I’ve been interested in commenting systems, comment spam in blogs, and whether it’s a good idea to have comments on this site. I once asked on Facebook:
What are the advantages and disadvantages of the various commenting systems used on websites? Some qualities one can pick out: whether to have nested comments, the size of the unit of language on which one can comment (e.g. comments for a whole article versus comments for each paragraph, or even smaller units), voting (and whether to display who voted), displaying who has seen the comments, which markup language to use, and so on.
It’s rather interesting how many different commenting systems one can find, and is natural to ask which one works best. And if that is an ill-defined question, then one can at least try to ask what one should use on one’s own website.
It might be worthwhile to list all the possible parameters (i.e. to extend the list I came up with in the Facebook post).
- Nesting/threading: Web Discussions: Flat by Design, Mark New Comments. There are several choices here. On sites like Reddit, Quora, and Hacker News, discussions both have a tree structure and are presented as a tree. On 4chan, discussions have a DAG structure but are presented as a flat list that is chronologically sorted. Facebook used to have comments that had a flat structure and were presented as such. In general the underlying data structure has to be more “complicated” than the format in which the discussion is presented.
- Unit of language relevant for replies
- talk about Comment Press
- Medium also has paragraph-level comments
Location of downvoted content. John_Maxwell_IV writes:
On reddit, if your submission is downvoted, it’s downvoted in to obscurity. On Less Wrong, downvoted posts remain on the Discussion page, creating a sort of public humiliation for people who are downvoted.
This is also true of a site like Quora, where heavily downvoted answers are hidden. (And in fact, only highly upvoted content is widely circulated through people’s feeds.)
- User accounts
- Markup language
- Permissions (e.g. ability to edit others’ comments)
- Version control (though I’ve never seen this before)
- Quotation (as on bulletin boards); see Web Discussions: Flat by Design again.
- Independence (i.e., each comment to a post could become a post itself, to which others could comment)
- Length (some sites restrict the length of comments; this is most extreme on Twitter)
Some questions to ask:
- Can a greedy selection (in the sense that, for each bullet point, one chooses the most preferable option) of the features optimize for the “best” commenting system?
- What patterns might we find if we classify many of the existing commenting systems (e.g. on Facebook, Quora, Wikipedia discussions, reddit, WordPress, Disqus, Discourse, etc.)?
Comments for personal sites
On social media like Facebook and Quora, and even on WordPress blogs, it is not necessary to think too much on commenting, since this is all provided upstream. When I first started using a static site generator to make my website, I followed gwern and used Disqus. However, currently this site has no comments. My reasoning is that, ever since I’ve become active on Facebook, I’ve realized that most of the useful discussions about an article happen there, instead of on the original article’s comments section. More directly, in his answer to “Why are people more willing to comment on a Facebook post sharing an article rather than in the comments section of the article itself?”, Andrew Ho writes:
To answer the question directly, it is because the comments section of the article is full of inane and worthless chatter that has zero intellectual value whatsoever. However, my Facebook friends are mostly fairly intelligent people, so usually they have something insightful to say about the article and/or participating in the discussion that they prompt is enjoyable and interesting.
This obviously does not happen on every site. In fact, I get the impression that small personal sites tend to have the highest-quality comments, while large news sites tend to have the worst, with blogs somewhere in the middle.
The other issue is with having to moderate the comments. Quora and Facebook are convenient because they require real names, and thus are able to keep out non-human spam (for the most part). Disqus does seem to have anti-spam measures, but verifying the identities of commenters may be more difficult since people can choose to post anonymously (and also because Disqus isn’t as widely-used as Facebook).
It’s probably worth thinking about why we have comments in the first place. Comments are useful chiefly because they are a means to provide feedback for the author. But this can be done via email or messaging (see my contact information for how); better yet, for minor things like pointing out typos, I have an “edit page” link at the top of each page that will allow people with GitHub accounts to fork the site repo and submit a pull request.
Another idea: having comments could mean more flame. When people on a site see that others are complaining about something, they might jump on that and add more flame1. (This, of course, relates to the problem of the burden of moderation.) Even seemingly innocuous content on an otherwise unknown site could become the target of a torrent of verbal attacks, and dealing with this seems highly stressful. Forcing people to email in their attacks could have a dampening effect. Of course, this doesn’t prevent public attacks happening on Twitter, Tumblr, and elsewhere.
One idea in favor of having comments is so criticisms are out in the open. But there are other ways to do this2, and in any case, how can one trust a site owner to be honest about the comments posted on their own site (since they usually have the power to delete any comment, or even edit them in the case of sites using WordPress)?
Another idea is that there is a natural presumption that commenting on a site is enabled. This site itself is the most natural place to have a discussion about the content on the site, and also provides an obvious place for readers to interact with each other; who am I to take this zone of exchange away?
Some convincing ideas for me are:
- Site-native comments are potentially easier to backup than comments posted to Facebook or Quora. Even Disqus does not allow the site owner to backup the comments for a site (only their own comments, last time I checked); gwern seems to backup Disqus comments though3.
- Some people don’t have an account on Facebook or Quora, and even emailing may be inconvenient. For this, I do have an anonymous feedback form (which works even with text-only browsers like Elinks), as advertised on my contacts section of the about page.
Having a comments section might be a good idea for an outsider (say, if they want to recruit bloggers). Vipul Naik writes in his answer to “How do you recruit writers for a group blog?” that he searched on Bryan Caplan’s EconLog when looking for bloggers to recruit to Open Borders:
I looked at the comment thread on Bryan’s migration posts on EconLog. I looked carefully for people who wrote long, detailed comments on migration issues, whose philosophy seemed broadly similar to what was being conveyed on the site, who were eloquent, and who didn’t seem to already have a very active blog presence of their own.
I also do have a GitHub repository for this site, where anyone can open an issue, which is viewable publicly.
A variety of people have discussed whether to have comments on their site. Here are some of the better ones.
- Matt Gemmell has:
Darren Steadman (through Gemmell)
MG Siegler (also through Gemmell):
Let’s be totally honest here: anyone worthwhile leaving a comment should do so on their own blog. Very few read blog comments anyway. I’m sorry, but it’s true. Commenting is a facade. It makes you think you have a voice. You don’t. Get your own blog and write how you really feel on your own site.
This is related to my idea of treating comments as their own post.
However, with this, it’s important to consider the counterfactual: by removing a place to rapidly provide thoughts, readers may never end up providing their thoughts. In other words, I suspect that for many, it’s comments or nothing; making one’s own site or blog isn’t an option (too much activation energy required).
If you’re going to put the work in to articulate your thoughts, to make an intelligent argument, and to bring something fresh to the conversation … you should be putting that work into your site, not ours.
The notorious John Gruber
No Comment by James Hague:
I think that initial knee-jerk “I’ve been looking at this for ten seconds and now let me explain the critical flaws” reaction is a common one among people with engineering mindsets. And that’s not a good thing. I’ve seen this repeatedly, from people putting down programming languages for silly, superficial reasons (Perl’s sigils, Python’s enforced indentation), to ridiculous off-the-cuff put downs of new products (such as the predictions of doom in the Slashdot announcement of the original iPod in 2001).
For years, comment boxes have been a staple of the online experience. You’ll find them everywhere, from The New York Times to Fox News to The Economist. But as online audiences have grown, the pain of moderating conversations on the web has grown, too. And in many cases, the most vibrant coversations about a particular article or topic are happening on sites like Facebook and Twitter. So many media companies are giving up on comments, at least for now. So far this year, Bloomberg, The Verge, The Daily Beast and now Motherboard have all dropped their comments feature.
The article links to “Why we’re killing our comments section”, which has:
[W]e suspect that many publishers will soon find that their existing commenting systems do not serve their readers as the conversation continues to move off websites to social media, where most of our content is discovered and consumed.
Jekyll can’t provide comments (obviously) and I am not interested in going back to Disqus (for various reasons). I also had the impression that comments were not doing it for me anymore. The ratio spam / useful comment was about 1000 to 1. Sure, Akismet took care of this and Disqus could, too. In addition, I’d get comments from other places (twitter, g+ or plain email) and since
I’m not a cool indieweb devit’s never that many, I manually added them to posts.
In other words, I started to feel like comments are just not that useful anymore (caveat lector: see below) and that having a special technology for it seems overkill.
At times I’ve been tempted to just turn off comments entirely on my blog, and just flat out avoid participating in comment threads on the web, but this feels like throwing the baby out with the bathwater. Despite the signal to noise ratio, there are conversations and exchanges that make talking about things on the web worth it to me.
I prefer to incorporate feedback into the original piece and acknowledge people for their contributions, similar to what’s done during peer review of an academic paper. […]
Finally, a lot of discussion of my pieces happens on Facebook, Reddit, or other places, which makes on-blog comments less crucial.
I don’t agree with the reasoning that having comments makes a site look less academic.
As of 2018, I still feel like commenting on websites is an unsolved problem. I want something that has low friction, revision history (or append-only), support for archival, no spam, and static. WordPress with Akismet for spam might be good, but composing posts on WordPress is unpleasant.
I wouldn’t compare this site to Slate Star Codex, but just look at any post there to see the problem of unending arguments.↩
gwern has an “External links” section for his page On Stress that links to a reddit discussion of the page. I can do the same with Facebook discussions or when I crosspost to Quora.↩
In response to “What about [archiving] the comments here?”, gwern writes:
Likely to die, I’m afraid. I do back up my Disqus comments on a regular basis, but there’s no guarantee that I’ll migrate them to the next commenting system. (On the plus side, I try to address any issues comments bring up, quote ones which add to the article, etc, so the losses won’t be too great.)