Kuro5hin.org: technology and culture, from the trenches
create account | help/FAQ | contact | links | search | IRC | site news
[ Everything | Diaries | Technology | Science | Culture | Politics | Media | News | Internet | Op-Ed | Fiction | Meta | MLP ]
We need your support: buy an ad | premium membership

[P]
ideas for a more democratic weblog

By dr k in Meta
Sat Dec 29, 2001 at 08:24:13 PM EST
Tags: Kuro5hin.org (all tags)
Kuro5hin.org

The author shares some of his ideas for a "next generation" weblog.


I've been tossing around these ideas for a while and thought I would share them with the k5 community. Let me preface this article with some of my stronger convictions, and I will develop the ideas from there.
  1. The sectioning of articles (including diary entries) should be more fluid - any diary entry should be a potential section or front page story, and rejected stories should move into the diaries.
  2. We need an open editing process that is immune to bait-and-switch tactics.
  3. Moderation and comment ratings should be based on a statistically valid mechanism, instead of the nonsense we have at present. [see here for further thoughts]
In my understanding, the whole diary concept was introduced as a way to keep excess verbiage out of the moderation queue. People wanted to record and publicize particular events, but the quantity of such events - and the often pedestrian nature the events ("I got a cat!") - produced too much noise. Now we have a situation where quality diary entries quickly fade from view, appearing only as footnotes to other stories.

So perhaps there should be a mechanism for bookmarking or promoting a diary entry to make it more visible. Something like hotlisting, but more powerful.

A simple interpretation of Robert's Rules of Order could be implemented. It would work as follows:

  1. A user makes a motion, a proposal for some action to be taken: "This story should be put in the Culture section."
  2. Another user - from a different IP - must "second" the motion. This is to prevent frivolous motions.
  3. The community votes on the proposal until a certain "pass" threshold is reached, or a number of votes (100 would be sufficient) have been cast, at which time the proposal fails. The votes should be a simple yes/no choice for all measures.
  4. If the measure passes, appropriate action is taken.
  5. The author of a motion may withdraw it at any time.
Proposals could include multiple actions to be taken, such as "Put this story in the Culture section AND put it on the front page."

article proposals

There are a number of potential proposals that could be made for the placement of articles:

  • Move a story into a section
  • Move a story out of a section (return to diaries)
  • Move a story to a different section (a combination of the first two)
  • Move a story to/remove a story from the front page
  • Highlight an old story (put it on the top of the list)

editing proposals

A proposal scheme could allow for a robust open editing process. Proposals could be made to:

  • Fix spelling or grammatical errors line-by-line
  • Change a story title
  • Revise active stories (without admin assistance)
  • Add yourself as a co-author to the story
  • Reset all vote counts for section proposals related to this story [this should always happen]
Special rules for editing proposals: unless there is an active section proposal or the story is already in a section, the author(s) of a story can apply editing changes without a vote.

meta proposals

I'm just going to toss these in, I'm sure some people will find them distasteful:

  • Add a new section
  • Rename section
  • Move all stories from section X to section Y and remove section X
Possible future expansion: have multiple front pages with independant sections.

some thoughts

Something like this should be possible:

  • Proposal:
    1. Change story title from "Cats are Neat" to "Cat Fancying"
    2. Remove paragraph 1 from introduction
    3. Edit line 6 of body:
      <had a persain cat> had a Persian cat
    4. Move story to "Pets" section
Certainly there are potential abuses of this system - that is why for each action the opposite action should be included. With no small effort, a group of people could maliciously re-edit a story and put it on the front page. But this could happen now and the community has no recourse at all.

With such a scheme, the different aspects of moderation would find their own champions. People who like to write articles would be less discouraged than they are with the current all-or-nothing process. Grammar and spelling Nazis could don their uniforms with joy. And those brave souls who scour the diaries for small gems of information would be rewarded for their efforts.

Thank you for your time.

Sponsors

Voxel dot net
o Managed Hosting
o VoxCAST Content Delivery
o Raw Infrastructure

Login

Poll
I propose that this story
o needs more poll options 3%
o needs fewer poll options 3%
o should be rewritten in third person 6%
o needs more links to porn 39%
o sounds great, but I want to play the role of Mr. Negativity and complain about who is going to do all that work 5%
o I like cats 40%

Votes: 88
Results | Other Polls

Related Links
o here
o Also by dr k


Display: Sort:
ideas for a more democratic weblog | 56 comments (40 topical, 16 editorial, 0 hidden)
Too Much Overhead (3.00 / 4) (#2)
by tudlio on Fri Dec 28, 2001 at 08:00:49 PM EST

I presume the non-democratic aspects of the current system that you object to are Rusty et. al.'s ability to modify posts?

Can't say that I see that as a problem, but I wouldn't object if you're proposal didn't obligate me too spend an awful lot of time administerinng and rendering judgment if I wantef to be a part of the community.

But your proposal does require some serious overhead, so I say nay.

I'll vote up the proposal, however.


insert self-deprecatory humor here
overhead (3.00 / 1) (#4)
by dr k on Fri Dec 28, 2001 at 08:25:57 PM EST

What overhead does this require? All of the actions would be implemented in code, the same way moderated stories move into sections when they are voted up. As far as community involvement, you don't have to concern yourself with anything that doesn't interest you.

I don't object to Rusty et al.'s ability to modify posts, I never established any kind of anti-non-democratic position at all. Perhaps you feel that my ideas are an attack on some imagined power elite of kuro5hin, a cabal if you will. If so - if there are indeed administrators out there who feel the need to keep "the power to edit" out of the hands of the masses - then I would certainly agitate against their rule. But I have better things to do with my time and energy.
Destroy all trusted users!
[ Parent ]

Power (5.00 / 3) (#11)
by rusty on Fri Dec 28, 2001 at 11:41:46 PM EST

My rule has always been the less I have to do the better. This site is Applied Laziness in action. :-)

Keeping power out of the hands of the masses has never been a motivation, to say the least. However, balanced against that is the KISS issue. How to put more control in the hands of the users without making it so unwieldy that no one can use it. For an example in this area, see HalfEmpty, which runs a system called Glasscode that's kind of Scoop to the Nth degree. It has a more complex rating system, and AFAIK anything can be rated. While it demonstrably more precise and accurate than Scoop's system, it is also alot more complicated. So much so that the author has confessed to me personally that it really is too complicated for people to use.

I'm not trying to say Glasscode sucks or anything. It's a cool idea, and I think it could have a niche, somewhere. But for general day-to-day use, it's just too complex. So another important goal, along with putting power in the hands of you guys, is making it easy enough to use that it's worth having the power.

So, that's the abstract theory. For specifics: I don't like the Diary <-> Section idea. Never have. Diaries are not part of the story system, as such -- they are an open place for people to post whatever they want. I don't think they should ever become part of the story system, because the line has a lot of value, I think. Consider the sections as the "product" of K5. We're a bunch of people, doing a little bit of work every day to produce this online magazine type thing. The diaries are the water cooler of the K5 production team. It's where we can hang out, talk about our offline lives, bounce story ideas off each other, and arrange a time to meet for drinks in some London dive.

I think the two should remain separate. Now, I think it could be possible to expand the queue into something more like an editorial desk, which seems like what you really want. I definitely want to make it possible for authors, at least, to edit their own stories before they get posted. My current thinking on that is as follows:

  • Add another vote: "Please Edit"
  • Make it so that any voter can change her vote at any time while the story is in the queue.
  • Make it so the author can edit the story at any time while it's in the queue
  • If a story is edited after someone votes, and their vote was "Please Edit", the story will again be marked "New" for them. Their edit vote will not be erased, nor will any other votes be erased, but it will be brought to their attention that it's been changed. (Note: I'm not decided yet if an edited story should be marked "new" for everyone, or just people who voted edit. Could go either way, but maybe marking it new all around would be a good idea)
This still leaves the possibility that someone will write a good story, wait till just before it posts, and then turn it into a great big goatse.cx link. I'm figuring that with the time voting takes, and the rate stories get posted, it would be easy for one of our four or five admins to just summarily drop any story to which that is done. I honestly don't think it would be a problem, considering the effort to get a story posted in the first place.

Better yet, but requiring more code, would be to save revisions, and let an admin roll a story back to the version that people actually voted for. This could be saved for version 2 of the editing features, probably.

Either way, the certainty that your l33t h4x0ring of your story would be gone before anyone even noticed it would probably serve as a pretty good deterrent to anyone actually doing it.

Story editing in the queue would also allow an author to take into account the broadest opinion of what the appropriate section/topic is, and change a story to fit. I think that would effectively solve that problem as well.

The creating new sections thing, I'd really like to do, but it's not much of a draw, in terms of payoff for effort. That is, it would be some fair bit of work to implement, but wouldn't make much difference in actual practice, meaning that on the long list of features to be added, it rates pretty low for us.

People would have to submit official "section" proposals, and justify the creation or removal of a section. There would also need to be various code that handles page layout without knowing the sections beforehand, and so on. It's certanly possible, it's just stayed pretty low on the list because I have things that are more directly in demand to work on, not to mention trying to bring in a paycheck every month.

Upshot: It's doable, certainly, but I'm not likely to code it in the near future. If anyone with some perl mojo takes an interest, we'll gladly consider patches to Scoop for that, or anything else, really. I'd love to have someone take the above editing idea off my hands and implement it too, since I'm probably not going to get to it for a few months.

Thanks for the suggestions. I look forward to seeing what others thnk.

____
Not the real rusty
[ Parent ]

On editing (4.00 / 1) (#14)
by MotorMachineMercenary on Sat Dec 29, 2001 at 12:00:12 AM EST

>This still leaves the possibility that someone
>will write a good story, wait till just before
>it posts, and then turn it into a great big
>goatse.cx link.

About 99% of this activity could be avoided by closing the possibility to edit a story when its score hits a certain number, let's say 60. A story that has hit that high a vote is likely to be posted even without the author editing it (unless a new editorial comment changes tides), so editing it wouldn't make much difference anyway. And those 20 or so points leave plenty of time for remaining voters to notice those gaping body cavities.

And come on, if that last 1% really makes it, I'll give kudos to the authors' ingenuity at luring me successfully to "Hey, I'm looking at gay porn!" at a net cafe :-D

My bodyweight is muscle and cock MMM
Tenured K5 uberdouchebag Herr mirleid
Meatgazer Frau gr3y


[ Parent ]
Yeah, but even that's too much (4.00 / 1) (#15)
by rusty on Sat Dec 29, 2001 at 12:03:02 AM EST

You could cut it off at a certain score, but all that would really do is force the would-be abuser to make ten new accounts instead of two. I don't think it'd help, or be worth the effort of adding it.

____
Not the real rusty
[ Parent ]
IP filters for abuse? (3.00 / 1) (#28)
by guet on Sat Dec 29, 2001 at 12:06:45 PM EST

I'm afraid I don't know much about scoop (did try a search...) but I assumed it'd do some sort of filtering for new users by IP to avoid someone from the same IP from setting up bogus accounts within an arbitrary (eg 2 hours) timeframe? (Wouldn't this avoid the majority of such attempts given the attention span of the perpetrators?)
If that was the case then closing editing at a threshold could be useful.
However, I'm not sure the idea of other users editing stories is workable, as most edits seen as constructive by others would be seen as destructive by the author, and aside from checks for spelling/typos, what else could other users usefully add? Is it possible to add ideas to an article without muddying the original intent/meaning? Surely this is what the comments on articles queued for submission are for?
If you allowed edits/authors to be added, I'll bet you'd have people who wanted their name taken off something that no longer represented anything close to their opinion.


[ Parent ]
IP filtering won't work (5.00 / 2) (#41)
by Rizzen on Sat Dec 29, 2001 at 11:45:49 PM EST

There are just too many people out there behind NAT firewalls and proxies. Granted, it wouldn't be that often that an entire office signs up at once, but it will happen. Especially for students in BC -- all HTTP traffic passes through a single proxy cluster in Victoria. Thus, all BC students who sign up from a school computer will appear as the same person.

IOW, you just can't do it. Even limiting it by time will agravate more legitimate users than abusers.
The years of peak mental activity are undoubtedly those between the ages of 4 and 18. At age four, we know all the questions; at eighteen, all the answers.
[ Parent ]
Depends on your definition of "work" (none / 0) (#46)
by rusty on Sun Dec 30, 2001 at 10:55:46 AM EST

As dram describes, we do IP filter. No, strictly speaking it doesn't work, but it works a little bit. That is, at least it forces someone trying to run a script to change IPs every two creation attempts, which makes it just a bit harder.

So far I haven't had any complaints from legitimate users about creating accounts, and it is limited to two per IP per day.

____
Not the real rusty
[ Parent ]

In short, yes. (none / 0) (#44)
by dram on Sun Dec 30, 2001 at 05:25:52 AM EST

There is a limit to how many accounts one can make per IP, per day. I know the default for Scoop is 2 per day, I would assume it is the same for K5, you can find out by making accounts for the fun of it, I'm sure.

But if somebody was going to go to the trouble of making a story like this they would need to plan it out. If they did that all they would need to do is make two accounts per day for the ten days prior to them posting the story. You might say nobody in their right mind would put that much effort forth, but I ask you, are all the people on the internet in their right mind? Doubtful.

-dram
[grant.henninger.name]

[ Parent ]

Question.. (4.00 / 2) (#16)
by eightball on Sat Dec 29, 2001 at 12:56:16 AM EST

You say: "I don't like the Diary <-> Section idea", and you give good reasons why diaries should not end up more public.1
You don't say anything about the possibility of sending failed stories to the user's diary. This would preserve the stories and see that if some users are discriminated against, they would at least have their say in diaries. Now, we can read comments from these stories, the only thing we are missing is the context.
You could respond to this saying that this is the sort of thing that pople should be using their diaries for. That is, stuff a user is interested in and perhaps noone else. I could say that you could say this about anything a user has an interest in, and whether the audience is 2 or 200, you don't know until the vote comes up. Obviously there are diary topics, "Got a cat",etc.. But what of stuff that noone is interested in, maybe because the user is ahead of the game..

1BTW, I had thought of a way to transition from 'all private' diary to a kind of syndicate, diary publishing thing. It could be an 'opt in' system where you would have to click a checkbox for it to be eligible for post-diary-dom.. I don't think this is so important, but it is not unworkable..

[ Parent ]
The story <=> diary thing. (5.00 / 1) (#27)
by scanman on Sat Dec 29, 2001 at 09:19:35 AM EST

I think the reason for the story <=> diary thing is that sometimes someone will post something to a diary that's really worthy of a front page article, but never gets much attention because it rolls off the diary page too quickly. Or, someone posts an unpopular opinion as a story, but some people want to continue the discussion, so it gets moved to their diary. I also think that the daily e-mail summary should include articles from all sections that I am interested in, including diaries - not just the front page.

"[You are] a narrow-minded moron [and] a complete loser." - David Quartz
"scanman: The moron." - ucblockhead
"I prefer the term 'lifeskills impaired'" - Inoshiro

[ Parent ]

Changing votes when editing (3.00 / 3) (#33)
by jasonab on Sat Dec 29, 2001 at 03:29:57 PM EST

If a story is edited after someone votes, and their vote was "Please Edit", the story will again be marked "New" for them. Their edit vote will not be erased, nor will any other votes be erased, but it will be brought to their attention that it's been changed. (Note: I'm not decided yet if an edited story should be marked "new" for everyone, or just people who voted edit. Could go either way, but maybe marking it new all around would be a good idea)
I think my preference would be to remove all positive votes, and leave all negative votes, while allowing negative votes to be changed. That way, a user couldn't make trivial changes to a story just to reset it's score and keep it in the queue longer. If a story happens to go down in flames even after an edit, it can always be reposted.

As an aside, I like the idea of stories voted down going into diaries. That way, I can always see what's I've missed in the queue.

[ Parent ]

How to preventing gotse.cx hax0ring (3.00 / 1) (#38)
by khym on Sat Dec 29, 2001 at 09:26:52 PM EST

This still leaves the possibility that someone will write a good story, wait till just before it posts, and then turn it into a great big goatse.cx link.
You could make it so that all edits need to be approved by either an admin, or by X trusted users. Even if you only required aproval by 3 trusted users, it would be rather hard for trolls to cultivate 3 accounts with trusted user status, and that status would be revoked once they were caught approving a goatse.cx edit, so they'd have to cultivate 3 new trusted accounts before doing it again.

--
Give a man a match, and he'll be warm for a minute, but set him on fire, and he'll be warm for the rest of his life.
[ Parent ]
Its not that hard (3.00 / 1) (#39)
by eightball on Sat Dec 29, 2001 at 10:02:28 PM EST

Its not that hard to create a trusted user account. I did it once so I could see who voted an unfair 0 vote on one of my early posts.

[ Parent ]
How about marking them "Edited" instead? (4.00 / 1) (#45)
by ZanThrax on Sun Dec 30, 2001 at 10:33:02 AM EST

If a story is edited after someone votes, and their vote was "Please Edit", the story will again be marked "New" for them. Their edit vote will not be erased, nor will any other votes be erased, but it will be brought to their attention that it's been changed. (Note: I'm not decided yet if an edited story should be marked "new" for everyone, or just people who voted edit. Could go either way, but maybe marking it new all around would be a good idea)
I don't know how much more work adding a new setting to the story status would be, but I think that it would solve your concern about marking edits "new" for people who will become annoyed at having stories they've already voted on showing "new" (And of course some would be. And they'd be vocal about their annoyance.)

Before flying off the handle over the suggestion that your a cocksucker, be sure that you do not, in fact, have a cock in your mouth.
[ Parent ]

Too much to decide, good point on comments... (5.00 / 2) (#21)
by tudlio on Sat Dec 29, 2001 at 04:26:48 AM EST

I was referring to decision-making overhead. Right now, all I have to decide is whether I think a story ought to be posted to the front page, posted to section, or dumped. In your proposal, I'd also have to decide which of an array of possible motions I support, and which I think should be overruled. You're asking me to do more thinking (even if it's just to decide to ignore a given motion). Granted, it's a small additional demand on my time, but I don't see the current setup as particularly anti-democratic and in need of change.

I guess I thought the "more democratic" bit in the title was your way of saying you thought the current system was less democratic than it could be or ought to be. Sounds like I misunderstood.

On re-reading your post, I have to admit I like the idea of making some changes to the comment rating system. I rarely look at the rating of any comments but my own, and that only to scratch the ego itch. It seems like it does a decent job of classifying people into the trusted, untrusted and everyone else groups, but it doesn't seem to do a very good job of identifying which comments are most worth reading. Slashdot's system seems to handle that a bit better, but I wonder if that's just because they have so many people commenting and rating that every article gets statistically valid attention? Or is it because you only get to vote +1, 0 or -1?




insert self-deprecatory humor here
[ Parent ]
A Question. (4.50 / 6) (#8)
by dram on Fri Dec 28, 2001 at 10:19:38 PM EST

If this was implemented what would the point of the submittion queue be? Why wouldn't people post all stories to their diary an hope somebody motions to move it to a section or to the front page? Howerver, giving users the ability to edit their stories while in the queue would be good, then they could address editorial comments on the fly instead of resubmiting the article.

Another change I would like to see is the ability to edit my votes on stories in the queue. But thats just my wishlist. :)

-dram
[grant.henninger.name]

KISS (3.00 / 5) (#9)
by Gutza on Fri Dec 28, 2001 at 10:41:20 PM EST

You probably already know the "Keep It Simple, Stupid!" theory on creating (and managing) public stuff. Your proposal obviously is 100% wrong based on this theory. However, I gave it +1 (FP) exactly due to the imperfections in the system you pointed out (I think moving a story from one section to another should be the reader's rather than the writer's option) and I'd like your opinion to be noticed.

Who's your vendor, who's your vendor? — Scott Adams
time is K5
Moving a story from one section to another (4.50 / 2) (#32)
by roiem on Sat Dec 29, 2001 at 02:56:15 PM EST

How about just adding "+1, resection to ..." for each section except the author's section? Then, every voter can either accept the submitter's section or propose an alternate one, and the story is placed in the section with the most voters, if it is accepted. This would be orthogonal to the "Section/Front page" vote.
90% of all projects out there are basically glorified interfaces to relational databases.
[ Parent ]
Author's Control (5.00 / 7) (#10)
by Blarney on Fri Dec 28, 2001 at 11:38:58 PM EST

I like your idea of "motions", but it seems to me that the author ought to have some supreme authority to decide what should be done with his post. We all all retain copyright of our posts, said copyright existing from the moment that the comment is composed, subject to the privileges that we grant to K5 by posting and are enumerated in the legal page. In other words, my comments are still mine for all eternity or until copyright expires (although it never will until that goddamned unfunny mouse finally kicks the bucket.).

I'm not sure that I want to write a diary entry on "How my girlfriend ruined my Christmas vacation", turn on my computer the next day, and find it posted in the new "Whiny Son-Of-A-Bitch" section recently created by a motion and majority vote. Nor do I want to write a serious article on "Imminent Death of the World Wide Web" with tons of footnotes and shit like that only to find it posted to the "Penis Bird" section in the dead of night by a dedicated crew of trollvoters. They're my comments, and I don't want them messed with.

With no small effort, a group of people could maliciously re-edit a story and put it on the front page. But this could happen now and the community has no recourse at all.

You're wrong about that. If somebody thought it was the height of humor to take other people's comments, insert "Also, I like butt sex" in a few places, and have all their buddies help them resubmit it to the Front Page, this would be plagiarism and would be against K5 policy, and the stuff might even be subject to deletion from the high authorities here.

However, were your system to be implemented, it would be necessary for K5 policy to explicitly ALLOW editing other people's stories without their consent. At this point, people might think it permitted to misbehave in this way.



Quite right (5.00 / 1) (#37)
by khym on Sat Dec 29, 2001 at 09:16:41 PM EST

I agree that the author of a story should have final control over the content of a story. You could require that the author vote "yes" on a motion/referendum/whatever for it to take effect, or you could allow content altering motions to only be started by the author.

--
Give a man a match, and he'll be warm for a minute, but set him on fire, and he'll be warm for the rest of his life.
[ Parent ]
I'm gonna disagree (5.00 / 4) (#17)
by 31: on Sat Dec 29, 2001 at 01:53:34 AM EST

One of the big problems that seems to come up as things get more democratic is that the process (or beauracracy, use the term of your choice) grows much faster than the individual's gain does.

Your ideas, while giving users much more power, seem to increase the process of being involved with the site to what seems to be too high of a level. Some people really like the process of making things work, I personally prefer that scoop takes some shortcuts so a fairly small amount of process gets us to the real conversation, rather than worrying too much about how we get to the conversation.

But hey, an interesting idea... if I actually end up with time over the next quarter I'd be willing to help with some code to see how it works (if anyone's in the mood to try it in the real world). I'm somewhat curious to see what audience it draws initially...

-Patrick
individual gain? (1.66 / 3) (#20)
by dr k on Sat Dec 29, 2001 at 04:06:54 AM EST

What the hell does individual gain have to do with democracy? Perhaps if you read my essay on influence you would understand my position on the ego vis a vis the mob. In short: you, as an individual, have very little meaningful power within a democratic group. Why do you need more power? Do you really think your influence should be more valuable than everyone else's, simply because you are more concerned with what is going on?
Destroy all trusted users!
[ Parent ]
Reply (3.66 / 3) (#22)
by qpt on Sat Dec 29, 2001 at 04:44:59 AM EST

Do you really think your influence should be more valuable than everyone else's...?
Well, yes.

Domine Deus, creator coeli et terrae respice humilitatem nostram.
[ Parent ]

much more than the alternatives (2.00 / 1) (#23)
by 31: on Sat Dec 29, 2001 at 05:40:17 AM EST

the more democratic (at least as the word seems to be used for this idea), the more power a user has. On slashdot, some users have the power to moderate others. More democratic than websites where the admins are the only control... the individuals gain some control (if nothing else, the potential to control is a gain). On k5, all users can moderate, and they control to a degree which stories appear. It may be way less control than if they were the sole admin of k5, but it's much more control than in situations where you read what's given you. It also requires more process to get there.

So when I say it gives the users more power, i don't mean more power over each other, but more power over their situation. It's not that my voice is more important than yours, it's that your and my voice together should be more powerful than rusty's, or you and rusty together should have more say than I do.

so i'll admit now i'm too tired to really read your essay (i skimmed it a few times, in the hopes that i don't come off too poorly), but it seems to be pointing out some of the big problems with individual's opinions in a democracy. I'm not gonna argue against it, instead, i think you're arguing against something different than i'm arguing for.

-Patrick
[ Parent ]
I would be willing to run that code on my site. (4.00 / 1) (#35)
by dram on Sat Dec 29, 2001 at 05:17:15 PM EST

Just to see what it was like I would be willing to try that out on my site, Ingenuitas.org. It might be fun, and something different, not just another Scoop site. I would really like to implement an open editing process, where each user can edit their own stories while they are still in the queue. That way if somebody makes an editorial comment that is good and I would like to implement the change I can. As it is now, if my story gets posted but somebody makes a comment on a change that I would like to implement in my story, I cannot. So that is a bit of code that I would like to work on, once I get a bit better at perl. :p

-dram
[grant.henninger.name]

[ Parent ]
If you code it, they will come... (4.66 / 3) (#25)
by Vladinator on Sat Dec 29, 2001 at 06:57:39 AM EST

And I will offer commercial hosting based on it. Try to get it into scoop itself rather than forking, as we're already doing scoop hosting commercially for people. Simmilarly, if any of you adequoids are reading, try to push your changes (as much as possible) into scoop too.
--
LRSE Hosting
Weblog as CVS tree (4.33 / 3) (#26)
by Scrymarch on Sat Dec 29, 2001 at 07:52:52 AM EST

So you want to generalise diaries and story submission, and give more editing power to users? How about viewing a community weblog like a CVS tree?

On the submission of a diary, there is an option, default off, "Consider this as a story".

The submission queue for a user is a filter view of diaries. Something like all the potential stories for the last week. Any user can vote on a story with the same options as now, and stories that pass a threshold reach the front page or section pages. Users can also withdraw stories from consideration at any time. Votes can be changed at any time.

Additionally, any user can submit an update to a story. The original author gets to vote yes/no on the change - if they vote to accept, it becomes a new version of the story. If they vote against, it becomes a fork. The changed versions also appear on the submission queue. They could be filtered out or displayed as connected with the original depending, again, on user preferences. Later versions are stories proper, and so they can be withdrawn by the author of the change.

Votes carry over for version changes but forks must be voted on anew. All changed versions are considered forks until approved by the author.

This assumes stories will be fairly polished before submission. k5 users generally have no compunction slapping down half-baked stories anyway.

Problems: lots of submissions in the queue. Potential bait and switch issues. Duplicate risk. Hmm.

More democracy on K5, eh? (3.00 / 1) (#31)
by quartz on Sat Dec 29, 2001 at 02:51:52 PM EST

I have a better idea: let's have users write petitions instead of motions and make them public so that anyone could read an sign them. Then, if a petition accumulates a certain number of signatures, it becomes a ballot on which people can vote, and if the ballot passes, it's submitted to the editorial mechanism for implementation. And you should also change the title of your proposal from "ideas for a more democratic weblog" to "Kuro5hin.org takes a new direction".

Now if only we could also replicate on K5 the level of inbreeding on LambdaMOO... Oh, wait. Nevermind.



--
Fuck 'em if they can't take a joke, and fuck 'em even if they can.
Author's Addendum (5.00 / 3) (#36)
by dr k on Sat Dec 29, 2001 at 06:20:14 PM EST

I want to elaborate a little more on the proper way to measure votes in an open-ended binary system. Everywhere you go people use X point rating systems without understanding the fundamental math they are tampering with. Careful study of the binomial distribution can teach you a lot. (Hmm, I sound like a college professor, which I am not.)

The first thing to look at is the mean deviation (sigma), which is going to be a function of the number of votes you have.

4 votes : sigma = 1
16 votes : sigma = 2
36 votes : sigma = 3
64 votes : sigma = 4
100 votes : sigma = 5

The way to calculate the strength of a vote is to take the difference between the total vote score and the mean score, then divide it by the above sigma. For 36 votes cast, if half the people vote yes and the other half no, you would have a score of 18. A score of 24 after 36 votes is 6 more than the mean, and 6 / sigma(36) = 6 / 3 = 2. A score of 24 after 36 votes is 2 deviations from the mean - not a lot, but significant enough.

The trick is to choose how many deviations you want to be the "pass" threshold. How about 5? If the vote score is ever 5 deviations from the mean, you can stop taking votes - the measure was successful. So we can figure out these various thresholds:

4 votes : sigma = 1, mean = 2, mean + 5 * sigma = 7
A proposal with 4 votes requires a score of 7 to pass - this is quite impossible.

16 votes : sigma = 2, mean = 8, mean + 5 * sigma = 18
Again, you cannot get a score of 18 with only 16 votes. But with three more yes votes this story would exceed the threshold - with the first 19 votes being unanimously positive, how could you not pass this measure?

36 votes : sigma = 3, mean = 18, mean + 5 * sigma = 33
64 votes : sigma = 4, mean = 32, mean + 5 * sigma = 52
100 votes : sigma = 5, mean = 50, mean + 5 * sigma = 75

As you can see, 5 deviations requires a strong majority.

The most common objection to this method is that it isn't "expressive enough" on the individual level. Well, if you were in a room with 50 people trying to decide on an issue, how much individual expression are you going to allow? How about a movie theater with 500 people?

The second objection is that there is a potential to stop taking votes after only a handful of people have voted. People worry that users might form voting blocks, or create 30 different accounts, or write auto-voting robots. These problems are not unique to this method - yet by severly limiting the expressive power of individual votes, you can take a lot of punch out of these abuses.

I've said most of this before in my crusade against bad math. If someone would calcuate the mean deviations for a 5 point scale I would be most appreciative.
Destroy all trusted users!

Open-ended (none / 0) (#47)
by rusty on Sun Dec 30, 2001 at 01:26:34 PM EST

Does the deviation method also guarantee that the system is "closed"? That is, the story voting system right now is not closed -- or, rather, it's closed artificially by capping votes at 350. Any proposed replacement would have to fix this problem, and be inherently closed in some way.

Also, I think allowing 19 +1 votes to decide it would be a bad idea. It's not hard to collect 19 different accounts without anyone noticing. But I see the principle-- I'd probably just set the deviation score a bit higher.

____
Not the real rusty
[ Parent ]

probabilities (none / 0) (#50)
by dr k on Sun Dec 30, 2001 at 06:05:36 PM EST

Using the deviation lets you make a decision much faster than a fixed threshold. While you would still need to cap the number of votes, this does not in any way affect your confidence in the current score - if the vote has failed to reach the threshold, it doesn't pass. (Keep in mind that there is the potential for, say, ten times as many proposals in the queue than there are currently. I would find it valuable to speed up the approval process, perhaps at the risk of more false negatives (failed submissions). So I would suggest stopping the ballot after only 100 votes. If you don't, I'm betting you'll end up with the beaurocratic backlog many people fear.)

The chance of a vote passing is the inverse of the deviation squared. 5 deviations gives you a 4% chance of success. 10 deviations gives you a 1% chance. That is very slim. Rather than making the system so prohibitive, make it more flexible - if a "bad" story gets posted, give people the chance to take it down. You can even raise the thresholds for each successive vote, so if a story was posted with the threshold set at 4 deviations, raise the threshold to 5 to "revoke" the story, and 6 to "reinstate" the story.
Destroy all trusted users!
[ Parent ]

Math musings (none / 0) (#54)
by Scrymarch on Mon Dec 31, 2001 at 08:43:56 AM EST

I've seen your comments on this stuff before. I have a feeling you're right within the way you've defined the problem, but missing the point. (My musings will not necessarily prove this argument :)

Are you assuming k5 users can be considered a homogenous, but randomly acting collection of agents? This might be a problem. What if voting patterns vary as the day goes on? People from different timezones tend to vote at different times, forming approximate timezone electorates. If you allow 19 votes to accept or reject a story, you'll amplify this effect.

Furthermore, in this sort of analysis you tend ignore the remarkable ability of people to adopt strategies to for maximising their impact on a system they have a stake in. Eg street urchins without math training running complicated currency exchange positions; people who flunk formal logic reacting correctly when exposed to structurally identical problems in real life.

[ Parent ]

Perfectly gruntled (2.66 / 3) (#40)
by gordian knot on Sat Dec 29, 2001 at 10:13:49 PM EST

I don't think I've ever run across a site that allows member input that didn't have to field the "new, improved" rating or classification ideas of people who seem to be permanently disgruntled. There is no such thing as a perfectly democratic system, online or off. The most we can hope for is something that works for most people, most of the time. I've been on sites that are real horrors of manipulation by trolls and revenge raters. K5 isn't one of them.

Any system can be manipulated by those determined to do so. The best thought out systems always have unforeseen consequences, not all of them good. I'm sure a few souls with nothing better to do would be glad to code these unnecessarily complex changes, but what would be the cost of implementing them--not only in presenting a new challenge to those intent on subversion, but in the downtime while unexpected bugs are worked out?

I'm not usually in the "leave well enough alone" party, but K5 runs a hell of a lot better than most sites I've been on. Is it likely to be improved by such nitpicking, bean counting changes as you propose? I don't think so.



counting beans (none / 0) (#43)
by dr k on Sun Dec 30, 2001 at 12:30:18 AM EST

"I do not suffer fools gladly."

Shall I point out the irony?
Destroy all trusted users!
[ Parent ]

Too much effort required by users (4.75 / 4) (#42)
by Rizzen on Sun Dec 30, 2001 at 12:00:59 AM EST

Nice ideas, but it would take too much effort by end-users. Not only would we need to read the existing stories, but now we'd have to wade through myriad motions and comments on those motions, decide which we support or not, vote on them, submit changes, table our own motions, and so on. IOW, very few would do it. A certain number of users will be all for it, and before you know it, 50 people are controlling everything as everyone else sits back and wallows in apathy.

A better idea would be to add the ability for authors to edit their stories while still in the queue, to add a "+1, please edit" or "+0, please edit" voting option, to add a "approved, but awaiting edits" status (has enough +1 votes to be added, but has too many 'please edit' votes to be posted at this time), and to have the 'please edit' votes reset after any editing changes.

That way, the story submition process is streamlined a little, and there's no extra work/time needed on the regular user's part.

If you see a diary entry that you feel should be a story, then submit an MLP story pointing out that diary entry. :)
The years of peak mental activity are undoubtedly those between the ages of 4 and 18. At age four, we know all the questions; at eighteen, all the answers.
The wiki way; good for a small weblog (none / 0) (#51)
by micheno on Sun Dec 30, 2001 at 06:57:39 PM EST

May be a way to make the blog more democratical would be to give the possibility to change the articles, the title or whatever by some other users (as proposed). Meatball wiki shows that it is possible to rely on users. But for a big big weblog the danger of spam and noise increases in a exponential way. So may be the way to keep the weblog 'clean of noise' is to keep it small but this is not so democratic...

Terrible Idea (5.00 / 2) (#52)
by br284 on Mon Dec 31, 2001 at 01:34:48 AM EST

-- start paranoid tyrrany of the majority rant --

The way that I see this is that you are co-opting the author's ability to say what they want and using the democratic process as a way of changing what the author writes. Granted, this is not the purpose of your proposals, but I see it as a potential way that the articles could be drastically altered to say something not intended by the author.

For example, let's say that I wrote a left-wing piece about something left-wingish. Now, suppose that a right-wing faction decided to rewrite the article for me. With a few accounts (this a is given because only a small percentage of the readership will be active enough to make this work), they can "democratically" cut and paste into my article at will. Before too long, I could wake up one morning and have an article attributed to me that I did not write.

Now, your approach may make sense if you are trying to produce articles for the outside community that are a reflection of the mean position of views within the k5 community. However, I personally don't use the site for that purpose. Rather, I use the site as a means of viewing alternative viewpoints and takes that I would not find elsewhere. I figure that spelling mistakes and grammar errors are well worth the gain obtained. Personally, I think I would find it to be more valuable to create a moderation system that provided more diversity in opinion rather than try to limit that diversity through democratic means.

-- end paranoid tyrrany of the majority rant --

-ck

democracy and critical mass (3.00 / 2) (#53)
by dr k on Mon Dec 31, 2001 at 04:26:28 AM EST

I'm going to interpret this as an expression of a fairly common sentiment: many people have no faith in any kind of democratic process. We saw this in the various outbursts after the US election in 2000, and we see it encoded into various moderation systems (like the "trusted user" concept here, which effective gives a bigger vote to, well, a pretty random group).

But let's make something clear: you, as an individual, have the ability to selfishly express yourself here on k5/the internet to a previously unknown degree. But you, as an individual, will always have an insignificant amount of power when it comes down to the decisions of a larger group. And it matters very little the degree to which "democratic" notions have been encoded into this or any other forum - people still participate in a more primate form of democracy by walking away from you, i.e. by voting with their feet.

To get back to your actual concern, that groups of people are conspiring to make you look like an ass: this is certainly possible if the community fails to be vigilant in their obligations. You need some kind of critical mass of active, er, patriots taking part in things. Someone to urge a complacent audience to vote one way or another. And yes, if this active group devolves into a Beavis and Butt-head administration, someone might change the name of this article to "Mmm, I like to suck big dicks." Since this would be done through an open process, it would be clear who really said what.


Destroy all trusted users!
[ Parent ]

democracy (5.00 / 1) (#55)
by br284 on Mon Dec 31, 2001 at 11:18:55 AM EST

While the idea that someone could alter my post is a bad thing, it is not the thing that I worry about most.

The purpose of a democracy (or similar type of goverance structure) is to gain the opinions and positions of a diverse group of people, "average" those positions and differences in opinion to a single point that is acceptable to the largest possible part of the membership. If you imagine a group of four people with opinions that could be thought of as points on a square graph, and each person's opinion in a corner all to themselves, what democracy will do is attempt to "average" out those differences and come to a position in the center of the graph.

Now, as I previously stated, if you view articles on k5 as a means of trying to get the "average" opinion and position of the k5 reader for an outside body, your system works finefor this purpose. However, I do not view k5 as a having nearly as much value for those outside as it does for those inside. And from the inside, I do not want a democratic process involved in the editting processes of a story that only tends to produce articles written for the k5 center group. Instead, the value k5 has is that it produces articles far enough away from that center to be interesting and that lead to more critical thought than just a democratic "me too" article.

Now, let me make something clear to you -- k5 has not increased my ability to selfishly express myself to the nth degree. Before k5 (and online forums), I could state my beliefs and ideas as selfishly as I wanted. I suspect should k5 disappear tomorrow, I will still be able to make myself out to be a hardheaded ass in public. What k5 has done is provide the best interested audience to whom I can spew those beliefs. And as an individual, sure I have little power in the picture of the whole group, but as I see it, on k5, I have enough power to do what I want -- I can express my approval or disapproval for a story and even if it gets posted, I can choose to not read it. This is the extent to which decisions are being made on k5. If it were some sort of online governance tool, I might feel differently.

So, I guess I must hold firm to my position that your scheme is still a terrible idea. It attempts to solve a problem that is not really a problem, and if implemented, it would water down k5 in the process.

-Chris

PS. Don't offend those not reading articles (walking away from those they don't like) by saying that they use monkey democracy.

[ Parent ]
ideas for a more democratic weblog | 56 comments (40 topical, 16 editorial, 0 hidden)
Display: Sort:

kuro5hin.org

[XML]
All trademarks and copyrights on this page are owned by their respective companies. The Rest 2000 - Present Kuro5hin.org Inc.
See our legalese page for copyright policies. Please also read our Privacy Policy.
Kuro5hin.org is powered by Free Software, including Apache, Perl, and Linux, The Scoop Engine that runs this site is freely available, under the terms of the GPL.
Need some help? Email help@kuro5hin.org.
My heart's the long stairs.

Powered by Scoop create account | help/FAQ | mission | links | search | IRC | YOU choose the stories!