I didn't put in enough effort to make either a good graphic or an ironic napkin graphic.

I implemented a new feature for this site, the footer of each new post is going to have three of these guys:

2007.09.24 Firepower and horsepower An example post preview element using my static site generator.

Minor detail: I don't show them in full-month view to keep things a bit cleaner. Autogenerating sneak peeks of other posts felt kind of dirty since that's what all the commercialized sites do, but it helps with navigating similar subjects.

You may have noticed that I just mentioned a feature but my post title is "Feature request". We've now arrived at that part. I want to do the post preview thing I just mentioned, but with other websites, especially from the so-called indieweb.

Rule 34 of internet development says this thing already exists and that's probably true, but I haven't found it. The closest analogue: a lot of sites provide links to content "around the web" that are just advertisements with context-adapted titles. I want this notional feature, but I want the other end of the href to be real content that is not selling anything. Two more asks:

• I don't want to get paid for linking to someone else's site.
• I don't want to have to discover the pages I'm linking to. A database should do that.

Architecture

Referring to the diagram above, the bottom half shows what I want to integrate into my publishing process. I submit my post information in one or more standards that already exist:

• RSS
• Page metadata tags
• Page text blob

The service gives me a few links to put at the bottom of my post that are timely and relevant (more on this later).

The top half of the diagram shows the service: a database of content aggregated via user submissions (e.g. indieblog.page) and subscriptions (e.g. RSS). In the diagram there are two partitions: community content and mainstream content. The latter isn't critical to this design, but when I was thinking about Reddit replacements and how accessible commercial feeds are, adding Associated Press links is a short putt. Similarly, with my recent geopoliticsposting, anyone that survives to the end of the post might want to see what legitimate information sources have to say.

As a publisher, I want:

• A simple mechanism to submit my post/feed to the big database and instantly get links I can add to my post.
• Link metadata (title, thumbnail, description, tags) in a generic format that I can present in my site's visual style.
• To have incoming links that point to the relevant content. If someone lands here while browsing garage floor surfacing ideas, I don't want them to be bothered by my 3,000 word review of the Superman 64 re-release.

As a reader, I want:

• Relevant links to the material I'm reading.
• Recent links, though in certain cases I prefer relevance to recency.
• Good content, not something autogenerated or overly terse or above/below my technical level.

That's it

It really is. It's a basic database, some undemanding subscription/discovery services, and some simple clientside html to have the whole thing going soup to nuts. Everything south of this sentence is just expanding on the 'why' and the 'how'.

Source.

I was going to use this ironically but then ended up following the format.

Strengths

• It doesn't need to be comprehensive to be useful. Getting feeds from 20% of the indieweb and 10% of mainstream media is sufficient to provide really good link material.
• It's self-sustaining. Submitting a page to gather recommended links organically grows the database.
• It's simple for developers. Post xml, get xml, append to post.
• It's mutualistic. It addresses the indieweb discoverability issue and is preferable to spamming links and other dirty SEO tricks. It even uplifts the entire indieweb since external references are one of the main metrics for search rank.
• It doesn't require mass adoption to work. The database can be built and grown with one subscriber or a million subscribers. The link quality is based entirely on the database and recommendation algorithm.

Weaknesses

• It doesn't facilitate discussion. It's not meant to, but if we're talking about the post-Reddit world, it's worth noting that this is closer to Reddit's link aggregator role than its discussion role.

Opportunities

• Search. The database could, of course, be linked to a search page.
• Link quality ratings. I mentioned link selection based on similarity and timeliness, there would certainly be a benefit to user-provided link scoring (active or passive).

Threats

• Garbage content/spam. Anything popular on the internet is a magnet for opportunists. Luckily we have opaque algorithms and shadowbanning.

nunez

[Twitter putting up a login wall] killed nitter. Fuck. I guess I'm done with Twitter. Reddit is in Eternal September. Twitter is login-walled. If HN is next, I'll probably be mostly done with the Internet. This version of the Internet is starting to suck. :(

"Let's go back to the old web" is a common refrain amongst geeks and people burned out on social media, engagement capture, SEO, influencers, astroturfers and all the other commercial nonsense. We're never going to actually go back though, the best we can do is branch and hope the old-new thing is better than the current thing.

Recapping the internet, from 14.4 to present day:

• Before September: bulletin boards and other stuff before my time.
• The old web: portals, blogs, forums, under construction gifs.
• Web2.0: social media. It briefly showed the power of the internet to connect people but quickly turned into walled gardens.
• Web3: like VR or something? It was a buzzword for a little while.

I've rolled the indiepage die more than a few times and can tell you that the blogosphere isn't all gold. Some notable issues:

• There are way too many web developers posting javascript. That's both strictly true and my euphemistic way to say that if I threw a dart at the blog board, I'd hit something that interests me only occasionally.
• Blogs, Discord/YouTube/etc channels, and the indieweb are sparse sources of information. They're like reading random articles in a newspaper rather than the front page.
• It's impossible to navigate the indieweb. Posts sometimes have links, sites sometimes have a link page. That's it.
• From the writer's perspective, discoverability amounts to spamming backlinks rather than letting content quality guide users to you. Site note: I don't spam backlinks, but that's what people have to do.

Last year I read a post about the indieweb being a dark forest; its issues are visibility and navigation. Quality and variety aren't an issue. LinkedIn hasn't gobbled up all of the off-hours posting by credible professionals. Hacker News is like 25% links to blogs or niche sites where people know what they're talking about. Just yesterday after a few blogroulette spins I saw an amazing, tongue-in-cheek, disappointed review of Far Cry 4 because it was a ten minute walking simulator. Many such cases!

Since this post started with the solution and ended with the problem, I needn't now recap why my humble feature request solves the dart board problem, the linking problem, and the discoverability problem. Fundamentally, the internet is about navigation and connections, the old web just lost some of that when Tom and Zuck took over.

Related / internal

Related / external