While I agree that originalworks is not very accurate, I can't imagine that a bot will ever be. Sure, simple copying can be detected with some effort. But with anything beyond that, you are inside a twilight zone already.
May be it would be a step in the right direction, if the originalworks comment clearly states that it is not a official seal from Steemit, if its that what some people assume.
You are viewing a single comment's thread from:
I don't think that's good enough. originalworks needs to not deem things as original. It should rephrase it's validation with a disclaimer to say something like, "I'm correct only 60% of the time--and I am not a Steem supported app" or whatever.
Oh I agree that there is no bot that can be 100% accurate, but at least the other plagiarism detection bot doesnt claim to certify posts as original. Thats why @steemcleaners are actual people and not robots.
I just come across way too many plagiarists who call originalworks to every post they make and I get tired of seeing people think it means something.
Yes, I see your point. But with cheetah its the same, only in the other direction. Lots of people got warnings ( may be its somewhat better now ) for no reason. And in some cases its still bugging people, for example if someone makes a frequent post, like a lotto every week which is basically the same text each time.
In such cases the problem lies exactly in the fact, that it gives its opinion without a request - or a guarantee for being correct. And it tends to scare people a little.
However, I cant see how this could be solved. There is no way that all posts could be checked by real people, not now and even less in the future, if Steemit keeps growing like it does now.
You make a good point with the "not being called" thing, on repeated posts. I wonder if the two could marry and make something more broad?
Right, but at least cheetah is doing what it is supposed to. It lets a person know if the content posted can be found somewhere else online. Cheetah doesn't flag, make any judgement on the originality, that is left for actual people to review and act on.
Well, I guess originalworks is also doing what its designed to do - sending the reply after being called. It doesn't seem to check anything at all.
But anyway, its just one of the many things around here that need to be observed - and not the most critical one either.
Exactly! it USED to flag and castigate, in the early days. But then changed the diction to be more friendly, since there is always that chance it IS original and came from another place.
It would really suck to post something original, only to get a message claiming it's been stolen from somewhere. I could see it causing people to turn cranky.
For the most part, when this happens, the poster knows why. It's because it was a contest post, or whatever, that repeated a significant chunk of text from a prior iteration. If folks respond to the bot, users can read the response and judge for themselves.