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

A Less Idle Front Page

By DoorFrame in Meta
Mon Apr 23, 2001 at 08:29:10 AM EST
Tags: Kuro5hin.org (all tags)

On Sunday, April 15th What can games teach us about human-computer interaction? was posted to the front page. Today is Saturday, April 21. I've been looking at that story for a week. I'm tired of it. We need to relax our standards a little bit (or better yet, change the voting system) so more stories are posted to the front page.

So this sort of story has been posted before (notably here and here). On top of that there have also been several stories asking for a customizable front page allowing users to choose what genre of stories they see on the front page, eliminating the need for the +1section v. +1frontpage battle (notably here and here).

Clearly something needs to be done. One front page story a week does not make for a very exciting weblog... I might as well just search the web for interesting information my self if that's going to be the pace (I could find things faster). So I propose that we either enable a system where users choose what stories appear on their own front page and ditch the +1frontpage option or merely change out voting habits so that we be SLIGHTLY more accepting of stories... give them a better chance to see the light of day.

That said, an additional problem (which is probably leading to this issue in the first place) is the burgeoning size of the submission queue. With more stories to consider evaluating, I feel that most of us are beginning to have less and less interest in fighting our way through the dozens that await us. Hence I suggest we take a suggestion which I believe was first posited by Signal 11 (although I cannot find the story right now) which said that there should be a limit on how many stories each individual can see in order to moderate. You wouldn't see the whole queue, only a few submissions making the moderation process less intimidating and less time consuming.

Just some thoughts.


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


How often would you like a new story to appear on the front page?
o Once a month (I've still got a 300 baud modem, it's all I can handle). 2%
o Once a week (what we've got right now). 9%
o Once a day (I like to read my story in the morning and move on). 63%
o Once an hour (about slashdot paced). 15%
o Once a minute (I just read the submission queue anyway). 8%

Votes: 247
Results | Other Polls

Related Links
o What can games teach us about human-computer interaction?
o here
o here [2]
o here [3]
o here [4]
o Signal 11
o Also by DoorFrame

Display: Sort:
A Less Idle Front Page | 128 comments (127 topical, 1 editorial, 0 hidden)
All Known Issues (4.80 / 15) (#1)
by Carnage4Life on Sat Apr 21, 2001 at 01:55:00 PM EST

This submission queue issue is hashed out almost every week and there are many potential solutions that have been discussed. Most notable are a current discussion on Meatball Swiki and this comment by rusty (in an article that was killed which seems a testament to the fact that the submission queue is flawed). I assume that rusty is still evaluating potential solutions and/or coding something up as we speak.

The customizable front page thing is something I was going to work on next month but rusty has told me that panner is working on it.


Click here to find out more about your fellow K5 readers. A bug in the code was fixed and over 100 names have been added.

Meatball Wiki K5 SubQueue discussion (3.66 / 6) (#12)
by kmself on Sat Apr 21, 2001 at 05:38:44 PM EST

Well, that answers the question of how long it takes material to get promoted from MeatballWiki to K5 ;-)

As C4L noted I've written at some length on problems with the submission queue. Most of these stem from or are related to the current voting system, which has the unfortunate charateristic of allowing fewer and fewer stories as K5 readership grows. There are a few other broken aspects of the queue (including topical comments) which are exascerbated by this problem.

I'd like to build on the ideas posted to MeatballWiki a bit before submitting them to K5 for wide open discussion, but am open on the issue. I've already heard from Rusty, and he agrees with most of the points and suggestions made.

And, for reference, I'm the guy who designed the moderation system here, if you're curious about credentials.

Karsten M. Self
SCO -- backgrounder on Caldera/SCO vs IBM
Support the EFF!!
There is no K5 cabal.
[ Parent ]

thanks for the link (3.00 / 4) (#26)
by mami on Sat Apr 21, 2001 at 08:57:47 PM EST

to the discussion going on in the past between you and Rusty and some others apparently.

To an outsider, who has no background about moderation features at /. and here, one comment in the older discussion makes much sense to me. At least I would believe that it helps to achieve a more balanced moderation.

I agree with you that one really good possibility is to make the rating system as transparent as the voting system. This has been an idea from the start, and is very likely to appear soon. Having a way to see who rated what comment and what the rating was would, I think, tip the scales in favor of objective rating.
Transparent rating of comments and transparent rating of stories (building averages) should help to get more considerate ratings and more explanations in comments, why someone rated another comment or story the way he did.

Right out of the top of my head I have something against "trusted" users. As soon as you give out this status people might want to find out what makes one a "trusted" person. This would already influence your commenting and rating behaviour.

Finally, I am not so sure, if I prefer a site which has continuously "fresh" stories with very little discussion and somewhat dubious quality at times, or if I would prefer a site which has less stories, but may be ones so thoughtful that a discussion could go on for days and days.

As awful I sometimes think it is to read /.'s comments, I, and I am sure many others, return to /., because they spit out more "news" stories. You just glance over the actual stories and read less the comments (at least not for articles which are less technical in nature).

I think k5 could be the site with less stories, but eventually better discussion over a longer period of time.

I think that's just a question of what you want this site to be.

[ Parent ]

Image whoring (4.00 / 3) (#90)
by Miniluv on Tue Apr 24, 2001 at 01:48:33 AM EST

As soon as you give out this status people might want to find out what makes one a "trusted" person. This would already influence your commenting and rating behaviour.
I tend to disagree. I object to trusted status for different reasons which I might enumerate later in this comment. But image whoring will happen regardless of the rewards. In a way, it is its own reward.

Image whoring is why streetlawyer visits, for example. He built his character, probably in a drunken stupor while trying to light his crackpipe, and now derives childish glee from acting it out in a way that provokes people into responding the way he expects. His character even has a built in response for this expected behavior.

I point him out because he's visible. Everybody knows who he is, and how he acts, and can understand the label when he is the example. There are many, many more people here who fit that. While many of them are trusted, so are others.

trhurler rubs at least 50% of the k5 population the wrong way, yet he is consistently rated well. I seriously doubt he gives even half a shit about trusted status, though he does give zeros where he feels appropriate.

The point I feel ought to be made in reference to your point is that "trusted user" status is derived on something meaningless compared to the "rewards" it confers.

Why is it assumed that because a person makes good comments they can recognize other good comments? Take me for example. I'm a raving asshole in many of my comments, yet I honestly believe I tend to rate fairly. So, if people dislike my opinion as presented in comments, my rating quality becomes suspect? How is that logical, or consistent?

Which brings up the fact that "trusted" status should only be conferred by a system which tracks rating habits. This is otherwise known as metamoderation, and is the last great hope for apocalypse. Metamoderation encourages paranoia, cliques, moderation circles, lack of moderation, and ultimately renders a community incapable of speech for fear of losing their coveted powers.

Meritocracy is a noble concept, but one that has yet to actually apply to K5. Well, it applies, but in a way I perceive as extremely warped compared to the way the term would logically apply. Throw in there extraordinarily subjective nature of the entire trusted status, comment rating, ad infinitum and you're completely fucked.

I say make 'em all equal and the hell with the consequences. They won't be any different than what we have now.

Come on, tell me how to moderate. I DARE YOU!

[ Parent ]

The Math is Simple (3.38 / 13) (#2)
by babylago on Sat Apr 21, 2001 at 01:58:08 PM EST

As the number of quality front page stories increases, so does the frequency of change on the front page. If it's not moving quick enough for you, try viewing less often. That way it will all be new when you come back!

Seriously, the answer is to increase story quality, not lower standards. You sound like you work in the educational testing industry.

[ Blog | Hunnh ]

secret knowledge (3.55 / 9) (#21)
by dr k on Sat Apr 21, 2001 at 07:27:52 PM EST

Do you have some secret knowledge about story quality you'd like to share with everyone? Can you show any correlation at all between the number of stories in the queue, their quantified quality, and the frequency of front page postings?

Because if you can't, I'd suggest you stop defending the status quo. People aren't trying to win the fucking Pulitzer Prize here. Nor are they writing stories to win a free Big Mac in the McDonald's "My Greatest Hero" essay competition. People write stories on topics they think are interesting, and with a little morderation magic and a lot of luck the stories get read.

What makes you think people vote based on quality anyway?
Destroy all trusted users!
[ Parent ]

The Fucking Pulitzer (3.66 / 6) (#27)
by babylago on Sat Apr 21, 2001 at 09:23:53 PM EST

And here I thought it was for journalism.

There are definitely some things that don't go over so well, including bad grammar, incredibly short links with no description, stories that don't have broad appeal, stories that are resubmitted three and four times, cluster stories where the author posts a lot of stories at the same time, and so forth. The part I find strange is that despite the (to me) fairly obvious pattern of rejection for these types of stories, people persist in posting them over and over.

You tell me. Without defining what story quality may be, do you think that we can arrive at some agreement that doing the same thing over and over and being surprised at the result is at least partially lacking in whatever that quality may be?

[ Blog | Hunnh ]
[ Parent ]

the writing process (3.60 / 5) (#30)
by dr k on Sat Apr 21, 2001 at 10:37:31 PM EST

Without defining what story quality may be, do you think that we can arrive at some agreement that doing the same thing over and over and being surprised at the result is at least partially lacking in whatever that quality may be?

Man, is that question tough to parse.

Should we equate the writing process with judgements about character, and then base some notion of quality on those judgements? No. If someone posts 16 versions of a story, is that person a cretin? Perhaps. Is the 16th iteration worthy of objective review? Yes, in an ideal system that story would be evaluated on its own merits. This goes against human nature a bit, but then human nature tends to involve a lot of shouting and intimidation.

I'm calling you out on the oft made claim that we don't see good stories because good stories aren't being written. Grammer and spelling aren't important in themselves, they are important for effective communication. But with this funky new moderation system thingy, we've exposed part of the editing process, a part of communication we normally leave to the care of editors and high school English teachers. We see bad spelling, bad phrasing, "improper" punctuation, "inappropriate" HTML. If you dislike the editing process, then you will probably dislike the moderation process, and the only "good" stories you will find will be those written by a rare breed of "natural" writer.

I believe the underlying complaint here is that "stupid" submissions waste a scarce resource - your time. Particularly offensive to you are stories that "lack broad appeal". The only thing I can recommend in that case is: broaden your interests.
Destroy all trusted users!
[ Parent ]

Wandering a bit (3.50 / 6) (#32)
by babylago on Sat Apr 21, 2001 at 11:14:54 PM EST

I believe the underlying complaint here is that "stupid" submissions waste a scarce resource - your time

Wasting my time is the least of my concerns. Participation is voluntary, so I would have no one to blame but myself if that were the case. My issue is that there are general characteristics that trigger -1 voting, but people continue to incorporate those into their stories. It's quite puzzling.

Also, it might help the discussion if you realize that none of this is really 'offensive' to me. I don't really care. I just find people and behavior interesting, especially when repetitive behavior is expected to yield different results. Your observations on the editing process are useful, I think, to understanding that people dislike finding common ground for communication, instead expecting others to make the effort to come around to their style and mode. Editing, then, implies that they cannot effectively communicate. In that sense, any critical comment leaves a sting in the person being criticized. Unfortunately a common reaction is to try the same thing again, the online equivalent of raising the voice.

[ Blog | Hunnh ]
[ Parent ]

news? (3.00 / 3) (#34)
by mezzo on Sat Apr 21, 2001 at 11:22:22 PM EST

Well. I thought it was for news (y'know, about technology and culture). Quality news would be nice. But after a few days of the same old headlines, even sucky grammar news would be welcomed.

"The avalanche has started, it is too late for the pebbles to vote."-- Kosh
[ Parent ]
The Math isn't that simple (3.50 / 2) (#87)
by Scrymarch on Tue Apr 24, 2001 at 01:09:21 AM EST

As the quantity of k5 users increases, so does the eclecticity* of their interests.

This is great for k5 as a community because it exposes a wider range of expertise while still roughly centring around the average k5 interests.

However, each article appeals strongly to a fairly narrow band of interested people. These are the ones that post FP. Essentially I'd assert that the number of users increases faster than the number interested in any particular article, independent of quality.

To change this situation, relate the proportion of users required to make a front page article to the number of users in the system. Maybe via a logarithm. You'd want to make it monotonic and recalculate it only on the creation of new users.

* If that isn't a word it deserves to be.

[ Parent ]

poll options (3.25 / 4) (#3)
by Speare on Sat Apr 21, 2001 at 02:11:34 PM EST

Another poll option:

    (+) There's a front page?

Seriously, I set my desktop's shortcut to jump into the moderation pool, not the front page.

[ e d @ e x p l o r a t i . c o m ]

moderation page... (3.00 / 1) (#33)
by mezzo on Sat Apr 21, 2001 at 11:17:38 PM EST

Tell me about it. I've ended up perusing slashdot and the moderation page just so that I get news. Those MLPs can be quite informative too..

<br><br>Anyways, I think that by the time a topic gets 'privileged' enough to make it to the FP, a lot of discussion has already happened on it during the moderation pool.

"The avalanche has started, it is too late for the pebbles to vote."-- Kosh
[ Parent ]
Story moderation brainstorm (3.50 / 4) (#4)
by Malicose on Sat Apr 21, 2001 at 02:20:45 PM EST

Instead of randomly choosing stories for each user, what about allowing people to select the topics that particularly interest them and even giving those individuals the possibility of -2 or +2 influence. In such a system, certain categories like MLP might need to be left out as universal categories because some might think it a bit of a stretch to be an MLP rating guru. Perhaps losing the ability to moderate another story (falling under a category) they just "0, Don't Care" about would allow specialists or those really (dis)liking a story to show an unusual amount of (dis)approval. This could speed things through the submission queue by swapping 0s for an additional 1 toward dropping or accepting.

Unclear (3.00 / 1) (#19)
by kmself on Sat Apr 21, 2001 at 07:14:13 PM EST

I'm not sure what you're describing here. Could you expand on that?

Karsten M. Self
SCO -- backgrounder on Caldera/SCO vs IBM
Support the EFF!!
There is no K5 cabal.
[ Parent ]

Story moderation revisited (3.66 / 3) (#31)
by Malicose on Sat Apr 21, 2001 at 11:04:06 PM EST

Heh, it must have gotten more confusing with each reading. I noticed your rating go from a 4, and then 3, before finally settling on 2. Though, my intertwining of a couple ideas certainly did not help matters. I hope what follows is a better explanation of my basic premise.

First, DoorFrame mentioned assigning only a portion of the submissions to each user. I disagree with this suggestion, and my first point offered an improvement to this in the way of preferred topics. Simply, it would be appropriate for people to critique and vote most often on selected issues of interest or specialization. Many just "0, Don't Care" about a lot of the rest, and it would be frustrating to have one's few voting opportunities wasted.

Second, I feel an implementation of übervotes (-2/+2) would both speed things through the queue and put more stories on the front page. It would also allow everyone's occasional strong attitudes to be voiced in a way other than by commenting, "I really want to see this discussed!" Such a system could allow for this power in many ways. One example might be an übervote pool that is incremented regularly and/or added to by sacrificing voting privileges on other submissions. As I put it before, story voting privileges for things people "0, Don't Care" about would gladly be traded for increased say in another article.

[ Parent ]

Why don't you just... (3.00 / 9) (#5)
by michaela on Sat Apr 21, 2001 at 02:27:39 PM EST

go to Everything instead and leave those of us who like it this way alone?
That is all
So Ironic... (4.00 / 8) (#6)
by ti dave on Sat Apr 21, 2001 at 03:03:57 PM EST

I just voted this +1 Front Page, but from my point of view, the Front Page is BROKEN.

I come here for fresh material, not necessarily up to the second breaking news. There are more appropriate sites for that (nakednews.com).

When I come to K5, the Front Page seems to be sort of "crusty".

I'll be changing my K5 link to the Submissions bin, in order to see some NEW stuff.

Note that this defeats the purpose of having a "Front Page".



"If you dial," Iran said, eyes open and watching, "for greater venom, then I'll dial the same."

I agree... (4.40 / 5) (#7)
by skim123 on Sat Apr 21, 2001 at 03:14:17 PM EST

The frontpage blows. I rely on the k5 email digest to let me know what stories are new to k5. I come directly to the moderation page, moderate, check my comments, check my diary, and leave. Hopefully rusty and crew come up with something innovative to get the front page worth viewing...

Sorta like being gay: you're walking around, you know something's up, you just don't know what it is yet.

[ Parent ]

Less idle? Please. Relax standards? No! (4.12 / 8) (#8)
by rikek on Sat Apr 21, 2001 at 03:37:10 PM EST

I get very worried when I hear the words "relax standards"... I know we all have differing opinions on what needs to go where; at least in my opinion, the front page should just be for articles that are so great that they deserve to stick out far above everything else.
Unlike Slashdot, where everything goes on the front page, kuro5hin has the ability to post articles in their section. Rather than encourage people to vote for +1FP more often, I wonder if we could just convince ourselves to read the section stories as often as we read the front page stories and moderation queue. Maybe I'm missing something, but isn't that the reason for having those little boxes on the left side?

slashdot (4.25 / 4) (#11)
by DoorFrame on Sat Apr 21, 2001 at 04:24:32 PM EST

Everything does not go on the front page of slashdot. If you look within the sections you'll see there are dozens of stories that don't make the front page every day, but are there in the sections for the more interested reader. Doesn't really negate your point, just a factual correction.

[ Parent ]
System, not standards (4.00 / 2) (#14)
by kmself on Sat Apr 21, 2001 at 06:07:28 PM EST

It's actually the voting system, standards have little to do with it.

The K5 moderation system as it stands is skimming off less and less of a bell curve or more posts. The bulk of posts, including both "average" (nobody's really thrilled by 'em) and controversial (there's a strong split of opinion) just don't get posted. And large-number statistics dictate that more and more posts fall into this category as time goes by.

First fix the voting system, then address quality issues that remain.

Karsten M. Self
SCO -- backgrounder on Caldera/SCO vs IBM
Support the EFF!!
There is no K5 cabal.
[ Parent ]

Say (3.50 / 2) (#63)
by Elendale on Mon Apr 23, 2001 at 01:19:53 PM EST

This may be offtopic, but check a diary of mine for some talk about the queue. I'm going to revise this tomorrow (April 24) hopefully and get it up. I'm stealing one of your bits off Wiki for this and am interested in your feedback- as well as making sure you don't want to kill me :)


When free speech is outlawed, only criminals will complain.

[ Parent ]
A better solution (3.78 / 14) (#9)
by RangerBob on Sat Apr 21, 2001 at 03:45:25 PM EST

A better solution for this problem would be for more people to actually post stories that are deserving of front page entries. Not the continual "Whine, why can't I download from Napster", not the "Whine, the RIAA and MPAA are a bunch of mean people", and certainly not the "Evil politicians are going to ruin us all boo hoo" stories. A lot of what's posted is honestly a rehash of topics that have been beaten on so much that the poor horse body can go straight to dog food now.

I do NOT think that lowering our standards is a better solution. Keep the standards the way they are now, maybe people will learn to do things better. Or they might post something that's of above normal interest so that more people would give it a +1 fp vote. Either way, if it's not interesting, then it's just not interesting, no matter how people want to mess with things.

It's more broken than that (3.00 / 1) (#15)
by kmself on Sat Apr 21, 2001 at 06:12:20 PM EST

The problem isn't story quality, it's the voting system itself. See my other posts here. This is a technical problem, not a content/social one.

Karsten M. Self
SCO -- backgrounder on Caldera/SCO vs IBM
Support the EFF!!
There is no K5 cabal.
[ Parent ]

Dunno, still disagree though (4.20 / 5) (#24)
by RangerBob on Sat Apr 21, 2001 at 08:24:25 PM EST

I don't think it's entirely the voting system. I've read how you think that fixing the voting system will fix the quality, but I disagree. Face it, a lot of things people try to get posted are really low quality. I may just be cranky today, but really, a lot of people online are just getting fed up with the constant whining. People can't keep posting more stories about how they hate their favorite corporate entities and expect anyone to care past a certain point. And reposting the exact same story over and over in the hopes that it'll someday get posted is silly also. Bell curves can't explain away all the utter crap that's posted online.

There comes a point in time where one must move on. People think that they're "fighting the good fight" by posting on places like K5 and Slashdot for God's sake. I'm tired of people equating downloading from Napster with fighting in the American Revolutionary war. From what I've seen, things that are different and interesting really do make it to the front page. I think the human interface game thing is interesting because it's really similiar to a research proposal that I'm writing now. Things like this are interesting and cool, and the make the front page.

Fixing the voting isn't going to automatically make people post better things here. It's kind of like saying that replacing someone's Beetle with a Mercedes will make them a better driver. No. I'd personally cry if it gets to the point that the front page is littered with stories like "I figured out a new keen way to download mp3's doodz" or "Multinational corporations and the MIB are turning the world into a bunch of fat mutants to work in the Martian salt mines."

[ Parent ]
Crap (3.66 / 3) (#28)
by kmself on Sat Apr 21, 2001 at 09:50:50 PM EST

I'm not disagreeing with you. There's a lot of sophomoric content on K5.

I more or less agree that the current cycle's driving bad content up. I'm not sure that any system is going to ensure high-grade content. I do feel that making the system less sucky is going to discourage it a hell of a lot less. That's aim #1 here.

Good tools won't make you better of themselves, but bad ones can definitely slow you down.

Karsten M. Self
SCO -- backgrounder on Caldera/SCO vs IBM
Support the EFF!!
There is no K5 cabal.
[ Parent ]

Why post quality content on K5? (none / 0) (#123)
by roblimo on Wed Apr 25, 2001 at 02:59:26 PM EST

I believe the K5 story moderation drives out quality content.

First, it gives a huge advantage to people like... well... me, who are well-known. I dislike "celebrity culture" no matter where it crops up, whether on book best-seller lists where a piece of "as told to" shit with a giggly TV actress's name on the cover gets taken more seriously than an important work by a non-photogenic researcher; or here on K5, where a "name" community member's material will at least get read in the submissions bin while other, equally valid work is ignored.

Second, the speed thing. I can submit a story proposal to most paying Web sites, newspapers, and magazines and get a faster response than the average K5 submission gets.

Third, the "K5 Community" tends to kill controversial stories -- or anything the least bit different from the norm, including humor or any kind of experimental style. I believe that K5 has collectively become more anal than many editor-controlled publications in its choice of front page material.

So, pray tell, why would a pro-quality writer (with or without an established professional reputation) want to submit quality work to K5?

Rusty and the gang are friends of mine, and I don't need to maximize my writing income because I get a substantial salary from OSDN, but if I was a stranger or if I depended on my freelance writing income to pay my bills, there is no way I would submit a story here, under the current moderation system, unless it had already been rejected by every other newspaper, magazine, and Web site in the world, paid or unpaid.

- Robin Miller

[ Parent ]

Topical comments in the submission queue (4.41 / 24) (#10)
by flieghund on Sat Apr 21, 2001 at 03:47:30 PM EST

Allowing topical comments in the submission queue is what kills the front page for me.* By the time that a story makes it out of the submission bin, be it to the front page or section-only, I've usually already read it at least once, made up my mind about it (like, don't like, don't care), and either posted a comment or chosen to ignore the discussion.

Disallowing topical comments in the submission queue would be a radical thing to do. It would likely lead to all sorts of complaints. But if people are that eager to post a comment to the story, wouldn't that be incentive to vote it up? (And if you don't want the story posted, doesn't that imply that you don't care about what the discussion might produce? Hence, why would you care if you couldn't post topical comments?)

A current example of what I consider to be the problem: this comment attached to the Ayn Rand parody currently resting in the queue.** The poster voted against the story (and specifically questioned the ability for the story to generate a discussion), but felt it necessary to post a comment (which happened to generate quite a bit of discussion) to the story anyway. Why? If someone votes against a story (-1 Dump It!), why would they want to comment on it? More specifically, why would they want to post a topical comment on the story? Would not an editorial comment be more appropriate for "Gah, what crap is this, there is nothing to discuss here"?

My impression is that people choose to post editorial comments as topical so that the comment doesn't disappear when the story clears the queue. Bah I say. If your comment is so damn important (and not merely an editorial note), let it wait until the story makes it to a section or the front page. If you could only post topical comments after the story clears the submission queue, I think stories would either live or die much faster. There would be fewer topical comments like "Gee, this was cool, +1FP" and (hopefully) more topical comments from people who actually give a damn about the story and want to add to the discussion rather than generate noise.

*I know, I know: I'm posting a topical comment while the story is in the queue. My admittedly weak justification is that this is a topical comment, and the current system allows topical comments to be posted while the story is in the queue. Cope.

**The specific comment, poster, and story are chosen almost at random for illustrative purposes only and should not be taken as anything other than a basic sample. (Read: No offense intended, thurler; that comment just happened to be the first one I found.)

Using a Macintosh is like picking your nose: everyone likes to do it, but no one will admit to it.
One of several fixes (3.00 / 1) (#13)
by kmself on Sat Apr 21, 2001 at 05:42:26 PM EST

As Carnage4Life indicated, I've written a long analysis and proposed fixes to the K5 SubQue problem at MeatballWiki.

Topical comments are a problem, they're exacerbated by the voting system. I agree topical comments should be eliminated -- their need, and their effects, are both bugs in the system.

Karsten M. Self
SCO -- backgrounder on Caldera/SCO vs IBM
Support the EFF!!
There is no K5 cabal.
[ Parent ]

But what about... (3.00 / 1) (#92)
by RadiantMatrix on Tue Apr 24, 2001 at 02:24:00 AM EST

My first instinct is to agree that queue comments should be editorial only. The only reservation I have is this: many people vote on a story based on the amount of discussion it has generated. If there are no topical comments in queue, then this criterion will disappear.

How do you propose to address that issue? Do you even see it as an issue?

never put off until tomorrow what can be done the day after.
Express Yourself

[ Parent ]

Think quickly (3.00 / 1) (#99)
by kmself on Tue Apr 24, 2001 at 03:42:41 AM EST

See this.

Karsten M. Self
SCO -- backgrounder on Caldera/SCO vs IBM
Support the EFF!!
There is no K5 cabal.
[ Parent ]

what is editorial? (3.66 / 3) (#18)
by dr k on Sat Apr 21, 2001 at 07:07:36 PM EST

If you only allow editorial comments, people will still write topical comments. There's no mechanism to stop then, unless you want to call the Editorial Police in to rank the comment down. But that doesn't help, because people don't rank comments based on their "editorialness" or "topicalness". (And people don't really rank comments anyway, it's such a hassle.)

Destroy all trusted users!
[ Parent ]

Part technical, part social (3.33 / 3) (#20)
by kmself on Sat Apr 21, 2001 at 07:21:55 PM EST

See the Meatball Wiki comments regarding why topical is bad, and why it's feeding on itself right now.

The fix is a three-step process:

  1. Disable topical comments in queue.
  2. Disable responses to editorial comments in posted articles (editorial are still available for review).
  3. Fix queue such that there isn't a pressing need to allow topical comments.

Effectively, people learn that:

  • Submissions post quickly enough that they can wait with their topical comments.
  • Comments made in queue don't transition to posted articles in any easy-to-access manner.
  • Topical comments in editorial queue are moderated down.

...at which point, no more problem. (I hope).

Another option is to do away with the queue altogether, but I don't think this is a good current goal, and it may not fit ultimate needs either. The submission queue is an editorial bullroom. Right now it just plain doesn't work (which is why I'm rationalizing making topical comments here).

Karsten M. Self
SCO -- backgrounder on Caldera/SCO vs IBM
Support the EFF!!
There is no K5 cabal.
[ Parent ]

regardless (3.66 / 3) (#22)
by dr k on Sat Apr 21, 2001 at 07:51:19 PM EST

I still say the editorial/topical distinction doesn't matter, you are making a rational distinction that ignores human behavior.

My fix is a two step process:

  1. get stories out of the queue as fast as possible
  2. trust the moderation system to make correct decisions about where stories go:
    • use statistically meaningful thresholds
    • no front page votes - the highest scoring stories go on the front page
  3. find a place for the divisive stories to live
My fix is a three step process.

Destroy all trusted users!
[ Parent ]

Largely agreed (3.00 / 1) (#25)
by kmself on Sat Apr 21, 2001 at 08:44:18 PM EST

Read the Meatball comments.

WRT topical/editorial. Point is we're shooting for a bullroom. Topical comments really shouldn't be there. We'll see where the other fixes leave us, but I think they'll reduce pressures a lot.

Karsten M. Self
SCO -- backgrounder on Caldera/SCO vs IBM
Support the EFF!!
There is no K5 cabal.
[ Parent ]

Topical etc (3.33 / 3) (#82)
by rusty on Mon Apr 23, 2001 at 07:36:31 PM EST

I agree with dr k here, that prohibiting topical comments in the queue is just going to run afoul of human behavior. Some people will just never post their comments anymore, and others will ignore the difference.

I think changing to a story rating system as you proposed will largely fix this anyway. IMO, the "too much discussion" is a byproduct of the fact that stories take too long to post, not necessarily a problem in and of itself.

As of now, I plan to make the voting changes, and then see how that works before changing how commenting is done.

Not the real rusty
[ Parent ]

Disagree (4.00 / 7) (#39)
by zephiros on Sun Apr 22, 2001 at 01:24:19 PM EST

I'm seeing a few problems with this:
  • It requires people to vote based on hand-waiving estimates of the type of discussion that will ensue. It seems to be pretty common practice to withhold votes until enough topical discussion has occurred. This is presumably because people want some small indication of the quantity and value of the discussion the article will generate. Banning topical comments from the queue requires each voter to simply guess how the other thousands of K5 users will respond.
  • It creates a gap between the initial presentation of the article and the ability to respond. IMO, it's fair to assume this would drive down the number of comments, as moderating users would need to read the article once, wait a day or two, then re-read and comment. While some might suggest this is a good thing, I think there's a lot of value in keeping a tight feedback loop in discussion (as per Amy Jo Kim).
  • It would prohibit discussion in doomed articles. A reasonable amount of fairly interesting discussion takes place in articles destined for the dustbin, usually because the comment authors (unlike the remainder of K5) have an inordinate amount of interest in the subject. Allowing users to only discuss "community approved" topics would put an end to these sorts of exchanges.
While I agree there is a problem, I think a solution will need to go a lot deeper than banning topical comments from the queue.
Kuro5hin is full of mostly freaks and hostile lunatics - KTB
[ Parent ]
And there's the problem... (4.20 / 5) (#44)
by Rhamadanth on Sun Apr 22, 2001 at 10:57:16 PM EST

If there's a fair amount of good discussion generated by articles 'destined for the dustbin', then maybe those articles shouldn't be headed to the dustbin, hmm? Articles that are good enough to be discussed when they're doomed are good enough to discuss on the front (or section) pages. For the articles that *do* make it to the front page, what percentage of K5 readers are actually posting. Is that percentage different from the percentage of readers that seem to be a fringe group that are posting to doomed stories? I'd wager not.

If you look at my other post to this thread, I also mention there how silly guessing how much discussion a story is going to generate is. Why not just let the discussion work for itself?

So again, I think we're back to there not being a reasonable method for rating articles that have decent interest value. If 200 of us vote to turf the story, and 125 vote to keep it, is it really a reasonable choice to kill the story, honestly? If 20% of voters end up posting followups to that article (an unrealistically low number?), then we have 25 people that don't get a chance to discuss something that was interesting to them, and there are 40 people that we've saved from having to waste time on the article. How many users are there on K5 again? Total? Enough to make 25 and 40 look like practically the same number.

It's poorly explained, but the point is that 125 people is a lot of people. Too many to exclude, even if 200 others think the story is drivel, IMHO.

-- The /bin/truth is out there.
[ Parent ]
Irrelevant (4.25 / 4) (#53)
by zephiros on Mon Apr 23, 2001 at 10:17:32 AM EST

Reread the last sentence in my original post. Banning topical comments from the submission queue as part of a larger rethinking of the submission process may or may not work. However, banning topical comments as an end in and of itself, within the framework of the current submission system, is a poor idea.

[ Parent ]
Missing the goal (4.16 / 6) (#91)
by kmself on Tue Apr 24, 2001 at 01:58:29 AM EST

I'd like to eliminate the current "Moderation Queue" entirely.

I'd like to replace it with a "Submission Queue".

The difference is that a submission queue is for ironing out the kinks in an article. Mostly structural issues -- spelling, grammar, broken links. Possibly some content issues -- there is a lot of, frankly, sophomoric crud spewed forth on K5 that's little more than uninformed opinion. Or maybe it is informed opinion, but the author has:

  • Grossly mismatched a title to a subject.
  • Failed to provide any external references.
  • Failed to organize the article at all
  • Gone hopelessly beyond his/her depth.

The submission queue is where stuff is submitted for a first, very rough cut, and slapped into technical shape. Really bad stuff -- spam, trolls, hopelessly lost articles -- are killed or remanded to diary. Everything else goes to section (possibly on the basis of what section the SubQue folks thought appropriate, possibly with a title SubQue folks thought appropriate). The only truly doomed content is spam and trolls. Everything else continues to exist and be accessible.

Once content hits the deck, it's still subject to moderation. Possibly also implicit measures such as how often it's read, how many comments are made, how highly rated these comments are. An article that generates a spew of flamebait doesn't merit front page placement. An article that generates a few, very highly regarded, comments, might.

The idea is to make the initial go/no-go decision quickly. Preferably within a few hours -- say anywhere from one to twelve. I'd prefer not to have anything hang in queue for more than a day. It's far less important to make critical decisions about quality than to promote a fair bit of reasonably good stuff.

Once on deck (in section), moderation and activity could further promote or demote an article. I'd like to see it be possible for an older article to be revived if it gets promoted high enough.

Periodically, highly-rated section content gets promoted to front page. Maybe once a day, maybe three times. Maybe once every other day. Scoop site manager gets to choose. This would include all sections, including best picks from diaries (it's really hard to find the good stuff in diaries right now). There would also be tools to provide lists of current, highly-rated articles within each section, etc. Scoop's an immensely rich data source, I just wish it would use what it's got.

The implicit assumptions are:

  • It doesn't take too many opinions to get a basic assessment (very rough cut) of quality.
  • No simple, computationaly unintensive, predictive metric is going to do a good job of predicting story performance.
  • A better assessment is to look at first-day's performance on deck. Note that this is less time than many submissions currently spend in queue.
  • A sufficient number of honest users exist to make this work.
  • Abuse significant enough to effect the system is significant enough to be detected.
  • Scheduling and other needs are handled outside the submission queue.


Karsten M. Self
SCO -- backgrounder on Caldera/SCO vs IBM
Support the EFF!!
There is no K5 cabal.
[ Parent ]

Sounds good to me. (none / 0) (#122)
by Another Scott on Wed Apr 25, 2001 at 01:19:33 PM EST

I agree the current MQ system seems to be trying to do too many things at once - decide whether a post is good and/or fixable, and deciding where it should go. As it stands, I'm forced to decide whether a poorly written article on an interesting topic should get the same score as a good article on an interesting topic, or whether it should be ranked the same as spam. I don't think they're fair choices.

Comments in a Submission Queue should be about whether the post is written well enough to stay or needs to be fixed. Not about whether one agrees with the topic or not.

Once an article is accepted, then it should be put in a Destination Queue/MQ. At that point the current Moderation Queue could take over, or something like you propose. As long as ratings still include the option to dump an article, I don't have a problems with either scheme. Also, promoting articles to the FP based on traffic or some other scheme seems appropriate to me. I don't think we should explicitly have to vote for Front early in the process (not knowing how the discussion will progress).

In short, I think a specific rating system which distinguishes between fixable problems and bad articles needs to be put in place. Front articles should be decided later in the process, IMHO.


[ Parent ]
responding should be +1 implicitly (3.75 / 4) (#74)
by speek on Mon Apr 23, 2001 at 03:59:11 PM EST

I wouldn't want to disallow topical comments for various reasons, but you could achieve much of what your talking about by making a response to a story give a +1 vote automatically (for the great reasons you mention). I'd do the same with comments and moderation if I were running the show....

al queda is kicking themsleves for not knowing about the levees
[ Parent ]

band-aid: lower the threshold (3.55 / 9) (#16)
by xdc on Sat Apr 21, 2001 at 06:36:06 PM EST

More stories would be posted, and more would probably make it to the front page, if the posting threshold were lowered. It seems to me that for many stories, it is too difficult to gain the level of concensus necessary to get posted -- particularly for controversial articles. Of course, lowering the threshold would only be a temporary band-aid until a smarter solution is implemented.

For stories voted up by the k5 community, how about making scoop automatically adjust the ratio of +1(FP) votes needed to make it to the front page, based on some anti-staleness parameters? IOW, scoop would notice that the front page is stagnating despite a large submission queue, and would compensate by allowing more articles to make it to the front page. Once the front page reaches a certain configurable level of volatility, scoop would decrease its leniency, and stories would need more FP votes to make it to the front page.

Also, FWIW, I believe that people should be able to see the entire submission queue, no matter how large it is. There is no need to vote on every story. Just pick interesting ones to vote on and don't worry about the rest. (IMHO)

Newbie baton (3.44 / 9) (#17)
by guffin on Sat Apr 21, 2001 at 06:52:40 PM EST

The front page is almost like a newbie baton. Example:

Newbie J. Jones hears about this great place Kuro5hin.org on Slashdot or Memepool or wherever. Newbie decides, "wow I'm pretty good at poking fun at people and occasionally trolling, this will afford me a new place to practice my antics". However, when Newbie shows up, he finds that there hasn't been anything posted to the site's main page in a week! What a bunch of losers, he thinks; no chance for pissing people off if there's no one around! So Newbie returns to Slashdot and no one here has to listen to his ravings.

My point is this: Why is it such a bad thing that all the action happens behind the scenes? I would expect that any intellegent person would be realize that if he's hearing good things about the site, and the main page is not all he expected, that there might just be some jewels hidden under the surface.

Hello Mr. Twain... (4.00 / 2) (#86)
by jtdubs on Tue Apr 24, 2001 at 12:17:14 AM EST

Wow. You're a cynic. I do see your point though. My only response is thus:

In User-Interface design, "jewels hidden under the surface" is equivalent to "the design sucks but I couldn't find a better way to do it."

Not that I have any great insight in to how better to work this site, but I do believe the front page to be more stagnant than it should be. The only logic I can imagine being added to the front page would be as follows...

The less active an article is, both in number of views over time and in number of new comments over time, the faster it should be replaced with something more active. I mean, this sounds obvious, and I'm sure it's pretty much what is already being attempted, but I think it needs to work better.

Justin Dubs

[ Parent ]
Cynicism? moi? (none / 0) (#119)
by guffin on Wed Apr 25, 2001 at 12:55:05 AM EST

Me a cynic? I just try not to disillusion myself about reality (well my reality anyway).

Anyway, I have to disagree about hidden jewels cooresponding to sucky design. Maybe this is tangential, but any sufficiently advanced program will be unusable to someone who is not versed in that area... would you want a neophyte administrating university's network? There's a reason the programs which regulate traffic are text based.. it's easier that way once you're used to it.

Maybe this has nothing to do with weblog interfaces, but if you're trying to keep Oog the Caveman out of your discussions... what's he going to think when he comes to the front page and lo-and-behold, a story hasn't updated in awhile? Hell, there's nothing to troll here, moving on...

[ Parent ]

Why don't we post EVERYTHING to the front page? (3.66 / 15) (#23)
by Rhamadanth on Sat Apr 21, 2001 at 08:13:24 PM EST

And I mean that. EVERYTHING. Well, almost.

One of the things that bugs me about slashdot is that after the day ends, that's it for the discussion. If the post is a day old, nobody is reading responses anymore. The huge retension on front page stories on K5 is nice. A discussion can take a couple of days. You don't have to monitor your posts every hour just to have a good argument with someone.

But what's weird is that we VOTE for a story in the submission queue based on how much discussion we THINK it will generate. Why don't we let the stories get posted, and SEE how much discussion they generate?

Would it be such a big deal to have stories dynamically juggled on the front page, based on how much discussion is currently active? Popular stories would sit on top of the heap. As discussion trails off, the story trickles down until it's done. When there's been no activity for a couple of days, it's gone.

I still think the submission queue should exist. But like many other people have mentioned, it should be for catching spam, and giving time to edit the story.

The discussions will take care of themselves. We shouldn't have to guess what everybody else thinks is an interesting topic.

-- The /bin/truth is out there.
The problem here... (4.50 / 4) (#45)
by TheLaser on Mon Apr 23, 2001 at 12:27:33 AM EST

...is that if you judge a story based on the activity, people's activity is going to be based on what they want the story to do.

That and everyone who wants to see a story make it to the front page will be posting useless comments. You can limit this by only counting the number of unique commenters per time interval or whatever, but it will still be a problem.

You could try and fix that by weighting the comments' effect by the ratings they generate, but then you get meta-meta-activity where people start to rate comments, even unconsiously, based on their opinion of the story, rather than the comment.

All in all, it's one big bag of worms. It might work though, I'm being quite cynical tonight.

[ Parent ]
Well... (3.57 / 7) (#29)
by kei on Sat Apr 21, 2001 at 10:10:43 PM EST

A submission gets FP status if a whole bunch of people think it's FP material, right? So obviously a whole lot of people don't see the need for more FP articles.
"[An] infinite number of monkeys typing into GNU emacs would never make a good program."
- /usr/src/linux/Documentation/CodingStyle
And yet... (4.00 / 3) (#36)
by Rhamadanth on Sat Apr 21, 2001 at 11:59:55 PM EST

...despite the fact that I realize that there is precisely zero statistical validity to the poll, it *is* indicating that quite a few people want more front page stories. Of course, the people that are reading this article are probably already predisposed to wanting more FP articles, and are just voting and posting comments to make sure that everybody knows it.

K5 is a weird place. Things are run in a democratic fashion, which people generally agree is a good thing. But there *is* such a thing as the 'tyranny of the majority'. Has our little democratic society become complacent? Do we no longer actually act in our own best interests, but merely work to maintain the status quo? I'm certainly not qualified to say one way or the other, but it's something to think about, I think.

-- The /bin/truth is out there.
[ Parent ]
an FP story a day (2.60 / 5) (#35)
by danny on Sat Apr 21, 2001 at 11:43:49 PM EST

If nothing has gone FP for 48 (?) hours, why not put up the story that got the largest fraction of FP votes but failed to make it there?

[900 book reviews and other stuff]

Because (2.66 / 3) (#70)
by aphrael on Mon Apr 23, 2001 at 03:09:46 PM EST

the fact that it failed to make it there means that a sufficiently large number of people thought it wasn't front page quality that it shouldn't.

That's the whole point, right? Who's going to bother to vote when their vote is meaningless?

[ Parent ]

The submission Queue as FP (3.50 / 4) (#37)
by GiTm on Sun Apr 22, 2001 at 10:52:15 AM EST

I personally tend to go straight to the submission queue rather than read what is on the front page.

That way you get to see (and get the chance to comment on) all stories. If a story strikes my interest and makes it through the queue I will continue to monitor comments on it - if not I pretty much ignore both the FP and section stories.

It wouldn't surprise me that a lot of other K5'ers do this.
--- I have nothing funny to say here.
Here's what I do (3.00 / 6) (#38)
by funwithmazers on Sun Apr 22, 2001 at 11:32:18 AM EST

First off, I have k5 load straight into the submission queue...I skip the frontpage completely. Then, I read everything there, as if it were frontpage material. Then I discard any useless or incorrect material received. I supplement this information by reading other weblogs.

What's the front page/section page ratio? (2.85 / 7) (#40)
by DoorFrame on Sun Apr 22, 2001 at 02:48:32 PM EST

What is the formula that decides whether a story will go to the front page versus the section page?

variable FP ratio (3.83 / 6) (#41)
by Puchitao on Sun Apr 22, 2001 at 05:16:20 PM EST

Even though I can assure you that I'm more tired of our current front-page selection than anyone here, I don't think a variable FP ratio is the way to go, as some have suggested. Maybe a lower FP ratio, but not a variable one. Sometimes, it's just a slow news week. Say that for 3 days, the only stories anyone posts are MLPs about 1337 mp3-sharing sites, meta articles about moderation abuse, and suchlike. These sorts of articles have their places, but I don't really want three front-page stories from them in three days because Scoop forces a story a day to the FP.

Perhaps, as someone's noted, the occasional +2 FP would help a bit. Perhaps everyone has a limited (repeat: limited) supply of these per day, per week, per month, or whatever. Something to be used sparingly, as sort of a "I really think this is FP material" vote.

Another thing that might help would be the ability to re-vote on submissions, just like we can re-moderate comments. A lot of editorial comments give constructive criticism like "-1, fix the links & grammar & resubmit and I'll +1FP it". And although a quick note to help@kuro5hin.org can fix a few broken links, there's no way for the well-meaning editorial poster to express his approval of the changes. (Of course, if the submission is going to make it through the queue anyway, the -1 helps it onto the front page almost as much as the +1FP, afaik, since it raises the ratio of FPs to total score. But I may be just misunderstanding the way the FP equation works.)

Perhaps we can do *snappy fun* with you everytime! -- Orz
No!!!!!!! (4.33 / 9) (#42)
by enterfornone on Sun Apr 22, 2001 at 05:50:10 PM EST

We need to relax our standards a little bit
That is the reason that there is rarely anything worthy of the FP. Standards have been "relaxed" to the point where people don't feel the need to proofread or take the time to write a decent story.

If people were willing to -1 below standard stories then people would be forced to write good stories. As it is people submit any old because they know people here will still vote it up.

efn 26/m/syd
Will sponsor new accounts for porn.

Agreed (5.00 / 2) (#69)
by aphrael on Mon Apr 23, 2001 at 03:08:55 PM EST

one of the most pernicious things that keeps appearing in the queue is (are?) political diatribes which show *zero* sign of any meaningful research into either the political theories which address the issue or the statistics related to the issue. Maddeningly, they keep appearing in sections even though similarly poorly researched stories would get flamed into oblivion if they were on technical matters.

[ Parent ]
Everything (4.18 / 11) (#46)
by theR on Mon Apr 23, 2001 at 12:58:31 AM EST

If you want to lower the threshold for front page, the easiest thing to do is go to the Everything page. I'm perfectly happy with the amount of stories that get to the front page. I don't want to lower my standards at the moment.

Thank you. (3.16 / 6) (#48)
by pb on Mon Apr 23, 2001 at 03:16:49 AM EST

I was going to post a link to precisely that, but I see that you already did. Thank you, TheR!

I'd like to see a more active front page as well, as in "User Customizable".

So that the story poster here could, say, set his front page to be all MLP and MLP Submissions and News and News Submissions, and have a MyKuroDot, and never bug us with a Meta Story again!

But I'm sure the Scoop Gurus are working on that. Until then, please link to Everything, like TheR said. :)
We need a new option in the Signature Behavior: "I don't have a signature, go *away*!"

P.S. I don't have a signature. :)
"See what the drooling, ravening, flesh-eating hordes^W^W^W^WKuro5hin.org readers have to say."
-- pwhysall
[ Parent ]
More than one threshhold (4.27 / 11) (#47)
by MmmmJoel on Mon Apr 23, 2001 at 01:08:36 AM EST

One thing I haven't seen suggested yet is changing the threshholds depending on what section the story is. The MLP threshhold, for example, should have a much lower threshhold than an Op-Ed. Some stories require more moderations than others.

Yes (2.00 / 1) (#95)
by kmself on Tue Apr 24, 2001 at 03:08:59 AM EST

That's one of my intents.

Karsten M. Self
SCO -- backgrounder on Caldera/SCO vs IBM
Support the EFF!!
There is no K5 cabal.
[ Parent ]

Shutdown (4.05 / 17) (#49)
by slaytanic killer on Mon Apr 23, 2001 at 09:27:44 AM EST

Please name something in the queue right now that should make front page.

dr k once made some very insightful comments on the nature of the mod queue that unfortunately people did not wish to take seriously. Now, as the population of K5 increases, the votes approach something like randomness; there are too many people with conflicting interests and ideas of what constitutes an FP article.

There are of course silver linings; to think K5 is dying is to forget the Anne Marie trolls, for instance. And plus, this is a low point during the academic and business years in terms of everyone having time to write something that we can feel proud about throwing to front page.

My main objection is this -- I would rather K5 temporarily grind to a halt rather than desperately put lower-quality stuff on the front page. We don't need that. We don't want to be a Slashdot, where people just talk even if there's nothing to talk about.

There might even be a benefit (3.80 / 5) (#68)
by aphrael on Mon Apr 23, 2001 at 03:07:17 PM EST

if we grind to a halt some people will get bored and wander away, and the resulting smaller group might be more cohesive, which would cause things to start posting to the FP again.

We're a dynamic community; this type of thing is going to happen from time-to-time and we shouldn't need to analyze it constantly. :)

[ Parent ]

improving the quality of articles (4.00 / 3) (#76)
by alprazolam on Mon Apr 23, 2001 at 04:40:24 PM EST

I'm in agreement with the majority who seem to desire better and fewer articles as opposed to more and lower quality. The question then becomes 'How can we improve the quality of articles?' This is something that should be the focus of all the meta discussion. I think that there are currently some ideas being implemented that should help:
  1. make the submissions editable
  2. come up with a way to incorporate comments into submissions (see 1)
  3. other stuff i don't remember offhand
In addition you could consider changing the way the threshold works (not make it easier to get on the front page though). You could possibly accomplish this by relating the threshold to the number of votes. For instance if only 50 people have voted, but 40 voted FP, the article should go ahead and be posted, so a bunch of discussion doesn't happen in the queue.

I have one more interesting idea. A pre-queue queue. An author who had one point of view on a subject could write a short article and submit. After some discussion, another person with knowledge of the subject pops up, and has some good debate with the author. The author could then invite the new person into some sort of dual authorship. This would really only be useful for a few subjects and probably wouldn't produce a whole lot of good articles but I think it may work some times.

[ Parent ]

one problem ... (2.00 / 1) (#100)
by blaine on Tue Apr 24, 2001 at 04:22:04 AM EST

1.make the submissions editable

There is one serious issue with this. What happens if I vote +1 on an article, but later on, the author changes it so much that I no longer feel it is +1 material? If editing is allowed in the submission queue, the voting would have to be reset every time an edit occurred. Otherwise it would be far too prone to abuse.

[ Parent ]
rusty posted a solution somewhere (3.00 / 1) (#106)
by alprazolam on Tue Apr 24, 2001 at 10:47:06 AM EST

that involved some checks to prevent this kind of abuse. i couldn't find the comment though. basically make it so you can change your vote if it does get edited, and then make it have to get the last few votes without being changed.

[ Parent ]
you want to improve quality? (5.00 / 1) (#108)
by Shren on Tue Apr 24, 2001 at 12:49:55 PM EST

You want to improve quality? Let people put micropayment tip jars with thier articles. Money has this funny way of motivating people.

[ Parent ]

No, micropayments are useless (2.00 / 1) (#116)
by Sunir on Tue Apr 24, 2001 at 09:27:30 PM EST

Money has a way of motivating people; micropayments are too small to be motivating. They don't count.

Moreover, money has way of motivating people to accrue money. This will change the site in indefinable ways. There are other motivations for contributing than making cash. If this site becomes a competition for material reward, then it may degrade my experience enough that I'd leave. Then again, it is becoming a competition for cash, so we shall see then.

"Look! You're free! Go, and be free!" and everyone hated it for that. --r
[ Parent ]

hit counters (4.00 / 1) (#127)
by Shren on Wed Apr 25, 2001 at 11:33:54 PM EST

A lot of people have hit counters on thier pages. They love to see the hit counter go up. The idea that people across the globe who they don't even know are visiting thier page is downright hypnotic.

Consider a tip jar as a hit counter on steroids. Not only do you know they saw it, but you know that they saw it, liked it, and thought it was worth a quarter. Dead tree published writers would laugh, but they know, just by the distribution levels, that thier article is getting read. For a web-only publisher, such feedback is hard to come by.

[ Parent ]

I'm not sure, but willing to explore it (5.00 / 1) (#120)
by kmself on Wed Apr 25, 2001 at 01:59:49 AM EST

Why don't you write this up as a Meta proposal. Hmm....looks like Rusty beat you to it (yeah, stolen from Sunir's post).

In general, I don't think micropayments work. They're IMO worse than useless when applied to systems over the web. In a situation like K5, though, we're talking more of an accounts infrastructure, the amounts are going to be a bit higher than are typically considered (pennies or fractions thereof for content). Supposing we had 10,000 paying members, US$0.10 would go a way toward making things worthwhile.

I disagree strongly with my friend Sunir's comments that money would corrupt the site. While I would agree that changing the entire site to a or for-pay basis would not be healthy, adding a component of paid-for and/or remunerated content could be good. Fact is that in print media, going rates are about $1.25 a word -- that's a nice sum for an afternoon or two's work. And neither K5 nor the rest of us survivie without money. I'm living off reserves myself at the moment (thank fnord I've got them).

I feel that certain payment models (advertising) skew content far more perniciously than others (direct payment, premium content, research services). The US popular newsweeklies, and The Economist are examples of both approaches. TIME is little more than a vehicle for advertising. The Economist newspaper is a sample of what can be bought through the Economist Intelligence Unit (EIU), which last I checked amounted for 30% of corporate revenues for the company. Not small change.

This could go somewhere.

Karsten M. Self
SCO -- backgrounder on Caldera/SCO vs IBM
Support the EFF!!
There is no K5 cabal.
[ Parent ]

great! (3.00 / 1) (#126)
by Shren on Wed Apr 25, 2001 at 11:30:38 PM EST

I'm glad that someone who cares more than I do is on the job.

[ Parent ]
Why not do it the way /. does it?? (1.58 / 12) (#50)
by Apreche on Mon Apr 23, 2001 at 09:33:18 AM EST

I say do it the way slashdot does it. There are a few people who go through and choose stories. Not everyone is a moderator, like here, either. Only certain folk. And don't tell me that it's a pain in the butt to root through all those stories, we're talking SLASHDOT here, they have more submissions than I have red blood cells.

I can't resist... (4.12 / 8) (#52)
by iGrrrl on Mon Apr 23, 2001 at 10:00:36 AM EST

I say do it the way slashdot does it.
Since this is your first comment, you may not know that Kuro5hin is notslashdot.org.

(Thanks for that, Sig11.)

You cannot have a reasonable conversation with someone who regards other people as toys to be played with. localroger
remove apostrophe for email.
[ Parent ]

But it really kind of is (3.00 / 3) (#61)
by Apreche on Mon Apr 23, 2001 at 12:59:22 PM EST

Kuroshin is so much like slashdot. It runs modified slashcode. It has links to freshmeat, sourceforge, and thinkgeek at the top of it. The only difference I can see between kuroshin and /. is the topic matter, the look of the site, the fact that everyone is a moderator, and the way front page articles are selected. Yeah Kuro5hin isn't slashdot, but it's the same kind of site. And considering how effective /.'s model is, why not be more like them?

[ Parent ]
I disagree (4.55 / 9) (#62)
by theboz on Mon Apr 23, 2001 at 01:16:25 PM EST

And considering how effective /.'s model is, why not be more like them?

Part of the problem is that most people here think slashdot's model is flawed. Moderation there is more of a popularity contest than it is here. The ability for everyone to moderate all messages is definitely a step in the right direction. It isn't perfect, but it is much better than slashdot's method of only giving moderator points to people that agree with the groupthink who moderate each other up and it all becomes a big wankfest.

Also, the editors on slashdot are not perfect. In fact they don't edit anything, they just post what they like or not, and have too many submissions to really be able to do a good job of that. It's very often that something is posted that later turns out to be bullshit. There are also duplicate stories because one editor can't be expected to remember every single story ever posted on slashdot.

Then, the speed that stories appear on slashdot is a huge problem as well. One of the reasons there are so many first posts is because stories come and go very quickly. You don't have much of a real discussion when things are going by that quickly. Remember those grey poupon commercials? Slashdot is like that. You only have time to maybe post a message or two and hope someone will reply and you have a minor discussion, then hurry and move on to the next article and repeat. You don't have time to introduce yourself and have a good conversation, you only have time for, "Pardon me, would you happen to have any grey poupon?"

Also with slashdot, nearly everything is MLP. If you ignore the crap posted by John Katz (it would be Op/Ed here) and the asinine "Ask Slashdot" (Dear Slashdot, I want to open an open source porno site to make a lot of money? What apache modules should I use?) crap everything is just a link to something else. No thought is required in posting articles or messages, it's just crap.

Kuro5hin seems to be different. I view it to be more like a cafe or a dorm. Lots of people come here for discussion, not current events and neat-o geek crap. The conversations are supposed to be mostly dealing with technology and posting an article about a neat new programming language is met with gladness, but you can also have a debate on philosophy. The main difference in the two sites is the discussion. That's why things take longer. If you have a conversation, it isn't very good to have a debate about religion for 30 seconds, then 30 seconds later talk about your favorite version of unix with 500 people all talking at once. Things should be slower here or else the site is useless.

Also, the real problem of stories not getting posted to the front page often is simply that most of what hits the submission queue is complete crap. It's mostly due to laziness and lack of inspiration I guess. We shouldn't just have a new article on the front page for the sake of having a new article. I think it should only be the quality stories that are posted there.

Also, I do think that there should be something done to adjust the thresholds to not include all accounts, but instead just the "active" ones...meaning the accounts that have posted a comment or story in the past 14 days or something. That would help some but it wouldn't fix the problem of the lack of interesting, high quality stories that are worthy of the front page. Perhaps you should write something good to post. Perhaps we all should.

[ Parent ]

Hear, hear (2.50 / 2) (#75)
by Kellnerin on Mon Apr 23, 2001 at 04:18:57 PM EST

And considering how effective /.'s model is, why not be more like them?

Kuro5hin vs Slashdot is like programming language evangelism. You need the right tool for the right problem. /.'s model works well for what they want to do. K5's goals are slightly different. You won't find timely "news for nerds" or the best mirrors to download RedHat 7.1, but you will (hopefully) find thoughtful articles and discussion on a wider range of subjects, less "FIRST PSOT"ing and hot grits. Kuro5hin is kuro5hin. There's no need to turn it into Slashdot; that site already exists.
--And is her power all in her club sandwich--

[ Parent ]

/. moderation (3.00 / 1) (#98)
by kmself on Tue Apr 24, 2001 at 03:32:10 AM EST

It doesn't work. Dig through my comments here to understand why. There are a number of serious problems.

I do give Slashdot a huge amount of credit -- the site pioneered the interactive weblog, made a huge number of mistakes for us (so we didn't have to), and have been tremendously supportive. Slash also scales pretty damned well, though it eats hardware to do it.

Slashdot comments work largely despite the moderation system. Sad, but true.

Karsten M. Self
SCO -- backgrounder on Caldera/SCO vs IBM
Support the EFF!!
There is no K5 cabal.
[ Parent ]

Apparently it works for some people (2.00 / 3) (#104)
by Kellnerin on Tue Apr 24, 2001 at 10:40:27 AM EST

I have read through your comments, all of them, and I agree in large part. I do prefer k5's model, imperfect as it is -- that's why I'm here and not there. I'm not advocating turning /. into k5 any more than I am the other way around, however. CmdrTaco et al are presumably happy with what they've created. Why not allow each one their strengths, and their distinctiveness? Surely that's the whole reason we have two separate sites.
--And is her power all in her club sandwich--

[ Parent ]
/.'s broken for reading thoughtful comments. (4.00 / 1) (#112)
by Another Scott on Tue Apr 24, 2001 at 04:19:35 PM EST


I greatly appreciate the work you've done here Karsten (and everyone else). I'm sorry to say I haven't figured out the moderation process yet, so I haven't contributed much to the site. I just occasionally read it.

I glance at /. a few times a week but rarely spend more than 5 minutes there at a time reading posts. There's just *too* much noise, even setting preferences to see only comments scored 3 and higher. It makes no sense to have a rating system which doesn't distinguish between thoughtful, on-topic comments and throw-away jokes and puns, IMHO.

/. would be a *much* better discussion site, IMHO, if they changed their scoring system or display system so that "funny" comments could be ranked and displayed differently (or not at all) for readers who've see enough juvenile humor.

On K5 - I'd prefer it if the site didn't require so much clicking around. It seems at times like a Biltmore House of a site with lots and lots of rooms which make it difficult to find one's way and feel at home. But I expect I'll feel a stronger connection once I understand the mechanics of using the site better, can "mark as read" (which I understand is coming soon), etc.


[ Parent ]
Thanks -- and UI improvements? (none / 0) (#113)
by kmself on Tue Apr 24, 2001 at 05:04:26 PM EST

Wait...you're one of them, aren't you?

First, the work credit goes to Rusty, Brent Metzler, Andrew Hurst, Dylan, and the other guys in the credits file who've done the coding. I've tossed in some ideas (and good ones IMO), but the original drive is Rusty's and implementation is theirs.

I share your assessment of /. Pity. Malda & co. know we're here though.

Clicking around -- how do you mean? Various sections and gewgaws? The interface is getting more and more complicated (I just discovered the comments preferences link under user settings -- about halfway through the right sidebar). What particular issues do you think could be improved? I find EZBoard to be a bit nav-heavy as well.

Karsten M. Self
SCO -- backgrounder on Caldera/SCO vs IBM
Support the EFF!!
There is no K5 cabal.
[ Parent ]

Yup. I'm a yammerer on IWeThey. UI issues. (4.00 / 1) (#114)
by Another Scott on Tue Apr 24, 2001 at 05:31:45 PM EST

I'm AnotherScott over there too.

On the "clicking around" I mean that it's just so flexible around here that it's tough for a newbie to know where s/he fits in. I applaud flexibility and crave it in my software, but it makes for a long, shallow learning curve. The lack of change on the Front page can make occasional visitors think that little happens here, so the idea of rotating popular threads from other areas of K5 to the Front, or an area of the Front, sounds good to me. Or at least a prominent pointer to those articles and discussions should be considered. But I agree that standards shouldn't be lowered.

IWeThey on ezboard has lots of problems too, but it's easy to come up to speed on the site. You quickly learn who the regulars are and who you can skip. Here there's so much *good* comments that it's overwhelming. You've got your Diaries and your OpEd and the Front and the novel Moderation issues and on and on. And lots of display options to sort through. I just haven't had the time to come up to speed on everything....

But the lack of a way to have the software skip over comments you've already read (or at least quickly note them with an LRPD or something similar) makes it really tough to keep up with what's going on here. I simply don't have time to visit more than a few times a week.

As I've said on IWeThey, I think web sites like this should take better advantage of discussion features which have a lot of history behind them, like Usenet. NNTP readers make it very easy for someone to read dozens of newsgroups and filter out wheat from the chaff. I dream of a site which has lots of thoughtful traffic, but has reader-defined score/kill files, is fast and offers minimal graphics for slow internet connections, and is friendly toward text-mode readers like Lynx. K5 seems to be heading that way, and I applaud the efforts.

I'll check around here more and try to give more thoughtful comments. Thanks.


[ Parent ]
Welcome to the House of Cheap Analogies (3.00 / 2) (#64)
by slaytanic killer on Mon Apr 23, 2001 at 01:58:12 PM EST

Ever noticed how in chess, one tiny move can change the entire position, forcing it to be reanalyzed from scratch? Think of the difference between Slashdot and Kuro5hin this way.

And take a look at this page, especially at the end, from the FAQ.

[ Parent ]
*sigh* (4.00 / 3) (#67)
by aphrael on Mon Apr 23, 2001 at 03:04:40 PM EST

The user moderation as opposed to moderation by a small cadre is one of the defining differences between K5 and /.. It was the reason a lot of the originalwave of immigrants came here, and remains important because it represents the distinction between an open, democratic site and a closed, autocratic one.

I may be overreaching here, but I think switching to a slashdot style of story approval would completely undermine everything that's cool about K5. There is no K5 cabal, and that's a *good* thing.

[ Parent ]

You're a troll, but I'll bite anyway... (5.00 / 1) (#77)
by avdi on Mon Apr 23, 2001 at 05:14:48 PM EST

First of all, Kuro5hin runs Scoop, not Slash. It's a completely different engine.

It has the same links across the top because both sites are affiliated with OSDN/Andover/VA Linux. It helps pay the bills.

And a great deal of the readers here are here because they feel that Slashdot's model is broken. Kuro5hin makes a serious effort to be a different kind of site thank /.. They serve different crowds. If you think that /.'s model is so effective, then I suggest you stick with that site. It's probably more suited to you.

Incidently, what do you mean by "effective"? I can't see that there is anything about Slashdot that the average K5 member thinks is "better" than K5. I think most people here would say that K5 is obviously more effective than /.. But maybe you have a different definition of "effective"?

Now leave us, and take your fish with you. - Faramir
[ Parent ]

Quality, rather than Quantity (4.33 / 9) (#51)
by Renegade Lisp on Mon Apr 23, 2001 at 09:34:47 AM EST

I come to Kuro5hin because I'm looking for Quality. It would be fine to see a new story or two each day on the front page, but if no such story exists, I would rather see nothing new, at least on the front page.

If the high standard of the front page eroded, it would place the burden on me again to wade through the articles and spot the real good ones among those that were just posted because there was nothing else.

Current system kills controversial issues (3.20 / 5) (#54)
by Tezcatlipoca on Mon Apr 23, 2001 at 10:22:40 AM EST

We have a situation where people vote 50-50 to post something, the think is something terribly devisive, there are good arguments pro and agaisnt posting and most probably bunches of comments.

Well, the current system will dump such an history. That is not good in my book.

The FP should be a little bit more dynamic: if there was no article worthy of FP under the current rules then the one with the bigger rarting should make it every 24 hours or so.

Currently the FP gives the wrong impression: that there is nothing going on.

Might is right
Freedom? Which freedom?
[ Parent ]
One reason (4.00 / 9) (#56)
by theboz on Mon Apr 23, 2001 at 11:51:50 AM EST

A lot of the controversial subjects (religion, politics, microsoft, etc) are presented in a very biased view that doesn't praise one side, but bashes the other and often is not true. While it could be a good subject to discuss, if the article is presented to me in a way that seems like it's just bitching about something without any facts or statements made in sweeping generalizations of ignorance, I have to vote them down. These stories still often get posted though so I'm not sure.

It's usually the controversial types of stories that are against the groupthink of "geeks" that get voted down. I hate those rotten no good scum-sucking pencil-neck geeks.

[ Parent ]

Perfect! Fine! We'll have one! (3.33 / 3) (#84)
by leonbrooks on Mon Apr 23, 2001 at 10:29:38 PM EST

A lot of the controversial subjects (religion, politics, microsoft, etc) are presented in a very biased view that doesn't praise one side, but bashes the other and often is not true. While it could be a good subject to discuss, if the article is presented to me in a way that seems like it's just bitching about something without any facts or statements made in sweeping generalizations of ignorance

Your idea of ignorance, or someone else's? (-:

My modest proposal: add a new section ``controversial'' which gets articles when the absolute value of the votes about them achieves a certain threshold.

You could include a mechanism to ensure that at least one story from ``controversial'' made it to FP every week or two. This would pretty much guarantee a lively FP, and help to prevent the kind of /. politics that Signal 11 despises.
-- If at first you don't succeed, try a shorter bungee
[ Parent ]

quality, sure, but how about interesting too. (4.00 / 5) (#81)
by G Neric on Mon Apr 23, 2001 at 06:14:11 PM EST

quality comes from smart, articulate, educated, clever, and funny people, not from laboriously researched and painstakingly documented fussy little pieces. A high quality discussion can ensue from any news story if people find it interesting, and 10 times a day if there are that many stories. Once a week, some long-winded philosophical piece (that is not topical to anything going on in the real world at that moment) will not particularly engender quality.

My point being, I agree with you about quality, but reject the idea that Kuro5hin is doing exactly what it takes to mine that vein.

[ Parent ]

God fscking bless you (3.00 / 1) (#89)
by kmself on Tue Apr 24, 2001 at 01:41:35 AM EST


I'd like to kiss you. And I don't say that to all the guys. Or girls. ;-)

You're seeking the same goal I am. Any suggestions on mechanism? Or comments on what I've suggested as fixes?

Karsten M. Self
SCO -- backgrounder on Caldera/SCO vs IBM
Support the EFF!!
There is no K5 cabal.
[ Parent ]

Boooooooring Academia (4.25 / 4) (#107)
by zephiros on Tue Apr 24, 2001 at 10:48:09 AM EST

There's nothing worse than tedious, well thought-out stories that make for little to no discussion. Personally, I think the following should be requirements for any story posted to K5:

Little to no research. Whenever possible, any actual understanding of the subject matter should be avoided. This insures a huge amount of "discussion" as subsequent posters point out factual errors, prior research, and naive conclusions within the article.

Oversimplification of complex issues. For example, "all drug users are dirty hippies who need to get a job." This insures lots of "discussion" as people cleverly articulate how narrow this view is.

Poorly supported extremist views. Article authors should be outlandish caricatures who tirelessly rebut anything that that falls outside the scope of their bizarre, myopic political theory. This allows sharp-witted K5 posters to play an endless game of "knock over the straw man."

Or, alternately, we could have well written, well researched articles that identify complex problems which require actual thinking to solve.

[ Parent ]

Thoughts on voting (4.37 / 8) (#55)
by ghjm on Mon Apr 23, 2001 at 11:19:50 AM EST

First of all, I'm not sure that giving people a choice of +1 FP / +1 Section is really working. I think most people simply treat +1 FP as if it were +2. In other words, if I like a story, it's +1 section; if I really really like a story, it gets +1 FP. This leads to dysfunctional results.

Let's take an example; imagine that a pretty good story gets posted. It isn't the best we've ever seen, but it's solidly worthwhile. Nobody (or a handful) votes it down or zero - essentially everyone votes +1. About a third of the people who vote feel that this story is really good, so they vote +1 FP to give it a bit of extra "oomph." The story doesn't make the front page because substantially more people voted section. Was this really the will of the people?

I guess what I'm trying to get at is that a vote of +1 Section doesn't necessarily indicate that the voter actively wishes the story wouldn't make it to the front page, only that they personally don't want to contribute to its getting there. Perhaps a vote of +1 Section should be neutral in terms of front-pageness? Or perhaps it would be best if users simply voted on the merits of a story, and front page status arose from some function of the voting behavior?

Personally, I'm in favor of the latter. I know when I vote on a story, if I want to give a +1 I'm always baffled which one to choose. I read kuro5hin on the "Everything" page so I really don't care if the story goes to the section or FP. However, if I want to express my preference for the story by awarding it a +1 vote, I am forced to choose between section and FP; so I pretty much choose at random. What I'd like is a way to say "+1, Whatever."

I also think -1/0/+1 doesn't provide quite enough range to fully express an opinion on the story. Crap is crap, I'm happy with -1 and 0. I don't think a -2 vote is a good idea because it makes it too easy to kill stories out of personal dislike for the author or what have you. But on the upside, there's a huge variation between a story that's just barely good enough to tip voters over into +1 territory, and the best story ever posted to kuro5hin. I'd like to see the possibility of +2 or maybe even +3.

There is a psychological issue with this: It violates the principle of "one person, one vote." Why should someone else get 3 votes to my 1? Of course this is silly: Everyone has 3 votes. But this sort of issue does bias behavior. So perhaps instead of allowing a +2 vote, we should allow a +0.5 vote. Now it's clear that everyone still has exactly one vote, but you can choose not to spend all of it.

The final twist I have to offer is that making topical comments in queue should bias your vote by +0.5. If you vote -1 but then comment, your effective vote would be -0.5. However, to maintain equity, you are still capped at +1 even if you vote +1 and also post. Of course this would only apply to the first comment; your second and subsequent topical comments wouldn't make any difference.

I know this is a bit complex but I think it solves the problems people have pointed out. A recap of my proposed system: People have five voting choices: -1, -0.5, 0, +0.5 and +1. They do not directly choose between FP and section. In addition, making one or more topical posts in queue adds +0.5 to your vote, to a maximum of +1.

  • Front page activity: We've taken away direct user participation in front page article selection. The highest rated articles go to the front page, based on a threshold chosen by the site administrators, the same way articles make it to section pages now. By adjusting these thresholds, the site administrators drive a pleasing number of articles onto the front page - though note that the site administrators control is long-term; there would still be fewer FP articles during slow times. (This is why I don't like "3 per day" type schemes.)

  • Topical comments languishing in queue: There would still be some potential for good topical comments to disappear, but it becomes much less likely. Stories that immediately generate a vast amount of topical discussion would tend to rocket out of queue. Note, however, that it is still possible for a user to post topically yet vote down the article - but it has to be much more intentional.

  • Topical "I voted -1 because" posts: People who want to vote -1 will have to make their complaints as editorial comments, to preserve their -1 vote. This is good. If you really think the article sucks, you should be able to say so editorially. If on the other hand you disagree with the article yet feel passionate enough about it that you *must* make your post topical - you must at least believe that the article could generate topical discussion, so the worst you can do to it is -0.5. Maybe it's flamebait, but you were hooked, so at least it's *good* flamebait. :-)

  • Lurkers: One of kuro5hin's current strengths is that it encourages participation by lurkers. People who for whatever reason don't feel compelled to write articles or comments can still be a part of the community by voting on things. I think it's important that this system doesn't penalize lurkers. You can still give an article +1 or -1, without posting. Those who post can't disenfranchise those who don't.

  • Abuse: I don't think this system changes the potential for abuse very much. There's no incentive to spam the discussions; once again, you can't do any more damage with a post than you can with a vote. We'll still have the problem of a user with multiple accounts voting his/her stories up, but I'm not trying to solve that problem. :-)

    Assume I could think of a witty closing paragraph and pretend it appears here.


  • Maybe it is a UI issue? (4.28 / 7) (#57)
    by mairas on Mon Apr 23, 2001 at 11:56:06 AM EST

    About the submission moderation procedure. As it currently is, it is somewhat clumsy and heavy, to the extent that I often just skim the headlines without moderating anything. There just is too much clicking and waiting for pages to load (I am located in Finland, Europe).

    I think the procedure could be quite easily flattened by making it possible to view the pending stories list (?op=modsub) so that the introductions would be shown in the list, and you could give votes on all pending stories at once. This would, IMHO, make the moderation procedure infinitely faster. You would have to wait only for a single (albeit longish) page to load for the whole voting procedure, instead of three pages per story at the moment (once the story listing, then the story itself, then the story again after submitting the vote, then back to the list).

    With this system in place, more people would probably vote, thus increasing the number of stories on the front page. Furthermore, it would automatically solve the problem with submission queue size mentioned in the story.

    Granted, this fix would not address other problems mentioned in this discussion, but at the moment it seems to me that a poor user interface is affecting the operation of the whole site.

    Re: Maybe it is a UI issue? (3.50 / 2) (#79)
    by SEAL on Mon Apr 23, 2001 at 05:29:45 PM EST

    This is an excellent point. Part of the problem is that there are SO MANY bad submissions to the queue that I get sick of wading through the junk. Many of these are obviously bad just from the title or the introduction. If there was a way to quickly vote on everything in the queue from a single page, I'd be a happy camper.

    - SEAL

    It's only after we've lost everything that we're free to do anything.
    [ Parent ]
    That's part of it (none / 0) (#96)
    by kmself on Tue Apr 24, 2001 at 03:16:54 AM EST

    I've had similar thoughts. I'd like to see the intro, current vote status, and score, of items in queue. There's a fair conclusion I can come to with just that data.

    Note that if intro is included in the queue display, that limits on length of intro text should be imposed. IMO intro should be no more than, say, 250-350 characters (about the length of the "Spamming is not tolerated here" message in the Post Comment dialog).

    The other side is that by reducing queue resident time, the length of the queue itself would be greatly reduced. Currently we've got a queue of 21 items, the oldest of which is four days old. Assuming four days is average, cutting this to, say, twelve hours, would reduce the queue to about three or four items with the same article load as we're currently seeing. The trick is that fewer people moderate each article through the submission queue.

    Karsten M. Self
    SCO -- backgrounder on Caldera/SCO vs IBM
    Support the EFF!!
    There is no K5 cabal.
    [ Parent ]

    Queue resident time shouldn't be less than 24 hrs (3.00 / 2) (#101)
    by mairas on Tue Apr 24, 2001 at 05:41:43 AM EST

    Note that if intro is included in the queue display, that limits on length of intro text should be imposed. IMO intro should be no more than, say, 250-350 characters (about the length of the "Spamming is not tolerated here" message in the Post Comment dialog).

    I agree. Actually, this way the article submission procedure would resemble that of many scientific conferences - you leave an abstract of limited length, and your paper gets either accepted or rejected based on the abstract. If it is sufficient for enforcing the quality of conference papers, it should do for k5 as well...

    The other side is that by reducing queue resident time, the length of the queue itself would be greatly reduced. [...] cutting this to, say, twelve hours, would reduce the queue to about three or four items with the same article load as we're currently seeing.

    If people would vote on articles they really don't care about or want to dump (as opposed to not bothering to take the time to do it), it should reduce the queue size by itself. The queue resident time could be cut as well, but probably not as radically as you suggest. If it is less than one day, there is a risk that people would miss the submissions because of time zone differences.

    [ Parent ]
    Appropriate dwell time, but less is more (3.00 / 1) (#111)
    by kmself on Tue Apr 24, 2001 at 03:03:02 PM EST

    Queue duration for news and MLP should be relatively short -- not more than an hour or so. You're not evaluating the final quality of the submission, just deciding if it's fit to print. For more details, see this comment.

    Other sections, less time-sensitive, could have a dwell of up to a few hours, but again, the primary concern is to identify issues with the post that can be repaired. The submission queue becomes just a place for evaluating and revising submissions, not for coming up with a firm rating for them. Any dwell past a few hours is going to lead to pressures for topical conversation -- not the purpose of the queue.

    Your "risk" -- that people might miss submissions because of time zone differences, is my feature. I don't want to see every bloody submission to the queue. Collaborative filtering means making inferences based on the judgement of others. In this case, a quick go/no-go decision in the queue, and a more sedate "how good is this article" concensus formed after it hits the deck.

    Short, quick, pointed feedback. That's the ticket.

    Karsten M. Self
    SCO -- backgrounder on Caldera/SCO vs IBM
    Support the EFF!!
    There is no K5 cabal.
    [ Parent ]

    two queues to solve a three-way battle... (3.57 / 7) (#58)
    by Puchitao on Mon Apr 23, 2001 at 12:08:47 PM EST

    For each article, there are two battles: Post it/Dump it and Front page/Section page. Right now, the system treats this as one big battle between three parties (FP/Section/Dump) and it appears that FP is losing.

    Perhaps we shouldn't conflate the two. Make it two battles: make there be two queues. The first queue is the ordinary queue we have now, minus the +1 FP choice. Articles sit in it, like they currently do, and once they pass the threshhold, they go onto their section page.

    Then and only then does the FP battle start. Once a story makes it through the queue, they're put in a second queue where people decide whether it's worthy of FP status. Same sort of voting system: +1 FP, 0 don't care, -1 Keep it where it is. (Or maybe just +1 FP and 0 Keep it where it is).

    It's a pretty simple process: first, we decide what articles to keep, and then we decide which among those we want to show the world.

    Perhaps we can do *snappy fun* with you everytime! -- Orz
    Three battles even (3.50 / 4) (#65)
    by aphrael on Mon Apr 23, 2001 at 02:59:37 PM EST

    because there's the 'dump it because this will never be worthwhile' vs the 'please rewrite this so it's better' debate.

    [ Parent ]
    Almost (none / 0) (#97)
    by kmself on Tue Apr 24, 2001 at 03:21:05 AM EST

    Read this

    Karsten M. Self
    SCO -- backgrounder on Caldera/SCO vs IBM
    Support the EFF!!
    There is no K5 cabal.
    [ Parent ]

    Easy way.. (3.75 / 4) (#59)
    by henrik on Mon Apr 23, 2001 at 12:44:01 PM EST

    Just implement a configurable per user fp/section vote threshhold where the users can choose how good the stories have to be before they end up on their front page (and leave the default for non logged in users).

    Scoop hackers - would this be hard to implement?

    After all - the K5 "motto" is "YOU choose the stories!". Making the threshhold configurable would just follow in that tradition.


    Akademiska Intresseklubben antecknar!

    Come to think of it.. (3.00 / 2) (#60)
    by henrik on Mon Apr 23, 2001 at 12:48:17 PM EST

    .. having configurable threshholds all around (how many points a story needs to be posted for example) would be really nifty. (it's probably not practical w/o beefy hardware and creates some other interesting problems)

    It would sure go a long way towards YOU choose the stories! though :)


    Akademiska Intresseklubben antecknar!
    [ Parent ]

    Interesting... (none / 0) (#117)
    by Canar on Wed Apr 25, 2001 at 12:14:13 AM EST

    I am in agreeance. Never having served dynamic HTML, or indeed, HTML at all from a server, I don't know the load amounts, but this would be a really interesting idea, and would give the functionality to do a lot of interesting things. For example, (although not entirely very feasibly) if we could spare the CPU cycles, maybe we could even match the voting patterns of different peoples with our own to better customize our display.

    On the other hand, this would take CPU. It would take a bit of time to implement this code into Scoop. But other than that, what drawbacks are there? A bit of community fragmentation, perhaps, a la the Diamond Age's newspapers? I think it's a good idea...


    [ Parent ]

    SO SAD (2.71 / 7) (#66)
    by farl on Mon Apr 23, 2001 at 03:00:37 PM EST

    This article is the perfect example of what SHOULD NOT get to the front page. The reason for the front page is for GOOD, WELL WRITTEN articles. Not gripes and complaints.

    Maybe we would have a better front page if people wrote better stories. Taking the ideology of "There are not enough front page stories therefore we should lower our standards just to stroke some illeterate writers ego" is pretty stupid IMHO.

    But that is just me. Personally I prefer to write articles with some substance. When I have made the front page, it has been with articles that took many days, if not weeks, of work on them. Not some slapstick drivel.

    Yeah, sorry about that. (3.60 / 5) (#73)
    by DoorFrame on Mon Apr 23, 2001 at 03:55:36 PM EST

    In all fairness I didn't intend for this story to get to the front page. I figured it would be looked at, maybe provoke a few people to write their own, better, pieces or maybe even do some scoop editing, and move on. If anything, I figured it would just be posted to the section. Don't be upset about the quality, I was just sad not to see anything on the front page for over a week.

    [ Parent ]
    Author option (5.00 / 1) (#94)
    by RadiantMatrix on Tue Apr 24, 2001 at 02:37:52 AM EST

    This is exactly why I wish, as an author, that I had the option to prevent a story from being FP if I so desire. Posting an editoral comment asking it not to be FP works most of the time, but not always.

    never put off until tomorrow what can be done the day after.
    Express Yourself

    [ Parent ]
    My thoughts. (2.66 / 3) (#71)
    by k5er on Mon Apr 23, 2001 at 03:40:23 PM EST

    Kuro5hin is not a news site (even though many use it as one). It is meant as a discussion site. If no posts are worthy of discussion, or worthy of being "front page news" then I don't want to see them posted. If people would write better articles, then the page would be updated more often, but I don't think Kuro5hins's standards should be lowered just because it hasn't been updated very much over the last while.
    Long live k5, down with CNN.
    make K5 available as XML? (4.00 / 6) (#72)
    by Mindcrym on Mon Apr 23, 2001 at 03:44:28 PM EST

    Regarding the slowness of the UI; would it be possible (if not already available) to make K5's content available as XML? This would allow other people to create frontends that could provide a more summarized view of the story submissions and perhaps streamline K5 viewing for people with slower connections.

    It is. (4.00 / 1) (#93)
    by RadiantMatrix on Tue Apr 24, 2001 at 02:30:26 AM EST

    There is K5 XML generated in order to function with ForumZilla. A simple parser in your favorite XML tool or programming language, and you could easily manipulate K5 XML data.

    never put off until tomorrow what can be done the day after.
    Express Yourself

    [ Parent ]
    The New New Economy (2.80 / 5) (#78)
    by r0cket on Mon Apr 23, 2001 at 05:20:18 PM EST

    All the slackers who used to spend the whole day at work moderating k5 stories have gotten laid off.

    expectations and standards (3.25 / 4) (#80)
    by mezzo on Mon Apr 23, 2001 at 05:34:30 PM EST

    It's all about what we expect kuro5hin to be.. if we consider it one of our primary source of news, then the FP is disappointingly slow. But if its mainly a discussion site, then it's fine (though still a bit slow methinks).

    The thing is that people still come to kuro5hin looking for news. And so they just skip the FP, and skim through the moderation pool. So yes, the standards for a story making FP is really high, but if it causes regular users to skip the FP and accept the lower standards of the moderation pool... then what's the point? It just makes our FP pretty for newcomers and visitors.

    "The avalanche has started, it is too late for the pebbles to vote."-- Kosh
    Once a story made it to FP, what happens ? (3.25 / 4) (#83)
    by mami on Mon Apr 23, 2001 at 08:38:50 PM EST

    How many more comments are made after it is out on the FP, or in the section ?

    My impression is that most of the topical discussion is already done in the queue and only some articles continue to be discussed after they are released. That's kind of odd and most probably not what was originally intended. Any idea to avoid it ?

    An associated fix (3.00 / 1) (#88)
    by kmself on Tue Apr 24, 2001 at 01:23:34 AM EST

    Stories will be coming out of queue faster than they are now, preferably with topical comments disabled while in queue. The fact that most discussion happens in queue currently is a bug.

    Karsten M. Self
    SCO -- backgrounder on Caldera/SCO vs IBM
    Support the EFF!!
    There is no K5 cabal.
    [ Parent ]

    Deep flaw in the system (4.33 / 12) (#85)
    by Global-Lightning on Mon Apr 23, 2001 at 11:50:08 PM EST

    As I've read through this topic and others in the <ahref="http://www.kuro5hin.org/?op=section;section=Meta">meta category, I've seen some very elaborate, well thought out changes and tweaks to the voting system. Some are simple, other complex, but they try to repair some weakness or another in the submission voting system.
    The one question that has yet to be asked is fundamental to Kuro5hin:

    "Does the concept of voting help or hinder the K5 community?"

    In this post I shall demonstrate that the voting system is fundamentally flawed and propose a radical, yet much simpler solution.

    The mission of K5 is a very simple one:

    Kuro5hin.org is a community of people who like to think.  You will not find garbage in the discussions here, because noise is not tolerated.  This is a site for people who want to discuss the world they live in.  It's a site for people who are on the ground in the modern world, and who sometimes look around and wonder what they have wrought. [emphasis added]
    If I may summarize this, this site promotes discussion of the world we live in. To this goal, participants are allowed four behaviors:
    • Posting Comments. This is the core behaviour. All other actions must support this
    • Rating Comments. This is a self-imposed pruning of comments to ensure the most insightful (or at least popular) ideas are given greater exposure for more comments
    • Submitting Stories. This supports discussion by giving us something to discuss!
    • Voting on Submissions This is another pruning action applied towards the submissions themselves. For each story, one may cast a vote against, move to section, or move to front page.
    Now, if voting does truly support the site mission of promoting discussion, we would expect to see a direct correlation between the number of votes a story receives, and the quantity and quality of discussion generated. This would be expressed by Front Page stories having a higher number of comments than other stories.
    Unfortunately, this is not the case. Revising the stories in the various sections, one sees numerous instances of: The relationship between voting and commenting is weak at best. What is needed is a system that directly ties a story's placement or abandonment to the discussion it generates. I thus propose the following system to replace voting:
    1. New stories shall be placed in the Submission Queue as they are now
    2. Once in the Queue, they shall be given a period of time to generate discussion (say 24 hrs)
    3. At the end of the evaluation period, if the number of people who post topical comments to the story exceeds a certain threshold, it shall be moved to the proper Section page. If it exceeds a higher threshold, it shall be moved to the Front Page. Otherwise, the story shall die
    4. Should the number of people who post comments to the story while it's in the Section page exceed the higher threshold, is shall be moved to the Front Page.
    This system is much simpler than the current voting system or any other proposed, and it directly ties the story moderation and placement process to this site's mission.

    more open to abuse, I fear (4.00 / 2) (#102)
    by Glacky on Tue Apr 24, 2001 at 06:21:21 AM EST

    Just to think of a few examples, how do you determine whether a story has achieved the FP based on the discussion? total number of comments? comment rating score? average comment score?

    Each of these methods are open to abuse by a (small) group of people who wish to promote or hide a particular item based on its content rather than its discussion-worthiness. at least the voting process is mainly democratic and not open to the same 'pressure group' tactics.

    Nice idea, in an 'if only people played by the rules' way :-)

    [ Parent ]
    One person, one vote. (2.00 / 2) (#105)
    by interiot on Tue Apr 24, 2001 at 10:41:22 AM EST

    What about... the number of people who have commented on a story?

    Which still has other drawbacks... namely that some people may think that the information in a story or discussion should be disseminated more widely, but can't vote for it without adding noise to the discussion. But it solves the aforementioned problem.

    A for the original comment, I contend that another of K5's goals is to disseminate unique insights of common people, because there are few other ways to do such a thing. K5 doesn't have narrow its stories to only controversial topics.

    [ Parent ]

    What I was thinking... (2.00 / 1) (#115)
    by Global-Lightning on Tue Apr 24, 2001 at 09:09:21 PM EST

    ... was counting total number of persons submitting comments, not count comments that had dropped below a certain comment rating as the "noise level".
    This way the current comment moderation process would eliminate the chatter of those trying to pump up a storie's viability.

    The system would have 4 variables:

    1. t = period of time a story is allowed in the queue
    2. Ps = # of persons commenting needed to move the story to the Section page
    3. Pf = # of persons commenting needed to move the story to the Section page
    4. n = noise level below which comments won't be considered
    Putting it all together:
    At time t after submission, calculate k=the number of people who have at least one comment rated higher than n.
    If k>Ps: move to Section page
    if k>Pf: move to Front page
    else delete story

    [ Parent ]
    What's a person (none / 0) (#118)
    by kmself on Wed Apr 25, 2001 at 12:45:06 AM EST

    There's no way to authenticate accounts to people at K5 or within Scoop. It's an inherently m:n relationship. We've already been through this. One reason for going to a rating rather than voting system is that it tends to reduce this bias. It's something that has to be dealt with.

    I don't like your suggestion largely as it ignores the editorial function of the submission queue. That's more-or-less what it was intended for, and what it's failing miserably at now.

    Karsten M. Self
    SCO -- backgrounder on Caldera/SCO vs IBM
    Support the EFF!!
    There is no K5 cabal.
    [ Parent ]

    Number of comments (2.00 / 1) (#109)
    by theR on Tue Apr 24, 2001 at 01:50:49 PM EST

    First, the story you linked to by me was never on the front page. It was posted on the 13th of April, and if it had gone to the front page it would still be there.

    Second, the voting process for story submissions works fine. Once customizable Front Pages are implemented (and I believe this has been planned), everybody will be able to make the front page look acceptable to their wishes, or at least moreso than now.

    The idea of basing a story's value on number of topical comments or number of people who comment makes little sense to me. I voted this story submission down because it is a gripe that is easily worked around at present (by going to the Everything section if you want a less static front page) and will be fixed at some point by rusty and the others. Yet I have commented to the story twice. Just because I write a couple comments, doesn't mean I think the story should be posted, and I bet others agree.

    Also, in my opinion, the amount of discussion in no way has to relate to the value of the discussion. From the FAQ, "You will not find garbage in the discussions here, because noise is not tolerated." By making the number of comments or commentators valuable towards getting stories posted, you will inevitably promote noise and the amount of noise will increase more quickly than with the current system.

    [ Parent ]
    Why I don't Want to Change the Front Page (3.50 / 6) (#103)
    by Morn on Tue Apr 24, 2001 at 06:56:21 AM EST

    I like K5 the way it is. I like that I can go away for a week when I'm busy, come back, and the most interesting stuff is still here for me to read.

    I like the self-imposed high standards.

    If I've got time, I'll have a look through the submission bin, but I often don't, and it's nice to be able to look at the front page and see the most important stories there. Changing this would, for me, turn K5 into 'just another weblog', and that would No Good Thing (tm).

    On a slightly different not, I've abstained from poll voting. I'd rather vote for 'Whenever a story is good enough'.

    That is what you get (1.00 / 1) (#110)
    by cable on Tue Apr 24, 2001 at 02:01:09 PM EST

    when you make it so that the "Groupmind" has to decide what articles to make the front page. Heaven help anyone who doesn't think like the "Groupmind" and post an article that interests them. Thinking differently than the majority is definately not rewarded here.

    Also rejecting a story for spelling and/or grammar is idiotic because an editor can fix that stuff! Or should we sh*tcan every article that has a typo in it?

    Change the frontpage stories every week, I am tired of looking at the same crap each day. If the new articles are not voted to become frontpage articles, I think the editors should pick one. Or maybe the users should vote on it?

    My article on Classic Consoles got over 100 votes, but failed to get the 123 needed in the timeframe it was being voted on. On the second revision over 71% of the voters said they used a Classic platform. Yet it gets voted down anyway? What is up with that? Most of you use them, but don't want to read an article on them?

    Only you, can help prevent Neb Rage!

    The irony.... (4.00 / 1) (#121)
    by ucblockhead on Wed Apr 25, 2001 at 12:18:10 PM EST

    In the 48 hours that this story hit the front page, complaining that only there'd been only one front page story in a week, four stories have hit the front page.

    That seems to me a pretty good indication that when people think a story worth being on the front page shows up in the queue, it'll get there.
    This is k5. We're all tools - duxup

    Another thing (none / 0) (#124)
    by theboz on Wed Apr 25, 2001 at 03:03:43 PM EST

    The lull coincided with Easter as well. Many people get spring break and such then, so I would assume less people were reading, posting and voting here.

    That did give me a good idea though. Why not make the thresholds reflect only "active" accounts that have been voted or posted comments or something in the past 2 weeks or so? Do you think rusty would go for that sort of thing? I'm not sure how difficult that would be to track when the person last did something, but it could be very useful.

    [ Parent ]

    Memory (none / 0) (#125)
    by ucblockhead on Wed Apr 25, 2001 at 03:18:32 PM EST

    I have a vague memory of that being discussed at some point, though I don't think that would have helped if the issue this time around were spring break.

    Perhaps instead of being just a percentage of the user base, it could require a percentage of the number of people who voted in the last month. A fairly high one, like 50%, perhaps. That might be a little more robust when dealing with variations in voting rates.

    This is k5. We're all tools - duxup
    [ Parent ]

    Why isn't this user configurable? (4.00 / 1) (#128)
    by hepster on Sun May 13, 2001 at 12:34:11 PM EST

    Is there any reason people can't choose their own thresholds?

    A Less Idle Front Page | 128 comments (127 topical, 1 editorial, 0 hidden)
    Display: Sort:


    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!