Just as we can filter tags here, is there a way to filter domains? Specifically, medium .com? I really dislike the reading experience there and would like to be able to filter submissions here. Vote with my wallet aka attention as they say.
There is no way to filter domains. If anyone was interested in adding this feature, you would need to add another scope to Story like we do for flags then update the StoryRepository.
Exactly this, lobste.rs is opensource, make a PR and done :D
There’s no point in making a PR if the admins will refuse to merge it afterwards. That happened to me when I tried to improve character escaping in Lobsters’ custom Markdown dialect, years ago. So it’s good to discuss and get approval for significant features before trying to implement them.
I’m generally okay with all such features. It’s a per-user thing and I think people should be free to have as many ways to control their experience as they want and use.
I use Firefox with uBlock Origin (for the ads) and the makemediumreadable.com (for the annoying medium formatting). Firefox also features a Toggle Reader View button (ctrl-alt-R) enabled on websites it detects doing too much text formatting (just like Medium).
I mostly use reader mode. But I specifically ment the submissions here on lobsters. Filter those for me just as I can filter any tag.
I have to agree.
a site like this that has no shame in making user experience awful shouldnt be allowed here.
if you are using ublock, add this to your ‘my filters’ list: lobste.rs##.story:has(a[href*=medium\.com])
There’s a userscript for that.
I’m not sure this is a good idea. I find medium.com repulsive, but once we have this feature what other domains will we filter? There’s already calls for imgur.com.
While domain filtering might start on sites with bad technical practices, what happens when it extends to sites with bad design, or where the author’s other writing is too controversial? I don’t think that’s a road worth going down.
I agree with this entirely. Going down the route of ‘ban everything’ tends to have people go off and congregate into far left or far right communities. I realize with Medium, it’s not the message but the … (ugh a pun)…medium. Yea it’s a garbage platform, but I don’t think creating user filters for websites/domains is going in the right direction.
This is just my completely unscientific opinion, but I do think certain mediums suggest/encourage/cultivate certain types or qualities of content.
This may be confirmation bias, but I perceive a strong correlation between Medium and superficial fluff pieces.
I think the idea is to implement domain hiding, the same way you can currently hide tags you’re disinterested in. I agree that automatic domain filters for the entire site is a very bad idea, but allowing users the ability to more finely curate their feed seems like a good plan to me. I imagine if it’s built the same way you’d be able to get a feed of only a domain if you wanted to.
On Reddit, I use a client that lets me filter subreddits, specific terms and sites. No politics, no BuzzFeed like stuff, no medium and a bunch of subreddits I dislike. Makes my experience wonderful. Using Reddit without my filters is just as awful as using the web without an AdBlocker. But, those filters are personal, just for me. I don’t ask to block it entirely for everyone. Just for my account.
If you’re really just looking to customize your own Lobsters experience, then I think some form of client-side filter is the best way to go.
I understand that others might use such a feature on the site, especially if it were convenient, but like @djsumdog and @endgame (IIUC) I do have some concern about the long term effects on the community.
This bookmarklet removes all sticky elements from the current page. It makes medium and many other otherwise unbearable websites bearable for me:
I recommend using uBlock to do that instead. I’ve already written a filter myself that does the job.
This is genius. Thanks.
I dislike Medium and avoid it myself. But like I once heard a wise old public librarian say:
Use your brain, the filter you were born with.
And when my brain figures out a better way to filter that offloads the work to a sub-system–like making a computer filter domains for me–I do that. I’m not sure your response amounts to much more than, “no, suck it up,” but with an added dose of condescension.
As a side effect, it keeps us from missing the few that are actually good. Some folks put good stuff on there due to large audience (aka higher impact).
I just tried Medium this week and republished three of my articles there. It certainly is not a magic audience source. Maybe it was at one point, but now I would consider it just as hard to find an audience as elsewhere. I’m back to my own site.
Personally (as somebody who uses medium as his default publishing platform & considers other platforms secondary), I don’t find that medium reaches a greater audience by default.
There have been progressive changes to medium policies since I started using it around 2015, each of which have made the distribution impact of any given post worse (starting with various misleading popups for non-logged-in users, followed by the marginalization and limitation of members’ personal feeds in favor of editor-approved content; now, all paywalled content gets submitted to human editors automatically and must be approved by them before it shows up as a recommended story for anybody but your followers!)
Basically, back in 2015 publishing on medium was a pretty good way to get more exposure than publishing on your own website, but today it’s a crapshoot. Users who haven’t registered get a bunch of misleading popups that seem vaguely like paywalls. For the occasional posts that actually are paywalled, lobste.rs by default removes friend links (which are the only reasonable way to share paywalled articles outside the core medium membership). Promotion of editor-approved stuff is good but not great: I’ve had one of my articles featured by Medium (for which they literally had one of their professional editors go over my whole piece with fact checking), and the promotion did reasonably well (getting hundreds of thousands of reads and netting me maybe a hundred bucks over the year or so that it’s been up), but that’s about a third of the number of reads (and about a third of the advance) of my second-most-popular article there, which was solicited and paid for by a publication (How We Get To Next, a tie in for Steven Johnson’s book How We Got To Now). My most popular article was not featured & owes almost all its popularity to having been posted to lobste.rs & HN.
Why do I bother with it?
It’s a reasonably good source of passive income, and one that doesn’t rely on advertising, ad-tech, or tracking. I’m an advocate of transcopyright as a stopgap measure (until we get something like UBI, at which point passive income will not require active monetization of creative work), & Medium’s revenue model is just about as close to transcopyright as we’ve gotten in well-known existing systems (substantially closer than patreon). I share my posts using friend links, and mirror them on my own website, so nobody is locked into medium or medium’s paywall even if they want to read them, but just the clicks alone from folks who happen to have already subscribed to medium’s member program make me between $4 and $20 a month for my backlog of articles, and I fairly frequently make between $40 and $80 if I’ve published a new paywalled longish technical essay that month.
The editing is fairly pleasant to use, most of the time. (Occasionally they’ll do an update and performance will tank, but the post editor is generally substantially smoother than the input box on twitter or google docs, & basic formatting hotkeys familiar from Word and Google Docs work out of the box.) I’m a long-time dedicated vim user for source code & a perpetual critic of web apps in general, but I actually slightly prefer composing prose on Medium, simply because they put an awful lot of effort into making it smooth even on old, slow machines like mine.
If you’re logged in (and especially if you’ve got one of the various readability extensions, which remove the floating bottom bar), Medium’s reading experience isn’t bad. Much like the kindle, it on one hand does a pretty decent job of trying to simulate paper (down to the kind of slightly fancy typesetting nonsense that’s usually done on professionally printed books but rarely on the web or in word processors), but on the other hand happily makes available useful hypertext features (not native to the web per-se) like user-supplied highlights (which can be shared, responded to, accumulated, made visible to other users on the document, or associated with private/personal notes). Being a web app, it’s heavier than a properly-made native version would be, but it’s substantially lighter than sites I use all the time like lobste.rs, laarc, twitter, facebook, every fediverse web client other than brutaldon, almost every news website, and all major webmail systems. (Meanwhile, reading a 15-20 minute article on medium gives me a whole lot more joy than spending the same amount of time on twitter, facebook, or even mastodon usually, so it has less to prove.)
Now, Medium has been on the wrong track for many years. With the exception of the open-paywall concept, almost every change they’ve made since around 2015 has been bad for the writer community there and bad for readers. There’s a good reason that I back up and mirror all my posts from there: I fully expect it to disappear with little warning as soon as the VC money dries up, and I fully expect VC pressure continue to push it into making choices that would boost the finances of a traditional blogging or social media platform but will kill everything worthwhile about Medium (leading to a financial downward spiral that causes that same pressure to increase). In fact, I was playing with the idea of writing a web-based publishing system with support for proper transcopyright under the hood, aimed at the audience of disgruntled ex-Medium users (although I probably won’t because running a company will inevitably suck, but if anybody wants to give it a go they can have my designs & code fragments).
But, for now, Medium doesn’t suck quite enough for me to stop using it as an editor & a side hustle.
Thanks for this, I appreciate the “insider” insight.
No problem! I feel the need to complicate the “medium sucks / is great” narrative whenever it comes up.
This is great insight, I appreciate it as well.
As someone already building a sustainable publishing platform (no VC, no ads), I’d love to know what you think a proper implementation would look like.
It is unfortunate Medium is built on these financial incentives that, I agree, mean they probably disappear one day. But I think we can build on some of their good ideas and make it open source, ActivityPub-powered, and long-term. So your perspective would help a lot.
I’ve seen write.as, though I don’t subscribe.
My main concern is basically the same as Medium’s: how do we incentivize high-quality content (which for me means longform content that people actually read all the way through). Medium’s solution is to distribute subscription fees based on recommendations plus vaguely-defined ‘activity’ (probably a weighted combination of other metrics, where they modify the formula monthly).
Another concern is how to appeal to the kind of writers who write stuff that people want to read. (I think if you appeal to writers, the readers will pretty much take care of themselves.) One way to encourage good writers to write a whole lot is to make the system feel like a genuine intellectual community – to put followed writers front & center in the UI, and encourage them to respond to each other and work off prior work in the community.
I think a transcopyright-like payment system makes sense for addressing these concerns.
Under a transcopyright system, content is bought once per user (possibly for no money at all, or possibly for a small fee, like one cent per kilobyte), after which that user has the ability to view that content in perpetuity or quote it in their own work; in the case of a quote, the quoted material (if not already owned) is fetched from the original owner rather than being embedded in the new context – in other words, this is an automation of the same model for avoiding complex rights negotiation as Rifftrax uses: a remix doesn’t need its components to be individually cleared and their value estimated, because the client fetches its own copies & assembles the bricolage itself. While a user might view an original work in its totality and therefore own all of it, more likely a user will see (and buy) tiny portions of many large works in the context of new juxtapositions.
This means several things:
Writers are incentivized to appeal to readers, so that (if content is fetched as it becomes visible) the reader views the whole thing.
Writers are incentivized to appeal to other writers, so that readers who wouldn’t read their work directly still provide fractional revenue through reuse of their work by someone else.
Writers are incentivized to work together and comment on each others’ work, because the impact of normally-present disincentives (rights difficulties & the potential for fights over too-long quotes or lack of proper attribution) are minimized, making the benefits of collaboration more clear.
There are some issues with moving money between readers and authors in a federated or decentralized system, particularly when rates are this low.
We can use traditional systems like credit cards (or stripe & paypal), but fees are high enough that we’d want to actually hold a subscription fee in escrow (like Medium and Flattr) or do all the transfers in big blocks (like Patreon) – both mean that any single node basically needs to be a corporation for liability reasons, which means that it needs to remain profitable (if only profitable enough to pay for the ~$1500/year registration fee for corporations – but to skim that much off means expanding enough that server fees become high too).
Alternately, we could use a cryptocurrency, but I’m not aware of any cryptocurrencies that have stable enough prices and low enough fees – we’d need to pick one where there were controls against speculation, since (as we saw with Bitcoin) once a cryptocurrency starts to be seriously used, speculators tend to step in and start creating bubbles.
Systems like token-coin don’t use a cryptocurrency per-se but just an internal centralized ledger. If you’ve got trust between nodes, maybe you can keep records of transfers between internal ledgers on a federated network & periodically flush those with ‘real money’ of some kind. From my experience on the fediverse, I don’t particularly trust nodes to make these transfers consistently enough, even when they are acting in good faith.
Transclusion, permission from remote oracles, ledger transfers, local document assembly, etc. can all be implemented on top of activitypub, though in some cases it would probably be awkward.
Basically, although I definitely prefer federated & distributed systems in general, implementing real-money micropayments in a federated system is difficult enough that if I were trying to write a Medium clone with transcopyright I’d make it centralized and have it use an internal ledger (and to the extent that federation might exist, it would be in the form of more concrete agreements with a smaller number of remote nodes running compatible software – contracts between corporations). The alternative is to have a new, non-fungible coin somehow tied into money escrowed into individual nodes.
It might depend on the content, keywords in their discovery algorithms, or even where it’s submitted. I have a feeling there’s a number of people reading these Medium articles who submit them to link aggregators with higher views. They might be amplifying the things they like. That crowd gets a feel for the kind of stuff that makes it. Becomes self-reinforcing. Untested hypothesis except for the general concept which I’ve observed anecdotally with submissions on HN, Lobsters, and Reddit.
Use your brain, the filter you were born with.
Sorry, but that’s a very unconstructive way to dismiss implementing/creating practically every tool known to man.
Have to carry tons of material from point A to point B, and maybe inventing a wheelbarrow would be helpful? Nope, use your hands/feet, the hauling device you were born with.
Have a disease/injury that might be helped by modern medicine? Nope, Use your immune system, the treatment system you were born with.
I wouldn’t consider someone who dismisses improvements that save people time, for example, for the sole reason that ‘you already have the ability, even if it’s not nearly as good’, as being anywhere near ‘wise’.
I think you are misconstruing my comment. I am suggesting that content filtering can be a poor substitute for tolerance and self-control.
What is this even supposed to mean? It sounds condescending
I didn’t intend to condescend. I took her to mean something like:
“The world will always be full of things that might offend you. You are generally better off learning to not be so easily offended than you would be trying to artificially restrict your perception.”
The more I encounter the self-induced misery of hypersensitive people living in perpetually-inadequate filter bubbles, the more I think she had it right back then, when content filtering was a relatively new idea. This seemed like an opportune moment to pass on some sage advice.
But I didn’t mean it as a blanket condemnation; I use ad filters myself. I don’t use tag filters here on Lobsters, but I don’t resent their existence. I generally appreciate the filtering work of moderators here, especially since I can see the mod log if I want to. I wouldn’t use a domain filter if one were implemented, but I don’t see it doing much damage to the community, and I certainly wouldn’t get in the way of anybody implementing it.
I don’t think medium is something that offends people based on its content, though. It’s more of an annoying and terrible user experience that people are trying to avoid. Brain filtering that doesn’t really work the same as ignoring someone saying something you don’t agree with
I use brain filtering on Medium. The popup and such are annoying. I expect it. I brace for it. I close it. I quickly skim the content. If it looks worthwhile, I continue reading. At worse, it makes me think I need to look for some plugin that suppresses all the bad UI while showing the content. Reader mode seems to do that, though, except for popup. Just gotta use it more often.
Whether you object to the content or the UX or whatever, the easiest way to avoid it is to simply not click on those links. They’re already tagged with the domain. If you don’t want to see them again, you can click ‘hide’. The only point of a domain filter is to not see them in the first place.
There was a similar request for an advertising tag some time ago, to identify all the shills and submarine posts and corporate blogs so that people wouldn’t have to decide for themselves what an authors motives are.
That’s pretty funny, when you think about who’s supposed to apply the tag.
As a side-note, hiding all of Medium is a bit tricky as many Medium weblogs use a custom domain. So filtering just “medium.com” might not be as useful as you’d expect.
As a second side-note, dev.to is possibly even worse, with their stupid throbbing “Join now!” thing in the sidebar. Massively distracting. How anyone can think that’s a good idea is truly beyond me.
I’ve blocked it in my /etc/hosts. 5 in 100 medium articles were worth reading. And after their greedy campaign, enough was enough already.
It seems that regardless of actual availability, this is a feature people seem to want. Is this something the mods are willing to implement?
See a mod’s response here.
It’s open source. Literally anyone here can implement it… Why is it the mods’ jobs?
Anyone here can implement it… on their own instance. For it to be implemented here the mods would presumably have to accept and deploy submitted changes.
It’s a good question whether the mods will allow this feature, but I would call that question whether they will “accept”, “merge”, or “deploy” the feature. In the usage I’m used to, the word “implementing” means writing the actual code.
FWIW you can use Firefox reader mode to make medium more readable.
…but why? Just don’t click it…
If you want a site where you do all the work of skipping things you don’t like, then Lobsters might not be for you.
Cool story. Nobody is here to talk about whether Lobsters is for me or not. Since we’re here though, filtering is optional. Since I can choose not to filter, looks like it’s just fine for me.
Thanks for the unsolicited advice, though. I guess?
Lobsters is fundamentally a sorting and filtering site for web pages. Even when you don’t enable tag filters, the home page of Lobsters is filtered to show only the highest-ranked stories (where ranking mostly depend on votes and time). So filtering is not optional, it is the goal of Lobsters. I think that’s why @nil said that if you don’t like filtering, you shouldn’t be on Lobsters.
Why do you assume that everyone only uses the homepage? There’s also a “/recent”, which is a pretty chill place. :)
Also, the problem isn’t what nil said. Like, off-topic unsolicited advice sucks, but unfortunately that’s the kind of attitude that tech culture spews. The issue is that telling people they don’t belong here is shitty, toxic behavior and in this case that it’s also completely off topic.
That’s some HackerNews level bullshit. If I ever don’t belong here, it’ll be for the same reason I stopped using HackerNews. Because it’s full of people like that.
Oh yeah… I forgot that the Recent page existed, because I never visit it. I thought that story lists only appeared on the home page and on searches. Now that I see that Lobsters has prominent support for viewing its links without filtering/ranking, I agree that the argument I ascribed to @nil is a bad argument.
Edit: I can see that you saw @nil’s reply as rude. That surprises me, because I interpreted “you” in nil’s comment to mean “one”, so that the comment was not specifically urging you to leave. I heard it more like “if one didn’t like filtering, one wouldn’t be on Lobsters – therefore, all Lobsters users, including you, should be okay with this filtering too”. I don’t actually know in which sense nil meant their comment, but it might not have been as aggressive as you are interpreting it.
@raymii this might not be the solution you’re looking for but a WebExtension can be written quickly to affect only lobste.rs and remove the offending entries from the HTML you view. This would work only on browsers which had the add-on installed and wouldn’t require any change to the lobster codebase.
I just went ahead and built the add-on, it is at https://git.sr.ht/~soapdog/lobsters-sans-medium
You can go to about:debugging in Firefox and select “Add Temporary Add-on” to test it out. If people think it is useful, I could publish it.
I would like to ask the same for imgur.com.
I suggest banning medium.com submissions from the site.
Yes please! Medium has a garbage UI, and usually garbage content. I’d enable the filter in a second.