Interoperability can be aided through enshittification too

Posted May 28, 2025 10:10 UTC (Wed) by mathstuf (subscriber, #69389)
Parent article: Cory Doctorow on how we lost the internet

> The first is interoperability; in the non-digital world, it is a lot of work to, say, ensure that any light bulb can be used with any light socket.

Ironically, we have a standard light socket because the Phoebus cartel[1] required compatible sockets for *their* enshittification enforcement mechanism: a rig that kept lights on and, if the bulb lasted too long, would trigger fines for the manufacturer.

[1] https://en.wikipedia.org/wiki/Phoebus_cartel

Interoperability can be aided through enshittification too

Posted May 28, 2025 12:21 UTC (Wed) by excors (subscriber, #95769) [Link] (5 responses)

Hmm, I looked into that and I don't think it's true. According to a book from 1927 (only 2 years after Phoebus formed), the Edison lamp base was already 70% of the (US?) market in 1900, and became pretty much universal within 5 years thanks to a marketing campaign and adapters being sold at cost: https://bulbs.2yr.net/porcelain-th-socket.php (quoting from https://archive.org/details/historyofincande00howe/page/182)

It's also *not* standard in the UK, even though Phoebus included most UK lamp manufacturers. (The 1951 report at https://www.gov.uk/government/publications/report-on-the-... has some details on the companies). The B22 bayonet fitting is the most common in the UK today, and as far as I can tell that's been the case since before the 1920s (though I can't find good evidence for that). There must have been some cooperation between lamp manufacturers to standardise on that, as with the Edison base in the US, but they evidently weren't aiming for a global standard (despite there being a lot of overlap and cooperation between the US/UK/EU manufacturers) and it was long before Phoebus.

It seems like they simply recognised that incompatible fittings were a pain for both consumers and manufacturers, and were holding back the industry, so standardising on the most common fitting within each market would help everyone.

(Bayonet isn't universal in the UK though, now that markets have become more international - it's easier for manufacturers to have one design for the whole UK+EU market, so E27 is quite widely available too, and many houses will have a mixture. And now we have expensive long-life LED smart bulbs, it's even harder to adopt a new standard, so I guess we'll be stuck in this transitional state forever.)

Interoperability can be aided through enshittification too

Posted May 28, 2025 14:20 UTC (Wed) by Wol (subscriber, #4433) [Link] (2 responses)

> (Bayonet isn't universal in the UK though, now that markets have become more international - it's easier for manufacturers to have one design for the whole UK+EU market, so E27 is quite widely available too, and many houses will have a mixture. And now we have expensive long-life LED smart bulbs, it's even harder to adopt a new standard, so I guess we'll be stuck in this transitional state forever.)

And it now gets worse because both Bayonet and Edison have their "small" equivalent so that's four ... and we have spotlights which are becoming more common which also come in two versions, GU10 and MR12 (plus, I believe, a bunch of minor players).

I'm trying to standardise my house on Edison (because I think it's better/safer than Bayonet), and GU10 (because I'm fed up with the grief of 12v transformers for MR12). More and more lamps and wall lights are ES or SES now, although it's still hard to find non-Bayonet pendant light fittings.

Cheers,
Wol

Interoperability can be aided through enshittification too

Posted May 30, 2025 9:28 UTC (Fri) by nim-nim (subscriber, #34454) [Link] (1 responses)

E27 means it’s not too hard to find a bulb with good CRI and white point choice. GU10 is second and E14 even further. If you make the mistake of buying anything with built-in led or any other kind of micro socket you’re condemned to 80 (or even 75) CRI and yellowish 2700K white.

Stable standard API/sockets trump “better” experiments that lack the reach to form a diverse ecosystem.

Interoperability can be aided through enshittification too

Posted May 30, 2025 14:26 UTC (Fri) by Wol (subscriber, #4433) [Link]

> E27 means it’s not too hard to find a bulb with good CRI and white point choice.

The number of people who don't even realise there is a white point choice (and why it's important), though ...

Our study had "cold white" bulbs, until we re-purposed it as a bedroom (at which point I was scrounging around for warm white bulbs to swap it with). Now it's pretty much just the room I use as an office that has cold white bulbs.

So I don't think most people would notice the lack of choice. But when you need it, you most definitely need it ...

Cheers,
Wol

Interoperability can be aided through enshittification too

Posted Jun 6, 2025 3:09 UTC (Fri) by zahlman (guest, #175387) [Link] (1 responses)

The cartel definitely existed, but the common narrative about conspiring to lower light bulb lifetimes is incomplete and not entirely fair. The short version is that incandescent light bulbs more or less directly trade off durability against the quality and amount of light produced; they can last more or less forever if they glow a dull orange that still lets you stare at the rest of the bulb, but even tungsten will eventually fail under the temperatures needed to make the light appear nearly pure white - and a bulb that produced the same amount of light at a lower temperature would still be orange (the concept of "colour temperature" comes directly from here) and would be much less efficient (drawing higher wattage and producing more waste heat).

As far as I'm aware, nobody ever came up with any meaningful improvements to the technology once we settled on tungsten filaments and argon (IIRC) filling the bulb; we had to move to completely different technologies (CF and then LED).

Technology Connections has a detailed video about this: https://www.youtube.com/watch?v=zb7Bs98KmnY

Interoperability can be aided through enshittification too

Posted Aug 9, 2025 23:16 UTC (Sat) by Rudd-O (guest, #61155) [Link]

Veritasium also has another video on it, also very interesting: https://www.youtube.com/watch?v=j5v8D-alAKE

Interoperability can be aided through enshittification too

Posted May 28, 2025 18:08 UTC (Wed) by Cyberax (✭ supporter ✭, #52523) [Link] (1 responses)

This is not what happened. And Technology Connections (I also support them) has a video about that, of course: https://www.youtube.com/watch?v=zb7Bs98KmnY

The TLDR; version - classic tungsten spiral incandescent lightbulbs were perfected in the early 1920-s, and the relationship between the lamp longevity and energy use became established. So the Phoebus association picked a reasonable compromise between them.

There is no magical pixie dust technology to make tungsten lamps last longer. The only way to do that is to run them at a cooler temperature, wasting more energy as invisible infrared.

Here's a nice article with numbers: https://pubs.aip.org/aip/adv/article/12/10/105116/2819829... - it's a recent one, Google Scholar isn't too great for articles from around 1920.

Interoperability can be aided through enshittification too

Posted Jun 6, 2025 13:27 UTC (Fri) by Wol (subscriber, #4433) [Link]

tl;dr in order to get brighter better light, you need to run more current which makes the wire glow hotter.

Smaller radiuses (thinner wire) glow hotter which means any defects in the wire will glow hotter than the rest.

Hotter wires evaporate tungsten, which will condense on the cooler (thicker) parts making them even thicker.

So the brighter the light, the hotter the wire, the quicker the defects grow, and the quicker it breaks.

Cheers,
Wol