Skip to main content
 

Bugs, breakthroughs and BlueSky

[Ghost]

"Last week we officially started the ActivityPub private beta and sent invitations to our first 3 publishers to start testing things out. This was the first big milestone we've been working toward for months, and it felt great to get there! šŸš€"

The Ghost team continues to build their ActivityPub integration in the open. It's really fun to see.

This update goes into the kinds of bugs you discover when you start showing your work to early adopters, and I love the joyul attitude here. I also particularly love the animated preview of the ActivityPub-aware profile viewer.

It's all coming together nicely - and it looks like it'll be one of the slickest Fediverse apps out there. I can't wait.

[Link]

· Links · Share this post

 

There's an election coming up and I can't believe we're still debating it.

Every vote counts

Heads up: this oneā€™s for American citizens. The rest of you can pass this one over, or peek at it for a shot of either schadenfreude or fear, depending on your predilictions and assumptions. Itā€™s your call.

The election, at the time of writing, is in ten days. Itā€™s on Tuesday, November 5th. If you havenā€™t made a plan to vote yet, you should do that! You might also be able to vote early, but if you canā€™t, your employer may be legally required to give you the time and space to go do it. I learned while writing this that the law doesnā€™t exist in twelve states; theyā€™re not even the ones youā€™re probably thinking of. Bananas. Anyway, Vote.gov is a great site that will give you the information you need.

There are two possible options in this election. And, to be honest with you, I canā€™t believe we are even having a conversation about it.

One of them is a convicted felon who 14 members of his prior administration, including former Chief of Staff John Kelly, call a ā€œfascistā€ who admires dictators and has praised Adolf Hitler multiple times. He seeks to mass-deport 15-20 million people by way of deploying the military against civilians and interning them in camps. In his last administration, he transformed the American judicial system, installing over two hundred judges and three Supreme Court justices who are loyal to his nationalist ideology. He will ramp up nuclear weapons proliferation, and has asked why we canā€™t use them, including against hurricanes. He is a proponent of Statesā€™ rights, a dog-whistle that speaks to a desire to avoid federally mandated desegegation, marriage equality, and reproductive rights. He has consistently demonized minority groups in increasingly-unhinged rallies that are reminiscent of a very dark era of the 20th century. He is a racistĀ fomer reality TV star who doesnā€™t pay his bills.

The other is Kamala Harris, who is running on a platform that has been described as ā€œpragmatic moderateā€. On the hard right, people complain, falsely, that sheā€™s a Marxist (oh, the humanity!); on the left, people complain about her focus on US military might and her lack of firm action around the ongoing suffering in Gaza. Voters like me would prefer a candidate who sits politically to the left of her, the very fact that any of the Cheneys, let alone the war criminal patriarch, feel comfortable standing anywhere near her makes me very uncomfortable, but she very clearly is not any of the things I just described about Donald Trump.

There are other candidates, but each of them, or submitting a blank or spoiled ballot, is, in effect, a vote for Trump.

So, look.

I do not think Biden is perfect, and he was not my preferred Democratic candidate in 2020 (that was Elizabeth Warren). For one thing, heā€™s tough on immigration in ways I donā€™t like; the number of deportations under his watch is on track to match the number in Trumpā€™s first term. (When people say Harris is soft on the border, it is not based in fact.) For another, heā€™s furthered American militarism overseas in all kinds of ways. I do not think Harris is perfect either, and there will be a lot of continuous work to do to pressure her administration to do the right thing both domestically and internationally. There is a lot to do, no matter which candidate, to undo the worst of the effects of American influence internationally. (She has actually been one of the most liberal representatives, while arguably not going far enough; both things can be true.)

But to say that the two candidacies are equivalently bad is bad-faith nonsense. One promises the same kind of American Presidency weā€™ve experienced, more or less, for better and for worse, for generations (the people calling Harris a Marxist are either idiots or out to mislead you; in my opinion we could use a great deal more European-style social democracy, which we simply arenā€™t going to get). The other is something that will take America to a darker, more authoritarian place for generations.

My ask is just this: that you take stock, decide what your values really are, and vote based on those values all the way down the ballot, from the President through to your local representatives. Iā€™m making no secret of how Iā€™m casting my vote or which values I think are important. Yours are entirely up to you.

· Posts · Share this post

 

The Washington Post says it will not endorse a candidate for president

[Manuel Roig-Franzia and Laura Wagner at The Washington Post]

"An endorsement of Harris had been drafted by Post editorial page staffers but had yet to be published, according to two people who were briefed on the sequence of events and who spoke on the condition of anonymity because they were not authorized to speak publicly. The decision to no longer publish presidential endorsements was made by The Postā€™s owner, Amazon founder Jeff Bezos, according to four people who were briefed on the decision."

What an act of absolute cowardice.

Later that same day, Donald Trump met with executives from Bezos-owned Blue Origin. Perhaps it's a coincidence, but the twin events illustrate the danger of this kind of ownership of a paper that is supposed to publish independent journalism.

Bob Woodward and Carl Bernstein's joint statement is pertinent:

ā€œWe respect the traditional independence of the editorial page, but this decision 12 days out from the 2024 presidential election ignores the Washington Postā€™s own overwhelming reportorial evidence on the threat Donald Trump poses to democracy. Under Jeff Bezosā€™s ownership, the Washington Postā€™s news operation has used its abundant resources to rigorously investigate the danger and damage a second Trump presidency could cause to the future of American democracy and that makes this decision even more surprising and disappointing, especially this late in the electoral process.ā€

[Link]

· Links · Share this post

 

The United States of Abortion Mazes

[Jan Diehm and Michelle Pera-McGhee at The Pudding]

"To illustrate how difficult it is to get abortion care, we built a maze for each state where the difficulty is calculated by the stateā€™s abortion policies."

What an incredible use of the web as a platform. These stories - even in more progressive, pro-human states like California - reveal that the process is harder and trap-filled than it should be. Of course, in anti-human states like Texas, it's significantly harder to the point of impossibility.

The Pudding is killing it. Just absolutely A-plus work for story after story. This one is a particular highlight.

[Link]

· Links · Share this post

 

"The Kids Are Too Soft"

[Anne Helen Petersen]

"The best indication of the health of an industry like journalism isnā€™t who excels there, because the answer is obvious: work robots who come from some sort of family money. To understand just how broken media is, look at who leaves the field ā€” or who dares not pursue it. Because this much I know is true: itā€™s not because theyā€™re soft."

Anne Helen Petersen makes some welcome, sharp observations about newsroom attitudes to work. In many ways, journalism is behind even tech in terms of reckoning with its own culture and having empathy for the people who push for better working conditions. The idea that they're too soft is absurd: they simply can't make ends meet and deserve to be supported at work, as everyone does.

Fundamentally, this needs to seep in - not just in practice, but in spirit:

"These media executives understand unions as a coddling mechanism, when what theyā€™re really trying to do is make the field sustainable. For the current generation of journalists, sure, but also for the journalists to come."

The advantages to producing a sustainable working environment are obvious and enormous. Inclusive, diverse environments with multiple perspectives that allow newsrooms to resonate with broader audiences aren't some kind of nice-to-have: doing this intentionally is good for business.

The system is broken. Younger entrants are showing how to fix it. Listen to them, for crying out loud. The goal is surely to speak truth to power and ensure everyone has the ability to make informed democratic decisions, not to preserve an industry as-is. Change isn't just inevitable: it's survival.

[Link]

· Links · Share this post

 

Bluesky Announces Series A to Grow Network of 13M+ Users

[Bluesky Announces Series A to Grow Network of 13M+ Users]

An important announcement from Bluesky:

"Weā€™re excited to announce that weā€™ve raised a $15 million Series A financing led by Blockchain Capital with participation from Alumni Ventures, True Ventures, SevenX, Amir Shevat of Darkmode, co-creator of Kubernetes Joe Beda, and others."

Bluesky is quick to point out that it will continue to not use blockchains or crypto, and that they will "not hyperfinancialize the social experience (through tokens, crypto trading, NFTs, etc.)".

Instead, this may be an indication that blockchain investors are interested in other forms of decentralization; Bluesky is talking about adding voluntary paths to revenue for creators, so there may be some way to make a return there. (I'd been wondering what the business model would be, in order to justify these funding rounds.)

Bluesky's CEO Jay Graber previously worked on ZCash, a cryptocurrency based on Bitcoin's codebase, so has some clout in that community, but this may have implications for other projects and companies that want to raise money. (Another investor is True Ventures, which previously heavily backed Automattic; those implications are also interesting.)

Another important note: Bluesky's had some flak in the past for not federating. But this announcement notes that there are over a thousand other personal data servers, which is a solid achievement.

[Link]

· Links · Share this post

 

The web and I

NCSA Mosaic

Mathew Ingram hasĀ posted some smart reflections inspired by Netscapeā€™s thirtieth birthday:

I donā€™t think an ancient serf seeing an illustrated manuscript for the first time in the 11th century would have been any more gobsmacked than I was at Netscape. Yes, there were things like America Online and Compuserve before that, and I had tried most of them. But I felt that they were like a childrenā€™s playground with 10-foot-high walls ā€” you couldnā€™t even see the real internet from there, let alone actually interact with it.

Thatā€™s how I felt too. I was an active CompuServe user and had connected to a bunch of the local Bulletin Board Systems by the time I touched the internet; they felt both easy to grasp and constrained.

The web and I grew up together.Ā 

Our family was friends withĀ John Rose, the proprietor of a local listings and classifieds broadsheet called Daily Information, who was a tech enthusiast on the side. Heā€™d turned the Daily Info office (a creaky Victorian house in North Oxford that smelled of photocopiers) into a part-time computer cafĆ© for the local students to use. My parents were both students at the University while I was growing up, and so Iā€™d hung out at Daily Info since I was small. We didnā€™t have much money, but because of John, I grew up around daisy-wheel typewriters, which became dedicated word-processors, which became Macs and IBM PCs.

John had become excited about the idea of BBSes (possible because heā€™d seen that I was excited about BBSes), so hired me as a fifteen year old to start one from him. We had a single line: one person at a time could dial in and look at apartments to rent or get todayā€™s movie times. Iā€™d come in after school for Ā£5 an hour and update the listings and make sure the BBS was working.

A BBS is a walled garden. You dial in, youā€™re presented with a menu (perhaps painstakingly built in ANSI characters by a teenager after school), and you can select a very small number of things to do. You might chat in a forum, upload or download a few files, or read some information. Thereā€™s no expansiveness: youā€™re logging into a limited information system thatā€™s designed for a small number of people to interact with, likely run from a single computer under a desk.

The internet, of course, is something else entirely.

While I was building text-only interfaces on the BBS computer in Daily Informationā€™s storage closet, the consumer internet was emerging. It wasnā€™t long before it entered my living room. My mother was a telecoms analyst for Kagan World Media, where she wrote a newsletter about the emerging internet, computer and cellphone industries. (Here she is quoted discussing CD-ROM penetration in Time Magazine in 1995, or in Communications International announcing the decline of the pager). Sheā€™d get to try out new tech from time to time, so we briefly got a very early version of commercial dial-up internet at home; I wowed myself with the Carnegie Mellon Coke machine and the Trojan Room coffee pot (the first IoT device and first webcam respectively). I found the internet much harder to use than BBSes, but it was clear that the possibilities were enormous. Family friends would come to our house to see it.

In that first year of running the BBS, John installed a 128kbps ISDN line at the Daily Info office. Iā€™d already played with the internet a little bit at home; here I had more time and bandwidth to try web browsers. Iā€™d been using NCSA Mosaic, an early web browser built at the University of Illinois at Urbanaā€“Champaign by student programmers Eric Bina and Marc Andreessen. When they graduated and started Netscape with Jim Clark, I eagerly downloaded every version: the one when it was still called Mosaic, before the University Illinois complained about use of the name; the version of Netscape with the boxy blue N in the top right that had a prominent role in the first Mission Impossible film; the one with the classy night sky logo.

It was a window into something entirely new. It was magic: a way for anyone to tell stories in practically any way they wanted. There was something about the slow speed which emphasized how special it was; a photograph that took a minute to download, coming into progressive focus or cascading down the screen line by line, felt like it was being delivered from half a world away. Thatā€™s been lost now that the web is instantaneous; itā€™s inarguably better now, of course, but itā€™s also easier to take it for granted.

With each Netscape release, I was also glued to every new feature that the web allowed. The HTML 2.0 release the next year introduced some major new ideas: a head and body tag, forms, inline images, a few basic styles. By the time I graduated high school, CSS had been invented, and people were beginning to add semantic details to the markup ā€” but HTML 2.0 was enough to get started with.

John bought us some web space, and we created a website for Daily Info. The BBS was still functional, but now any number of people with an internet connection could view the listings simultaneously. It was very basic ā€” this was 1995 ā€” but it was possible for someone to see the listings and pay to add their own to the site on the same day, albeit with a real human dealing with it. The PageMaker files for the paper version of the sheet were still the primary source of truth, so ads were added there first, and then extracted back into files that I could convert into HTML and upload to the server.

I realized years later that the Daily Info website was online before either Craigslist or eBay, which are usually credited as being the first web classifieds sites. It was certainly more basic (built, as it was, by a teenager in a closet), although we progressively built more interactivity through Perl scripts. That fact speaks one of the most powerful things about the web: anyone can do it. You donā€™t need permission to publish. You just need to have something to say.

My excitement about the internet at Daily Info led to us finally getting the internet at home, through Demon, an early dial-up ISP that literally connected you to the internet with a static IP whenever you dialed in. It was the first to give every customer free web space, which felt like freedom: even though Iā€™d been building at my after-school job for a while, having web space of my own meant I could do anything I wanted with it. I began to experiment with my own homepages, and narrate my life through a kind of online diary (we have a different word for that now). All the while, I continued to update the Daily Info website, which is still running today, with a very different codebase.

I thought I was going to be a writer; experimenting with the web meant that I chose to take the computer science route and learn more about building software. It radically changed the course of my life. Iā€™m still a writer at heart ā€” my love of technology stems from my desire to tell stories with it ā€” but Iā€™ve also been a developer, a startup founder, an advisor, and a CTO. So much of what Iā€™ve been able to do, the people Iā€™ve met, the things Iā€™ve experienced, the work Iā€™ve been privileged to take on, has been because of the magic of those first Netscape releases. Iā€™m grateful for all of these influences ā€” Netscape, John Rose, my mother, the permissionless experimentation that the web itself made possible. That spirit of magic and possibility is still what Iā€™m chasing, and, despite the exploitation of big tech and the corrosive nature of unequal funding and the politics and everything else, is still what I think is magical about the web.

· Posts · Share this post

 

Supernatural Detective's Field Guide

[Jon Hicks]

I love this sort of thing:

"Usborne's The Detectiveā€™s Handbook and The Guide to the Supernatural captivated my imagination in equal measure. This site is an imagined combination of those books ā€“ an engaging thematic prompt for me to overcome the quandary of 'what to draw?'. The order and frequency of new chapters will be random - the intention is to have a bit of fun with it!"

The web needs more whimsy, and this is so incredibly well-executed. (Well, maybe not incredibly - Jon Hicks's work is reliably amazing.)

Absolutely I plugged this into my feed reader so I can follow along. I'm already delighted.

[Link]

· Links · Share this post

 

It's Now Illegal to Post Fake AI-Generated Product Reviews by People Who Don't Exist

[Maggie Harrison DuprƩ at Futurism]

File this under "good, but I can't believe this wasn't already banned":

"Sweeping changes to Federal Trade Commission (FTC) guidelines aimed at cleaning up the polluted, confusing world of online product reviews went into effect on Monday, meaning the federal agency is now allowed to levy civil penalties against bad actors who knowingly post product reviews and testimonials deemed misleading to American consumers."

Regardless of the fact that they should obviously have never been allowed, fake reviews, including AI-generated reviews, are now definitively not. This also includes people who buy star ratings and followers (which, as a practice, is I think far more prevalent than we might realize).

Because this is a US law, and the internet is what it is, we can probably expect a lot of these activities to now take place overseas, on other platforms.

[Link]

· Links · Share this post

 

Inside the U.S. Government-Bought Tool That Can Track Phones at Abortion Clinics

[Joseph Cox at 404 Media]

Without needing a warrant, police can track ordinary peoples' smartphone locations - including people who travel out of state to get abortion procedures. The implications are troubling:

"ā€œWarrantless law enforcement access to digital information related to reproductive health care, including location data, threatens reproductive freedom,ā€ Ashley Emery, senior policy analyst, reproductive health and rights at the non-profit the National Partnership for Women & Families, told 404 Media. ā€œIf law enforcement can bypass court approval needed to obtain sensitive data and instead use this new surveillance tool to track pregnant people and build cases against them, the implications for abortion and pregnancy criminalization are alarming. This risk is especially salient for Black women, brown women, and low-income women, who are already over-surveilled and over-policed.ā€"

The tracking crosses states and is made possible by the cellphone networks themselves as part of what are shockingly lenient data sharing policies overall. Because of the jurisdiction, and the complicated way this data becomes available, the only surefire way to solve this problem is with a federal privacy law that protects our data.

At the very least it should need a warrant - but really, this sort of tracking shouldn't be possible at all. Without strong technical and legal protections against sharing, all our cellphones (this problem is not limited to smartphones) can be used as tracking devices to understand our whereabouts, who we're gathering with, and potentially more. We're all highly-dependent on them at this stage, but it's worth questioning whether we should be.

[Link]

· Links · Share this post

 

How elderly dementia patients are unwittingly fueling political campaigns

[Blake Ellis, Melanie Hicken, Yahya Abou-Ghazala, Audrey Ash, Kyung Lah, Anna-Maja Rappard, Casey Tolan, Lou Robinson and Byron Manley at CNN]

"More than 1,000 reports filed with government agencies and consumer advocacy groups reviewed by CNN, along with an analysis of campaign finance data and interviews with dozens of contributors and their family members, show how deceptive political fundraisers have victimized hundreds of elderly Americans and misled those battling dementia or other cognitive impairments into giving away millions of dollars ā€” far more than they ever intended."

Some of these are for Democrats, but most are for Republicans, who use an array of dark patterns including pre-selected checkboxes and misleading UI elements to convince donors to pay far more than originally intended.

The problem is most acute for elderly donors, and particularly for those with dementia, but there are plenty of other people who have been misled. It's a giant problem that stems from something everyone who's worked in tech will be familiar with: a focus on pushing success metrics up and to the right above all else.

There needs to be stronger regulation here, but of course, politicians aren't necessarily incentivized to push it. The best option would likely be for dark patterns overall to be more highly-regulated - after all, these same techniques are often used by lenders, insurance providers, subscription services, and more.

There's an even sadder story lurking here, too, which is more to do with a lack of the support and infrastructure for elder-care that these politicians should be providing:

"Forensic geriatrician Kathryn Locatell said what Richard Benjamin felt each time he received a ā€œthank youā€ message or made a donation is the same ā€œdopamine hitā€ a lot of elderly Americans are seeking. And the solicitations are crafted in a way that intentionally suck elderly donors into their web, providing ā€œa feeling of belonging to a thrilling, special club.ā€"

In other words, if these people weren't so lonely and isolated to begin with, they might be less susceptible to this and other scams. That feels like an important problem worth solving, too, and one that should be tackled universally, for every person who needs it, regardless of means. Instead, the people who claim to want to help them end up persuading them to part with sometimes tens of thousands of dollars they can't afford to spend. It's nothing short of an abuse of power.

[Link]

· Links · Share this post

 

Twelve Million Deportations

[Timothy Snyder]

Timothy Snyder on the seriousness of the Trump-Vance deportation plans and their implications:

"Such an enormous deportation will requires an army of informers. People who denounce their neighbors or coworkers will be presented as positive examples. Denunciation then becomes a culture. If you are Latino, expect to be denounced at some point, and expect special attention from a government that will demand your help to find people who are not documented. This is especially true if you are a local civic or business leader."

The proposal itself has echoes in some of the worst policies of the worst governments of the past. To see arenas of people giddily waving "mass deportations now" signs is genuinely chilling, and it's not reasonable to dismiss this as electioneering. (Even as electioneering, it establishes a despicable us-and-them division that is unabashedly fascist.)

But Timothy Snyder is right here to go a get step further and ask what the impact would be on communities. Some people will inevitably be willing collaborators; others will not want to make a scene or hurt their own community standing and will become de facto collaborators. And the effect will be to establish a new normal that will be incredibly difficult culturally to turn back from.

"The deep purpose of a mass deportation is to establish a new sort of politics, a politics of us-and-them, which means (at first) everyone else against the Latinos. In this new regime, the government just stokes the fears and encourages the denunciations, and we expect little more of it."

It's sickening to think that this is an America that some people actively, intentionally want. If they win, I genuinely don't know what happens next.

Personally, I can't wait for an election that's fought on tax policy or infrastructure or the nuances of government. Right now, here in 2024, it seems like a big ask.

[Link]

· Links · Share this post

 

Online Safety and the “Great Decentralization” – The Perils and Promises of Federated Social Media

[Samantha Lai and Yoel Roth at Tech Policy Press]

"Decentralized social media platforms offer the promise of alternative governance structures that empower consumers and rebuild social media on a foundation of trust. However, over two years after Elon Muskā€™s acquisition of Twitter sparked an exodus of users seeking new homes on the social web, federated platforms remain ill-equipped to meet the threats of abuse, harassment, coordinated manipulation, and spam that have plagued social media for years. Given the porous nature of decentralized services, these limitations will not just affect individual servers, but reverberate through the social web."

Most major decentralized and federated platforms don't have the necessary tooling "for scalable management of harmful content and conduct ā€” or even the enforcement of their own rules."

For some, of course, this is by design: the same version of "free speech" which animates Elon Musk and in effect prevents speech from anyone except for in-groups and the loud and powerful. To have truly free speech - where people from vulnerable communities can have a voice and real debate can be held without threat of violence - there must be trust and safety and moderation.

The piece rightly calls out IFTAS for the great work it's doing in this area. More must be done - which in part means convincing federated communities that these ideas are important.

Unfortunately a common attitude is that "we don't have these problems" - a common refrain when your bias makes you blind to your lack of inclusion. As many Black users found when they joined Mastodon and were asked to hide the details of their lived experiences under content warnings, or when people told them that these were American-only experiences (which, of course, they aren't), a predominantly white and male Fediverse that seeks to maintain the status quo rather than learning and growing can be quite a conservative place.

This is an important piece, and an important finding, which everyone working on decentralized tech should pay attention to.

[Link]

· Links · Share this post

 

Helping to build the open social web

A very literal illustration of a construction site

As regular readers know, I care a lot about growing the open social web: the rapidly-growing decentralized network of interoperable social platforms that includes Mastodon, Threads, Ghost, Flipboard, and many other platforms, both emerging and established. This is for a few reasons, including but not limited to:

Support for strong communities

  • Support for niche interests and diversity: Smaller, independent communities can flourish without the pressure to appeal to mass audiences, leading to richer, more diverse conversations and interactions. But these arenā€™t silos: any member from one community can easily follow someone from any other.
  • Community-driven moderation: Instead of top-down moderation, communities set their own rules and guidelines, which can lead to healthier and more relevant interactions. Community health isnā€™t subject to a single corporationā€™s policies and motivations.

Better developer experience

  • An easier way to build social apps: Shared libraries, tools and protocols let developers get started faster. And developers no longer have to worry about their social products feeling empty: every new product can plug into communities of millions of people.
  • Developer stability: Developers donā€™t need to ask anyone for permission to build on open social web protocols. Nobody will suddenly turn off the open social web and charge developers to access it: just like the web itself, itā€™s open and permissionless, forever. The result is a less risky playing field for new entrants.

Respect for users

  • Decentralized governance: Users have more control over their data, identity, and interactions, without reliance on a single corporation or platform.
  • Freedom from corporate algorithms: No algorithm-driven feeds prioritize ads or engagement-maximizing content, allowing for more authentic and community-driven interaction (and significantly less election interference, for example).
  • Data ownership and portability: Users have greater control over their data and are not at the mercy of corporate interests. The open social web has the potential to connect every social platform, allowing anyone to be in conversation. And users can move from provider to provider at any time without losing their communities.
  • Reduced surveillance: Federated systems are often less focused on advertising and surveillance-based business models, reducing targeted ads and invasive data collection.
  • A more ethical ecosystem: Itā€™s far easier for developers to build ethical apps that donā€™t hold user data hostage.

Iā€™d love to be more involved in helping it grow. Here are some ways Iā€™ve thought about doing that. As always, Iā€™d love to hear what you think.

Acting as an advocate between publishers and vendors.

Status: Iā€™m already doing this informally.

Open social web vendors like Mastodon seem to want to understand the needs of news publishers; there are already lots of advantages for news publishers who join the open social web. Thereā€™s some need for a go-between to help both groups understand each other.

Publishers need to prove that thereā€™s return on investment on getting involved in any social platform. Mastodon in particular has some analytics-hostile features, including preventing linked websites from knowing where traffic is coming from, and stripping the utm tags that audience teams use to analyze traffic. Thereā€™s also no great analytics dashboard and little integration with professional social media tools.

Meanwhile, the open social web already has a highly engaged, intelligent, action-oriented community of early adopters who care about the world around them and are willing to back news publishers they think are doing good work. Iā€™ve done work to prove this, and have found that publishers can easily get more meaningful engagement (subscriptions, donations) on the open social web than on all closed social networks combined. Thatā€™s a huge advantage.

But both groups need to collaborate ā€” and in the case of publishers, need to want to collaborate. Thereā€™s certainly work to do here.

Providing tertiary services.

Status: I built ShareOpenly, but thereā€™s much more work to do.

There are a lot of ways a service provider could add value to the open social web.

Automattic, the commercial company behind WordPress, got its start by providing anti-spam services through a tool called Akismet. Automattic itself is unfortunately not a wonderful example to point to at this moment in time, but the model stands: take an open source product and make it more useful through add-ons.

Thereā€™s absolutely the need for anti-spam and moderation services on the open social web (which are already provided by Independent Federated Trust And Safety, which is a group that deserves to be better-funded).

My tiny contribution so far is ShareOpenly, a site that provides ā€œshare to ā€¦ā€ buttons for websites that are inclusive of Mastodon and other Fediverse platforms. A few sites, like my own blog and Tedium, include ShareOpenly links on posts, and itā€™s been used to share to hundreds of Mastodon instances. (I donā€™t track links shared at all, so donā€™t have stats about that.) But, of course, it could be a lot bigger.

I think thereā€™s potential in anti-spam services in particular: unlike trust and safety, they can largely be automated, and thereā€™s a proven model with Akismet.

Rebuilding Known to support the Fediverse ā€” or contributing to an existing Fediverse platform.

Status: I just need more time.

My publishing platform Known could be rewritten to have a new, faster, cleaner architecture that is Fediverse-first.

Itā€™s not clear to me what the sustainability model is here: how can I make sure I continue to have the time and resources to work on it? But I do think thereā€™s a lot of potential for it to be useful ā€” particularly for individual bloggers and smaller publishers ā€” once it was built.

And of course, there are many other open source Fediverse platforms (like Mastodon) that always need extra hands. The question remains: how can I find the time and resources to be able to make those contributions?

(Iā€™ve already tried: funding as a startup, consultancy services, donations, and a paid hosting service. If youā€™ve got other ideas, Iā€™d love to hear them!)

An API engine for the Fediverse

Status: idea only, but validated with both experts and potential customers. Would need to be funded.

ActivityPub, the underlying protocol underneath the Fediverse, can sometimes be hard to implement. Unlike many web apps, you often need to set up asynchronous queues and process data in potentially expensive ways when both publishing and reading data from other instances.

So why not abstract all of that away? Here smaller communities and experimental developers can rely on shared infrastructure that handles inboxes and queues automatically behind a simple RESTful API with SDKs in every modern language. Rather than have to build out all that infrastructure to begin with, developers can start with the Fediverse API, saving them a bunch of time and allowing them to focus on their unique idea.

It would start out with a free tier, allowing experimentation, and then scale up to affordable, use-based billing.

Add-on services could provide the aforementioned anti-spam, and there could be plugins from services like IFTAS in order to provide real human moderation for communities that need it.

Suddenly, developers can build a fully Fediverse-compatible app in an afternoon instead of in weeks or months, and know that they donā€™t need to be responsible for maintaining its underlying ActivityPub infrastructure.

A professional open social network (Fediverse VIP)

Status: idea only, but validated with domain experts.

A first-class social network with top-tier UX and UI design, particularly around onboarding and discovery, built explicitly to be part of the Fediverse. The aim is to be the destination for anyone who wants to join the Fediverse for professional purposes ā€” or if they simply donā€™t know what other instance to join.

There is full active moderation and trust and safety for all users. Videos are supported out of the box. Images all receive automatic alt text generation by default (or you can specify your own). There is a first-class app across all mobile platforms, and live search for events, TV shows, sports, and so on. Posts can easily be embedded on third-party sites.

You can break out long-form posts from shorter posts, allowing you to read stories from Ghost and other platforms that publish long-form text to the Fediverse.

If publishers and brands join Fediverse VIP, profiles of their employees can be fully branded and be associated with their domains. A paid tier offers full analytics (in contrast in particular to Mastodon, which offers almost none) and scheduled posts, as well as advanced trust and safety features for journalists and other users from sensitive organizations. Publishers can opt to syndicate full-content feeds into the Fediverse. This becomes the best, safest, most feature-supported and brand-safe way for publishers to share with the hundreds of millions of Fediverse users.

Finally, an enterprise concierge tier allows Fediverse VIP to be deeply customized and integrated with any website or tool, for example to run Fediverse-aware experiments on their own sites, do data research (free for accredited academic institutions and non-profit newsrooms), build new tools that work with Fediverse VIP, or use live feeds of content on TV or at other events.

What do you think?

Those are some ideas I have. But Iā€™m curious: what do you think would be most effective? Is this even an important goal?

Iā€™d love to hear what you think.

· Posts · Share this post

 

US startup charging couples to ‘screen embryos for IQ’

[Hannah Devlin, Tom Burgis, David Pegg and Jason Wilson at The Guardian]

Quite a disturbing new startup coming to light in The Guardian:

ā€œThe footage appears to show experimental genetic selection techniques being advertised to prospective parents. A Heliospect employee, who has been helping the company recruit clients, outlined how couples could rank up to 100 embryos based on ā€œIQ and the other naughty traits that everybody wantsā€, including sex, height, risk of obesity and risk of mental illness.ā€

Eugenics is a discredited, troubling idea, and the startupā€™s claims are akin to junk science, even if the underlying data was drawn from UK Biobank, which seems like a great resource when used for good. Still, the startup is clearly out there offering its services, while using a regulatory arbitrage strategy (operating between jurisdictions to exploit legal differences and finding ways to exploit loopholes in the law) that isnā€™t a million miles away from techniques used by startups like Uber, and throwing up all kinds of ethical questions in the process.

A major figure in the startup is Jonathan Anomaly (his real name), who has been advocating for ā€œliberal eugenicsā€ for some time:

ā€œAnomaly is a well-known figure in a growing transatlantic movement that promotes development of genetic selection and enhancement tools, which he says should not be confused with coercive state-sponsored eugenics. ā€œAll we mean by [liberal eugenics] is that parents should be free and maybe even encouraged to use technology to improve their childrenā€™s prospects once itā€™s available,ā€ he told the podcast The Dissenter.ā€

Of course, eugenics isnā€™t controversial or unethical solely when itā€™s forcibly done by the government. As the article notes:

ā€œKatie Hasson, associate director of the Center for Genetics and Society, in California, said: ā€œOne of the biggest problems is that it normalises this idea of ā€˜superiorā€™ and ā€˜inferiorā€™ genetics.ā€ The rollout of such technologies, she said, ā€œreinforces the belief that inequality comes from biology rather than social causesā€.ā€

Enough ink has been spilled on science fiction stories that describe the effects of exactly this startupā€™s mission that the founders should have understood they were building a biotech torment nexus: something that was described in fiction as a technology that must never be built for the good of humanity, lest we fall victim to both intended and unintended consequences. Regardless, if someone can build it, they eventually will, and here we are.

Thereā€™s a related ethical question raised here, which related to who, exactly, should have access to biological research data. It turns out that UK Biobank also gave access to its database to a race science group. Should it have? Or should there be ethical safeguards on these databases? Iā€™m more inclined to say that knowledge bases should be as open access as possible, but the implications for use by eugenicists and racist groups are pretty dire.

[Link]

· Links · Share this post

 

Republicans, young adults trust news on social media about as much as news from national outlets

[Kirsten Eddy at Pew Research Center]

The lede is a little buried here behind some pretty shocking top-line stats:

"Today, 37% of Republicans and Republican-leaning independents say they have a lot of or some trust in the information that comes from social media sites. This is nearly on par with the 40% of Republicans who express this level of trust in national news organizations."

"[...] Adults under 30 are now nearly as likely to have a lot of or some trust in the information that comes from social media sites (52%) as from national news organizations (56%)."

Okay, but what's fascinating is that both groups trust local news outlets a great deal more. These have been systemically underfunded and are often run on a shoestring, but there's something about the local voice that really matters.

My suspicion - which is really just a hunch, so take it with a pinch of salt - is that it's because local news outlets don't tend to deal as much with abstract partisan politics. They're not going to comment on what Trump said now, or perceived shortcomings in the Harris campaign.

But, of course, local politics really matters. So it's interesting to think about what might happen if there's more investment in the space - something that initiatives like Tiny News Collective, the American Journalism Project and The Lenfest Institute are already thinking hard about. We need diverse, mission-driven outlets like Open Vallejo and Mission Local to spring up across the country.

My question as a technologist is how platforms, and more pointedly, open protocols can support these newsrooms. How can technology help great local journalists find the reach and make the impact they need, on their terms? And how can journalists, technologists, and product thinkers work together to shine a light on local politics and improve life in communities across the country?

[Link]

· Links · Share this post

 

You should be using an RSS reader

[Cory Doctorow]

Cory Doctorow discusses how he reads writers like Molly White:

"This conduit is anti-lock-in, it works for nearly the whole internet. It is surveillance-resistant, far more accessible than the web or any mobile app interface. It is my secret super-power."

I agree. I start every day in my RSS reader (I maintain a very simple live list of my subscriptions over here) and it's one of the best tools I use. I rarely miss a news story from a publisher I care about - whether that's a newsroom, an individual, or an organization. And nobody's getting in the way to try and predict what I should be interested in.

RSS is free, open, well-established, and easy to use. More people should be using it. Even you.

[Link]

· Links · Share this post

 

Far-Right Extremists Embrace Environmentalism to Justify Violent Anti-Immigrant Beliefs

[Abrahm Lustgarten at ProPublica]

"For a generation, conservatives ā€” not just the far right, which Crusius appeared to identify with ā€” had propelled the notion that climate change was a hoax fabricated so the government could impose new restrictions on the economy and society. Yet Crusius hadnā€™t denied climate change at all. Instead, he seemed to claim its impacts were themselves arguments justifying his violence."

Abrahm Lustgarten has written a sobering piece about the far right's embrace of climate change as a root for anti-immigrant and eugenicist sentiments. We can see this playing out among conservative groups across the country: in a world where resources are becoming more scarce, preserving "white European ideals and beliefs" becomes a part of "preserving the purity of [America's] ecology".

Ecofascism has been with us for a long time, and unfortunately has long been a subset of climate movements. But as the planet heats up and climate discussions become less hypothetical and more immediate, these conversations are becoming louder, too.

[Link]

· Links · Share this post

 

My solar-powered and self-hosted website

[Dries Buytaert]

"I'm excited to share an experiment I've been working on: a solar-powered, self-hosted website running on a Raspberry Pi."

Lovely!

The key seems to be a Voltaic 50-watt panel and 18 amp-hour battery, which run to around $300 in total. That's not a lot of money for something that can theoretically run in perpetuity.

I've been wanting to make my own website run on fully green energy for a long time, and it's hard to find a web host that does this directly rather than through trading carbon credits, which I'm deeply suspicious of. (The exception is Iceland, where geothermal energy is common.)

I wonder what it would take to productize something like this and make it an all-in-one home server solution? Or to put your wifi router and modem on solar? (Assuming your whole house isn't on solar, that is, which mine sadly isn't.)

This also seems fair:

"It may seem unconventional, but I believe it's worth considering: many websites, mine included, aren't mission-critical. The world won't end if they occasionally go offline. That is why I like the idea of hosting my 10,000 photos on a solar-powered Raspberry Pi."

I feel the same way.

[Link]

· Links · Share this post

 

Underrepresented journalists most impacted by layoffs, says new report

[James Salanga at The Objective]

"Layoffs in journalism since 2022 have disproportionately impacted people of marginalized genders and people of color, according to a new report from the Institute of Independent Journalists (IIJ). It collects data from a survey with 176 journalist respondents who had undergone a layoff or buyout since 2022."

This mirrors the impact of layoffs in tech, and likely other industries. In 2023, Prism reported that:

"Recent surveys have found that women, people of color, disabled workers, and other marginalized groups are disproportionately affected by mass layoffs in tech despite being underrepresented in the industry. According to Harvard Business Review, companies rely heavily on position and tenure when deciding on cuts, which translates to wiping out ā€œmost or all of the gains theyā€™ve made in diversity.ā€"

This is damning in itself, but also suggests that many diversity gains were in positions closer to entry level than management level.

The irony for journalism is that it's the diverse members of newsrooms who can help them find broader audiences by ensuring that diverse perspectives are represented both in coverage and in management decisions. For a declining industry, it's a self-sabotaging thing to do. But, again, it says a lot about the demographics of the people who make the decisions.

[Link]

· Links · Share this post

 

I joined Dot Social for a conversation about the future of media

Ben Werdmuller and Jason Koebler on Dot Social

I was lucky enough to sit down withĀ Mike McCue, CEO at Flipboard, and 404 Media co-founder (and former Motherboard Editor-in-Chief) Jason Koebler to talk about the future of media and its intersection with the future of the social web.

Savvy journalists at forward-thinking newsrooms are not letting this happen to them. Instead, theyā€™re doing the work that arguably has been most critical all along: building direct connections with their audiences. Itā€™s common to do this through email lists and subscription models, but the open social web offers a new, more equitable ecosystem for quality journalism to thrive.

Two people on the frontlines of this movement are Jason Koebler, a journalist and co-founder at 404 Media, and Ben Werdmuller, the senior director of technology at ProPublica. In this episode of Dot Social, the two talk about their fediverse experiences so far and why theyā€™re hopeful for publishing in the future.

I loved being a part of this conversation. You can watch / listen over here.

· Posts · Share this post

 

To be clear, I'm not in any way saying that they should - just that I can see it as a possibility!

· Replies · Share this post

 

Is Matt Mullenweg defending WordPress or sabotaging it?

[Mathew Ingram]

Mathew Ingram's overview of the WordPress drama continues to be updated with new information. The hole just seems to be getting deeper and deeper. As he says: it's a mess.

"It's pretty clear that Matt sees what he is doing as protecting WordPress, and forcing a no-good corporation to cough up some dough after years of taking advantage of the community (he says he has been trying to negotiate with WP Engine for more than a year now, while WP Engine says it gives back to WordPress in a number of ways.) To some observers like me, however ā€” and to some other longtime members of the WordPress ecosystem ā€” it looks like Matt has dragged the WordPress community into a legal mess with a variety of unforeseen and potentially serious consequences."

I still don't fully understand what prompted this sea change in how Matt has addressed the wider community, including WP Engine. I have this ongoing sense that there's another shoe left to drop, whether it's relating to stalling revenue at Automattic and pressure from its board (pure conjecture on my part, to be clear), or something else entirely. Without some strong motivating factor this just seems to be self-sabotage.

At this point I'm really curious to see what's next. All this drama has also made it clear that for the kind of CMS WordPress is - more of a framework than an out-of-the-box solution at this point, but with a strong ready-made editing and administration experience - there aren't many alternatives. That's not to denegrate other projects like Drupal, etc, because I think they're different kinds of CMSes. Ghost is much more narrowly focused, too. I think if WordPress had a real competitor in its space, this might all be playing out differently.

(If I was WP Engine and had run out of options to de-escalate, I'd be going ahead and forking WordPress right now. And what a mess that would be.)

[Link]

· Links · Share this post

 

It feels like 2004 again.

[Anil Dash]

Anil Dash makes a pertinent observation about the current state of the web:

"At the start of this year, I wrote The Internet Is About To Get Weird Again, which began by calling back to the Internet of 2000. In thinking more about it, though, we more closely resemble the Internet of a few years later, where the crash of the dot-com bubble and the stock market had the same effect that the popping of the crypto bubble did: the casuals who were just trying to make a quick buck are much less likely to jump in the pool."

I agree.

The way I've been thinking about it is: There's everything to play for. We understand what can go wrong. We understand many of the needs, although we should always go out and learn more. But for the first time in a long time, the internet isn't calcified: there isn't a sense that the platforms people use are set. Anyone can come along and build something new, and it's absolutely possible for it to catch on.

And, as Anil says, the spirit of the web is more intact than it has been in a long time. Gone (hopefully) are the Wall Street-esque folks who are here to make a bunch of money; instead, we're left with the people who genuinely care about connecting and creating and making something good. That's what powered the web's heyday, and that's what has the potential to make a difference now.

Let's go make good stuff.

[Link]

· Links · Share this post

 

Revisiting Known

The original Known mockup image

I thought it would be fun to revisitĀ Known, the open source publishing platform that powers my site.

How it works

Known allows a team or community to publish news on any topic to a single, searchable stream of content thatā€™s easily accessible from any device. It is not a full CMS, and nor is it designed for independent publishers to sell subscriptions; instead, itā€™s optimized for publishing to a single feed.

Every Known site is a single feed of content that any number of users can publish to. You can have one user, as my site does; you could have thousands, if you wanted.

The stream can also be filtered by hashtag, author, or content type ā€” so you can choose to only view content on a certain topic, or only photos, or some combination thereof.

Each stream, filtered or not, is shown as a standard web page by default. These can be themed, but itā€™s also easy to view different interfaces. RSS and JSON are available for every screen you can view as a web page, and it would be easy to add low bandwidth HTML, for example. (I once added an interface type that displayed everything as a Star Wars crawl. It got old fast.)

The Known menu bar

When you log in, you get a little menu bar that lets you publish different kinds of content. Itā€™s a little bit like Tumblrā€™s bar, but here, every type of content is powered by a plugin. You can download new content types created by other people, or you can write your own. On my site Iā€™ve created a kind of blog post called an ā€œasideā€, which Iā€™ve decided to make a distinct content type.

Hit the button, and you can compose right on the page.

Known status update composer

Known supports an idea called POSSE: Publish on your Own Site, Syndicate Elsewhere. You can elect to syndicate a post to a third-party site by enabling the toggle for that site below the compose window. In this illustration I have two example webhooks, but people have written plugins for Mastodon, etc. (In the beginning, Known had plugins for Twitter, Facebook, and so on, but all those APIs locked down over time. The promo image, which you can see above, includes Foursquare and Flickr as options, which is a clue about the era it originated from.)

You can also compose using any application that supports the Micropub standard. I tend to write all my blog posts in iA Writer.

Known supports Webmention, so when you publish a post that links to a site, that site will be notified. You can even use webmention to respond to someone elseā€™s post elsewhere and have a conversation across the web.

Itā€™s free and open source, and intentionally runs on the same LAMP stack as WordPress. Be warned though; as the screenshots suggest, itā€™s now a little old.

A little history

Known was originally called Idno. (ā€œWhat does it stand for?ā€ someone once asked me. ā€œI dā€™no,ā€ I replied. This is the level of humor you can generally expect from me.)

I wrote the first version of it when my mother was recovering from a double lung transplant: she was in need of community but absolutely didnā€™t want to discuss her condition on Facebook. Iā€™d previously written Elgg, an older open source social networking platform, so I decided to think about what a social community platform might look like in the era of the mobile, ubiquitous web. What would it look like for a community to publish to a place where it could continue to own its own content, on its own domain? (It seems like a quaint exploration now, but remember that this was 2013.)

I became friends with the indieweb folks, and met Erin Richey at an IndieWebCamp. We decided to collaborate on the project. It was her idea to submit it to Matter, where we took part in the third accelerator class. Along the way, we did some focus group testing (Erinā€™s instigation) and chose Known as a permanent name.

Known at Matter Three Demo Day

It was a startup for a couple of years; there was a paid, hosted version; a Known-powered site even won an award for KQED. But it wasnā€™t the kind of thing that excited investors, and we werenā€™t making enough money for it to be sustainable. Ultimately, I allowed myself to be acquihired by Medium, which allowed us to pay Matter back, and we both settled into new jobs. The day before my first Medium paycheck, I spent my last five dollars on gas. (Erin and I welcomed our actual child ā€” a human one ā€” two years ago. So thereā€™s a coda.)

But there are still users out there, myself included, and the open source project is still alive. Itā€™s been slower over the last few years, because I havenā€™t had much time to devote to it. (The main thing Iā€™ve been looking at is a command line exporter to allow people to more easily take their content into WordPress, as well as some experiments with ActivityPub.) But it remains a core part of the operating system that powers my identity online, and the identity of others.

Lately Iā€™ve been thinking that thereā€™s a place for this model of publishing. The internal architecture needs to be overhauled; the Bootstrap-driven default template needs to go; but I think thereā€™s really something to the model of letting communities publish to a simple, queryable feed of content that syndicates out to the world.

Perhaps itā€™s finally time for Known 2, with an easy upgrade path from the original? If youā€™re intrigued by the idea ā€” or if youā€™re a Known user ā€” Iā€™d love to hear your thoughts.

· Posts · Share this post