[_wmd]

Pretty unfortunate design flaw, and hopefully it can be corrected, but this post is horribly finger-pointy and the tone generally stinks. Would hate to make a mistake while working around this guy

Can the link be changed to the apparent source article? It's not as bad. https://www.scorpia.co.uk/2019/06/28/pi4-not-working-with-so...

[arghwhat]

It's a bit finger-pointy, but it is also a pretty darn stupid mistake.

That USB-C is complicated is all the more reason for just taking the pre-provided mandatory design for things, and to test with at least a few cables. Having worked in hardware before, this seems very careless.

[koala_man]

>and to test with at least a few cables

I'm sure they did, but I can easily see beta testers missing this. I have a rat's nest of no-name USB cables, and if I pulled out one that didn't power a known working device from a known working adapter, I'd probably just chuck it in the bin.

Someone trustworthy should enumerate and tag all the possible USB-C cable configurations and sell them as a bundle for testing purposes.

[MobileVet]

Seriously. Considering what a cluster the spec is and the budgetary goals for a project like the Pi, this feels really harsh. The focus is on the mistake and how trivial it was without any thought of why or how.

Pointing out mistakes can be done without being such a jerk about it.

[joelhaasnoot]

They spend months if not years on the design and crank out millions. Getting USB-C right seems like a no-brainer!

[MobileVet]

I highly doubt they spent years on the design... But regardless we all make mistakes.

What is a no brainer to you may not be to everyone.

Besides, you are fundamentally assuming this was a mistake and not a deliberate design decision. Maybe they prioritized X over following the spec 100%. I may not make that decision, but that doesn’t mean they didn’t.

[arghwhat]

This is exactly why the presence of this mistake appears very careless.

When you are met with a complicated spec with known issues, do you test more or less? I know which I'd pick, and if they had tested common cables like those from Google or Apple (I'm sure someone has a macbook), they'd see it not work.

[pas]

The spec might be bad but then why deviate from The Reference implementation without thorough review and testing!?

[sctb]

Thanks! We've updated the link from https://medium.com/@leung.benson/how-to-design-a-proper-usb-....

[snvzz]

Yet there was nothing wrong with the old link.

[sctb]

> Please submit the original source. If a post reports on something found on another site, submit the latter.

https://news.ycombinator.com/newsguidelines.html

[snvzz]

Was the whole medium article really pointless regurgitation?