Dead Startup Toys (2021) | Hacker News Reader

Dead Startup Toys (2021)

deadstartuptoys.com

26 pointssegasaturn18 comments

Some of these are obviously stupid and deserve to be ridiculed, but I don’t think this site needs to be as hard on the “one laptop per child” one as they are. That seemed like a well-intentioned product designed to do good in the world, not everything ends up working out.

I think the assessment, harsh as it might be, is fair. Morgan Ames' excellent 2019 book on the OLPC project, "The Charisma Machine" [1], does a really good job looking at all the things that went wrong in the project. There was a lot of hubris in this project from top to bottom -- a lot of Ivy League intellectuals who believed they knew how to best teach the developing world and that somehow, this device would be the one to do it. It wasn't.

Well-intentioned or not, I think its broader impact is probably overstated in many circles (the notion that we wouldn't have sub-$300 laptops without OLPC is just silly), especially since many of the promises behind the device (the price, the crank, the way it would "reshape" education") were just untrue.

Is it a blind waste of investment looking for a problem to solve like Juicero? No. Is it a scam like Theranos? Also no. But given the poor-execution of the project, the imperialist nature of its whole raison d'être, and the negative effects its failure had on the EdTech movement as a whole, I think it is definitely worthy of critique.

[1]: https://mitpress.mit.edu/9780262537445/the-charisma-machine/

SugarOS was really interesting and still exists [1], but the hardware was always a solution looking for a problem. Sure the hardware might have been kind of neat, but it was always too expensive and obsolete and never worth the huge investment. [2]

[1] https://www.sugarlabs.org/

[2] https://cacm.acm.org/research/one-laptop-per-child-vision-vs...

The hardware served two purposes, as I see it:

1. It was obviously a "toy"/"educational tool" making it less desirable to divert from the intended use case. If you had a charity unload pallets of refurbished Thinkpads to the Global South, plenty of them won't end up in the classrooms. If you visit the local petty functionary and he has a bright green toylike machine on his desk, it's obviously been misallocated.

2. A standard reference platform provides a uniform target for third party devs. Think about developing for the C64, versus a contemporary PC which could have one of four or five different video cards and a dozen memory sizes.

I had the same reaction. The product also existed and represented some interesting developments for the time tbh.

I think their goal was interesting for the time, but its release coincided first with the whole 'netbook' craze which meant, all of a sudden, you could buy a regular laptop for $100-$200, rather than the maybe $700-$800 low-end laptop price that had prevailed before that, and then with smart phones becoming ubiquitous and dirt cheap all over the world.

One laptop per child is why there was a netbook craze!

They basically showed it was possible to get a machine at that lower price point and then capitalism and mass market manufacturing did the rest.

Is this so often the case innovation requires someone to prove what's possible and then going down the same path is much easier for those who come later.

One thing those netbooks didn't/don't do that OLPC aimed to do was mesh networking. Internet connectivity is still... spotty, at best, in a lot of countries - even in some of the so-called "developed" ones.

I watched a video¹ today about the ongoing civil war in Myanmar, and while the video focuses on the rebels' use of 3D-printed firearms, there was a smaller point about how a lot of Rohingya villages ended up entirely caught off-guard during the most recent wave of genocidal purging because news would travel too slow from village to village; as the Junta forces would descend upon one village, there was no effective way for that village to warn its neighbors.

First thing that popped into my head: "ain't this something OLPC and other mesh network attempts would've been able to address?"

A lot of mesh networking experiments, including that of OLPC, seem to have failed - but some have shown some recent success. Maybe it's time to have another go at deploying mesh networks to the masses at scale, learning from those failures and successes?

----

¹ https://www.youtube.com/watch?v=l0oXupwf2D4

Aren’t AirTags a mesh network of sorts? Or do the tags just piggy back on nearest device.

My understanding is that they just piggyback on the nearest Bluetooth-enabled Apple device, which then pings Apple's servers; non-Apple-specific equivalents like Tile work the same way (just with different host-device-level software and different centralized servers).

That said, Bluetooth mesh networks are absolutely a thing (https://en.wikipedia.org/wiki/Bluetooth_mesh_networking), and it'd be neat if low-power devices could capitalize on that.

At the time I believed the proliferation of netbooks was caused by OLPC. I don't think it was a fundamentally bad idea either.

The project identified a legitimate need for a small, durable, and cheap laptop, but failed on a number of specifics: it did worse as a not-for-profit than it would have done as a company, the design was child-centric instead of something adults could use too, and Linux on every machine was a design decision that only made sense in the flawed context of a not-for-profit.

The MSCHF message of "press [F] to pay respects to capitalism" is incredibly ironic when applied to OLPC, since it arguably failed due to not being capitalist enough and was succeeded by for-profit companies who remade a similar product for a wider and less charitable audience.

The laptop also was not called "the one laptop per child" or "the OLPC". That was the name of the project. The laptop was the XO.

The website getting this wrong makes it harder for me to take them seriously.

If nothing else, I liked the page design. The Elizabeth Holmes cherubs are a nice touch.

Everything is sold out, but best I can tell from a light skimming is that these toys have actually been produced at one time, or am I just falling for the joke?

EDIT: ah ha, they did actually produce the toys:

https://news.ycombinator.com/item?id=40980942

I don’t understand the juicero. If you are “juicing” what is basically canned fruit, in what world could that be better than blending and straining fresh fruit?

And now you know why it failed (well, that and product reviews that were rivaled only by the ones for the Humane Pin for their scathing take). I don't know how much you know about the product, or how much of the blurb you read, but VCs threw money at a company making a complicated hydraulic/electric military-grade internet-connected industrial press suitable for stamping out automobile doors. Turns out, though, you could get equal results just squeezing the bags with your hands, skipping the $800 hardware piece.

MSCHF drop from (2021)

Discussion then: https://news.ycombinator.com/item?id=27811047

This is from a MSCHF drop from 2021. I bought them all, of course, and they are some of my favorite stupid little toys ever.

Scrolling is a bit janky on Safari on iOS. The page “jumps” occasionally.

Reply on news.ycombinator.com