|
|
Log in / Subscribe / Register

On moderation

By Jonathan Corbet
February 18, 2015
Back in the early days, there were no user-posted comments on LWN; that feature was added, after a great deal of thought, in 2002. This was not universally regarded as a good move, but, over time, comments have become an integral part of the LWN experience. Most of the time, the comment area is interesting, respectful, and polite. But there are always exceptions.

Those exceptions lead some people to ask for more "moderation" of the comment stream.

Those of us who work on LWN have put a great deal of ourselves into it for many years. LWN is, in many ways, a labor of love. When somebody posts a comment that is offensive, abusive, spammy, or just a troll, it feels like they have painted obscene graffiti onto our houses. But we remain highly reluctant to "moderate" (or "delete," or "censor") those comments for a number of reasons. At the top of the list is the simple fact that we believe that "freedom" means more than the ability to share software with each other. To silence a voice with editorial force feels like an attack on freedom; it is not something to be done lightly.

Let there be no mistake: we do not support those who post abusive, bigoted, or demeaning comments, and we feel that there is no place for them on LWN. Our desire and goal is to run a site where that kind of comment is simply not present. But chances are that we will continue to use the heavy hand of "moderation" toward that goal rather less frequently than some would like.

Trying to control the direction of a comment stream is a slippery slope in a number of ways. In a heavily controlled environment, it is hard to tell whether the opinions on offer reflect the community, or whether they simply reflect what the moderator thinks is OK to be said. We want LWN to reflect our community, not our view of what that community should be.

What are the objective criteria by which moderation would be performed? If we try to rid the stream of anything that might be offensive to anybody, we will have a bland discussion indeed. And why stop at limiting offense? Consider that the recent call for more moderation quickly led to a request to cut off a discussion that, while admittedly long and repetitive, was focused on an area of great concern to many developers. The desire to see moderation used in such a situation shows how slippery this slope can be.

Also relevant are the facts that moderation is simply expensive and a source of legal liability. Putting together an edition every week is a serious grind; we have nobody with spare time on their hands to patrol the comments. The only way to make that kind of time available would be to reduce the number of articles we write for the site. That would be a heavy cost, and it still doesn't address the fact that unpleasant personalities feel free to post outside of LWN's business hours.

It is not surprising that a number of publications have recently taken the step of eliminating comments altogether. That is not a step we would like to take, but we will if that appears to be necessary.

But such an extreme step should not be necessary. We as a community are better than that.

Experience shows that simple requests to tone down or close off a discussion are often successful. We may well be making more such requests in the future. But the success of LWN's comment areas depends mostly on the commenters; we do not have the time, energy, or desire to play "comment nanny" all the time. If we could all make an effort to recognize when a discussion has run its course, avoid repeating ourselves, and refrain from feeding trolls, things would work better for all of us.

At the same time, we can respond to overtly offensive comments by stating, politely, that they are not welcome in our community. That, in your editor's opinion, is more likely to lead to a more respectful community in the long run than trying to moderate our way into a world where such people just do not seem to exist.

As a last resort, we may implement the ability to close down specific threads if the need arises — but we will endeavor to almost never use it. We already have the ability to block comments from specific accounts; we almost never use that either. The use of such tools is indicative of a failure at a higher level; it is a sad day when they must be put into action.

This is not the position that some people would like us to take, but it is what we feel that we are able, morally and economically, to do. And please bear in mind that comments here have gotten somewhat better over the years, even if they are still not where we would like them to be. Those with strong stomachs can look at some previous discussions for evidence of that. "Things used to be worse" is no justification for an unacceptable state of affairs now, but it does show that progress can be made. We are a community of adults, and we are certainly capable of behaving that way.


to post comments

On moderation

Posted Feb 19, 2015 1:54 UTC (Thu) by JdGordy (subscriber, #70103) [Link] (9 responses)

Is it fair to say most of the bad comments are from drive-by guest accounts?
If so, why not try disabling guest comments for some time after an article is posted? Trolls would be less likely to bother if their post is at the bottom of a long thread already.

On moderation

Posted Feb 19, 2015 4:25 UTC (Thu) by dashesy (guest, #74652) [Link] (1 responses)

no, some valuable comments actually come from guests, landley posts as guest I have seen for example.

On moderation

Posted Feb 23, 2015 17:35 UTC (Mon) by nix (subscriber, #2304) [Link]

Quite. Some of the worst posters on certain flammable topics are subscribers, also :(

On moderation

Posted Feb 19, 2015 4:41 UTC (Thu) by zuki (subscriber, #41808) [Link]

> why not try disabling guest comments for some time after an article is posted?

That might work. It would narrow down the circle of comments a bit, but I don't think this would such a bad thing. If someone really needs to comment, they can always get a starving hacker subscription.

On moderation

Posted Feb 19, 2015 10:15 UTC (Thu) by epa (subscriber, #39769) [Link] (4 responses)

You could enable guest comments but only show them with a 20% probability. Any logged-in user would then be able to click 'this comment is not troll' to promote the comment to full visibility. I suggest this as the simplest possible system that provides some moderation of guest comments without going all Slashdot.

On moderation

Posted Feb 20, 2015 17:46 UTC (Fri) by felixfix (subscriber, #242) [Link] (3 responses)

If that 20% varies by each page load, then what happens to replies?

Consistency is pretty important in a lot of things. I'd hate to think that pages change on the whim of a random number generator.

On moderation

Posted Feb 23, 2015 8:18 UTC (Mon) by epa (subscriber, #39769) [Link] (2 responses)

I imagined that you hash the comment id with the id of the viewing user (or IP address, if not logged in) and then see if the result is divisible by 5. So it's essentially random, but deterministic from one page load to the next.

On moderation

Posted Feb 23, 2015 10:11 UTC (Mon) by vonbrand (guest, #4458) [Link] (1 responses)

I for one read LWN here at home and at work, and sometimes on my phone. You are saying I should see different selections of comments? And left wondering what some random comment is answering to?

On moderation

Posted Feb 23, 2015 11:45 UTC (Mon) by epa (subscriber, #39769) [Link]

Yeah, my proposal illustrates a general problem with moderation systems: an invisible or lower-scored comment could be replied to by a user with higher karma, so that the reply can be seen but not what it is replying to.

I envisaged that the state of a comment being shown with 20% probability would not last long. When any logged-in user views the page then click, click, click - and the comments are promoted to full visibility. If a comment hangs around in ghost state for more than a few minutes then it is probably a troll and an LWN admin might choose to delete it.

As a logged-in user you would always see the same selection of comments on a page.

On moderation

Posted Feb 22, 2015 3:53 UTC (Sun) by Trelane (guest, #56877) [Link]

There is now a filter for guest accounts so that you can view guest comments if they seem relevant by clicking on the +. Hilariously, this elided the entirety of the offensive comments in the the article's link. It seems to be a very nice feature, but it's demand instead for supply side so might not be what you'd like.

On moderation

Posted Feb 19, 2015 4:33 UTC (Thu) by dashesy (guest, #74652) [Link] (9 responses)

Why moderation at all? One can just ignore to read (or view) what they do not like, self-censorship and political correctness has become really annoying. If one does not like an invited guest they can avoid going to their meeting, if one has strong allergy to caricatures, well it is easy do not look. I for one appreciate the direction here in LWN that assumes everyone here is an adult and can deal with some random trolls without tantrum, it is a waste of their valuable time and energy. Last filter (after the one you can apply on users) is the eyes.

On moderation

Posted Feb 19, 2015 12:04 UTC (Thu) by fandom (subscriber, #4028) [Link] (2 responses)

"One can just ignore to read (or view) what they do not like"

The problem is that that's exactly what happens, many forums end up being nothing but trolls shouting at each other because anyone with any sense has given up.

On moderation

Posted Feb 23, 2015 17:36 UTC (Mon) by nix (subscriber, #2304) [Link] (1 responses)

This is basically what happened to USENET, alas.

On moderation

Posted Feb 23, 2015 20:49 UTC (Mon) by pizza (subscriber, #46) [Link]

...except USENET clients were well structured and had good killfile support, unlike $random_web_forum. :)

On moderation

Posted Feb 19, 2015 14:48 UTC (Thu) by ovitters (guest, #27950) [Link] (5 responses)

If someone is behaving in a meeting I rather kick them out of the meeting.

More practical: I give trainings every so often. It's my show, if you're just e.g. there to annoy (never happens) or to play with your phone (happens.. urgh) then you can leave.

Calling everything politicall correctness is a cop out.

On moderation

Posted Feb 19, 2015 15:14 UTC (Thu) by rriggs (guest, #11598) [Link] (2 responses)

Bad analogy. This isn't "your" meeting. This is a community forum. It is more akin to have a conversation in a public square.

On moderation

Posted Feb 19, 2015 17:30 UTC (Thu) by raven667 (subscriber, #5198) [Link] (1 responses)

This isn't a public square, it is a private house owned by our gracious Editors who have ultimate authority on how conversation is conducted here. They are free to enforce whatever standards of civility they like, you are free to accept this or take your comments elsewhere.

So far if they have erred they have erred on the side of allowing anything-goes and people are free to look at foolish as they actually are with only gentle suggestions that people stop if conversations get too far afield, which have been surprisingly effective. The downside from a social-justice perspective is that by providing a safe-space for people to say stupid, hurtful things this is most likely to cut against any kind of disadvantaged group, who are often the target of the stupid and hurtful things people say and have a heavier burden to carry in this area.

On moderation

Posted Feb 26, 2015 12:45 UTC (Thu) by Wol (subscriber, #4433) [Link]

That was PJ's standard. If it wasn't acceptable in her living room, it wasn't acceptable on GL.

Caused the odd problem, for example, when a discussion turned to race issues my perfectly acceptable English was to her unacceptable American, but hey ho - it happens. "The English and Americans are two people divided by a common language" :-)

Cheers,
Wol

On moderation

Posted Feb 19, 2015 15:27 UTC (Thu) by dashesy (guest, #74652) [Link] (1 responses)

If someone is behaving, first time I would talk to them in private, second time I talk to HR to deal with them. I am not a primary school teacher to ask everyone to behave, nor is it respectful (to other attendees) to assume so. If an obvious troll lurks in a forum, it might make sense to just delete them, but going after their posts and deleting them is just wasteful.

On moderation

Posted Feb 24, 2015 21:04 UTC (Tue) by dlang (guest, #313) [Link]

> If someone is behaving, first time...

why is it a problem if people are bahaving? Isn't the problem when they are MISbehaving? :-)

On moderation

Posted Feb 19, 2015 5:02 UTC (Thu) by karath (subscriber, #19025) [Link] (3 responses)

The commitment of the editorial team to freedom with all its benefits and the occasional wart is what keeps me coming back. And the comments are valuable. Among other things, it often adds a great deal when a prominent member of a project reviewed in an article takes the opportunity to add their response. Or when an uninvolved but interested expert takes the time to provide further value.

For my part, I am often disturbed enough by some comment or thread that I start to draft a reply. More often than not, the catharsis of just drafting a reply is sufficient, there is no need to feed the trolls by actually posting.

On a related note, I am saddened (but understanding) that the editorial team appear to have been browbeaten into not posting on what is happening in the systemd project as there seems to be a lot of great technical interest going on there. Regardless of whether one feels that systemd is right for one's (or even anyone's) computers, this should not be a reason for commenters to shutdown any article or discussion of what systemd does or how it does it.

Systemd coverage

Posted Feb 19, 2015 15:10 UTC (Thu) by corbet (editor, #1) [Link] (1 responses)

As far as I know, we have not actually suppressed or avoided any systemd coverage out of fear of the comment stream. Though we may have been tempted once or twice... We can't cover everything in depth all the time, and nothing from the systemd camp has risen to the top of the list recently. But there's nothing more serious than that going on there.

Systemd coverage

Posted Feb 20, 2015 2:15 UTC (Fri) by karath (subscriber, #19025) [Link]

I'm happy to hear this. Thank-you

On moderation

Posted Feb 19, 2015 21:35 UTC (Thu) by jhhaller (guest, #56103) [Link]

Not to steal Jonathan's thunder, but I believe LWN would like to see well written articles about technology within it's purview. If you think there are significant technology developments happening in systemd, please write an article to enlighten the rest of us.

On moderation

Posted Feb 19, 2015 6:14 UTC (Thu) by dune73 (guest, #17225) [Link]

Thank you for your reasonable approach to comments.

It would be sad to lose valuable comments and it would be sad if the bad comments would affect the experience of the audience at lwn.net.

It's a thin line between Scylla and Charybdis. But if anybody can steer us along this line, it's Jonathan and the editorial team of lwn.net.

Flag system

Posted Feb 19, 2015 8:42 UTC (Thu) by skissane (subscriber, #38675) [Link] (4 responses)

How about a system for "flagging" comments. A link to flag a comment as inappropriate. A confirmation popup that has a little blurb about when comments should be flagged, to discourage people from doing it unnecessarily. If a comment gets more than a certain number of flags (from distinct users), it could be automatically hidden (but with a button to expand it to visible again so people can see what it was.) You can keep track of who is flagging who, and if you find a user who keeps on flagging inappropriately, you can disable their ability to flag comments (or just tell the system to silently ignore their flags in deciding which comments to autohide). Maybe you could also limit the ability to flag comments to subscribers (or even higher level subscribers) to minimise the abuse potential; alternatively, you could make the autohide algorithm weight flags based on subscription level.

Flag system

Posted Feb 19, 2015 11:02 UTC (Thu) by SimonO (guest, #56318) [Link] (3 responses)

I just thought of a system based on nonviolent communication (http://www.cnvc.org); if you allow readers to indicate their subjective response (degrees of offended/annoyed or enrichedby/enjoyedthis/etc.) It could be made configurable how much hurting you can take ;-)

So if you want conflict free browsing, select to see only comments which have been enjoyed by a few readers and where nobody was offended or annoyed.

Or you could filter out comments which a few or a lot of people found annoying/offensive, this way you can see all the unrated comments, but not those which were offensive to some.

I'm sure this would take some tweaking to get right, but it doesn't censor anything, unless you want to read censored comments.

When you select to be censored, you won't see comments until enough people have rated them.

/Simon

Flag system

Posted Feb 19, 2015 15:22 UTC (Thu) by ortalo (guest, #4654) [Link] (1 responses)

Thanks for raising the idea.

Flag system

Posted Feb 19, 2015 15:25 UTC (Thu) by ortalo (guest, #4654) [Link]

Alternatively, though that's not so aesthetically satisfying IMHO, should I say:
<enrichedby comment="Thank you."/>

PS: Hopefully, rate-limiting control is not triggering on me?

Flag system

Posted Feb 19, 2015 15:58 UTC (Thu) by kvaml (guest, #61841) [Link]

BaseballProspectus employs simple + - voting flags. Subscribers are limited to a single increment/decrement effect and guests can't vote. Baseball fans (fanatics) can be every bit as boorish as software developers. A threshold setting filters comments that fall below the threshold. The site provides an option in the comment steam where the filtered comments would have appeared to view the filtered comments.

In my experience, this has worked very nicely. The commentators "police" the comments, but the degree of censorship is under user control. I use a slightly negative threshold, but that's a matter of personal taste.

On moderation

Posted Feb 19, 2015 9:14 UTC (Thu) by tdz (subscriber, #58733) [Link] (6 responses)

Did you ever consider limiting the number of comments per commenter per article?

In the few cases where I commented, I rarely felt a need to write more than two or three comments. Just posting an opinion and maybe replying to someone was always enough.

On moderation

Posted Feb 19, 2015 9:33 UTC (Thu) by dgm (subscriber, #49227) [Link] (4 responses)

This is a very interesting and apparently simple idea to explore, but it would not work on all cases.

Take for example articles about software projects. It happens with some frequency that one or more of its developers are LWN readers/subscribers, and they usually enrich the article sensibly with their comments.

On moderation

Posted Feb 19, 2015 11:46 UTC (Thu) by njd27 (subscriber, #5770) [Link] (2 responses)

Rate limiting might work.

On moderation

Posted Feb 19, 2015 15:19 UTC (Thu) by ortalo (guest, #4654) [Link]

I agree on the rate limiting idea too. I just had the same idea and think event simple number-by-article-by-day limitation would be okay for 90% of cases; at least for me.
In the minority of cases where fast numerous comments maybe desirable by the commenters, maybe there could be some additional account-based rate limiting before escalation to peer-based validation and even finally editors moderation for the ultimate heated, but polite and fruitful debate between Ms.A and Mr.B about e.g. ;-) including X11 in the kernel.

Looks neat and nearly functional no?

On moderation

Posted Feb 19, 2015 17:43 UTC (Thu) by raven667 (subscriber, #5198) [Link]

I don't think rate limiting really solves anything worth the additional complexity and effort. You will thwart heated discussions regardless of whether they are offensive trolling or just a heathy exchange of ideas, so throwing out good and bad alike, while racists, misogynists and fanatics can effectively derail a conversation with a few posts, or even just one. Committed trolls might just start using sock-puppet accounts if the difficulty of using a persistent identity is made too high, which limits the ability of the existing filtering mechanisms to function.

I think the approach being taken right now, where filtering is under the control of each individual subscriber on what they want to see, is a good direction to take. If anything is to be done I would suggest expanding the user-controlled system by having additional hints about bad actors/comments that one can choose to filter on and by making the filtering more thorough, by collapsing the entire thread history after a filtered user even on the /Comments/unread view when you don't have the full history visible.

On moderation

Posted Feb 20, 2015 17:53 UTC (Fri) by felixfix (subscriber, #242) [Link]

Combine that with +- rating buttons -- if your current rating sum for a thread (or day, or week, or ...) is negative, no more posting.

One problem with all these systems is the varieties possible. Do you have simple +- overall buttons, do you have buttons for different categories (troll, informational content, insight, ...), do you have meta rating to discover the saboteurs, how do you handle robots, .....

No matter how simple or complicated you make it, any rating system invites in self-proclaimed experts who know how to do it better. Having no rating system eliminates most of that crowd.

On moderation

Posted Feb 19, 2015 15:59 UTC (Thu) by rgmoore (✭ supporter ✭, #75) [Link]

Limiting the number of comments per commenter per article just encourages people who want to be disruptive to use sock puppets, while limiting what constructive posters can do. If there were actually an easy way to control trolls, the internet would be a very different place.

On moderation

Posted Feb 19, 2015 11:10 UTC (Thu) by sdalley (subscriber, #18550) [Link]

I have to say, the editors deserve a prize for being so broad-shouldered. Putting up with the din, only now dying down, in threads like http://lwn.net/Articles/631319/ is quite something.

Looking at that thread, I'm inspired to pontificate that that the contentiousness of the multiplicity of some contributors' protestations is only matched by the indefatigability of their self-justifications.

Ah, well...

On moderation

Posted Feb 19, 2015 13:40 UTC (Thu) by vonbrand (guest, #4458) [Link]

Perhaps a middle-ground to outright filtering would be to have some articles with no comments, or which just allow comments by subscribers.

Way too much theory

Posted Feb 19, 2015 15:06 UTC (Thu) by ovitters (guest, #27950) [Link] (3 responses)

LWN seems to avoid certain topics due to the type of comments that follow. It's really difficult to be a moderator. Where to draw the line is not easy. You might change the line. Replying that certain comments are not ok is already part of moderation. That said, in various cases people do you way overboard. Not doing anything because a possible eventual slippery slope is too easy.

On LWN from time to time it seems way too acceptable to behave badly. As a result, people behave badly. In the article Jonathan states that a simple reply to behave more nicely is enough. I've done that various times, it sometimes works but it is not always enough. Then it is the responsibility of LWN to intervene.

I don't believe everyone here behaves like an adult or the "We as a community are better than that". Most do, but not all. Further, on this site it is way too easy for one person to steer the direction of the comments in one way. Just make a big rant as the first or second comment.

Then the strange notion of freedom. When as LWN you really think someone did the equivalent of "painted obscene graffiti onto our houses", then in normal life you want the person caught by police, fined and pay to remove it. But you don't say that it is someones freedom and leave the graffiti on your house. The minimal thing you do is clean it up!

I do appreciate that LWN will step in more often in future. However, the article in full makes me go: "ehh???".

There are options which doe not require LWN to read every comment. And IMO I'd prefer if the decision is solely taken by LWN and by default that LWN doesn't really participate. This to avoid group think.

Way too much theory

Posted Feb 20, 2015 5:41 UTC (Fri) by elvis_ (guest, #63935) [Link] (1 responses)

You do realise you are one of the main offenders in pointless long threads? And everyone doing the same probably thinks it's not them, it's the other bloke?
How about less time thinking up censorship systems and more time thinking "do I really need to say this" before you post?

Way too much theory

Posted Feb 24, 2015 8:15 UTC (Tue) by jond (subscriber, #37669) [Link]

Pointless long threads are not what moderation is for. Moderation is to address the (often just one post long) outright hate posts which are very upsetting to those targeted (by gender for example). I don't think ovitters is ever guilty of that type of post.

Way too much theory

Posted Feb 20, 2015 13:40 UTC (Fri) by henrikb (guest, #58898) [Link]

"Further, on this site it is way too easy for one person to steer the direction of the comments in one way. Just make a big rant as the first or second comment."

I think this hits the nail on the head. If there was some way to sort the comments by relevance it would be a lot easier to ignore the trolls. Some +/- buttons for relevance and sorting maybe?

On moderation

Posted Feb 19, 2015 15:38 UTC (Thu) by njd27 (subscriber, #5770) [Link]

I take it that going back to "Letters to the Editor" is not an option?

On moderation

Posted Feb 19, 2015 19:11 UTC (Thu) by tom.prince (guest, #70680) [Link]

One compromise in moderation, that has worked at various places is to remove the vowels[1] from offensive comments. That way, the comment is still readable, and not censored, but makes it easy to skip over.

[1] https://en.wikipedia.org/wiki/Disemvoweling

On moderation

Posted Feb 19, 2015 20:23 UTC (Thu) by MrWim (subscriber, #47432) [Link]

I'd like a way of thanking people for making positive contributions to a discussion without having to write a reply myself which wouldn't add anything to the discussion itself.

You could have a "thank you" button which increments a counter that only the author of the comment can see. It could be a nice gentle way of encouraging polite behaviour while avoiding the gamesmanship that is present when the "scores" are public.

As far as undesirable comments go I think people are already responding to unpleasantness by clicking "Reply" and that mostly works.

With the assumption that the majority of people aren't malicious, one thing that might be interesting is finding out roughly how many people have added you to their black-list. That would give a little indication that you're not meeting the social standards that lwn.net expects and might make commenters re-evaluate their tone.

This might be better than having a way to hide individual comments as any feedback that way could be interpreted as a "disagree" or "controversial opinion" indication rather than "you're being unpleasant to the extent that people don't even want to read your comments anymore".

On moderation

Posted Feb 20, 2015 15:05 UTC (Fri) by cdmiller (guest, #2813) [Link] (6 responses)

I've proposed before, and still think it's a good idea, the ability to collapse any thread from a default open state would greatly enhance navigation of the comments. Sometimes, having to scroll down through a subscriber thread which is of no interest is a pain. Default collapsing of guest threads is a great feature, I can take a quick look and collapse it again if I see no value in reading further.

On moderation

Posted Feb 20, 2015 15:45 UTC (Fri) by rghetta (subscriber, #39444) [Link]

+1, and collapsing any subthread too.

Collapsing subthreads

Posted Feb 20, 2015 16:53 UTC (Fri) by corbet (editor, #1) [Link] (4 responses)

This seems like a reasonable feature, and probably not too hard to implement. The only downside is that it increases the reliance on javascript, which is something we've tended to resist, but, as an optional feature, it's not so bad.

Collapsing subthreads

Posted Feb 20, 2015 17:22 UTC (Fri) by kpfleming (subscriber, #23250) [Link] (1 responses)

Even though I understand the reluctance to consider Javascript-based solutions, nearly every proposal I've read in this thread (including reader-generated flagging, allowing the community to police itself) has already been implemented in Discourse.

Collapsing subthreads

Posted Feb 20, 2015 17:49 UTC (Fri) by gerv (guest, #3376) [Link]

I appreciate the comment policy here (well described by one commenter as "freedom with the occasional wart") and am glad the editors are resisting calls for change. I'd say the current policy is "the worst one except for all the others".

Gerv

Collapsing subthreads

Posted Feb 26, 2015 6:48 UTC (Thu) by Mook (subscriber, #71173) [Link]

If a checkbox can be massaged to be a sibling of .CommentBox, you can use CSS selectors to hide all following sibling .Comment elements; that would mean no JS would need to be used (but would be harder to use for text-mode and visually impaired browsers due to the extra focus-receiving element). For now, I just manually scroll past the deep threads to nowhere.

As a guest, I appreciate the current ability to post comments; if at some point you decide that limiting guest comments is better for the site, ah well, I started out reading without an account. In the mean time, though: thank you for maintaining this useful resource, whether or not I can comment.

Collapsing subthreads

Posted Aug 9, 2015 0:25 UTC (Sun) by sethml (guest, #8471) [Link]

You can avoid JavaScript and get most of the benefits of thread collapsing by simply providing per-comment links for next-in-thread, previous-in-thread, and parent. They'd just be links to anchors.

On moderation

Posted Feb 20, 2015 23:56 UTC (Fri) by flussence (guest, #85566) [Link] (6 responses)

I recognise two names of note in that last old thread link:

- "mjg59", known for the "fart fart fart" meme, infamous for his self-labelled affiliation with online harassment groups, and anyone paying close attention may remember some of his recent commentary here (in particular, telling one user to F off over mild criticism).
- "chromatic", who single-handedly managed to create a chilling effect around the topic of Perl 6 on Reddit through years of targeted FUD and trolling (of the same sort grsecurity is known for around these parts). The consensus in the Perl community is language-related discussion on that site is now to be avoided, solely because of him, at a loss to everyone else.

And so I find it quite surreal to see the same identities attached to level-headed, agreeable statements speaking out against *exactly* the sort of thing they've become, eight years later. How did things go so wrong?

On moderation

Posted Feb 21, 2015 13:44 UTC (Sat) by spender (guest, #23067) [Link] (4 responses)

FUD and trolling? Should I not be permitted to speak at all about security on this site because a handful of users (dlang, nix, vonbrand, etc) for some reason take personal offense to the claim that Linux's security isn't all it's cracked up to be?

I don't see you mention the problem of these people completely making things up, repeatedly accusing others of claiming things they never said, commenting as authorities on subjects they lack any experience or knowledge in -- are those the kinds of people that have a place on LWN? Because that's the real problem with this site, and it wastes everyone's time.

Ah right, here was your grand contribution to the "fresh kernels" thread: http://lwn.net/Articles/631707/
You're of the camp that places pleasantries above learning and sharing of actual knowledge. That you would place us in a killfile (the only people with actual knowledge on the subject providing a wealth of evidence to people too stubborn to accept it) and not these other arm-chair experts says to me you come to LWN just to have your opinion validated and don't want your fragile sensibilities hurt when it's not.

BTW, the idea that we're digging our project into a hole, a project responsible for the NX, SMEP, and SMAP in your processor and the PIE and ASLR in your userland, purely due to offending you -- is laughable. Last I checked, I didn't care whether you used grsecurity or not. I'm not fretting at all here over here about how many additional freeloading users who don't care about security aren't using our software. I appreciate the Darwinian nature of your decision.

-Brad

On moderation

Posted Feb 21, 2015 14:45 UTC (Sat) by vonbrand (guest, #4458) [Link]

I know for a fact that Linux (the kernel; but also complete distributions) security isn't up to NSA standards. Nobody claims it is (or aspires to it). That is not part of the design objectives of Linux, and would seriously impair it. The developers' attitude is in line with this, that is all I'm saying.

On moderation

Posted Feb 23, 2015 17:45 UTC (Mon) by nix (subscriber, #2304) [Link] (2 responses)

For the record, I do not claim Linux's security is not all it's cracked up to be. I claim that you have no evidence that it is a conspiracy, that your conspiracy theorizing is baseless lunacy, and that a far more plausible reason for everyone on l-k ignoring you is the appalling way you and PaXTeam present your arguments, massively pedantically condescending and virulently offensive conspiratorial madness by turns.

Hint: if people with actual diagnosed deficits in their social modelling can tell your methods of social interaction suck, they probably *do* suck. (And having methods of social interaction that suck worse than mine is an accomplishment indeed.)

If you actually cared about Linux's security, rather than claiming credit, you'd try to fix this, since it would lead to the security of the upstream kernel improving, and thus similar improvements in those trees that depend upon it (more or less all of them). But your past antics (posting hashes of descriptions of exploits rather than the descriptions and the like) make it quite clear that getting the credit is far more important to you than the security of the systems people actually use. (Sure, if everyone used only your kernels, they might get improved security -- but they'd have to deal with *you*. That's a *major* downside, though it seems you don't realise how major.)

On moderation

Posted Feb 23, 2015 19:00 UTC (Mon) by PaXTeam (guest, #24616) [Link] (1 responses)

> I claim that you have no evidence that it is a conspiracy,

and it's a well-known strawman of yours, noone claimed that (except for you and vonbrand but then you're arguing with yourselves, and i'm sure you have the intelligence to save us from it in the future ;). but if you have evidence to the contrary, post it.

as for the rest of your petty ad hominem... surely you can do better than that and instead present valuable arguments as to why there's security@kernel.org when all bugs are supposed to be treated the same way?

On moderation

Posted Feb 24, 2015 8:18 UTC (Tue) by jond (subscriber, #37669) [Link]

none of this security-related discussion (really, old vendetta point scoring) is relevant to the topic at hand.

On moderation

Posted Feb 24, 2015 8:21 UTC (Tue) by jond (subscriber, #37669) [Link]

Even if we take it as read that your characterisation of those people 8 years ago is true (and I don't): why should behaviour 8 years ago influence a position on hate speech today? You are free to consider such people hypocrites if you wish (or previously unenlightened if you are being more charitable) but it doesn't change the objective reality of the argument today.

On moderation

Posted Feb 22, 2015 8:23 UTC (Sun) by thoeme (subscriber, #2871) [Link]

Hm, I think the current mechanism where you can blank out comments written by certain serial offenders is sufficient. I only had to use it twice in the recent years, and it worked well for me. After all, the signal-to-noise-ratio is still high here on the LWN comment sections.

And no, guest comments should be allowed: As other have stated, valuable input has come from guest commenters, which would be a real loss.

Being able to collapse sub-threads would be a nice thing to have: To have to scroll down and down to see where a possibly interesting sub topic has been opened, doesn't play well with my laziness ;-)

On moderation

Posted Feb 26, 2015 10:29 UTC (Thu) by jdub (subscriber, #27) [Link]

LWN team, I love and respect your work. Have for years.

It's really this simple:

LWN is your house. If you don't want abuse happening in your house, under your roof, under your watch, you need to tell the abusers to be quiet or to go away.

There's no technical solution to this social problem. Just leadership.

Moderation is critically flawed.

Posted Feb 26, 2015 18:01 UTC (Thu) by ksandstr (guest, #60862) [Link]

Looking at the "request for moderation" in the first link of this article, and the "request for closing a thread" in the second, there's a common theme that has gone unmentioned so far: that of calling for moderation, and further instructing the moderator(s) on how to go about the task. In my opinion this is an undesirable attempt to permit administrative control of discussion such that the result matches certain activist organizations' PR goals.

Not to make too fine a point of this, these requests call for a policy where censorship is performed by a moderator, and that a particular ilk of advocate would have their hand firmly up said moderator's skirts. Then it would be (say) Mr. Corbet protecting the protectable and punishing the irreverent, and not his handlers. All in good faith of course, not that we plebs could tell -- evidence of mis-moderation having been erased by the act.

At the same time it's significant that despite previous attempts to move LWN into the "fluffy hugbox" sphere (as seen in the article's third link) by an implied threat of becoming a cess-pit of rumoured Internet horrors otherwise, it's neither been coerced to be the former nor degenerated into the latter. It follows that the current argument, "... and it'll drive away the minorities; and your site will be doomed!" is likely to appear wholly bunk in seven years' time like the "you'll drive away half the user-base (demographic)!" is today, judging from the foundational similarity between the two.

As such, it surprises and disgusts me that such requests aren't being dismissed summarily, as the intransparency point alone would give sufficient reason to.

[As for anyone itching to point out that their view of censorship is different from mine, usually one that permits great numbers of non-governmental people to engage in righteous de-facto censorship: that viewpoint is wrong in theory because of the outsourcing argument, and wrong in practice because the effect is exactly the same regardless of whether it's Big Brother blacking out your letters, or Little Sister giving your wife a headache for your being such a horrible non-team-player.]

Reddit-style up/down votes?

Posted Feb 26, 2015 20:24 UTC (Thu) by jrw (guest, #69959) [Link]

Perhaps a reddit-style up/down vote (for subscribers only), a filter threshold (view all or view only above a certain threshold), and the ability to reveal any below-threshold post would be sufficient to let the community handle its own moderation.

On moderation

Posted Mar 1, 2015 11:32 UTC (Sun) by tux3 (subscriber, #101245) [Link] (1 responses)

I just registered and subscribed because I wholeheartedly agree with this article. Perhaps this is noise and if so I apologize but I'd like to take the opportunity to thank LWN.
It seems to me that the comment section is working very well, as far as Internet Comments go, and I don't currently see the need for radical change.
Some of the solutions brought forward so far are interesting, but I am not yet convinced that 'bad commenters' can be so easily turned into 'quality posters' or kept away without imposing significant limitations on others.

I would think that encouraging quality would be a better approach than heavy handed moderation to maintaining a healthy and interesting discussion space.

On moderation

Posted Mar 1, 2015 17:23 UTC (Sun) by vonbrand (guest, #4458) [Link]

As a very longtime LWN reader, I can attest that our esteemed editors have done an outstanding job at encouraging quality discussion, and have kept the comments remarkably civil with very little intervention, mostly by gentle remainders. Not much space for improvement I can see.


Copyright © 2015, Eklektix, Inc.
This article may be redistributed under the terms of the Creative Commons CC BY-SA 4.0 license
Comments and public postings are copyrighted by their creators.
Linux is a registered trademark of Linus Torvalds