1. 25

Hey folks!

So, for a while now we’ve had the science tag, and it’s brought some interesting things our way:

At the same time, it’s had quite a few less than stellar submission recently:

Now, I love science as much as the next lobster, but there are already some rumblings about getting rid of it.

For what it’s worth, my observations about the sort of articles we like on lobsters are:

  • They are useful as references for best practices or techniques (primarily engineering or practical material, reversing, debugging, security)
  • They provide deeper insight into fundamentals of the engineering in our field (compsci articles, math articles)
  • They are very simply delightful and edifying (the dispensation for the art tag)
  • They cover how to build something (hardware) or new hardware releases
  • They help elaborate on one of the above by showing better visualizations (visualization)
  • Grudgingly, they talk about new software releases (software)
  • They talk about relevant legal and employment practices for developers (law, finance, privacy)

science article submissions tend to:

  • frequently be press releases or news
  • lack useful technical detail. haskell articles aren’t my jam, but the submissions are usually useful as a reference.
  • rely on pretty pictures and handwaving
  • focus on topics that are almost entirely non-practical–the overwhelming majority of current users will never build a rocket or leave the planet (unfortunately)
  • be covered well by other sources (Reddit, HN, Wired, Ars Technica)

All of that said, I think we should open the floor and talk about this. Some guiding questions:

  • What do we like about the science tag, if anything?
  • What did @jcs envision for the tag originally? Has it strayed?
  • Would we want to retire the science tag, or replace it with something a bit tighter?



  2. 23

    As a scientist, I disagree with this on general principle. I think a technically focused community shouldn’t be afraid to have discussions about science as well as more focused technical topics.

    That said, maybe lobsters really needs a “clickbait” flag option? We should be filtering out articles that are just press releases or are low on information.

    1. 11

      I don’t think the problem is “science”, but that it usually means pop-science.

      1. [Comment removed by author]

        1. 5

          I’m not sure “pop”-anything belongs on Lobsters.

      2. 8

        I agree that some way of signalling “Hey, this article is clickbait” would be nice–I currently use the spam tag for this purpose, and less often the off-topic flag. I currently also post in every thread I notice that I think fits this definition–and I’m sure that everyone is as sick of reading those posts as I am of making them. :(

        To elaborate on my position with science:

        The first big problem is that we have is that the tag lends itself to popsci submissions, which in turn are churned out by the hundreds by all kinds of businesses and institutions who use that to get marketing views (this is literally Popular Science magazine’s reason to exist). There is big money in saturating the airwaves with popsci nerd bait, and allowing those submissions here tends to invite that sort of saturation.

        The next big problem is that the science tag is just not very specific. Compare:

        • Javascript–is the story about the Javascript programming debacle^Wlanguage?
        • debugging–is the story about figuring out how a system failed?
        • software–is the story about the release of a software package?
        • cogsci–is the story about cognition or modeling cognition?

        Meanwhile, for the science tag, we have:

        • Is the story…um…space? Animals? Rocks?
        • …maybe electronics?
        • …physics, that’s a big one with the kid’s these days?
        • …or wait, academia itself?
        • …maybe health?


        The obvious answer is “Well, let’s add tags for subareas of science, and then people can filter what they want”. The reason the obvious answer is incorrect is that it results in:

        • the creation of a bajillion tags when people can barely manage to properly use the ones we do have.
        • the balkanization of the community into different filter bubble that are completely disjoint (e.g., we’re all basically developers right now, but in the future one could imagine biologists without any interaction with the rest of the community).
        • the continual broadening of what sorts of submissions kinda fit, and hence the regression to the mean of submission quality and topicality.


        That’s why I brought the retirement of the science tag up, not only because it specifically is a source of trouble, but because it exposes some existential questions about the culture of Lobsters and how focuses we want to be as a community.

      3. 8

        What if we just try to focus on getting submissions that are reflective of the original source?

        For instance this posting on your “shit growing in a box” videos does a very good job of explaining why its interesting, while also being approachable.

        1. 14

          I agree with this. It is very, very common that the most popular article about a science news item is the least informative one. I encourage people to trace things back to researchers explaining their work in their own words, rather than journalists talking for them, and submit those stories. When that doesn’t exist, I suggest submitting any published papers on the topic instead.

          I tend to imagine this audience usually has a lot more to discuss about the underlying research than about the popularization.

          1. 8

            That’s a decent mitigator, though I don’t think it fully solves the problem. Tracing back something that hit the pop-science press to the original article still gets us basically the pop-science discussion but with the polite fiction that we really independently dug up the paper, read it, and found it interesting separately from the hype, when mostly people ran across it from the hype and are using it as kind of a stand-in (I see this happen on HN a lot). And often, unfortunately, the original paper is only slightly better, as sensationalizing starts with the researchers. Especially in certain areas like neuroscience there’s a huge overselling / overinterpretation of results, starting right from the paper title and abstract.

            I’d personally prefer people only submit papers in areas they know something about, and found through some means other than popular-science hype leading them back to the paper.

            edit: It’s a slightly different point, but I think I also tend to like lobste.rs best when it’s less correlated with what every other outlet is talking about, with somewhat unique articles that people dug up on their own, rather than stuff that’s currently hot elsewhere that people then cross-posted. And the pop-science stuff rarely fits that category, even if the original paper is dug up.

            1. 1

              That all makes sense, as well.

        2. 7

          It seems like we’re conflating two separate issues: (1) the value of postings that are not related to computer science and related technologies and (2) the posting of clickbait, press releases and the like instead of links to high-quality “original” sources. Given that one of the top-ten articles on the first page right now is essentially a press release about Google’s burrito delivery drones, problem (2) doesn’t seem to be contained to the ‘science’ tag. More generally, many of the original observations about why science articles seem problematic could easily apply to other tags as well.

          I would personally like to see better moderation and screening of clickbait/press release style posts. If we’re going to keep art posts because “they’re simply delightful and edifying”, that should be justification enough to keep the science tag as well.

          1. 6

            Up voting because this, rather than leaving sour grape comments on every article you don’t like, feels like the right way to handle this issue. I personally enjoyed the “shit growing in a box” video, but when you lay out the argument like this I see what you mean and agree with you.

            1. 9

              I am all for removing the science tag, for essentially the seasons listed here. If we don’t remove it, there needs to be some clear rule banning pop-sci. Not because pop-sci is inherently bad (although I think much of it is actually bad), but because pop-sci is non technical and therefore not appropriate for this site. I come to Lobsters for interesting, well put technical discussions in a variety of areas outside of my own. The things posted with the science tag do not prompt these sorts of discussions, and are therefore not useful to me.

              1. 5

                I try not to submit articles with the science tag unless it pairs with another tag - this way, I keep it in scope of the site.

                1. 5

                  I think the science tag attracts pop-science articles because many of us are not scientists, so most aren’t frequently wading through in-depth articles looking for gems to post here, so Lobsters only sees science articles when the topic bubbles up to the clickbait realm. Just like I’m not involved in many fields that we have tags for, I rely on people that are in those fields to find neat things for me.

                  I think the tag is still worth keeping, but I will try to apply a heavier hand in moderating pop-sci articles when I see them (which relies on you folks calling them out, because I’m not here all the time). If it eventually becomes a barren wasteland of year-old articles, I can just mark it inactive.

                  1. 5

                    Yes, and cogsci and finance, for every reason you’ve listed. art seems to tag-team with other, more relevant, tags.

                    @angersock, you seem to have been on a campaign to define what is and isn’t lobste.rs. Had any talks with the mods?

                    1. 9

                      I haven’t been party to any private discussion of redefining lobste.rs, which is as it should be. Open threads that anyone can participate in are the correct place for any such campaign.

                      The corollary is that any criticism of such efforts should also come from all of you, not from above.

                      1. 4

                        I’m glad to see I’m taking the correct approach.

                        For something like this, it’s a community thing–what’s the point in summoning mods (assuming such a thing were possible!) when the community itself would prefer some other course of action?

                        Hearts and claws–er, minds.

                        1. 3

                          I didn’t intend to summon the mods, as if that was a thing. But @Irene answered the question I was trying to ask: if there was private discussion and context missing here.

                          1. 2

                            Yeah, I’m here anyway, of course I was going to comment. :) I wasn’t sure where you were coming from so I defaulted to just providing my own perspective; I’m glad it answered your question.

                        2. 1

                          So happy to hear all that!

                      2. 3

                        An existential crisis? Are we that old yet? :)

                        I’d like to raise a deeper question. Why do we have [hiding by] tags?

                        [Hiding by] Tags allow us to filter out stuff we find generally not relevant/interesting to us. Tags is how we mold lobste.rs to our own sphere of interest. From that point of view, @angersock, why not just filter that tag?

                        Perhaps what you are railing against, @angersock, is the balkanization of our, now, not so little, community? The end result of a proliferation of tags and of filtering by tags, the ability to hide articles, means lobste.rs is just another big city, where you don’t know your neighbors and never say “Hi”, rather than a small village where everyone is in everyones' business.

                        I don’t know, even for me, which is better. Lobste.rs is still low volume enough, that I would suggest removing all tags and leaving the “hide” option.

                        “Hide” is a way for me to leave a conversation without leaving for ever which is less isolationist than hiding a whole community of people which is actually what a tag kind of represents.

                        Yours, in extreme solutions.

                        (Edited to clarify that I meant filtering and hiding by tags indiscriminately, rather than using them as a search tool)

                        1. 6

                          in theory, tags are also an indicator that you know why your post is relevant to lobsters - if there’s no good tag, maybe it doesn’t fit in here. so in that sense the [science] tag is an attractive nuisance; there are off-topic posts getting submitted because they fit the tag.

                          1. 5

                            Not just in theory, this is definitely used in practice.

                          2. 3

                            I do use tags to filter sometimes (I filter two of them), but I also find them useful on the positive side, to find articles I’ve missed in areas I’m especially interested, since I don’t always read lobste.rs often enough to catch everything when it’s fresh. So I use a few tags like art, games, lisp, and historical as something like RSS feeds, to see if anything I might like got by me recently.

                            1. 4

                              I use a few tags like art, games, lisp, and historical as something like RSS feeds

                              I use some tags as literal RSS feeds. I subscribe to eg. https://lobste.rs/t/openbsd.rss in my news reader to make sure I don’t miss anything on particular subjects.

                              1. 2

                                I’ve been using the lisp tag a lot too, though I often just search for “lisp”. We can keep tags, but not filter, for example.

                              2. 2

                                The other benefit of tags is for discovery, especially searching. It’s easy to gloss over that in conversations about filtering, because they are separate concerns…

                              3. 3

                                As the original submitter of “shit in a box” and as a computer scientist, i found that video compelling. I’ve worked on genetic/evolutionary algorithms and was fascinated to see how analogous the real biology is to “simulated” biology. And how different.

                                Too bad not everyone cared for it. Some people upvoted it, so not a complete waste of everyone’s time.

                                I ran across shit-in-a-box randomly surfing. Thought i’d share. Probably should have looked up the original paper.

                                Thinking about it more based on above comments, though: I used to submit more often years ago when i first came from hn. I’ll be sure, though, to be more careful now to only post things that interest everyone.

                                1. 4

                                  You’ve posted some really great stories–I don’t mean to imply that that’s not the case. Please continue submitting things that we have tags for, and the rest will see to itself. :)

                                  1. 2

                                    Yep. It takes a certain amount of “courage” to submit on a small site like Lobsters. There aren’t that many submissions, so it’s possible to see them all. Basically, the submitter is saying: “I think this is cool, and is worth people’s time.”

                                2. 1

                                  Could we make the science tag work the way the pdf tag works, where you also need another tag to submit it? That way we could keep the “computer-y” science and keep filtering / searching for it neatly, but stuff that’s just science can see itself out?