Journal cjpez's Journal: Damn 5
A feature I really want for Slashdot is the ability to automatically mod posts down that bitch about moderation. And anyone who's got some lame-ass .sig about "ooh, the /. editors are evil! Just Evil!" Give it a rest. Jesus. It'd be like those "anyone on my 'friends' list gets +1" options, or whatever. Except that it would scan the content of posts and just be able to tell. It could be put under the "People Modifier" section, under "Idiot." Maybe I should submit this into /.'s sourceforge account as a feature request . . .
Update: Oh, and anyone who bitches about some kind of slashdot hypocrisy, too.
One problem... (Score:2)
How in the heck would you automate the detection of mod-complaints? And if it wasn't automated, how long do you think it would be before someone figured out a way to abuse the system?
A better idea (when we get another order of magnitude processor speed up) might be to let each user define a list of regular expressions and associated +/- mods (sort of like a spam filter). That wouldn't be much harder, and it would be a lot more general.
-- MarkusQ
Re:One problem... (Score:2)
I was actually just jokingly suggesting that Slashdot create an interpreter capable of accurately parsing and understanding the English language (we'll deal with other languages later, I guess. <g>), which would be able to (again, accurately) determine what the post was about. Which is, of course, at least as far as I'm concerned, impossible with today's hardware and current languages and programming methodology in general. So the point is somewhat moot. But to answer the question of automation, I had figured that Slashcode would just filter the posts real-time as registered users pulled up the comments page, so it really wouldn't be automated per se. And abuse of the system wouldn't be possible, given the accuracy of the program's English-parsing algorithms (I mean, the program would almost have to be sentient to be able to do this). So no problems there . . .
Although the regexp idea is quite intriguing, and sounds like it could actually be doable! Except I feel that it'd be quite prone to error, plus in order to be really careful about it, you'd have to get fairly specific with the regexps, so you don't mod down posts that you actually wanted to see, which would allow variations on wording and spelling to make it through the filter unscathed . . .
No, I think we're going to have to wait for the intelligent language-parser for this to become a reality. For what it's worth, I did end up posting the Feature Request [sourceforge.net]. We'll see what they do with it.
Now I'm just hoping that people appreciate the irony of me linking to a Journal Entry bitching about people linking to Journal Entries bitching about moderations in their sig, in my sig. :)
Thanks for the interest!
Re:One problem... (Score:2)
*** SYSERROR 413 ***
IN PARSING THE TEXT:
PROCESS markusq THREW A Stack overflow exception WITH THE META-TAG "Self referential humour recurrsion stack depth exceeded."markusq HAS BEEN RESTARTED BY ai-spawnd
Re:One problem... (Score:1)
I was actually just jokingly suggesting that Slashdot create an interpreter capable of accurately parsing and understanding the English language...
This would also have to wait until Slashdot gains an editor capable of the same feat.
Re:One problem... (Score:2)