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

[P]
Popcorn queue

By delmoi in Meta
Sun Apr 29, 2001 at 09:34:16 PM EST
Tags: Kuro5hin.org (all tags)
Kuro5hin.org

Ladies and Gentlemen: The queue is full. It has spilled over onto another page. And I think that might be bad for kuro5hin. The last time the queue reached a size anywhere near this, Rusty wrote some code to expedite the processing. Of course, no one wants to take 'due process' away from stories, that is, cut them off before a reasonable sample of votes can be taken. I think I have a solution.


It seems to me, that after an initial flurry of votes, the number of votes a story gets per unit time gets to be less and less. And if the more voracious voters don't come to a consensus, a story could end up being stuck in the queue for a week. The simple solution might be simply to lower the hide/post thresholds, or lower the total-vote processing threshold. The problem there is that people who might want to vote on a story might not get a chance, they may not even be able to see it. I think that's the reasoning for doing such, anyway.

Another solution would be base the post/drop/process thresholds based on the number of K5 users who regularly vote, rather then the total number of users.

Then, there's my solution, what I call the "Popcorn queue". The popcorn queue works the same way you cook popcorn in a microwave or on a stove. When you cook popcorn, you want to try to maximize the number of popped-kernels without burring anything. So you wait until the popping slows down enough that you can be sure you won't loose to many kernels.

Ideally, kuro5hin should allow everyone who wants to vote an opportunity. Of course, we can't wait for everyone to vote and still have a usable site. I think the metaphor of cooking popcorn applies well here, and I think it's a metaphor we can extend to get a better voting system.

Basically, what I propose is this. Record how many votes a story gets each hour (or some other arbitrary unit of time). If its less then a certain threshold, The story gets processed (either dropped or posted, based on the system laid out here).

I don't think this approach would be very computationally challenging, you just have to store a timestamp along with each stored vote, and calculate once in a while. Depending on how the database is laid out, this could probably be done with just one SQL statement (something like: select from count(*) storyvotes where story=thisone and timestamp > one time unit ago )

Of course, this isn't taking into consideration programming time, but it doesn't seem like something that would require too much work.

Before everyone jumps on me with the obvious criticism: "Well, if the queue is so full, why are you just adding more stuff to it." Well, it took me a long time to, and I'm worried that maybe people are holding off posting new stories (and not just meta stories) because the queue is so full. Obviously I don't have any real data on how the queue.

as for the other obvious criticism, scoop.kuro5hin.org is down :P

Sponsors

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

Login

Poll
Can code be art?
o YES! 25%
o Yes. 21%
o eh 9%
o No 0%
o NO!! 2%
o I (the reader) am superior to you in every measurement imaginable. 41%

Votes: 108
Results | Other Polls

Related Links
o Scoop
o Kuro5hin
o here
o Also by delmoi


Display: Sort:
Popcorn queue | 46 comments (40 topical, 6 editorial, 0 hidden)
Submission Queue Story of the Week (3.40 / 10) (#1)
by Carnage4Life on Sat Apr 28, 2001 at 07:35:02 PM EST

The submission queue is broken, you know it, I know it, rusty knows it. There have been various solutions have been discussed in numerous articles (now dead) and we still haven't seen any progress. Until rusty tells us what's up I don't see the point in Yet Another Submission Queue Article.

I havn't seen a story like this in quite a while (3.00 / 2) (#4)
by delmoi on Sat Apr 28, 2001 at 07:54:40 PM EST

And the queue really is overloaded. What I'd really love to see is having a averaged score (like comments) and then being able to set thresholds (Say, show me all the stories above 3.4 or whatever)

But I don't see that happening. I think this solution would solve a lot of the problems, without requiring much work to implement, or significantly altering the way kuro5hin worked.
--
"'argumentation' is not a word, idiot." -- thelizman
[ Parent ]
Stealing Ideas? (4.50 / 2) (#10)
by Carnage4Life on Sat Apr 28, 2001 at 08:19:24 PM EST

And the queue really is overloaded. What I'd really love to see is having a averaged score (like comments) and then being able to set thresholds (Say, show me all the stories above 3.4 or whatever)

Is this sarcasm? This is exactly the same idea rusty proposed 3 weeks ago..

[ Parent ]
Yes I know (2.50 / 2) (#14)
by delmoi on Sat Apr 28, 2001 at 08:35:54 PM EST

I was reiterating.
--
"'argumentation' is not a word, idiot." -- thelizman
[ Parent ]
Err, sorry (3.00 / 2) (#15)
by delmoi on Sat Apr 28, 2001 at 08:42:11 PM EST

Actually, I wasn't reiterating, as my last post said, although I had read Rusty's comment before I made mine. The solution I outlined there is actually a bit different then rusty's, in that you can set your threshold to whatever you want (very low or very high), rather then having a pre-set threshold for every one (in this case, 1). I also didn't mention anything about story editing, etc, etc.

I had also thought it up a long time ago. (I had only heard of rusty's suggestion today). I don't think I am 'stealing' his idea.
--
"'argumentation' is not a word, idiot." -- thelizman
[ Parent ]
No difference (5.00 / 2) (#16)
by Carnage4Life on Sat Apr 28, 2001 at 08:52:20 PM EST

The solution I outlined there is actually a bit different then rusty's, in that you can set your threshold to whatever you want (very low or very high), rather then having a pre-set threshold for every one (in this case, 1).

From rusty's comment:
Registered users would be allowed to set their own preferred thresholds, replacing the defaults. Want to see evrything? Just set your threshold to 1. Only the best? Set all thresholds to 4.


[ Parent ]
Oh, ok. (4.00 / 3) (#17)
by delmoi on Sat Apr 28, 2001 at 08:55:22 PM EST

Well, that would be what I would like to see. But the question is how long it would take before that arrives. And whether or not we should just let the site 'slide' in the meantime. Just because the site is going to be overhauled eventually doesn't mean that tweaking it could hurt.
--
"'argumentation' is not a word, idiot." -- thelizman
[ Parent ]
Blah (2.50 / 4) (#9)
by regeya on Sat Apr 28, 2001 at 08:09:48 PM EST

It's mindless navel gazing! You can't tell me how to vote! Just let the site grow, man! If you resection to MLP, I'll vote it up! It's rusty's site so let him figure out on his own how to fix it since we certainly won't have better ideas than rusty!

[ yokelpunk | kuro5hin diary ]
[ Parent ]

Better ideas have been proposed and are on the way (3.50 / 2) (#12)
by Carnage4Life on Sat Apr 28, 2001 at 08:24:10 PM EST

It's rusty's site so let him figure out on his own how to fix it since we certainly won't have better ideas than rusty!

Since rusty already proposed an idea superior to his 3 weeks ago and there have been better suggestions in articles long killed I see no reason why we should waste time discussing what is a band aid hack as opposed to a proper redesign of the system.

PS: Your immature comments and rating style as are humorous as ever, I still can't believe you are a grown man with a wife. Keep it up, you're hilarious.


[ Parent ]
because it /is/ a band-aid (3.50 / 2) (#20)
by _Quinn on Sat Apr 28, 2001 at 11:50:59 PM EST

   How long will it take Rusty to finish the closed-skull brain-surgery he's performing on scoop to implement his idea? I'm not oppossed to bandaids in the interim.

-_Quinn
Reality Maintenance Group, Silver City Construction Co., Ltd.
[ Parent ]
Of course. (1.00 / 2) (#27)
by regeya on Sun Apr 29, 2001 at 02:32:52 AM EST

I'll respond with immature comments, so long as people keep posting immature comments, I'll respond in kind.

[ yokelpunk | kuro5hin diary ]
[ Parent ]

Oh no! Not again! (3.75 / 4) (#2)
by Anonymous 6522 on Sat Apr 28, 2001 at 07:38:10 PM EST

JCB hides under a rock.

This should be implemented imediately, IMHO. It might finally keep the queue at a reasonable level without having to rewrite the queue code every couple of months, to account for the greater number of stories/users.

time limits (3.50 / 4) (#5)
by Delirium on Sat Apr 28, 2001 at 08:01:03 PM EST

While rusty implemented some code a while back to force a decision after a certain number of votes when the story was still lingering somewhere between "post" and "dump" range, I think a decision should be forced after a certain time period as well. A story shouldn't be able to sit in the queue and then be posted 8 days later (there's a few in the queue right now from 7 days ago that are near posting levels). On stories like that, usually the discussion has started, progressed, and ended before it even gets posted; the discussion should be taking place while the story is posted and available either on the front page or a section (otherwise we might as well just make the entire site a big queue and ditch the rest of the stuff). I'd say after 4 or 5 days a decision should be forced (with some sort of algorithm taking into account vote totals and comment activity) and the story either posted or dumped.

Quick Fix Alternative until an Overhaul (3.00 / 3) (#6)
by MrAcheson on Sat Apr 28, 2001 at 08:04:44 PM EST

We know when the story was posted to the queue. Apply the final judgement used to determine where the story goes either when a given number of votes are counted or when a given time in the queue is reached (say a week). IANA Scoop Coder but I would imagine this would be a simple patch that could be used until the submission queue gets the complete overhaul that it needs.


These opinions do not represent those of the US Army, DoD, or US Government.


Pretty cool idea... (3.75 / 4) (#8)
by Giant Space Hamster on Sat Apr 28, 2001 at 08:08:28 PM EST

But would you need to take into account what the time is?

For example, I'd guess that there are more votes during 9-5 EST than any other time. Or that more people vote during the week than during the weekend. How would a large holiday (Christmas, for example) affect the queue?

-------------------------------------------
The whole problem with the world is that fools and fanatics are always so certain of themselves, but wiser people so full of doubts.
-- Bertrand Russell

Time Zones (2.50 / 2) (#25)
by Mabb on Sun Apr 29, 2001 at 12:30:33 AM EST

Not forgetting time zones...

Those on the other side of the world come in to K5 6-12 hours after y'all in the US and UK.

I've seen the effect of this very recently, where a story was posted and dumped while I slept...



QuiltBlog: WIP, SEX, WOW, MQ, LQS, HST...

[ Parent ]
Two words: (3.80 / 5) (#11)
by Global-Lightning on Sat Apr 28, 2001 at 08:23:37 PM EST

Time Limits

The oldest submission in the queue was posted a week ago, has a score in the 40's, and the last comment made was four days ago. It's dead; give it a decent burial.

No need to add any extra data, just an extra variable t and one line of code that executes:
At time t after submission, if votes < threshold, delete story from queue.

Sometimes you have to sacrifice a little fairness for the sake of efficiency

rusty's comment on this 3 weeks ago. (4.14 / 14) (#13)
by Carnage4Life on Sat Apr 28, 2001 at 08:26:27 PM EST

http://www.kuro5hin.org/?op=comments&sid=2001/4/6/234524/8153&cid=23#23

I think you're right (5.00 / 6) (#23)
by rusty on Sat Apr 7th, 2001 at 07:44:55 PM EST
(rusty@kuro5hin.org) http://www.kuro5hin.org/


I've been pondering this all night, and, while I love the concept of stories giving birth to other stories, I can't see this working in practice. All your criticisms are right on, and I came up with a couple more as well. Basically, it's too complicated, too confusing, and unlikely to actually fix anything.

But. I'd like you input on another idea I'm chewing on. I think even you might actually like this one. ;-)

An author submits a story. It goes into the queue, just like now. However, instead of cumulative voting, readers instead rate the story, on the same scale as comments are done now -- 1-5. The only difference is everyone has a '0' rating as well, reserved for spam, just like comments.

While the story is in the queue, the author may edit it at any time. People can post editorial comments suggesting changes, people can also post topical comments, if they want. And (this is important), people may change their rating of the story if and when an author addresses any concerns about it that they have.

There would be a default number of ratings required to promote the story out of the queue. Say, 100 people must rate the story, and it must have a score higher than one to be posted. The queue now is mainly for editing and revision, and to act as a spam trap. Also, authors may optionally choose to "hold" a story longer than the default time, in order to do more work on it. Once out of the queue, they can no longer edit. Also, the "hide below one" rule would apply to stories as well. If an author wrote a decent story, waited till it had 99 ratings, and then made it all goatse.cx links and posted it, people can immediately change their rating to '0'. The average drops below one, it goes away. This may happen at any time during the life of a story.

Once a story has passed editing, it is posted. Readers may continue to rate each story, and may still change their rating at any time. The display of stories on the section pages and front page would be threshold based -- that is, for example, the default sections might show everything rated 3+, and the front page might show all stories from any section rated 4+, or 4.5+.

Registered users would be allowed to set their own preferred thresholds, replacing the defaults. Want to see evrything? Just set your threshold to 1. Only the best? Set all thresholds to 4.

I think this could be a winner, for a few reasons.
It preserves the "author-centrism" that you correctly identify as an important part of the site.
It should dramatically lower the amount of time a story spends in the queue, and make "stuck in the queue" a thing of the past.
It makes story rating and comment rating similar enough to not confuse people. If you understand one, you understand the other.
It actually makes author-editing feasable, and finally provides some utility for all those editorial comments.
Less story droppage, more choice.
Basically, I think our current voting system suffers from a lot of thge problems that Slashdot's comment moderation suffers from, because of it's additive nature. Comment rating works pretty well, why not give it a shot with stories too?



I like this idea a lot (none / 0) (#31)
by johnny on Sun Apr 29, 2001 at 07:20:30 AM EST

I have a story in the queue (self-publishing. . .) that I would like to revise (for reasons summarized in my own editorial comment: basically I wrote an annoying post about what most k5ers consider to be a worthy topic).

I would welcome the chance to rework the post based on feedback. I learned a lot from the reaction to my post, including from people who called me a whiney asshole. I don't think I'm a whiney asshole, but I see how my post reads that way. So I would like to change it now, before its rating gets from 115 to 129 and a million more people get a chance to call me a whiney asshole. (WA's have feelings too! :-)

There, now I'm feeling more like a member of the community.

jrs

yr frn,
jrs
Get your free download of prizewinning novels Acts of the Apostles and Cheap Complex Devices.
[ Parent ]

The Web (none / 0) (#46)
by Mitheral on Tue May 01, 2001 at 08:43:24 PM EST

You know this is the web; you could just link to the comment.

[ Parent ]
Couple of questions (3.00 / 2) (#18)
by Tachys on Sat Apr 28, 2001 at 09:21:30 PM EST

Why is the voting result hidden unless I vote myself?

Is there a "time limit" on how long a story has to reach a threshold?

Time limit (4.00 / 1) (#19)
by delmoi on Sat Apr 28, 2001 at 09:23:06 PM EST

Right now, there is no time limit, (there is a story here that has been in the queue for 7 days). I don't know why you can't see the voting score before you vote, but presumably it's to keep people voting honestly
--
"'argumentation' is not a word, idiot." -- thelizman
[ Parent ]
hidden vote (4.00 / 2) (#24)
by radar bunny on Sun Apr 29, 2001 at 12:30:26 AM EST

the reason is to keep people from voting a story up or down just because it's goign that way. This way you vote on the story solely based on weather youfeel it should be voted on, not just how everyone else has already voted. So, the question someone reverses itself to be "why do you need to see the current score before you vote?"

[ Parent ]
True Neutral (5.00 / 2) (#39)
by Mad Hughagi on Sun Apr 29, 2001 at 11:06:39 PM EST

In following the allegiance to my alignment, I must try to balance the good and evil in the universe. Knowing the score would help once in a while

;)


HUGHAGI INDUSTRIES

We don't make the products you like, we make you like the products we make.
[ Parent ]

You aren't the only one (4.00 / 2) (#21)
by Elendale on Sat Apr 28, 2001 at 11:54:35 PM EST

A few other K5ers have been working on a few proposed fixes for this. Check them out here. I'm planning on submitting a meta story on this soon, but you beat me! Grar!

-Elendale
---

When free speech is outlawed, only criminals will complain.


I have a simpler solution (2.50 / 2) (#23)
by John Milton on Sun Apr 29, 2001 at 12:00:39 AM EST

Change it so that the threshold takes into account the amount of stories in the queue. Also, automatically drop stories that have been in for longer than a certain period of time.


"When we consider that woman are treated as property, it is degrading to women that we should Treat our children as property to be disposed of as we see fit." -Elizabeth Cady Stanton


Only counting active users? (4.25 / 8) (#26)
by YesNoCancel on Sun Apr 29, 2001 at 02:06:02 AM EST

Your idea to calculate the post/hide thresholds based on the number of users that regularly vote is brilliant, IMHO. The more users kuro5hin has, the higher the percentage of inactive users.

I'd say every user that hasn't voted on a story for three days should be considered "inactive" and not included in the calculation of the post/drop thresholds. Once he votes on a story, he would be active again and the thresholds would be recalculated.

The way the thresholds are calculated now (based on the total number of accounts) doesn't make much sense. Consider a real-life election in a town with 10,000 eligible voters. There's an election, and only 1,000 of them actually vote: 600 for candidate A, 200 for candidate B and 200 for candidate C. Now candidate A has clearly won (60%), but if you base the percentage on the number of _total eligible voters_, he only gets 6% and thus can never actually win. Doesn't make much sense, does it?

I like this idea... (none / 0) (#32)
by Sairon on Sun Apr 29, 2001 at 08:51:47 AM EST

probably because it is much similiar to my own. I wanted to see it based on "active" users, yet had no idea how to determine who was "active." That seems like a terrrific metric to me.

JPM

[ Parent ]

If the queue isn't going to be fixed soon... (3.00 / 1) (#28)
by Anonymous 6522 on Sun Apr 29, 2001 at 04:16:26 AM EST

Rusty might as well put a "(x comments, n new)" next to the stories in the queue. It would make tracking any discussion that's going on much less annnoying.

Heck, they should be there anyway. Even if the delay until posting gets shorter, there will still be a delay of at least a few hours. A lot of discussion can go on in a few hours.

Good idea (5.00 / 2) (#41)
by enterfornone on Mon Apr 30, 2001 at 12:30:14 AM EST

Actually, a better idea would be to have a queue page that is formatted just like the front page (with the intro showing etc) for people who just read K5 from the queue.

--
efn 26/m/syd
Will sponsor new accounts for porn.
[ Parent ]
Even more radical idea (5.00 / 2) (#43)
by delmoi on Mon Apr 30, 2001 at 01:36:06 AM EST

Post everything, but let people vote untill a threshold is reached. That way, a bad story would be removed from section, whereas a medicore story won't sit 'in limbo' for forever.
--
"'argumentation' is not a word, idiot." -- thelizman
[ Parent ]
I know this music... (5.00 / 2) (#34)
by acestus on Sun Apr 29, 2001 at 09:48:38 AM EST

This is nearly the exact same idea, including the popcorn analogy, that cyphergoth and I described in our article here.

Acestus
This is not an exit.
Good article, recommended reading. (none / 0) (#38)
by kmself on Sun Apr 29, 2001 at 01:43:16 PM EST

Thanks for posting that link. It's strongly recommended.

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

Occasional users (3.80 / 5) (#35)
by xdroop on Sun Apr 29, 2001 at 11:18:06 AM EST

As an occasional participant (I read daily as an annonymous visitor, but only moderate/comment occasionally), I've watched the thresholds climb as the number of registered users climb, and I think that the problem is that you have no way to account for those users who sign up, participate briefly, then vanish back into the ether from which they came (or are abandoned when they grant 'trusted user' status to the holder's 'real' account :)

There probably needs to be a way to determine on a day-by-day basis how many users can be considered 'active' (moderating, rating, commenting, submitting, etc) and make the thresholds a function of that number, rather than the total number. Of course, the definition of 'active' could be left to the community -- or ultimately, rusty.

Of course, since I'm only an occasional visitor, this idea has probably been floated before.
---
xhost +

Totally broken (4.28 / 7) (#36)
by Sunir on Sun Apr 29, 2001 at 12:32:13 PM EST

This idea is totally broken. The failed assumption is that voting will procede regularly until interest wanes. But this is not true. Voting happens by people in the real world, and they live on real world schedules.

We already know empirically that weekend voters are more likely to front page material than workday voters. We also know that net usage decreases around 8-9am UTC as that time neatly corresponds to slack time around the world. We also know network usage is completely erratic around the Superbowl, and nearly non-existant around Christmas to New Year's.

Simply put, the voting rates are going to fluctuate anyway, and not along a simple curve, and there is no way of measuring the baseline in any simple or consistent way.

Moreover, it is a bit of magic that people won't understand. "Wait, why did that story get posted all of a sudden?" Magic algorithms suck. See OpenProcess for the story of Badvogato.

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

Baseline (4.00 / 3) (#37)
by delmoi on Sun Apr 29, 2001 at 12:50:46 PM EST

Simply put, the voting rates are going to fluctuate anyway, and not along a simple curve, and there is no way of measuring the baseline in any simple or consistent way.

Actualy, that isn't true. There are a number of ways to mesure the overall baseline. The first would be to mesure the total number of votes on the whole system, but you would run into a problem if most of the stories in the queue were 'old'. Another method of mesuring the baseline would be simply to mesure pageviews to the main site, or pageviews to the queue. Simple experimentation could be used to make sure that the numbers are linked (look at the delta of pageview frequency and the vote frequency on new stories)
--
"'argumentation' is not a word, idiot." -- thelizman
[ Parent ]
Non-correlated measurements and the Commons (4.00 / 2) (#40)
by Sunir on Sun Apr 29, 2001 at 11:57:01 PM EST

No, your measurements are still wrong because they are not directly correlated to the level of interest in the story. The same phenomenon that posts more front page stories on the weekend will skew this false correlation you made. The proportion of people who are casual readers vs. people who will actually vote will change non-uniformly given the demographics of the site at the current time.

Even measuring against different stories doesn't work. Some stories take a long time to accrue the voting threshold, and others take a shorter time. For instance, any story by Rusty about changes to the site software will make it to the front page very quickly. However, from the point of view of kuro5hin, both stories have equal merit once they are posted. You cannot preserve this change if you tie all voting together. In fact, the destruction of independence of polls is a serious reason not to do this.

Some may argue that you only want the relatively good stories to post, but I doubt that's the goal of the site proprietors. As I understand it, they want truly good stories to post, not just the best of the lot. If this means the site moves slowly relative to the rate of submissions, that's fine.

The alternative would be to allow spam or low quality postings to weight the system down in a serious way. Also, since it's easier to post crap, it will also be downward force on the quality of stories.

Since there is no good way to measure baseline interest, there is nothing to compare against except for magic numbers. Those magic numbers are likely to be very wrong on a routine basis.

The real solution is to find some way of expanding the size of the Commons (the submission queue page). The best recent solution was the addition of diaries. Diaries significantly reduced pressure on the queue by allowing authors to write material of limited interest somewhere without burdening the rest of the site. The other major advance in increasing the size of the Commons was the addition of sections to split the front page (the other major Commons).

The floated suggestion of disallowing topical comments while in the queue has no direct effect on this solution, but it has an indirect result of encouraging the quick promotion of material in the queue. This increases the space in the queue along the temporal axis (more stories in the same amount of space over the same period of time). For the short term, this will work famously. And it's also a way of encouraging people like myself to move out of the queue onto the rest of kuro5hin.

In the long term, a much more radical solution is needed. The floated suggestion of having an editorial queue in addition to the submission queue will reduce the size of the Commons. Consequently, since the editorial queue will be happening shortly, a serious rethinking of the submission process is in order. Perhaps splitting the queue into one queue per section? But even that is limited.

Actually, I think the One Face to Kuro5hin goal is the limiting pressure. If you continue to have the kind of traffic growth as experienced here over the last year, kuro5hin will no longer be able to maintain a single Commons. It will need to break into smaller Commons with at best a larger Commons where people can congregate. This would make kuro5hin a hierarchy of individual boards, and it would disrupt the common community feeling here. The other option is to thwart the pressure to grow. (Hey, growth is bad.)

The other option is to think more about social dynamics and organizational structure to come up with a more scalable solution.

"Haven't thought of anything, have you? No, neither have I. Think, think, think." -- Winnie the Pooh from The Honey Tree by A.A. Milne

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

Non-correlation to suggested measures (none / 0) (#44)
by kmself on Mon Apr 30, 2001 at 01:48:34 PM EST

The solution Rusty and I seem to be aiming at is not to introduce two queues (editing and rating), but to replace the current moderation queue with an editing queue, in which gross errors in an article can be addressed, and an assessment of spam / not spam made. Everything else floats out, possibly to diary if the initial moderation rating is low enough, otherwise to a section page, where wholesale moderation takes place.

Periodically, new material from sections, which is highly moderated, is promoted to the front page. How often will depend on the section -- news and MLP might rotate or promote several times daily. OpEd, Columns, or Meta might remain resident for a week or so. Other material rotating every several days. Readers will go to specific sections for more detailed and fluid content. It might also be possible for material to be promoted from section to diary if it is considered lower quality.

Note that the editing queue should be easier to get through than the current moderation queue -- it's going to be a faster cycling, with fewer individual inputs requred. Once a submission is through the editing queue, it is posted, for all intents and purposes, and what may change is its relative placement on the section page, possible placement on the front page, or possible demotion to diary.

Scheduling issues -- when a story is to be posted, how long it remains in queue -- currently handled by virtue of the broken voting system, would be addressed explicitly. An author could either accept default time slots, or could request an earliest possible, orlatest possible, or both, dates, allowing posts to be "windowed".

Edit and yank capabilities are also required.

I believe these ideas address your concerns (which are valid, just not reflected in the likely solutions) of reducing the efficacy of the story moderation system.

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

Time Limit (3.00 / 2) (#42)
by logiceight on Mon Apr 30, 2001 at 12:55:36 AM EST

I think a time limit should be used instead of a threshold. At end of time limit it is posted or dumped.

What should the time limit be? 24 hours.

This would make sure each time zone gets a chance to rate the article.

One more idea (4.00 / 4) (#45)
by spacejack on Mon Apr 30, 2001 at 11:29:27 PM EST

How about this (apologies if this has been suggested, I haven't seen it). What if posting a topical comment automatically voted +1 section for any story you haven't voted on. i.e., if you're going to post a reply but haven't bothered to vote, then we must assume you want to participate in the discussion. Could speed things up.

Popcorn queue | 46 comments (40 topical, 6 editorial, 0 hidden)
Display: Sort:

kuro5hin.org

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

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