4 ms·
A lot of this boils down to: it's hard to follow links. Someone's massive Twitter following is simply a large list of links to follow up on...and determining t
by danso 14y ago
A lot of this boils down to: it's hard to follow links.
Someone's massive Twitter following is simply a large list of links to follow up on...and determining their fakeness is algorithmically but who has the time to check each follower's activity level? Most social media "professionals" I know have no clue except to trust a paid-for black box analysis service.
And many well-accepted Wikipedia articles have plenty of dead links in their footnotes, even though such cruft should be easy to red flag. What's harder to diagnose are footnotes that refer to information in print. If I told you that, among my accomplishments, the story of how I built my fortune is referred to on pages 106-118 of one of Malcolm Gladwell's early, more obscure titles...how many Wikipedia editors would bother to look that up? Or maybe I just make up a book and enter in a random ISBN.
It's still all-to-easy to avoid examination by promising that the original source of the info is just one click away. As long as you don't inspire anyone to vet you out of jealousy or spite, you could go undetected for a long while, despite the power of the hyperlink
- Wingman4l7 14y agoFWIW, there is at least one bot on Wikipedia that crawls for dead reference links and marks them as such.
- eridius 14y agoHow often does it crawl? Just this week I tried to look up a reference on a page and found it dead, probably for quite some time (I forget the page I was on).
- Wingman4l7 14y agoWell it turns out I may have been mistaken; the bot that I thought did that is defunct, and I cannot easily find another that does that. After thinking about it, I can see some reasons why you might not want a bot automatically marking links as dead. There are these, which I think are a bit more manual in nature: http://en.wikipedia.org/wiki/Wikipedia:Link_rot#Bots http://en.wikipedia.org/wiki/Wikipedia:Link_rot#Bots
- emiliobumachar 14y agoPlease share those reasons. The only one I can think of is false positives of temporarily down servers, which could be mitigated by timed retries.
- Wingman4l7 14y agoThat is the main reason. Another would be some sort of "Content not here anymore" message that doesn't conform to HTTP status code standards, giving a false negative.
- ChuckMcM 14y ago"A lot of this boils down to: it's hard to follow links." Its not that hard. We have a search engine which does it all day and all night, billions and billions. But it does make people less willing to engage. I've speculated for a while now that you could do a twitter equivalent of hellbanning where people who were hellbanned saw their robot followers but people who were considered 'legit' saw the number that twitter actually thought were people. If you just blow away the bot followers then the people who have them will be disillusioned and not actually use twitter or point people at it, so it is in Twitter's interest to have them out their pushing the story 'follow me on Twitter, everyone else does!' even if that is fake. Hopefully those people will find someone real to follow and become customers.
- nano111 14y agoit probably is more profitable for twitter not to get rid of most fake accounts