If you're interested in recent innovations on this front: (rather than just retreading the same Web as before)
* Beaker (https://beakerbrowser.com) the peer-to-peer browser just released their beta release - and it has some exciting features. Particularly the built-in editor, meaning you can edit, serve and read your pages all from the browser. (Blogging in Beaker is as simple as visiting: hyper://a8e9bd0f4df60ed5246a1b1f53d51a1feaeb1315266f769ac218436f12fda830/. And the posts are stored locally.)
* https://special.fish/ This completely low-tech social network has taken off. The innovation here is that it's all just focused on profile pages - not feeds of random posts.
* There's a growing subculture of public Tiddlywikis (philosopher.life, sphygm.us, etc) - rather than focusing on protocols and APIs, they are much more focused on how to organize and style personal hypertext.
* As for RSS, well, as HN custom insists, I am also commenting to plug my own fraidyc.at. See, you knew it was here.
* There's also a forum on tiny personal link directories that's been forming at https://forum.indieseek.xyz. The idea here is to use Yahoo! or DMOZ style link directories at a smaller scale, to catalog corners of the Web. (Note that this whole comment itself is a kind of small 'directory'. Rather than an algorithm stepping in to show you 'related' stuff, I have.)
Thanks for this post. I checked Fraidycat out and it solves many issues that I consider broken with social medias these days. Your personal website made me incredibly nostalgic and almost depressed, longing back to a simpler time in my life.
Both of your projects will serve as inspiration for myself. Your personal site especially as it feels creative and liberated, things I strive for in my personal life.
Oh don’t be depressed - my messy site took five years to make and you would be disgusted at the guts of it. And besides, there are many many days still in the future - I just flicked through my calendar and couldn’t seem to reach the end - so there is time to unlock more for you. Your site looks good, so I think you are underestimating yourself, which is lying Sami - it’s lying.
I'm not sure I understand the special fish or the Tiddlywikis - what are they? I clicked around, but I don't get it.
Fraidyc.at is very cool though - I remember seeing that a while back and think I'll likely end up using it at some point when I have some time to play with it.
Hi zalberico, I'm h0p3. I can't say I get it either. My wiki is a place to store stuff and to tell myself linked stories. Maybe it is a hypertext castle, a rabbithole, a garden, a mind map, an exobrain, a Zettelkasten, an existential mirror or conduit. Despite significant effort, I'm not sure how to describe it well enough. It's a site I like to see. It's where I do muh thinkin, I tellyawhat.
I love Fraidycat - it gave me the final push to delete my twitter account I had since 2007: let's me keep following the half-dozen accounts I actually cared about. Thanks for your work!
I just tried downloading the Windows executable. Initially it said it will take 1.5 hours, but then the download just failed. Tried a few more times with the same result. This is ironic on so many levels.
Hey – Installed Fraidycat (FC) and tested with DaringFireball (DF) feed. I see it shows the DF title and links to whatever 3rd party link DF is reporting on.
Title: The Unicorns Fell Into a Ditch
Link: bloomberg article
In this example the title is pretty much meaningless so I need DF's take on the topic before I make a decision to go off and visit Bloomberg. My existing RSS feed provides DF's blog content (text).
So I just wanted to understand whether I'm missing something here, or in fact FC is doing exactly what you designed it to do; title and link to 3rd party page only. Pretty much like how Hacker News (HN) operates?
Prehaps it's just the way I personally navigate. I never click on HN titles to go to 3rd party links. I always hit the comments first. The top comments usually give me a good overview before I decide to vist the 3rd party page.
Probably FC isn't intended for me, but I just wanted to sure I'm not missing something, because I'm genuinely excited to use. Regardless, many thanks for making FC.
P.S. I love the retro look of you personal site and FC's home page.
That's cool - thanks Pixxel! So Fraidycat literally just grabs the title and link from each post in the feed - so it has to do with how DF formats its feed. It appears that the JSON feed works correctly tho: https://daringfireball.net/feeds/json
Also BTW, I joined Fish yesterday. Today I went to login and I couldn't. My password/email was definitely correct. So I tried resetting. When adding a new password the error message kept telling me that my new passwords didn't match.
I used a special character (&) in my password. I removed the ampersand and was able to reset password, and log back in.
Sorry if this isn't the correct channel for bugs (feel free to redirect me).
Interesting! These sites reminds me of the various scenes that flourished around the nettime mailing list and the net.art movement[1]; and it also reminds me of the Web 2.0 days when microblogging sites such as Jaiku and others where popular.
Any interest in breaking fraidycat out into an app you can self-host and visit via the browser? I'd love to be able to use it from any device via a browser, and have the hosted version be the single source of truth.
> As for RSS, well, as HN custom insists, I am also
> commenting to plug my own fraidyc.at. See, you knew it was
> here.
This is quite a nice implementation and the back story is cool.
I have also been thinking about this for some time, but what I want to achieve is decentralized RSS feeds. Servers go down over time, are blocked or suffer some other issue. I want to be able to help keep alive the content I consume, possibly without the need for centralization.
Ideally the aim would be to maintain backwards compatibility and be as decentralized as possible, but I'm still yet to solve this problem without resulting to just using torrents.
Have a look at the beaker browser suggested in the parent post. It uses the hypercore protocol (new name for the dat protocol), which allows one to build collections of data, update them, and have them mirrored over a DHT, kind of like IPFS does. You can choose to contribute to mirroring some content you like, if you want to :)
Of course, ideally, the websites themselves would publish articles on the hypercore network, and be the source of trust here, but I don't see why you couldn't do it from RSS feeds.
I am not sure whether the data is content-addressed on hypercore? If so, multiple people could archive the same content to make sure it is identical, and still enjoy the benefits of distributed hosting :)
Completely different. Look up privacy or anonymity on beaker's site? Crickets. It's a browser, apparently with all the dangerous bloat (javascript...) with a bolted-on p2p network that apparently does nothing for privacy or censorship resistance.
Freenet is a p2p network with a focus on anonymity and resistance against censorship. It's not a browser, but for freesites it's got protections in place to filter out dangerous active content.
Beaker came to my mind immediately as well when reading this article :) It sounds so much like the perfect fit for the author.
The new beta has been quite depressing for me, as all my personal apps stopped working due to their complete rewrite of their API. I don't know if I'll ever be willing to start it all over again.
But for anyone who hasn't already played with it, I totally recommend it. It feels like the web we should have had.
Really sorry about breaking your old apps. If you ping me on twitter (@pfrazee) or IRC, I can spend some time helping you convert them over. Most of the same capabilities are there (or on the way) so we'll hopefully just need to update the code.
EDIT: for anybody curious, we had to make breaking changes to the p2p protocol and used that as a chance to bundle a lot of improvements -- mainly with performance and reliability. It sucked to break existing content though.
Beaker looks very interesting indeed! Download it as I write this. Reminds me a little of what I've read about HyperCard. Looking forward to testing it.
Thanks for the heads-up! There are so many links to follow up on here in this thread.
Wow, thanks for this list of things to check out. This is exactly the sort of thing I was hoping we'd be talking about and share.
I downloaded and am playing with fraidyc.at and am so far really liking the idea -- the whole idea made more sense after watching your video and then actually testing it. Thanks for making it.
Oh hey! Really honored you took the time. I love that your blog has has had some attention these past few days. You write very thorough articles and I love that you have a somewhat quirky design, Parimal. It’s classic!
My favourite portal for discovering sites like this is something I discovered just recently here on HN: https://wiby.me/
It never fails to amaze me how much amazing stuff is out there online, hidden by a thick layer of top search results, and even more than that, the sheer amount of individual and collective effort that has been put into each of these sites. Someone mentioned the word "niche" and there is certainly some weirdly (or wonderfully) specific content you will find in the Wiby.me index. Lots of sites that haven't been updated since 1998, but still have an enormous and encyclopedic list of everything related to some topic (like the characteristics of different types of tomatoes, or how to build a motorcycle from spare parts or whatever). Some of it may be a little out of date, but a lot of it has been submitted for indexing precisely because of its timelessness or continued usefulness.
Whenever I feel hopeless about the current state of the web, I find this is the perfect antidote!
I've been thinking about this for ages, and I want my own contribution to this to be a simple webring service.
If you're unfamiliar with the concept, a webring was a simple circular linked list. You had a link on your knitting-themed site to the "next knitting-themed site", that site had a link to the next one, etc.
To join the ring, you just emailed someone and said "hey, I, too, have a knitting-themed site, can you add me to your webring?", they looked at your site, and changed their link to your site, you added the link they previously had, and the ring continued.
I want to build something simple that'll serve a small widget with previous/next/random site buttons, it'll work like the webrings of old regarding the curation aspect, so to get added you'll need to be referred to by someone.
Would you use something like that? You'd basically just drop a bit of HTML on your page and it wouldn't load heavy JS/analytics/crap, just whatever was necessary to paint a few links.
I'm personally more into personal directories and blogrolls than these random clicks - but they still seem to be a good way to put together a small community. (And this post isn't meant to discourage you - but rather to encourage you to form your own.)
Absolutely. Working on this for a rural area near me with many local artisans. A simple "Our Friends" section that rotates and encourages users to explore artisans in the area.
I'm working on a webring type project right now, with a central site that pulls in RSS feeds of all the ring members to create a central news feed (HN style interface), to help with discoverability too.
You add the webring to your site, allowing your visitors to discover more sites you specifically like, then you can opt-in to have your RSS feed fed back to the main site.
When I picture it in my head I think of the early web as more of a library. Over time it has transitioned into a shopping mall.
If I continue with this thought exercise, a lot of the big indoor shopping malls around me have been knocked down and replaced with standalone outdoor stores (walled gardens?).
I think there have been two main changes that have hurt the web:
1) The shift from spammy shit content as something to squash to something to allow and even promote over better content, provided it follows certain (Google's) rules. This shift (in terms of Google's behavior) happened ~2008-2010 and we haven't seen a period of spammy crap content getting heavily downranked since then, like we used to when they were still trying to stay ahead of it rather than give it a "legitimate" avenue as a method of control. Google's still being the most important search provider to appease has left the rest unable to direct behavior toward anything better than what does well on Google, so their results aren't much better.
2) A move away from actual or de facto open systems & protocols to deliberately carved up communities. The only thing keeping chat, Twitter-like services, and other social media—hell, even Youtube, so far as some kind of format for hosting video with metadata—from being standards or protocols is that business incentives reward "owning" a userbase (so you can better spy on them, and to keep anyone from providing a better, perhaps less-spying-laden client and "stealing" ad-viewing eyeballs)
Both of these are fundamentally problems of the spyvertising economy taking over the Web and I think a lot of the issues would go away if we could (legally—I don't think tech will do it) permanently and completely break that. More specifically a big part of the problem is Google, though of course the rest of the Web giants are gleefully following similar bad incentives.
I don't think search engines are promoting spammy $#!+ content per se; what they're doing is heavily promoting newer content that's relevant to the most common search queries, as this gives them the only real hope of staying ahead of the spam. Of course the "small", long-lasting, independent Web is heavily disadvantaged by this shift.
One development that would be good for small web sites to look into is schema.org linked-data formats. Those might simply be too effort-intensive for the spammers to adopt (at a high level of detail) and perhaps too much of a commitment to quality and transparency (they would have to actively forge the info, which would leave them open to bans given the lack of plausible deniability), so they might become a viable signal of quality and lead to higher visibility in SERP.
(Similar for things like proper separation of style from content, that have always been advocated for in the web-standards community but are not really commercially viable.)
I'm not quite sure if others have experimented with this stuff already, but it seems worth trying.
I thought the exact same thing about the web as being a library. It's nice hearing someone else say it, it's validating.
However, I feel like it's moved more in the direction of being like broadcast television: lots of content that's designed to be consumed once and then forgotten. Maybe the television analogy oversimplifies the matter. Still, I think the more that content creators view their content as going into a permanent library, the better the quality.
Author here. That's an interesting analogy — especially given that small book stores and libraries (to a lesser extent) still do exist, albeit in much smaller numbers, attracting fewer patrons.
A lot of web traffic today transits through applications and platforms. This isn't necessarily a bad thing; although I'd hate to see even more walled gardens. My hope is only that the small and independent web not go forgotten/ignored.
It's still mostly a library, except with commissioned salespeople everywhere and books full of vapid content that scream, fly at you from the shelves, and turn their pages as you try to read them.
It's a search engine that removes the top million domains from your search results (or top 100,000 or 10,000, etc). I find it useful sometimes to discover things on more obscure sites.
It’s been going well so far, only a small solution to a big problem, but it’s been fun to discover a lot of interesting blogs from people in this community.
If you missed the initial post feel free to join and add yours too.
There are 2 major factors that will power a resurgence, that could use better tools:
1. Discoverability - self-reinforcing webrings, blogrolls, directories, ad free-search etc.
2. Creation - next-gen Dreamweaver. A low-code site creator app that exports a static website as a folder of readable CSS/HTML that people can tinker with by hand (and learn), instead of being locked into one of these cloud WYSIWYG site generators. Hosting is solved. No need to tie the one to the other...
This is a great article, the kind of stuff I find interesting on HN.
Some of the "old school" web style clashes with my aesthetic sensibilities these days -- a lot of words to say I find it somewhat ugly! -- but I miss its hobbyist, non-commercial aspect. A lot of hobby style content I find interesting has moved to YouTube or Facebook these days, and everyone who's been reading HN is aware of the lack of control authors have over those platforms...
I found myself nodding in agreement to a lot of what the author was saying.
Indeed, I enjoy that æsthetic in certain contexts, admittedly because it makes me nostalgic more than anything else. I had to exercise restraint! :) I wanted the site to remain readable and accessible in text mode/lynx/screen readers and respect basic typograhical conventions.
But I agree with you that the thing that I miss the most is the hobby-ist, non-commercial aspect. But I'm discovering a lot of great links on this HN thread!
Readit News
* Beaker (https://beakerbrowser.com) the peer-to-peer browser just released their beta release - and it has some exciting features. Particularly the built-in editor, meaning you can edit, serve and read your pages all from the browser. (Blogging in Beaker is as simple as visiting: hyper://a8e9bd0f4df60ed5246a1b1f53d51a1feaeb1315266f769ac218436f12fda830/. And the posts are stored locally.)
* https://special.fish/ This completely low-tech social network has taken off. The innovation here is that it's all just focused on profile pages - not feeds of random posts.
* There's a growing subculture of public Tiddlywikis (philosopher.life, sphygm.us, etc) - rather than focusing on protocols and APIs, they are much more focused on how to organize and style personal hypertext.
* As for RSS, well, as HN custom insists, I am also commenting to plug my own fraidyc.at. See, you knew it was here.
* On a related note, I've also been working on an RSS/Atom extension to handle ephemeral posts: live streams, "stories", pinned posts, etc. https://github.com/kickscondor/fraidycat/wiki/RSS-Atom-Exten...
* There's also a forum on tiny personal link directories that's been forming at https://forum.indieseek.xyz. The idea here is to use Yahoo! or DMOZ style link directories at a smaller scale, to catalog corners of the Web. (Note that this whole comment itself is a kind of small 'directory'. Rather than an algorithm stepping in to show you 'related' stuff, I have.)
Both of your projects will serve as inspiration for myself. Your personal site especially as it feels creative and liberated, things I strive for in my personal life.
Thank you.
Fraidyc.at is very cool though - I remember seeing that a while back and think I'll likely end up using it at some point when I have some time to play with it.
Also, Fraidyc.at works with Tiddlywikis.
I just tried downloading the Windows executable. Initially it said it will take 1.5 hours, but then the download just failed. Tried a few more times with the same result. This is ironic on so many levels.
Title: The Unicorns Fell Into a Ditch
Link: bloomberg article
In this example the title is pretty much meaningless so I need DF's take on the topic before I make a decision to go off and visit Bloomberg. My existing RSS feed provides DF's blog content (text).
So I just wanted to understand whether I'm missing something here, or in fact FC is doing exactly what you designed it to do; title and link to 3rd party page only. Pretty much like how Hacker News (HN) operates?
Prehaps it's just the way I personally navigate. I never click on HN titles to go to 3rd party links. I always hit the comments first. The top comments usually give me a good overview before I decide to vist the 3rd party page.
Probably FC isn't intended for me, but I just wanted to sure I'm not missing something, because I'm genuinely excited to use. Regardless, many thanks for making FC.
P.S. I love the retro look of you personal site and FC's home page.
Also, there is a redesign underway that will give richer post view: https://github.com/kickscondor/fraidycat/issues/30
I used a special character (&) in my password. I removed the ampersand and was able to reset password, and log back in.
Sorry if this isn't the correct channel for bugs (feel free to redirect me).
Cheers.
[1] If anyone wants to know more about it, see https://en.wikipedia.org/wiki/Net.art
> commenting to plug my own fraidyc.at. See, you knew it was
> here.
This is quite a nice implementation and the back story is cool.
I have also been thinking about this for some time, but what I want to achieve is decentralized RSS feeds. Servers go down over time, are blocked or suffer some other issue. I want to be able to help keep alive the content I consume, possibly without the need for centralization.
Ideally the aim would be to maintain backwards compatibility and be as decentralized as possible, but I'm still yet to solve this problem without resulting to just using torrents.
Of course, ideally, the websites themselves would publish articles on the hypercore network, and be the source of trust here, but I don't see why you couldn't do it from RSS feeds.
I am not sure whether the data is content-addressed on hypercore? If so, multiple people could archive the same content to make sure it is identical, and still enjoy the benefits of distributed hosting :)
More about hyperdrive/hypercore: https://news.ycombinator.com/item?id=23180572
https://freenetproject.org/index.html
Freenet is a p2p network with a focus on anonymity and resistance against censorship. It's not a browser, but for freesites it's got protections in place to filter out dangerous active content.
The new beta has been quite depressing for me, as all my personal apps stopped working due to their complete rewrite of their API. I don't know if I'll ever be willing to start it all over again.
But for anyone who hasn't already played with it, I totally recommend it. It feels like the web we should have had.
EDIT: for anybody curious, we had to make breaking changes to the p2p protocol and used that as a chance to bundle a lot of improvements -- mainly with performance and reliability. It sucked to break existing content though.
Here's my first commit: https://github.com/kickscondor/duxtape/commit/55dbde9519aedb.... Can't really call it a rewrite - more of a search-and-replace. (Though I had a bit of code that used the old peer sockets - and that code works differently now.)
Thanks for the heads-up! There are so many links to follow up on here in this thread.
Deleted Comment
Thank you so much for all the hard work.
I downloaded and am playing with fraidyc.at and am so far really liking the idea -- the whole idea made more sense after watching your video and then actually testing it. Thanks for making it.
It never fails to amaze me how much amazing stuff is out there online, hidden by a thick layer of top search results, and even more than that, the sheer amount of individual and collective effort that has been put into each of these sites. Someone mentioned the word "niche" and there is certainly some weirdly (or wonderfully) specific content you will find in the Wiby.me index. Lots of sites that haven't been updated since 1998, but still have an enormous and encyclopedic list of everything related to some topic (like the characteristics of different types of tomatoes, or how to build a motorcycle from spare parts or whatever). Some of it may be a little out of date, but a lot of it has been submitted for indexing precisely because of its timelessness or continued usefulness.
Whenever I feel hopeless about the current state of the web, I find this is the perfect antidote!
If you're unfamiliar with the concept, a webring was a simple circular linked list. You had a link on your knitting-themed site to the "next knitting-themed site", that site had a link to the next one, etc.
To join the ring, you just emailed someone and said "hey, I, too, have a knitting-themed site, can you add me to your webring?", they looked at your site, and changed their link to your site, you added the link they previously had, and the ring continued.
I want to build something simple that'll serve a small widget with previous/next/random site buttons, it'll work like the webrings of old regarding the curation aspect, so to get added you'll need to be referred to by someone.
Would you use something like that? You'd basically just drop a bit of HTML on your page and it wouldn't load heavy JS/analytics/crap, just whatever was necessary to paint a few links.
Hotline Webring: https://hotlinewebring.club
XXIIVV Webring: https://webring.xxiivv.com
Weird Wide Webring: https://weirdwidewebring.net
I'm personally more into personal directories and blogrolls than these random clicks - but they still seem to be a good way to put together a small community. (And this post isn't meant to discourage you - but rather to encourage you to form your own.)
You add the webring to your site, allowing your visitors to discover more sites you specifically like, then you can opt-in to have your RSS feed fed back to the main site.
If I continue with this thought exercise, a lot of the big indoor shopping malls around me have been knocked down and replaced with standalone outdoor stores (walled gardens?).
I'm not sure where things are going next.
1) The shift from spammy shit content as something to squash to something to allow and even promote over better content, provided it follows certain (Google's) rules. This shift (in terms of Google's behavior) happened ~2008-2010 and we haven't seen a period of spammy crap content getting heavily downranked since then, like we used to when they were still trying to stay ahead of it rather than give it a "legitimate" avenue as a method of control. Google's still being the most important search provider to appease has left the rest unable to direct behavior toward anything better than what does well on Google, so their results aren't much better.
2) A move away from actual or de facto open systems & protocols to deliberately carved up communities. The only thing keeping chat, Twitter-like services, and other social media—hell, even Youtube, so far as some kind of format for hosting video with metadata—from being standards or protocols is that business incentives reward "owning" a userbase (so you can better spy on them, and to keep anyone from providing a better, perhaps less-spying-laden client and "stealing" ad-viewing eyeballs)
Both of these are fundamentally problems of the spyvertising economy taking over the Web and I think a lot of the issues would go away if we could (legally—I don't think tech will do it) permanently and completely break that. More specifically a big part of the problem is Google, though of course the rest of the Web giants are gleefully following similar bad incentives.
One development that would be good for small web sites to look into is schema.org linked-data formats. Those might simply be too effort-intensive for the spammers to adopt (at a high level of detail) and perhaps too much of a commitment to quality and transparency (they would have to actively forge the info, which would leave them open to bans given the lack of plausible deniability), so they might become a viable signal of quality and lead to higher visibility in SERP.
(Similar for things like proper separation of style from content, that have always been advocated for in the web-standards community but are not really commercially viable.)
I'm not quite sure if others have experimented with this stuff already, but it seems worth trying.
However, I feel like it's moved more in the direction of being like broadcast television: lots of content that's designed to be consumed once and then forgotten. Maybe the television analogy oversimplifies the matter. Still, I think the more that content creators view their content as going into a permanent library, the better the quality.
A lot of web traffic today transits through applications and platforms. This isn't necessarily a bad thing; although I'd hate to see even more walled gardens. My hope is only that the small and independent web not go forgotten/ignored.
https://twitter.com/earthboundkid/status/1095385048008798208
It's a search engine that removes the top million domains from your search results (or top 100,000 or 10,000, etc). I find it useful sometimes to discover things on more obscure sites.
I created a subreddit recently to help with the discovery problem and posted it on HN earlier this week.
Show HN: https://news.ycombinator.com/item?id=23287286
Subreddit: https://www.reddit.com/r/hnblogs/
It’s been going well so far, only a small solution to a big problem, but it’s been fun to discover a lot of interesting blogs from people in this community.
If you missed the initial post feel free to join and add yours too.
I'm a fan.
[Edit]: I also caught a tiny typo:
> "But the web is not always "profit-oriented" and it certainly does not need be "user-centric" (and I say this as a UX consultant)."
I think you wanted "does not need to be"
I created a PR for you: https://github.com/parimalsatyal/neu/pull/2
There are 2 major factors that will power a resurgence, that could use better tools:
1. Discoverability - self-reinforcing webrings, blogrolls, directories, ad free-search etc.
2. Creation - next-gen Dreamweaver. A low-code site creator app that exports a static website as a folder of readable CSS/HTML that people can tinker with by hand (and learn), instead of being locked into one of these cloud WYSIWYG site generators. Hosting is solved. No need to tie the one to the other...
Some of the "old school" web style clashes with my aesthetic sensibilities these days -- a lot of words to say I find it somewhat ugly! -- but I miss its hobbyist, non-commercial aspect. A lot of hobby style content I find interesting has moved to YouTube or Facebook these days, and everyone who's been reading HN is aware of the lack of control authors have over those platforms...
I found myself nodding in agreement to a lot of what the author was saying.
I don't miss the "geocities look" though!
But I agree with you that the thing that I miss the most is the hobby-ist, non-commercial aspect. But I'm discovering a lot of great links on this HN thread!