On Sun, Oct 15, 2017 at 12:50 PM, Keegan Peterzell <[email protected]> wrote:
> On Sun, Oct 15, 2017 at 1:04 PM, Todd Allen <[email protected]> wrote: > > > Keegan, calling people names isn't helpful here. > > > > > I didn't. I'm calling out the tone. > > I care if someone's right or wrong, not their tone. If we want to talk about that, we certainly could discuss calling someone "sexist" when they're calling attention to a potential problem, but I really don't care much about that either. The point is whether this is a good, well-conceived idea. > > > We've already had horrible projects to write tons of stubs before, like > the > > "place" bots. And in those cases, we'd know at least roughly what they > > would do and how. > > > > Yes, the horrible place bots like User:Rambot on the English Wikipedia. > The bot started almost every place stub in the United States, and almost > every one of those seeds has generated a more fully formed article. > > And would've anyway. It's not like editing them would have been forbidden if they'd been redlinked until someone was ready to work on it. Every article started out as a redlink. But we're probably a bit off topic with that here. > > > > > This project is going for 100k articles. There are as of this writing 118 > > editors signed up. That is, even if we presume 100% participation (which > is > > generally wildly optimistic), nearly 1000 articles per editor to reach > that > > goal. If somehow that does happen, there are four judges who would need > to > > review, if the goal is reached, 25000 articles each. Those are not > > realistic numbers. > > > > Add into that financial incentives for being the most prolific, and we're > > setting up for a very foreseeable disaster. > > > > Risk management is one thing. A foreseeable disaster is quite another. > Overblown hyperbole. > > Call it "risk management" or whatever you want, but those numbers are unrealistic by orders of magnitude. Now, that in itself wouldn't be a substantial concern, but then you've got the issue of offering money to hit them. That's absolutely a recipe for disaster, and many years of experience both on Wikipedia and otherwise would tell me that. Offering money for just being prolific, quantity over quality, is absolutely a bad idea. > > > > > I have no problems with editing initiatives focused on underrepresented > > areas. But they need to have realistic goals, numbers actually run during > > planning, and most importantly, no financial rewards. This project is > not a > > good idea. > > > > Mmmhmm, and who should be the ones to set the goals? The ones that "know > better"? > People can set whatever goals they want. But what they're proposing to do affects the entire project, not just their corner of it, so everyone on the project should be involved in that. That absolutely includes people who have been around the block more than a few times. There's a reason we shut down the "reward board" type systems, > > I'd advise you all who'd like to tell people what they're doing wrong, > instead focus on helping people to do things right. > > Otherwise, this is just patronizing. > You start toward fixing a problem by saying it is a problem. After that, you decide what should be done about it. And I did offer suggestions, thank you, that being, choose realistic numbers and keep money out of the equation. Money, like it or not, is a powerful motivator, and to some it's a motivator to cut corners, like using poor sources or plagiarizing. I'm not saying any individual editor would do that, mind you, but I'm saying it is part of what we know about human behavior. Even relatively trivial amounts of money can have substantial impacts on people's behavior; I'd be happy to find you the sources for that if you'd like to see them. Todd _______________________________________________ Wikimedia-l mailing list, guidelines at: https://meta.wikimedia.org/wiki/Mailing_lists/Guidelines and https://meta.wikimedia.org/wiki/Wikimedia-l New messages to: [email protected] Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/wikimedia-l, <mailto:[email protected]?subject=unsubscribe>
