User talk:Sycamore/PLS/Scoring
Jump to navigation
Jump to search
Bread Golem? You mean that article from 2006? --
02:37, February 2, 2011 (UTC)- That's an even bigger fail the Mrthejazz's rambling + Timecube copypasta or my redirect to A wizard did it. Hahaha! MegaPleb • Dexter111344 • Complain here 02:48, February 2, 2011 (UTC)
Whatever happened to in-depth scoring? What ever happened to You and I? --
20:59, February 6, 2011 (UTC)- Whatever happened to the judges finishing ahead of time? Whatever happened to Robot Jones? MegaPleb • Dexter111344 • Complain here 21:29, February 6, 2011 (UTC)
- Whatever happened to predictability? The milkman, the paperboy, evening T.V.?--(lol.) 07:08, February 9, 2011 (UTC)
- Whatever. -- The Zombiebaron 09:58, February 9, 2011 (UTC)
- I'd forgotten how bad the Star Trek Enterprise theme was. Sir Modusoperandi Boinc! 14:52, February 9, 2011 (UTC)
- Whatever happened to predictability? The milkman, the paperboy, evening T.V.?--(lol.) 07:08, February 9, 2011 (UTC)
- @TKF: I did some in-depth reviews for my judging, and then had the reviews criticised and the fact that they we're put in as reviews criticised. Why on Earth would anyone want to do in-depth reviews - especially when you are looking at 10+ articles and a time frame if only a week to judge, then what is the point of reviewing in depth? Pup 03:59 10 Feb '11
- Well I mean you don't have to go in depth in depth (I forgot people take those two words so literally around here) but just doing what Thinker, Skifan, etc have done by keeping short little tablets of what worked and what didn't in their userspace/talk pages (or, in your case, the mainspace, which is somewhat unorthodox) --
- Aye, there is a difference between in-depth judging and in-depth reviews, anyhow. With judging, it's simply a matter of having something to explain the results. With reviews, on the other hand, it is more of a matter of picking out just what needs improvement and how, something about being more helpful to the person who wrote the article. *shifty eyes* ~ 05:12, 10 February 2011
- Six to one, half a dozen to the other. Pup 05:25 10 Feb '11
- When I judged what I did was randomly pick how many points each page got, randomly organize them based on that, then randomly declare a winner. It was considered to be quite the groundbreaking technique at the time. Lots of torches and pitchforks. Sir Modusoperandi Boinc! 05:33, February 10, 2011 (UTC)
- Six to one, half a dozen to the other. Pup 05:25 10 Feb '11
04:54, February 10, 2011 (UTC)
- Aye, there is a difference between in-depth judging and in-depth reviews, anyhow. With judging, it's simply a matter of having something to explain the results. With reviews, on the other hand, it is more of a matter of picking out just what needs improvement and how, something about being more helpful to the person who wrote the article. *shifty eyes* ~ 05:12, 10 February 2011
- Well I mean you don't have to go in depth in depth (I forgot people take those two words so literally around here) but just doing what Thinker, Skifan, etc have done by keeping short little tablets of what worked and what didn't in their userspace/talk pages (or, in your case, the mainspace, which is somewhat unorthodox) --