Deletion Debate

[zinshin]

There has been some really novel ideas that could help the Decision Engine expose more tasks. One idea said that we should automatically delete tasks that are extremely disapproved, or even implement a time limit for a task to reach a certain amount of approval.

I'm not a fan of a time limit, while I do have faith in a voting deletion tool. For example, we could add a side vote for tasks with less than 5 "Do it Dan!"s, and if there are enough votes, BAM, it's gone.

I want to know your thoughts.

Don't leave, we need everyone's two-cents.

Should Revision3 use a time-limit, an automatic tool, or a vote-for-deletion tool? How many votes is enough? How long a time-limit? Should we give tasks immunity if they reach a certain approval level? Well?

[qralex]

Well, there should at least be an option for the creator of an task to delete his it.

As for the automatic thing... maybe something like, if after a task has 10 votes or more and only 10% of those votes are approvals (that is, 90% or more is "No, don't do it!"), it gets removed from "mainstream" lists and is put on a special page for disapproved tasks. Similar (or maybe even the same as) the "rejected" tab for tasks that Dan rejected himself. That way they'd still be there (and could possibly be revived) but would not clutter other lists.

[griffiegrrl]

I like the idea of there being a marriage between a time limit and a voting system.

Such as, if a task has 75% disapproval rate, and it stays at that for more than say 24-48 hours, then it gets deleted.

[cheekat]

I like the idea of there being a marriage between a time limit and a voting system.

Such as, if a task has 75% disapproval rate, and it stays at that for more than say 24-48 hours, then it gets deleted.

Seconded.

[mrhandshake]

I guess at the time being tasks tend to dissapear on their own, as long as they aren't either extremely successful or often commented, but I see that the huge amount of "task-waste" will be a problem to handle as soon as upcoming possibilities to highlight or radomly display submitted tasks take effect.

Regarding the fact that ideas failing that hard could probably be submitted a second time as long as there is no way to approve if they have ever been there, and also that I'm not a friend of deletion, I would rather propose highlighting- and task-displaying-functions not to include old and unsuccessful tasks.

[ulrichthegreat]

The tasks need to be deleted at some point, because otherwise the tasks would get piled up and noone would read 1000 pages of tasks.

But I would eliminate the time limit and limit the number of pages

The time limit wouldn't work because it would be dependable of the viewership in the website, ex. if there would be a lot of people on everything would get jammed again cause the time limit would be too slow too keep deleting the exess of tasks.
However the limit of pages needs to be set up just by the needs of the users (how many pages of tasks you are wiling to go trough).

So, if the pages would be jammed, when a new task would be made, a task from the bottom of the list (the least commented/rated/oldest tasks) would get deleted thus freeing space.

I wrote a bit to much eh?

[ulrichthegreat]

Oh and I think imageboards (4chan work in a similar way)

[mrhandshake]

Isn't finding Tasks more about the way they are sorted and displayed and not really dependint on their amount as long as you have the tools to find and present to others what you are looking for? I can hardly imagine anybody reading more than 3 to 5 pages of the task-list they're displaying, but there are (as I believe) more than five pages of nice ideas per category.

Also, do they have to reach a certain page-number in each sorting, or do you think of tasks being deleted only by their position in the popularity/date/whatever-list ?

In case we decide to start a deletion routine I would rather prefer the time/popularity-check, which looks way safer to me. (I still think theres no need to delete anything, as long as sorting-mechanisms work fine.)

[ulrichthegreat]

@mrhandshake
1. Yes and no. Of course the tasks should be sorted correctly, but the tasks that are not active/rated are just waisting space, noone will find them anyway if there are 1000 pages.
It just gets confusing.

2. I explained that. The max. pages number is set, when all the availabe pages are full, the tasks that are on the bottom of the list get deleted as new ones come up.

The sorting of the tasks that decides what tasks should be on the first page, last page or anywhere in between, it is up to debate. I think we should put into consideration ratings, numbers of votes, num. of comments, date added ... up to debate.

I recomend you check how 4chan works, its the same idea.

[mrhandshake]

@mrhandshake
1. Yes and no. Of course the tasks should be sorted correctly, but the tasks that are not active/rated are just waisting space, noone will find them anyway if there are 1000 pages.
It just gets confusing.

2. I explained that. The max. pages number is set, when all the availabe pages are full, the tasks that are on the bottom of the list get deleted as new ones come up.

The sorting of the tasks that decides what tasks should be on the first page, last page or anywhere in between, it is up to debate. I think we should put into consideration ratings, numbers of votes, num. of comments, date added ... up to debate.

I recomend you check how 4chan works, its the same idea.

I get the point you're making in your first part and would propose an task-archivation as a compromise. If searching the tasks they can still be found, which may help preventing double-submitions as long as people care, but the main task-pages won't get too crowdy.

Though I doesn't seem to me as if you got what I wanted to say in my second paragraph (probably because my language-skills are... let's call them limited), however you do answer my question in the last part of your explanations (It was rather focusing on which criteria exacly lead to task-deletion) and I do support your Idea, exept that I still think it would be better if we do keep outsorted tasks, but only store them at a different, seperated place.

[ulrichthegreat]

I get the point you're making in your first part and would propose an task-archivation as a compromise. If searching the tasks they can still be found, which may help preventing double-submitions as long as people care, but the main task-pages won't get too crowdy.
I doubt anyone would go searching for tasks, it is easier to just make a new one.
Besides the archived tasks still couldn't get many attention to get accepted just by people having the same idea and (in the best ase scenario) searching for it. You need to show people the idea, make it easily acessable, not archive it to somevhere noone would care to look.
Bring the idea to the people not make them look fi it.

I hope I answered correctly :)

[mrhandshake]

The general idea is of course to allow people to access interesting ideas fast and easily. However there will be ideas that won't gain the popularity they needed in order to prevent deletion.

The idea with the archive is not supposed to affect the successful tasks, but the other ones, that shall be removed because they make things messy. If now someone decided to create a new task, he can first search for it and find out if it already has been created. I think of a results-page where results are divided between regular ones, and results in the archive. Afterwords he can decide, if it's worth resubmitting this task and how successful it was. I could also imagine a possibility to "reactivate" old tasks if you have an idea that already has been submitted, but didn't have enough people voting on it.

The archive is ment as an alternative to the deleted tasks, would of course be a crowdy thing, but it isn' really ment to find anything manually. Rather to allow people a better planning of their task-submission.

(I hope I got what you wanted to express there :S)

[crackedstraw]

If you think the following is a good idea, please vote on it at http://revision3.com/dan30/task/good-bad-crap-filter, as it's already been proposed. Thanks! :D

Right now, the way the "popularity" filter works for viewing tasks only ranks tasks by the total number of positive votes, not taking into account the negative ones. This is fine, but it does not give enough room for new ideas to show up and squeeze themselves into the "competitive voting zone" of the Dan 3.0 decision engine, as new ideas will always be in "last place" unless they become EXTRAORDINARILY popular. I.e., a good idea can have 10 good votes and no bad ones, but a crappy idea will still be at the top of the current (cruel) popularity hierarchy, even if it receives 100 good votes and 2,000 bad ones.

I propose that there is instead a percentage-based filter. This would guarantee that new ideas start out at the top of the filter, presumably since the writer would give one positive rating out of one. All ideas therefore start out at 100%, and slowly get voted down towards the back of the line, giving every idea a fair shot to stay at the front of the pack based on each task's merits as a good idea, not on how long it has existed and thus how many votes it has received, even if the majority of these votes are bad.

There would obviously need to be some sort of algorithm also in place based on the time that the task has existed and the number of votes it has garnered - a "handicap" if you will - since a fresh idea that receives one bad vote and one good vote in the first few moments of its existence will automatically get booted to the middle of the line with a quick 50%, never to be heard from again.

A compensating measure would need to be in place to allow ideas with a certain lifetime to survive in the front, until an acceptable cross-section of votes had been gathered and a task had had an fair enough amount of time to be voted on as good or bad. At this point, the handicap would slowly fade as the time alive and number of votes increase, allowing the percentage to take over and empirically determine rank.

There should also be sub-categories for ideas, based on their percentages. For example, tasks with a 90% positive rating would go into a "good" category, those with less than say 50% go into a "bad" category, and those with less than 10% go into the "crap" category, and these outliers could be subject to review by Dan himself, at which point he could simply remove them from the site, be they the product of trolling, stupidity, impossibility, etc. Finally, the "good", "bad", and "crap" tasks could be removed from the main filters, already having achieved a certain general rank. They would then only be viewable by selecting the sub-category to which they belong, and the competition in these sub-categories would pit these gladiator ideas together, instead of against the mousy, newer tasks. This would allow the middle-percentage ideas to receive more opportunities to be voted upon, until they, too, get filtered into one of these categories. These ranked hierarchies would allow for a more fair voting system where the rich only get so rich, and the poor only get so poor, and the crap only gets so stinky.

[skyfire_esoteric]

I think there should be a seperate list for low rated tasks. Similar to the 'accepted tasks'.

We should also have a flag option for spam tasks or tasks we think Dan can't do (really questionable tasks). If many people flag it a certain amount of times or Dan sees a few people flag it for 'questionable' tasks he can determine if its a troll/spam task or one he's unable to do. There really should be an archive so we can see what tasks were already suggested so people are not repeating the same task over & over again. I often see the same tasks posted by different users, obviously it's not efficient because it could be a good task but yet only gets 20-30 votes for each user.

A better organized task list that has 'voting tasks', 'accepted tasks' 'rejected tasks' (for extremely low votes) & 'veto tasks' if it ever arises.

[skyfire_esoteric]

If you think the following is a good idea, please vote on it at http://revision3.com/dan30/task/good-bad-crap-filter, as it's already been proposed. Thanks! :D

There should also be sub-categories for ideas, based on their percentages. For example, tasks with a 90% positive rating would go into a "good" category, those with less than say 50% go into a "bad" category, and those with less than 10% go into the "crap" category, and these outliers could be subject to review by Dan himself, at which point he could simply remove them from the site, be they the product of trolling, stupidity, impossibility, etc. Finally, the "good", "bad", and "crap" tasks could be removed from the main filters, already having achieved a certain general rank.

This is a good idea as well. Being able to see tasks that are bad would be only viewable (not able to vote)

[icrimbo]

The side section of disaprovals seems good, but i've seen members like sic just posting nonsense. if this goes through, he'll just set up the amount of people it would take to get good ideas disapproved, and we'd fuck the whole system over ((excuse the french))

[safiregurl32]

I get the point you're making in your first part and would propose an task-archivation as a compromise. If searching the tasks they can still be found, which may help preventing double-submitions as long as people care, but the main task-pages won't get too crowdy.

Though I doesn't seem to me as if you got what I wanted to say in my second paragraph (probably because my language-skills are... let's call them limited), however you do answer my question in the last part of your explanations (It was rather focusing on which criteria exacly lead to task-deletion) and I do support your Idea, exept that I still think it would be better if we do keep outsorted tasks, but only store them at a different, seperated place.

so, um, i was wondering if maybe they could have a list that sorted them by popularity, and then if it was new it could have like 2 days or something to get up over like 10 votes or something more. also, the no dan dont do its should minus one from the score, so say if i got 20 yes's but 7 no's it would be changed to 13 yes's. anybody who gets into the negatives is automatically deleted, unless your initial 24 or 48 hour time period isnt gone yet, but itll go faster so you'll only have like a fifth of the time you had left left. this might be confusing and complicated, but i think it would work well if implimented properly.

[ihateyouokkkkkkk]

i think there should be an abuse button that will automatically delete comments after a certain amount of abuse reports.