McFilter

Interesting Things on the Internet: April 17th 2023 Edition

• Smoke Screen. An essay by Mandy Brown on the narrative power of AI or machine learning.
• The age-appropriate rubber dinghies of the sunlit uplands. The Online Safety Bill is a mess and already being abused by this Government. It should be thrown out.
• On Snowballs, Napoleons, and sharks. This matches my experiences. I've never understood the "business networking" events, or the multi-day huge "business" conferences that we seem to have annually in Liverpool, where the folk (tangentially) involved in putting them on exhort me to attend because somehow spending a few days alongside a bunch of other folk trying to sell me things I'm not interested in will somehow help me sell them things they aren't interested in. I similarly lament the loss of the late 2000s tech meetup/unconference scene; and at some point I'll start doing something about it.

And via Alex, a great video of bike culture in London. We often gain groups of scallies on mountain bikes when we're out Joyriding. Maybe we need to start running some video-editing or bike-pimping workshops round here too...

Interesting Things on the Internet: April 10th 2023 Edition

• Unpacking the suitcase words. So many empty words being (ab)used to hide and obfuscate technology.
• Hacking the Cis-tem. An excellent essay from Mar Hicks on how the dawn of computer bureaucracy also ushered in more trouble for trans folk. And a lot of that was only because the state had different rules for men and women, which shouldn't have been there in the first place (and we can and should fix!)
• How Mondragón Became The World's Largest Co-op. I see, in the age-old tradition of Silicon Blah, that Merseydragón isn't too much of a stretch ;-)
• Statement from the listed authors of Stochastic Parrots on the “AI pause” letter.
• How Paris Kicked Out the Cars. More cities should be doing this, and investing more in public transport.

It Was Twenty Years Ago Today

It was twenty years ago today... I didn't teach any bands to play, but I did start writing this blog.

I didn't have any real plans for it then, grand or otherwise; it was basically a form of commonplace book—somewhere for me to make notes of things. The fact that it was public wasn't a big consideration. I'd had an RSS reader and had been reading plenty of other blogs for a while by then, but I chose it more as a convenient readymade that I could bend to the job of taking notes. And I might as well make it public, on the off-chance that I write something useful for someone else.

I don't know how useful it's been for others, but it has been extremely good for me. For a few years it did at least provide the answer for folk searching for a basic sponge cake recipe; when Google was a fan of blogs, over half the traffic to my site was to that page. I doubt anyone gets pointed there now, in the age of content-farms and ads, but I never paid much attention to site stats and removed the Google Analytics code a while back to stop tracking visitors on their behalf.

I can't quantify how useful blogging has been, but writing things down is a great way to help me work out what I think about something. It's useful, as mentioned, to have somewhere to note down recipes and the like. It's a handy way to write things once and then just send folk a link the next time it comes up. It's good for putting things into the public record, noting down what I think, rather than what anyone in power thinks, even if that's just the tiniest nudge to our collective culture. And without the practice writing however many blog posts, I wouldn't ever have agreed to write a book. (Hmm, seems I never got round to writing about the book on here. I was pretty burnt out of writing by the time it was finished!)

Lots of the posts aren't worth revisiting, but I'm often pleasantly surprised when I end up re-reading something from years back. Usually that happens when I'm searching for a link to something I've written in the past. This is far, far from a considered list—I'm amazed that I've actually managed to write this on time, usually I'd forget the date and realise a few weeks late!—but here are a few selected posts from over the years:

• A pair of posts from my 2004 trip to Liverpool. Noteworthy because this was when I decided to move back up North.
• Remember, They Drive On the Wrong Side of the Road Over There. Although before I managed to make it back to the North West of England, I spent a year in the North West of Italy.
• What is the Point of Liverpool? was my talk from Barcamp Liverpool 2011. The first Barcamp Liverpool, in 2008, was more important as an event, as that's where what became DoES Liverpool was born; but this was the better talk.
• Moderately Messed Up. Not all of the blogging is good news, but it's important to share the downs as well as the ups. Everything is good right now (in case anyone wonders).
• There are also a bunch of posts marking anniversaries of my sister's death, but I'll Do My Crying Here Again is a better stand-alone post.
• There are a load of Blog All Dog-eared Pages book reviews, such as this one on the Lucas Plan.
• And also a fair few on the culture of and thoughts around innovation or regeneration and the follies of the usual suspects' attempts.

Anyway, read more blogs. Get an RSS reader (I still use Thunderbird, because that's still where all my email goes and it's nice to be able to manage blog posts in the same way). And it's never too late to start writing one yourself.

Interesting Things on the Internet: March 27th 2023 Edition

• Degrowth: A Call for Radical Abundance. What do we want? A greater commons!
• Detached from reality. What’s wrong with the way we design the welfare system. Fascinating insiders view of how we should be fixing the benefits system.
• The Exhibition of People's Technology, 1972. A really interesting reflection on The Exhibition of People's Technology, 1972. Packed full of nuggets of wisdom over what the Alternative Technology folk got wrong back then, or what they tried that sort-of worked and why it wasn't a good answer. It should be required reading for modern makers, particularly those looking to making to help with climate and/or equitable society.
• Disclosure. This is a great idea. It should be illegal to impersonate humans without notifying them.

Thinking about AI

Tim Bray has a good, balanced post about AI and large-language-models (LLMs). He seems a bit more neutral on it than I am, but I agree with what he says.

Elsewhere, I've been disappointed to see a bunch of people whose opinions I respect getting excited and cheerleading for it uncritically. They're (I guess, almost by definition, given they're folk who've done well in tech) all comfortable, (mostly) white men. They're not the techbros; they're people I'd expect to have better critical faculties and to do a better job of tempering their enthusiasm with potential risks. They've lost some of my respect as a result. (Just for clarity, Tim Bray isn't in this group)

Their interest shows that there's some utility in the technology, unlike blockchain. If anything that makes me more, rather than less, nervous about it. I don't have a clear view of my concerns, and maybe they'll be mitigated, but I think it's useful to share them and try to work out my thinking in public.

Who owns the tech? The tech world has benefitted massively over the past decade or two from open source greatly reducing the barriers to innovation and development. At present the AI tech is locked behind APIs and privately owned. There do seem to be moves to opening that up though, so I'm least worried about this. Opening up access will, however, risk removing what little guard-rails are in place, which will no doubt make my next concern even worse.

Companies will use the tech to drive down costs, which will drive down the acceptable quality floor. Think about how annoying interactions with call centres are at present; it's about to get more annoying and soul-destroying. Look at how search results are pages of poorly-researched content farm output, rather than the expert's blog post they're all based on; that blog post will be further buried under AI-generated sites confidently telling you things you'll have to hope are correct.

And this is my biggest concern. We've already got a huge problem with the output of computers and algorithms being taken as some sort of objective truth. The Post Office scandal shows how much suffering this can cause; even driving people to suicide. Over-confident AI will be deployed by people who don't understand it, or don't care about the risks because (they hope) they won't be affected by the downsides.

Interesting Things on the Internet: March 6th 2023 Edition

• Those meddling kids! The Reverse Scooby-Doo theory of tech innovation comes with the excuses baked in. "he’s doing this as part of a political project to stymie regulators and public institutions so the tech sector can get back into the habit of moving fast and breaking things. (It’s 2023. They have broken enough already.)"
• What are consultancy companies doing to our governments? The end-game of appearance-over-substance and avoiding taking responsibility and covering your arse.

Interesting Things on the Internet: February 27th 2023 Edition

• Smart technologies for disciplining the poor. "Prepayment meters don’t protect customers at all. They protect suppliers and humiliate customers." All of this. The gas meter in my flat is a pre-pay one. I wish I'd kept the leaflet that I got at the start of using it—it was full of patronizing copy explaining how it made my life better and was for my benefit. People with well-paid jobs sat around and wrote that copy. It's written to make them feel better, not the customer who reads it. Thankfully I don't need to worry about how to pay for it, but it's still no end of annoyance: I'll wake up to a cold flat because it's run out; I have to leave the flat and go into the basement (where the meter is) to turn on the "emergency credit"; it can only be topped up at a handful of places, which require a special trip as they're not particularly convenient; you have to top-up in cash, you can't pay by card; the maximum amount you can top-up is £99 (it's gone up since the cost-of-living crisis, it was £49 before then), not £100... It's full of things like that, seemingly designed for the customer's inconvenience.
• Britain is screwed. "On most measures, the [UK] has the most limited welfare state of any developed country, including the United States"
• Don't believe ChatGPT - we do NOT offer a "phone lookup" service. You best hope that you don't end up the target of ChatGPT's plausible bullshit. Sigh.
• How Clean is Hydrogen, Actually? Interesting discussion about the challenges of using hydrogen as a fuel.
• Gas industry paid lobbyists £200,000 to get MPs’ support for ‘blue hydrogen’. The MPs and areas mentioned in this article are around Teeside, but we have similar large petrochemical plants and plans here in the North-West. We need alternative employment options, to let the workforces transition as well as our energy sources.
• BBC Radio 4 - Seriously…, The Privatisation of British Gas. Not related to the last two links. As Denise notes: "Tell Sid he already owned British Gas." Late on in the podcast they note that big, nationalised industry was good at the start and drifted into bureaucracy; and that privatisation shook things up but then suffered from the same state. They wonder if that's inevitable and just a cycle that will repeat. How about we try finding a way to keep the energy companies smaller and in public ownership? How about we acknowledge the tendency towards stasis and try to design a system that allows for change?

How to Respond to Tech Layoffs

Reading through Ed Zitron's Tech's Elite Hates Labor, I was reminded of the Lucas Combine's battles over job losses with defence firms in the 1970s.

Faced with factory closures and downsizing of the workforce, the workers themselves drew up plans for other technology that could be developed. They proposed things like wind turbines, fuel cells, and more—which could open new markets and provide for employment whilst also benefitting society more than tanks and bombs.

In a book written about the Lucas Plan (the name given to the proposals drawn up by the group of unions at Lucas Aerospace) we learn:

The reasons for these cutbacks have not been "technical". Rather they were a result of corporate priorities and financial objectives. This helps us to rule out at least one explanation of why negotiations did not take place over the Combine's Plan.

and further on:

Lucas Aerospace, however, is not in the habit of looking for new markets. In fact probably one of the main reasons why the company wants to stick with military aerospace is that it can then forget about "marketing", relying instead on profitable cost-plus defence department contracts.

Management asserted its power over the workers by refusing to pursue these ideas; even though in some cases there were already orders for the work on the table. There's more background on the plan in my dog-eared pages notes for the Lucas Plan book.

Of course, history doesn't repeat itself, but it rhymes. Tech workers should definitely join a union, particularly if they're going to continue working for somebody else.

There are a few bright spots of hope for ways that things could be better this time round.

One is the work that the CoTech community are doing to encourage and help tech folk set up as co-operatives.

And the other is the fact that digital tech workers tend to own the means of production. It was hard for the Lucas workers to set up their own production facilities, given the capital cost of the machinery required. I think hackspaces and makerspaces are starting to shift the balance there; but it hasn't really existed in the digital world for my entire multi-decade career, especially with the continued rise of open source. That said, there are always companies looking to reverse that, with things like Github's Copilot, or AWS's myriad of services and APIs which try to lock you into a single provider.

Still. Opt out of closed tools and languages; make careful decisions (at least considering your escape route) about platforms; and resist the siren song of easy VC money. More co-operatives, indie manufacturers and indie developers please!

Interesting Things on the Internet: February 6th 2023 Edition

• Yeah but…I don’t really have anything worth sharing. "Follow your passions, your frustrations, your intrigues, and see where they take you. Then share how you see it, in your world and in your own words. You might believe everything has been said by voices more expert than yours but, trust me, your perspective is just as valid as theirs. And, if your context and lived experience is underrepresented in what’s written about your topic, then your contribution to our collective understanding will be more valuable than most. "
• The sky is falling. Why do groups of well-meaning, nice people make such wrong (and harmful) decisions? This is excellent. And chiming with me particularly as watch the status quo of the regeneration cycle reassert itself. Not that I really expected anything else, I'm just laying down markers and continuing to play the long game. Maybe it's a gyroscope, not a cycle.
• You wise up. Are we seeing the beginning of the end for the Online Safety Bill? Let's hope so.
• Why the super rich are inevitable. Fantastic interactive exploration of economic models to show why a meritocracy isn't the best economic approach.

The Platform Cooperative Cloud

Tom Critchlow has written a great essay about The Magic of Small Databases.

We’ve built many tools for publishing to the web - but I want to make the claim that we have underdeveloped the tools and platforms for publishing collections, indexes and small databases. It’s too hard to build these kinds of experiences, too hard to maintain them and a lack of collaborative tools.

Towards the end he talks about how part of the problem isn't really a technical one:

Actually - maybe this isn’t a real technology problem. Or at least not completely a technology problem. I think Substack is a good analogy here. Substack’s innovation comes in two flavors: firstly it is a lovely UX for creating, publishing and maintaining a paid email list. But secondly it’s also creating social validation and educating people that running a paid email list is a viable business.

That reminded me of a post I wrote fifteen years ago (almost to the day) - Let a Thousand Niches Wither.

In that I was grappling with how to release a tiny web service to let authors track their book's rank on Amazon. In the end I put the effort in to turn it into a service and ran it for a few years; but it wasn't ever profitable and eventually died from API-rot.

These days I'd likely just release the single-user version as open-source code, but that doesn't help most of the potential users if they don't have the technical ability to spin up their own version.

Heroku offers a solution to that, where the developer can add a fairly simple JSON file to give Heroku the instructions on how to run a version of the code. Then non-technical users could—without too much know-how—run a version of it from their Heroku account.

That used to let you run tiny sites on the free version of Heroku, but now there's no such thing. Five dollars-a-month isn't too bad, but soon adds up. More importantly, you'd want the ability to run it on a number of different services; not be tied to a single supplier.

I wonder if the Heroku JSON file would allow the bootstrapping of open-source cloud that supports such services? Ideally there would be a platform co-op (or even better, a number of platform co-ops) which would be owned by the users and the maintainers and developers. Maybe this is what co-op cloud will become?

Let me have a one-click "run this on a co-operative cloud" button to add to my open-source projects!