/search

On this day ten years ago, I returned to the blog after a two-and-a-half-year break and set myself on the path that would ultimately lead to where I am today.

It was a true landmark moment.

I hadn't been happy with what I was doing, so stepped away to explore other things.

A year with the #write365 project on Google+ left me burnt out and it took another year to feel ready to return, but on new terms. It was a fresh start with a minimal theme ¹ and all older posts moved to an archive.

I slipped back into some old (bad) habits. It took another year for me to really adjust, allow myself to be more casual and imperfect.

Stumbling (in the dark) across the Kickstarter for micro.blog was like a lightbulb turning on, showing me where I could take things. Ironically, it looked a lot like when I started back in 2003.

It just goes to show that change doesn't always equal progress.

Six years ago, I wrote about how my blogging journey was like a series of books:

• Book 1 (parts 1 & 2) are all about social media - it's two parts because I took a hiatus

• Book 2 - ownership, control and the indieweb, setting the scene

• Book 3 - more personal, more about me and how I'm dealing with things

I find it strange that I've never really returned to this metaphor, especially with all the changes that have happened.

Book 3, as I called it, continued for some of 2020, but I was becoming increasingly frustrated with WordPress. The writing was on the wall when I started creating a "custom layer" that let me do everything ¹ without having to use wp-admin.

As I have said on many occasions, that layer formed the basis of the current blog. The nudge I needed was discovering how easy it was to make a "non-WordPress" page that looked virtually identical, with cleaner code, and pulled posts from the database.

The custom layer probably forms the later chapters of Book 3 (rather than being a book in its own right) with the new site being an immediate sequel.

While my own CMS was driven by similar goals and ideals, the switch-over was a defining moment in my journey.

Book 4 has already gone through many chapters over the past five years. There have been twists and turns in the plot, threads that were never explored, storylines seeming to go back to a previous point in time — maybe we can call them flashbacks.

It's obvious, however, that, as all good series progress, the latest installment builds on those before. Ideas that began in earlier volumes start to pay off as the story continues.

I don't know how many more books there will be, after all, I don't know how the story ends. Do I have it in me to take a different approach or do something truly distinct? I've toyed with other concepts, but they've really just been a new coat of paint.

I'm just making it up as I go along.

Still got a headache and fuzzy after yesterday's migraine and didn't sleep well. Our grandson has a temperature and I woke in a sweat this morning so did a precautionary Covid test. It's negative 👍

# As the daily feed is now the only feed for the site I decided to remove the "no posts on this day" feature. It will now only include days with posts.

Dave Winer has added the <source:self> channel level element to the source namespace as a way of indicating the canonical URL of the feed without having to include the Atom namespace. I've also added this to the daily feed.

# In her post Social norms of the indieweb, Tracy discusses how communities and their norms cross platform boundaries and the subsequent risk of context collapse.

Even within the indieweb (where a goal is to allow sites to communicate with each other) opinions differ on what information sound be displayed after receiving a webmention.

When I stepped back from the blog recently I decided to undo the changes that allowed me to cross post to Bluesky and back feed replies. I also removed my RSS feed from my micro.blog account so that posts wouldn't show in the timeline there. As micro.blog supports webementions replies would show as comments on the blog.

Returning to blogging again, I have decided that this is how I'll keep it. While I am active on Bluesky (albeit fairly minimally) I have decided to keep that separate. Although Bluesky is designed as a decentralised system, and as the network expands and more "people" run their own servers or compatible systems, the places posts spread will obviously increase beyond the Bluesky app. That's a given but, even though my bio says that replies could show as comments on my blog, many won't be expecting it.

I feel that intentionally sending a webmention from your own site to another is a different matter. It is understood that webmentions allow this cross talk between sites so I don't feel the need to hide the names, links or reply content of those sites sending them. Micro.blog is an exception as, while it supports webementions, not everyone who sets up an account might understand what they are and how they work.

Some prefer to only publicly show the number of mentions or likes, keeping the details for their private consumption. The webmention spec details how they work but not what one should do upon receiving them. A social norm needs to develop here so that those implementing them have a reasonable expectation as to their handling. If that norm becomes not publicly showing any details then I'll be happy to follow it.

I've been saying for years that indieweb technology needs to be simpler to implement. That's indieweb - one word.

A number of people have linked to a post by Giles Turnbull saying Let's make the indie web easier. That's indie web — two words.

There is overlap and the same principles apply. However, the indieweb is part of the independent web but not all of the independent web is indieweb.

The indieweb website states it is a people-focused alternative to the "corporate web" and:

is a community of independent and personal websites connected by simple standards, based on the principles of: owning your domain and using it as your primary identity, publishing on your own site (optionally syndicating elsewhere), and owning your data.

Great, isn't that the same thing? Well, not exactly no. To be 'IndieWeb' (one word) means to follow the basic principles of the independent web (and more) but ideally using indieweb standards as listed on the site: ¹

• IndieAuth
• Webmention
• Micropub
• WebSub
• Microsub

My previous posts have been about how difficult it is for people to implement these standards.

Giles goes way further by saying that it needs to be easier to set up a website period. Never mind the fancy bells and whistles, never mind the nerdy standards and protocols, just setting up a simple self-hosted site is … too. fucking. hard.

We need more self-hosted platforms for personal publishing that aren’t Wordpress …

Why not build static website generators that people can just unzip, upload to the shared hosting they've just paid for, and start using via a browser?

Jeremy Herve goes even further saying that it's not just the tools that are the problem:

The minute we're talking about "unzip", "upload", we’ve already lost folks.

He argues that the onboarding flow matters more than the tools. There absolutely needs to be more platforms that provide a simple 'point and click' style of installation and more hosting providers that support such installations. That over 40% of the web runs on WordPress tells a worrying story.

Update Kev has some thoughts and is collecting links to others as well.

James let me know that yesterday's post on RSS had hit the front page of Hacker News.

I was expecting a lot more push back on my statement that "people inventing new 'standards' are deliberately trying to make them hard" but (so far) only one comment mentions it:

This blog post feels like a subtweet, making accusations of malice about... something? ActivityPub? If it flat out said what it was talking about, we could have a more productive discussion.

I've long said that various standards are too complicated for the average person. ActivityPub is certainly beyond me and a lot of indieweb building blocks are also out of reach.

Things don't have to be so complicated.