Hacker News Reader

Comment by e28eta | Hacker News Reader

And in the age of Google, where you can just put quotes around the first sentence and find the original source, it shows a complete lack of care.

View on HN

Once Google indexes Borges' Library of Babel, though, you will find that any sentence is already used, and every article is already written.

On a more serious note: it would be interesting to add a check to CMS systems to automatically look up all sentences in an article before submitting and apply information theory to compute a "plagiarism" coefficient. No doubt these kind of systems already exist to detect student plagiarism.

The software should already be there, I've seen it mentioned many times over the years.

If one of those software producers would provide an API (something similar to Automattic's Akismet for spam) I'm sure there's money to be made in selling that service outside the academic realm.

Indeed, my university uses a service called Turnitin to rate the percentage of coursework that is plagiarised, along with giving where from etc.

It's pretty nice actually, can recognise when things are quoted and not passed off as your own.

My high school actually used this too - my teacher once showed my a sample report, and it was actually very good at detecting plagiarism even with words changed, sentence order mildly munged, etc. (with percentage probabilities). They seem to have a fairly good model of what kind of changes writers make when copying material.

Good point! Specifically, reputable journalistic outfits should implement a policy of running their materials through TurnItIn before publishing. This would allow editorial control over plagiarism. Any journalists want to comment on whether or not this is existing policy and why?

If you have to have your writers run their material through something to make sure they're not stealing it from somewhere else, shouldn't hire them in the first place. Really, most journalists would never do this.

We run our writer's material through spell checkers and grammar editing. We run it through fact checking processes. In theory, a good writer would never need any of these, but in the real world it is a useful step. If we do THESE things, then why not a step that requires nearly no effort (just use an automated tool) and which WOULD catch some cases that have affected even major publications with excellent reputations?

Reply on news.ycombinator.com