Comment by ISL | Hacker News Reader

Comment by ISL | Hacker News Reader

Parent Full thread

Agreed in part (and upvoted), however, I understand the frustration as an open web advocate when interacting with walled gardens like Instagram and Facebook.

Rather than competing on the product/network alone, Instagram employs dark patterns like allowing only a single link ("link in bio") to the wider web. Links in comments don't work, nor do older links-in-bio persist with their original post.

Facebook Groups is cannibalizing a dearly beloved two-decade-old backcountry ski forum ( turns-all-year.com ) because the user-onboarding is effortless, familiar, and integrates with other familiar walled-garden features. With the FB interface, the discussions are shorter-lived and the archiving of trip-reporting information is no longer effectively searchable. I don't know how an open-web forum with limited resources can compete against FB effectively -- suggestions are very welcome.

Strategies like these are designed to gently carve users away from the open web and gather them within the garden. While not impossible to combat, they are a stout challenge for sites in the modern era.

The degree to which Instagram or Pinterest bleed into the "old web" is hurting the surfing experience of those who want to stay outside of those walled gardens.

There are millions of entry points into those walled gardens, which stop me right at the gate. I can't see a timelapse of a solar flare, because the photographer [1] has been lured into using Instagram as his publishing platform in return for a better interaction with "followers". From a technical viewpoint, there is no reason for this restriction. This is breaking the old internet.

[1] https://www.instagram.com/cosmic_background/

"Log in to view posts from cosmic_background"

I love the modern web :)

This is why we have to fight against it at a much lower level. HTTPS is a prime example of this. It effectively "locks" the data channel or pipe from the source (FB/Instagram/Google servers) to your viewing device (screen, and in a lot of cases, locked-down web-browser) making it effectively a dumb terminal. If MITM-type techologies were allowed to exist safely and controlled by the user, we could be seeing a proliferation of an ecosystem geared around breaking and integrating-with the walled-gardens. I'd go out on a limb and say that HTTPS is one of the biggest hurdles we face when it comes to keeping the web open. With it firmly in-place, we are forced to beg for "scraps" in terms of these walled-gardens giving us API/Integration points.

Edit. Grammar.

This argument is silly in the presence of open-source web browsers that can accept extensions.

> proliferation of an ecosystem geared around breaking and integrating-with the walled-gardens

Youtube-dl works just fine. What kind of things did you have in mind that people might want but are unachievable because of https?

> This argument is silly in the presence of open-source web browsers that can accept extensions.

The only browser like that with any meaningful adoption is slowly dying, as its parent organization is spinning like a headless chicken.

I'm sure many would happily put an end to Youtube-dl (and other such tooling), and for some the copyright infringement argument is just a cover for their issue with youtube-dl basically punching a hole in their garden's wall.

As a side note, imo the state of the copyright laws also does no good to the openness of the web.

I don’t think instagram’s link policy has anything to do with dark patterns. Rather it is a distinct design choice for a social network to avoid spam. Another key decision instagram made was preventing shares/retweets, with the intended effect of promoting original content. These two consequential decisions (both made very early on in Instagram history, likely even before the fb acquisition) made original photos the focus. It’s an opinionated app and has succeeded because of it — I don’t believe either decision should be called a dark pattern.

> Rather it is a distinct design choice for a social network to avoid spam

It's not clear to me how blocking read access prevents spam.

Don’t forget that Instagram was originally just an app with no website. The fact that you got a website to see some posts technically is a step towards openness. =]]

I have similar feelings about the move away from native apps that others have about the move away from the “open web”: I like being able to control the update cycle of the applications I use and not accept new features I don’t like.

I was referring to the fact that they don’t make urls clickable, to disincentivize spammers sharing urls (at least make it a less effective spam tactic)

I never thought of the blocking of links as spam prevention but it makes complete sense.

Now that they’ve progressed a lot they have swipe up links for various accounts that integrate with ads and other publishers so they are loosening the grip.

At first i hated not having links and being able to click but now i rather appreciate it because anecdotally it definitely reduced spam.

How the hell does it prevent spam? Spammers just bake bit.ly links into the image which I doubt have a much lower hit rate than usual spam.

Meanwhile, every time I find a new account I like and want to see what they're all about, I have to use a search engine because all of their posts just say "link in bio" for the context and additional content, but their bio link has long since changed. This is a clear dark pattern to keep people in their walled garden and I really can't see any other explanation.

(they do allow links for ads though, which are usually just spam anyways, so that reinforces this even further)

I think making urls non-clickable absolutely reduces the effectiveness of spamming urls. How the hell wouldn’t it?

I figure that people who are gullible enough to believe spam like "Triple your investment at our new exchange: cryptomatron.xyz" wouldn't be stopped by having to type the URL in manually.

Instagram made a lot of changes recently while their success came much earlier. I wouldn't say they succeeded because of those later decisions.

Perhaps some of the recent developments in technology are too effective in moving people towards where others want them. Look at how many of us are actively trying to disengage from smartphones. Not having a smartphone is now unusual to a lot of people, but their use could be affecting us in ways that we don't yet understand on a scale of decades.

Cynicism reminds me of Brave New World, but even for walled gardens like Pinterest that I would rather not put up with, thousands of people outside the tech sphere have taken them up and used them for purposes which are materially advantageous to them. Imagine all of the Facebook-exclusive events that went on to connect people to marriages a decade later, or ideas for wedding registries sealed away in Pinterest. "If it weren't for Facebook/Pinterest/etc., then my life wouldn't have turned out like this." These platforms are becoming melded with millions of people's lives in inextricable ways. We have seen what is possible with enough ingenuity and progress, and we probably won't unsee it. People are happy to use these things, and imposing your will on them to change is usually what people would deem poor social etiquette.

Unless civilization collapses, I doubt it will be any time soon that technology regresses to before this point or a significant movement to move off these platforms emerges.

Agreed about the dying online communities. Part of all online communities are network effects: a forum is only as good as the users on it; and if all the users are on Reddit and Facebook, it makes it far, far harder to find interesting independent communities to join

Anyone can start their own subreddit and attract users. And you get high quality UI for free. Or I could spend weeks setting up my own forum software. Why would I do that?

Facebook et al won because they made things easy. I don’t like FB’s approach to privacy or dark patterns, but those aren’t why they won. Running your own website is a demanding hobby for tech nerds. Running your own FB group, anyone can do. This is because FB makes money from making it easy. “The web” doesn’t make money from making it easy, so it isn’t.

In addition, the open web is open to spammers. That’s why Google’s search results are degrading. Walled gardens cut down on these negative externalities.

I get it, but besides making it easy, FB made online communities worse. They have absurd limitations that you only realize once you're in too deep. For whatever reason, for example, nudity is banned even in private groups on FB, so several photography groups I know are trying to go back to forums and blogs, but can't migrate their content, nor get most people to cooperate, so many of them are juggling multiple FB accounts because they keep getting temp-banned (and I'm not talking porn here - these are actual photographers that are regularly featured in big galleries).

There was also a game dev group that died out because FB kept flagging their users for spam. Turns out posting executables makes it think you're spamming viruses, even though it's just a gane prototype you want feedback on.

Finally, the quality of discussion is worse no matter what the content is, as FB forces the post>comments relationship, with comments being way less visible and usable than a top-level post.

And this isn't just one grumpy programmer screaming into the void - I only know about that because friends and relatives (from young to quite old) keep complaining to me and asking me for workaround to these limitations. Of course, the answer "just use something other than FB" is rarely accepted.

I wonder if a modern discussion platform with Facebook as the login provider (and some deeper integration of content, perhaps?) would help bridge this gap?

> Anyone can start their own subreddit and attract users. And you get high quality UI

Wait, what?

Any old php forum has a much better UI (faster, easier to read) than reddit's.

Do you actually use reddit? It's an extremely cancerous experience on mobile, they actively work on making the ui as inconvenient as possible. Every iteration of reddits ui is even slower than the last one, which I consider quite an achievement considering how slow it was months ago already.

Reddit also has the "gold" thing going which is probably not what you'd want for your own forum, where some users are more equal than others because they paid money to an entity you have a questionable relationship with.

Unrelated: Google's results decline in quality because google values engagement more than quality. It's been like this for years. When you are the monopoly, you can easily trade quality for more money without negative consequences.

I spend a lot of time there, but the day "old.reddit.com" stops working is the day I walk away. The newer interface is unusable in comparison to the original and I will not be forced to take such a massive step backwards.

Deeply missing the ability to search past year trip reports with the shifts in TAY. I think the challenge is once the door is opened to moving a community partly to FB it’s a slippery slope to the lowest common denominator (FB).

Seeing the same unfortunate pattern with some other small camping and car forums.

At least individual users can push back against the slide on TAY by posting trip-reports. If enough of us continue to post, then as GP suggests, the open web (and TAY) will survive just fine :).

One must also balance the utility of the TR with the now-outsized impact of TRs. It was a different world in ~2006, where everyone knew everyone else and a TR meant that at most a few friends and a friendly internet stranger might follow your tracks tomorrow. It is a delicate balance.

I gotta say, it always makes me furious whenever I see "Facebook and Instagram". Not because of the person writing the comment (you in this case) but because of the basic feeling I have that they SHOULD NOT BE ALLOWED to buy up all their competition. Zuck's "innovation yay, rock on" is a bunch of empty platitudes and he's actively criminal

Reply on news.ycombinator.com