for the data, well, if you fly into Bermuda, you'll land in St. George's, which is absolutely not the larger of the islands, though if it's cupmatch (https://www.gotobermuda.com/article/cup-match-time-bermuda), some say it is the greater.

These projects are successful, yes, and they're mostly reasonable data sets, but if you are a motivated adversary, or an innocent fallible, you can use these projects, directly or indirectly to make significant change. While this story didn't have much more than a cosmetic effect, some of the folks prodding me at the time predicted worse outcomes had the situation continued. Bringing the issue closer to home for most here, integrate this example with the social political strife and active participants, and you may question whether these projects are even a good idea. Perhaps we could instead once again employ people to write encyclopedias, and fact check? We sure need the jobs, and there's plenty of money around.

> Perhaps we could instead once again employ people to write encyclopedias, and fact check?

We have a couple of centuries' worth of evidence that this system is subject to exactly the same kinds of flaws, along with the downside of them being much harder to correct.

As always, there's a relevant XKCD

[1] https://xkcd.com/978/

And funnily enough that problem also happens for encyclopedias. Check the article about airplanes/flight in any old encyclopedia and you will probably see this incorrect theory about how airplanes can fly:

https://www.grc.nasa.gov/WWW/k-12/airplane/wrong1.html

I suppose most of the people writing encyclopedias check their facts in other encyclopedias.

Also a lot of people working for NASA check their facts from those:

https://www.grc.nasa.gov/www/k-12/UEET/StudentSite/dynamicso...

The irony... I guess this proves that inter-team-communication is more difficult than rocket science...

True, but i feel like central repos like wikidata are quite a bit different from the original vision of many small websites all interoperating into a semantic web.

Then again the same thing happened with the normal web. We have wikipedia instead of thousands of people writing articles on their personal homepages.

But Wikipedia isn't a primary source and presumably it could (and does) link out to those personal sites.

Unfortunately you then run into "reliable sources" and "notability" battles with other wikipedians over citing personal websites.

For reference, official rules are https://en.wikipedia.org/w/index.php?title=Wikipedia:RSSELF

> True, but i feel like central repos like wikidata are quite a bit different from the original vision of many small websites all interoperating into a semantic web.

One of the main roles for Wikidata is to act as a directory for these "many small websites" that enables them to interoperate seamlessly. You can look up a real-world entity on any website that Wikidata supports and use a third-party service ("Entity Explosion", see https://www.wikidata.org/wiki/Wikidata:Entity_Explosion ) to get links to the same entity as it appears on Wikidata and other sites.

What is IPFS though if not a glimpse of the future? Distributed makes more sense to me and P2P concepts are being heavily researched with all the blockchain hoo-ha!

A proof-of-concept that's generated a lot of buzz but not a lot in terms of concrete or unique usage?

How do people remember specific years for such ordinary events as lunch, so long ago? My mind simply doesn't tag things that way.

Usually there are a few life events to anchor it to (school graduation, marriage, child birth, moving, changing jobs, etc.)

now: google calendar, search for 'lunch with peter norvig'

2002: maybe iCal

Not to mention schema.org. These are used heavily for Google search integration (reviews, business information, show times).