vroom's Journal: Comments & Moderation Improvements Under Way

While reading through your responses to the reader survey from a couple months ago a couple things were clear. You both love & hate comments. You love the insightfulness of our readers, you hate trolls, and there are a number of things currently getting in the way from you being able to navigate through discussions as easily as you might like to.

Load All Comments

The number one thing that kept coming up in comments was an annoyance at not being able to see all the comments right away when you loaded a page. We'd previously made a design decision to show 50 comments to make for a fast page load, while allowing you the user to load more comments. (It should be noted if you login you can change your default view to 250 comments). The major nuissance many of you noted was that if you wanted to load all comments before reading a discussion you'd have to scroll down to the bottom of the discussion & mash on the button multiple times to load the entire conversation & then scroll up to the top to begin reading the discussion. One change we added a week ago or so was a 'Load All Comments' button at the top of the discussion to get around this annoyance.

Number of Initial Comments Shown

Another thing many of you noted was not liking only seeing 50 comments for your initial view. We bumped that number to 80 for an incremental change, and could be further increasing it as we monitor changes to page speed, and other comment dynamics.

Fix for Mod Point Allocation

About a week ago we corrected a bug with mod point allocation which brought a lot of new people into the moderation pool. The number of people moderating, and number of moderations done has gone up significantly as a result of this.

Comment Preview Speed

As part of our comment posting (and submission) processes we have some security checks that take a lot of time. Unfortunately for users many of you spent more time than you needed to waiting for the comment preview to come back while this check was occurring. We recently made a change to do this processing in the background when you trigger a reply form, as a result you should spend less time waiting to preview or submit a comment. Instead of making you wait to preview a comment we're doing processing in the background while you're typing up your comment. We made similar changes to the submit process. There may be further speed improvements later but this is a start, and should be fairly noticeable to frequent commenters.

Comment Threshold Inconsistencies

Another problem that was fixed recently was that comments above your threshold, that were children of lower rated comments didn't always appear. As a result the numbers on the slider for 'Full', 'Abbreviated', and 'Hidden' didn't necessarily reflect what was in the discussion. As of today that should be corrected, surfacing all the comments that should be surfaced by your selected threshold.

More to Come

We're by no means done with improvements to the comment system. Comments are in many ways one of the things that sets Slashdot apart so we want to continue working to make discussions easier to navigate, make interesting comments easier to find, and surfacing the insights within our community that set our community apart from many other places on the web.

Troll

[Jane Q. Public]

While it is true that people dislike trolls, another fact is that the "troll" mod is grossly overused by people who simply disagree. I am not sure what is to be done about that, but some kind of change is called for.

Re:

[Anonymous Coward]

Yeah, I got modded "troll" just for posting an incendiary and off-topic comment. I mean, come on!

I know. It should have been modded "flamebait"

Re:

[Jane Q. Public]

"Yeah, I got modded "troll" just for posting an incendiary and off-topic comment. I mean, come on!"

I am not referring to just myself, although I have had my share of inappropriate "troll" mods. I see it all the time. If it weren't a real problem, why do so many people have the "Troll is not a substitute for 'I disagee'" quote in their sig tags?

Re:

[poity]

I made the suggestion that negative moderation should require the mod to both highlight the problem text and to have his/her username shown publicly.

Re:

[Jane Q. Public]

I don't know if I would modify much about how the the main page works, but I would like to see a stats area for people who are registered, showing what comments were modded, what the mod was, and by whom. That way some of the most egregious real trolls could be caught at it.

Re:

[Qzukk]

So it's your fault we had those stories where people were given the opportunity to annotate other people's text.

It [degraded into full-on [bullshit]** after a [few minutes]***.]*

*: Poster was saying "derp derp I'm a retard"
**: oooooo you said a bad word
***: banana banana banana banana

Re:

[Captain Splendid]

You can easily change your comment preferences. Troll mods make comments highly visible for me, and I wouldn't want it any other way.

Re:

[mcgrew]

I think they should go back to the old metamoderation, where you moderated a moderation as "fair" or "unfair", and too many "unfairs" took away your moderation priveliges.

"Troll" isn't the only misused mod, either. Yesterday's /. had an article about viruses, and folks were calling all malware "viruses". I explained (courteously, I might add) the differences between viruses, trojans, and worms. It was modded both "informative" and -- get this -- offtopic.

Re:

[Jane Q. Public]

"I think they should go back to the old metamoderation, where you moderated a moderation as "fair" or "unfair", and too many "unfairs" took away your moderation priveliges."

Yeah, I liked that system. At least it had some way of flagging the worst of the abuse.

""Troll" isn't the only misused mod, either. Yesterday's /. had an article about viruses, and folks were calling all malware "viruses". I explained (courteously, I might add) the differences between viruses, trojans, and worms. It was modded both "informative" and -- get this -- offtopic."

Yeah, it happens. "Overrated" gets abused too sometimes. And I have gotten modded "off-topic" for answering direct questions before. :o)

Other terms that often get misused or conflated with other things:

It is impossible to know anymore what someone means when they say "global warming", because you have no idea whether they are referring to the general warming trend, or anthropogenic global warming, which is a differe

Just when I got out... they pull me back in...

[snowgirl]

(Screw you, I haven't actually watched the Godfather movies, I just know this line from popular memetic propagation.)

Anyways, horray! Just as everyone gets relatively used to the current comment system, they're going to tweak it some more, and people are going to bitch about how the NEW new way sucks, and that it's driving more people from slashdot than ever before...

Trolls

[PCM2]

I don't dislike trolls.

I, and a lot of other people like me (based on comments I've seen in the past) browse at -1 because we enjoy both the insightful and informative comments and the trolls. /. has a unique trolling culture (it even once had its own Wikipedia entry [wikipedia.org]) and some of us enjoy reading about tech here rather than somewhere else for that reason, among others.

Re:

[igreaterthanu]

For the most part, I agree, however there really needs to be a -1 Goatse mod to distinguish clever trolls from pathetic trolls.

Re:

[Qzukk]

What gets me is that people keep clicking on that boredgeek site. Does everyone have the [domain.com] tag turned off, or do they just have that poor of a memory?

Re:

[PCM2]

Not at all. Men vs. boys, and all that.

We want trolls.

[sidragon.net]

They are the characteristic that defines Slashdot [slashdot.org]!

Go away, troll.

[mcgrew]

Bullshit, you're trolling right now. There are trolls in every single messageboard on the internet. Hell, my former brother in law used to spend time on geezer forums blasting Social Security and Medicare just to piss people off -- and that's all trolls are good for.

From your link: More moderator points are being used to mod posts down than up.

False.

Furthermore, when modding a post up, every moderator seems to follow previous moderators in their choices, even when it's not a particularly interesting or clev

What about metamods?

[ElectricTurtle]

Are there any plans to reform metamoderation so that it isn't double jeopardy for posts? The old metamoderation system was much better since it rated the moderations, not the posts.

The other changes are good, I suppose. I don't really care as I use D1, browse at zero, with max posts loading by default, and don't use preview.

Re:

[Qzukk]

Yes! Restore metamod and I'll go back to bothering with it. The current instructions of

Clicking the + and - indicates that you think that a comment is good or bad.

is makes it just mod, only 100% more pointless. I went there to copy and paste that line and this post [slashdot.org] is 4th in my queue, with a score of 1 in the firehose, no moderation history. Go back to asking us to agree or disagree with the mods themselves, so we know exactly what we're being asked to do (and preferrably getting comments that actually

Re:

[vroom]

That is one of the things further down my list to revisit and reconsider. Your complaints are definitely not alone and it was something that was mentioned on the survey several times.

I'm not 100% sure what our decision will be but it's a problem I was considering before delving into reader comments much, and it's certainly something that's been voiced by many community members too. There's certainly improvements which could be made, and one of those on the table is going back to the old way of doing thi

Items I should have noted in my survey response

[Khopesh]

This is mostly a set of feature requests that is probably better suited as enhancement bugs for slashcode.

Viewing thresholds by number of comments shown

I would like to have a dynamic threshold contingent on how popular the article is. Ideally, I want 15-20 full comments with roughly 30 abbreviated comments. This means that an article with thousands of posts would get a full/abbrev threshold of 5/4 while a low-traffic article like this journal entry would be considerably lower (as of this writing, I