Anil Dash reflects on two decades of blogging.

Some quotes that resonate:

I also do still strongly believe that someone who really has a strong point of view, and substantive insights into their area of interest, can have huge impact just by consistently blogging about that topic. It’s not currently the fashionable way to participate in social media, but the opportunity is still wide open.

Yes, maintaining a sustained online identity and presence is an opportunity, as it provides agency. The open web is an open invitation to do so, but it takes time to blog. Time that will not immediately result in dopamine triggering likes and retweets, so you will need to find the motivation for keeping up blogging elsewhere, likely within yourself. Even if you don’t have ‘substantive insights’ in your areas of interest but still consistently blog, there will be impact. I once had a client who hired me after reading my blog archives and realising from its tone and content I would bring the right attitude and outlook to the project. Over time any blog is a body of work.

If your website’s full of assholes, it’s your fault.

Platforms have always maintained they are just platforms, and not responsible for its content. That argument has been severely eroded by the platforms itself, because their adtech business models depend on engagement, and so they introduced addictive design patterns and algorithms that decide what you see, based on likelihood of sparking engagement (usually outrage, as it works so well). A platform that decides what you see in order to sell more ads, makes conscious editorial decisions, and is no longer a platform. Roads and their maintainers generally aren’t responsible for the conduct of drivers, but in this case the roads over time have been deliberately increasingly designed to make you speed, reward repeat offenders by reserving the fast lane for them, and road maintainers get paid by car repair shops based on a metric of a steady rise in car crashes so road rage gets encouraged to raise revenue. It’s why federation of very distributed nodes is important to me, it strongly reduces amplification of the things that we’ve come to loathe on the platforms. That e.g. Gab, having been deplatformed, moved to federated servers is a good thing. Now anyone can round around it as damage, and its content doesn’t get the amplification and recognition by being on general platforms, it otherwise would. I am the only one on my website. I am the only one on my Mastodon server. At that level moderation is extremely easy, while it doesn’t reduce my interactions in any way.

I’m not someone who thinks there was a “good old days”; social media has always been too exclusionary, and too dependent on systems and infrastructures that replicate the injustices of society as a whole. It is possible, though, to make new systems that are a little more equitable, and I still haven’t given up on that hope at all.

Me neither, there’s huge potential for increasing agency, especially for groups in specific contexts and around specific issues. A networked agency emerging from lowering technology and process thresholds. It means taking ourselves as the starting point, not the platform or its business model.

Bonus pic: my friend Paolo blogging, 13 years ago in June 2006.

blogging
image by Paolo Valdemarin, license CC BY-NC-ND

Blogging usually doesn’t involve a pipe, sitting outside, prosecco, or a sea view from the Ligurian coast. Blogging is totally mundane, this the exception. It might be a good addictive design pattern for blogging though 😉

Read 20 Years of Blogging: What I’ve Learned (Anil Dash)

This week marks the 20th anniversary of this blog. I thought the best way to observe the milestone, and to try to pass along some of the benefits I’ve gained from keeping a presence online all these years, would be to share some of the most important things I’ve learned since I started this site.

Do y’all understand how easy it is to make a fake tweet from a screenshot? Like by inspecting the browser and changing the text? …. I don’t trust posts I can’t search up on archives. And if you do have a link, archive it (not in an image but using an reputable archiving service).

Jacky Alciné’s words are true, so I thought I’d illustrate.

The general principle here is: if you make a statement about someone or something other than yourself or your personal opinions, you need to back it up with a link to supporting material. “X said on Twitter” needs to be linked to that tweet. Leaving googling for your source as an exercise to your readers isn’t just merely convenient to you, it is actively destructive of the web. The web is links, and they’re a key piece of information for your readers to judge if what you tweeted/said/blogged might be signal or noise. No links means it’s likely noise and it will degrade your standing as a source of signals. No links is aiding and abetting the bots, trolls and fakesters, as it allows them to hide in more noise.

Adding a screen-shot as Jacky Alciné says is not enough ‘proof’, as they can easily be altered directly in your browser. An example:

Yesterday I posted my first Tweet from my recent brain implant. It was awesome! So awesome in fact, I made a screenshot of it to preserve the moment for posterity.

In reality I posted from Indigenous (see there’s a link there!), a mobile app that provides my phone with IndieWeb reading and publishing capabilities, which I syndicated to my Twitter account (see there’s another link!). Also awesome, but much less awesome than blogging from a brain implant.

The difference between those two screenshots, getting from true to fake, is that I altered the text of the Twitter website in my browser. Every browser allows you to see a website you visit in ‘developer’ mode. It is helpful to e.g. play around with colors, to see what might work better for your site. But you can also use it to alter content. It’s all the same to your browser. See this screenshot, where I am in the process of changing ‘Indigenous’ into ‘brain implant’

But, you say, tweets might have been deleted and grabbing a screenshot is a good way of making sure I still have some proof if a tweet does get deleted. That’s true, tweets and other content do get deleted. Like self-congratulatory tweets/VK/FB messages about the downing of MH17 by separatist supporting accounts, before it became clear a regular line flight was shot out of the air, and those accounts were quickly scrubbed (See Bellingcat‘s overview). Having a screenshot is useful, but isn’t enough. If only for the reason that the originator may simply say you faked it, as it can so easily be done in a browser (see above). You still need to provide a link.

Using the Web Archive, or another archiving site, is your solution. The Web Archive has preserving as much of the web and other online content as possible as its mission. It is a trustable source. They save web pages on their own initiative, but you can submit any URL for preservation yourself and it will immediately be saved to the archive. Each archived page has its own URL as well, so you can always reference it. (Many links in Wikipedia point to the archived version of a page from the point in time it was referenced in Wikipedia for this reason).

I submitted my tweet from yesterday to the Web Archive, where it now has a web address that neither I, nor Twitter can change. This makes it acceptable proof of what I did in fact send out as a tweet yesterday.

Fully in agreement with you Eli. To provide agency to potential IndieWeb adopters, it is needed to start from purposes and what things a person wants to achieve. Build a pathway from there, and provide the building blocks. Not start from the tech specs. Myself I am frequently lost in the IndieWeb woods, even if I have some tech knowledge, and am not easily thrown off by the need to hunt down clues in obscure fora for a fix to an issue. Too often too much knowledge is assumed on all things IndieWeb if you seem to have some knowledge about a tiny part of it.

Replied to Dear IndieWeb, it may be time to start considering the user, not just the technical spec. by Eli MellenEli Mellen

I’ve been working on a series of walkthrough posts that outline how to IndieWebify a WordPress site. I presumed the initial setup would be fairly straightforward because a) I have a vague idea of what I’m doing, and b) a suite of plugins already exists. Boy-howdy, was I wrong. (ಥ﹏ಥ)
I’ve…

HTML Tag
HTML graffity tag, image by Markus Tacker, license CC-BY-ND

In developing for the web HTML is the very frontest of the front-end, and if you’re a front-end person, you do need to know your HTML. It helps keeping things simple and allows people like me to hit ‘view source’ and figure out how something is done, so I can use it on my own site. I started out writing HTML decades ago in simple text editors like notepad. I still write on my blog in text mode exclusively, never in visual or wysiwyg mode, and add a lot of my html in postings by hand (sometimes aided by keyboard shortcuts that make things easier and avoid repetition)

HTML is the web. And it is useful and powerful in its own right. Without embellishments through scripts etc.
It is in part why I like the IndieWeb, as it seeks to use HTML itself to make webpages machine readable, and to add things that take the best of the social media silos, without all the ajax stuff for instance. So that it works, because it is made of the web, on the web, for the web.

When organising the IndieWebCamp Utrecht last month I realised how little connection I still have to coders and developers for the web in my network. Many people I approached with an invitation to participate told me ‘I don’t develop much for the web really.’, they’re more into all kinds of frameworks and work on things like algorithms, machine learning and data analysis. Cool stuff I heartily agree, but ultimately it mostly ends up being shown in a browser. In HTML. So in a way it is disappointing to encounter a certain disdain here and there for HTML.

For me, I need to dive more deeply in the various ways HTML is currently used to add machine readability to web pages.

Thank you Chris for pointing out your work on your own blogroll, and how WordPress itself might be of use here.

Adding images is a nice feature. I added faces in my blogroll in 2003, because I generally subscribe to people not sources, and showing them in my blogroll was a nice way to visualise my blogging peer network, and make blogs look more like the social tools they are.


My blogroll in 2003

Bringing that back would be cool. Especially if relying on gravatars where possible.

So if I understand your postings correctly, the Links manager in WordPress also creates a separate OPML file. Now if this OPML file could e.g. be automatically loaded into a microsub server like Yarns, that would be even better. Then it would all be under the same WP roof.

I notice that the Links Manager allows categories and multiple at that, but tags next to categories would be even better. To do ‘Berlin coders into gardening posts this week’ type of searches in a reader. Having all the tags as categories would look cluttered in WP. I have little use for the defined XFN fields, I’d rather have tags that concern various facets of a blogger’s profile (tech, Drupal, infosec, parent, Barcelona, French, Arabic, rock climbing) to enable fast and detailed cross sections of my feeds. Having those tags here would presumably more easily allow me to carry them over into my reader somehow. Basically trying to figure out if WP Links manager could be the source of such data.

In terms of my ideal feedreader lots of the other features could then happen in a microsub/pub client.

One other question to explore: is there a way to bulk load links into the link manager. It is likely easier to build a spreadsheet with all relevant info for my current 200 feeds or so first. Do you add link by link by hand, Chris?

Replied to a post by Chris AldrichChris Aldrich

I’ll see you your blogroll and add in images and descriptions as well! … Perhaps what we really need is to give some love to that Link Manager in core to update it to OPML v2 and add in the rel attributes from XFN microformats to the links?