Hacker News Reader

Comment by jacobobryant | Hacker News Reader

Heh, I published a post in a similar vein just today[1]. Lately I've been thinking the way to do it is have everyone publish a (paginated) Atom feed with all their public content interactions. So each item in the feed would be something like "published this blog post", "liked this tweet", "thinks this comment was written in bad faith", etc. And "follows this person" with a link to that person's atom feed. Then if you want to build a search engine/social network/recommender system, you can get started by crawling and indexing a bunch of feeds.

I'd like to make a web app (probably on replit since it makes self-hosting dead easy) that has plugins for a bunch of other sites. so you connect to all your other accounts, and then this app imports your data and publishes it to a single feed for you.

Obviously starting audience would be programmers, but I bet it could expand to a general audience too.

[1] https://news.findka.com/p/we-need-user-centric-data

View on HN

I've considered this perspective as well from a lower level. I'm a huge fan of gossip-style communication, and I've wondered why all my data is shredded across various databases when it could be all located within a single giant JSON on my phone, and then I could just gossip a version to my friends.

The key is privacy at the data-layer, and I have a start with my programming language for board games: http://www.adama-lang.org/

Have you seen ActivityPub? From a very quick glance, it seems like it might fit this, and it's relatively actively used: https://en.wikipedia.org/wiki/ActivityPub

It's not RSS (which is a bit of a shame), but it scales better since it bakes in pushing / pulling patterns, where RSS only pulls.

I have seen it but haven't looked into it too deeply. I think it's more complex/sophisticated than what I have in mind. i.e. it's meant to be a protocol for an entire social networking app, whereas I'm interested in a much smaller scope: just a raw event log. I figure if that's in place, everything else can be built on top. Heck, you could even use it as a base for an ActivityPub implementation! Write a tweet -> gets imported into your RSS event log -> gets published via ActivityPub.

Also, there is another protocol called Pubsubhubbub (yeah, I know...) which adds push on top of Atom/RSS. I know of at least one company, Superfeedr, that runs an implementation as a service (they were acquired by Medium): https://superfeedr.com/

I’ve been noodling this idea for a few years. I think it’s a great idea and it being based on Atom/RSS is a great move.

The problem, I have found, is no one wants to pay for hosting. No one wants to self host. And by no one I mean no one that is using Facebook /Twitter / Insta / etc. and for this to really work, I think you need everyone hosting their own “log”.

Interested in seeing more work done in this space and am glad to see others feel the same as I do.

That's a big reason why I'm excited about repl.it. You can self-host a github repo in about 2 clicks (not including account setup). I could see it increasing the reach of self-hosted apps a lot.

Actually I made a little proof of concept for this a week ago: https://github.com/jacobobryant/Feedstuff

I don't think the issue is hosting the app itself, it's hosting each user's data. For the vast majority of users, hosting their own data is not even on the radar. They assume that whatever app they are using is hosting their data, and the only price they will accept for doing that is "free".

The app would include a database (perhaps sqlite), so that shouldn't be an issue unless I'm misunderstanding you. you can host it for free on replit already (albeit with sleeping/cold starts)

> The app would include a database (perhaps sqlite)

Meaning an sqlite database hosted with the app's code? How will "host it for free" scale when your app goes viral and you have a million plus (or a billion plus) users?

It'll scale because it's self-hosted. You don't serve everyone with the same app instance; everyone runs their own instance.

> It'll scale because it's self-hosted. You don't serve everyone with the same app instance; everyone runs their own instance.

So everyone has their own free hosting account with repl.it and runs an app instance on it? Yes, technically that's "free" as in "no cost", but it's not "free" as in "no effort". The latter is the kind of "free" that Facebook users expect.

It's hardly more effort than making a Facebook account, and replit is still young--it's only going to get easier.

> It's hardly more effort than making a Facebook account

I'm not talking about the effort involved in opening an account. I'm talking about the effort involved in managing one's own instance of an app. That is not zero, and nobody, not repl.it or anyone else, can magically make it zero. But it is zero for Facebook users, because they aren't managing anything; Facebook is.

If, OTOH, your idea is that repl.it does all the managing (what happens if the server hosting the app goes down? how is the data backed up? etc., etc.), then we're back to the scaling problem: how do you scale that to a huge number of users without either not being free, or adopting the same business model that Facebook has (the avoiding of which was supposed to be the reason for doing all this in the first place)?

my mother won't sign up at github.

>> The problem, I have found, is no one wants to pay for hosting. No one wants to self host.

Self hosting needs to be as easy as plugging a box into your router. This means some type of DNS or other way to find your node given all the ISPs failure to provide fixed IPs. Also a super easy management method and access controls (public, private, friends).

I have a lot more thoughts on that.

Do you think there needs to be the infra to support the self hosting or the “killer app” first to drive the early adopters?

I think an internet that is mostly self-hosted is a healthy one and unlocks a lot of things I’ve been noodling with over the years. I’m hopeful someone smarter than me is close to getting us there.

A few gaps in infrastructure need to be plugged first. We have most of the constituent parts, it's just that they lack administration layers suitable for modern internet users needs. It's work that requires overcoming inertia and tedium more than any particular smarts. (It'll likely go financially unrewarded too.)

indeed, what do you think about an onboarding like at https://demo.mro.name/shaarligo/o/p/ac2gthz

I love the core idea too. Also static hosting keeps getting cheaper, free-tier can take you pretty far, and self-hosting keeps getting easier.

This kind of thing might never hit mass-market (read Facebook-level) appeal, but I think if one branded it as an open-web network for creators (as opposed to Facebook, which is for mere “consumers”) you might reasonably avoid the hosting problem.

A “creator” will figure out how to host, be it a Wordpress.com site on the casual end, or something more specialized for the savvier creator.

Just of out curiosity, would you consider hosting your friends and families logs if there were a low friction way to do it?

sure I would.

Interesting concept for sure. Is there a revenue model to attract top influencers? Maybe just links to a collection of Patreon-likes?

hm, no idea. I haven't thought about revenue models much since I've envisioned this more as a grass-roots thing than a startup opportunity. I think the main thing would be just making it dead easy to get started. You wouldn't necessarily need to jump start a new network since the whole point of this is to be interoperable with existing networks.

A good first step might be making a personal site generator. Put in links to your twitter/youtub/substack/medium/whatever accounts, and then you get a nice looking site with all your content in one place. Maybe throw in a subscribe button so people can get email digests of all your content.

and then, almost as a side effect, anyone who uses the site generator also is publishing it all as an atom feed.

How about Facebook, which requires an account to view the person posts? How would those be read? Besides maybe requiring user & password and logging in with a headless browser. Not sure if Instagram/Pinterest require a login too.

Entering in username + password is probably the only feasible way. Suboptimal for sure, but worked for mint...

In practice, this would probably start out with plugins for only sites that make data public. Not perfect, but still useful.

I think the money is in the aggregator. Everyone creates their own “log” so the noise is crazy. You follow “influencers” or go to aggregate sites that only vet info, not produce it.

You could have the aggregate behind a paywall, you could have donations or ads as well.

Web Monetization standard. Yes. https://webmonetization.org

I'm doing paginated atom feeds as microblog targeting laypeople: https://demo.mro.name/shaarligo - still quite rough.

Reply on news.ycombinator.com