Social bookmarking. Seems a new user-driven content site pops up every day. And it seems that now we have 100,000 social bookmark sites, which are not only repeating each other’s content, but the same dumb mistakes over and over again. Everybody wants to be the next Digg, Reddit, or Netscape.

What we need is one quality bookmarking site. A site which:

Automatically checks other bookmark sites for updates. The same link gets grabbed from site to site anyway, often with the submitter too lazy to even change the headline or the description. So it might as well aggregate all the RSS feeds and run them for you. Yes, we already scraped the 5000 other news sites. Do you have something new to show us?

Automatically check the URL for dupes. The dupe thing is serious. No matter if it’s been 1000 years ago, the link gets submitted once in history and once only. An attempt to submit a second copy automatically transforms into a vote for the first time it was submitted. Ditto for linkjacking; check the URL linked from the post in the submitted URL. Thanks, we saw it already.

Automatically check the headline for duplicate stories. You know how some B.S. meaningless celebrity gossip event happens and the entire Internet has just one story repeated over and over for 24 hours? It’s easy, just compare the words in the headline and the first paragraph from the linked content. Too much redundancy = rejected link.

Automatically check through Copyscape.com for blogspam. The story you submitted has copied content from here, here, and here. Perhaps offer to post the original site that first had the content if it has never appeared before. But this would be one site that doesn’t support plagiarism.

Automatically check through the Wayback machine for ancient history. Thank you, we saw the dancing hamster site back in 1995. Next?

Automatically check spelling and grammar. Also have a gibberish filter. Type in ALL CAPS, 13375|)34|&, or spel lyk ai morron and your post will not go through. Fix it, learn to type, or let somebody else submit the link.

Automatically crawl and copy the linked-to site to a mirror server BEFORE it gets posted. This would mean no more stories with thirty people able to see it before the server crashes and leaves 5000 more people to each check and then post a comment “The site is down.”

Each of these filters would be quite possible to implement with the tech we have today – I can’t see any of these that would be beyond a LAMP server and some not-particularly-advanced coding. Why hasn’t anybody thought of them yet? I’d do it myself if I could afford the bandwidth.

All of the above modifications are based on years of observing the comments on bookmark sites. People are sick and tired of seeing a site get linked and immediately go down. They’re fed up with duped stories. They’re tired of “OMG! LOOKA TH1S!!!!!” titles. There’s a fortune waiting for the first web entrepreneur to solve all of these problems…

Popularity: 2% [?]

Related Posts:

• Laws of Web 2.0
• Digg Friends
• The Most Exotic Computer and Internet History Links
• Top 10 Reasons Not To Permit Duplicate Submissions On A Social News Site
• Anatomy of an Internet Hoax – Your Thumbs Work Fine for the iPhone