Profile cover photo
Profile photo
Christian Werner
3D printing, projector building, 28 hour day Universalist
3D printing, projector building, 28 hour day Universalist

Christian's posts

Post has attachment

Post has attachment

Post has attachment

Post has attachment

Post has shared content
Seems like valid logic.
"Fuck the Community" - working in deep infrastructure

Work in infrastructure is a strange thing. "Hey, I flipped a switch and the light actually went on" said nobody ever. But flip a switch and it stays dark only once, and it's the fucking GE again, which is ruining the country and the reason why we can't have nice things.

Infrastructure is invisible. When infrastructure people are doing their job right, nobody will thank them, ever, because they do not exist.

The is also true for the Linux kernel. It is a given, invisibly at work on your phone, on the switches that carry your packets from your phone to your provider, and on the servers that your phone is talking to. But the Linux kernel is only talked about when xyz-FS has eaten your draft, or your laptop fails to wakeup from deep sleep when you open the lid.

So while infrastructure developers are using the same tools that other people are using - languages, test frameworks, DCVS'es and so on, the operate under completely different success metrics, and consequently under a completely different value system than feature developers.

Feature developers are doing just that. They build new features = new best cases. Infrastructure developer are concerned about failure and avoiding breakdown = worst cases.

If you show up on the LKML with a patch containing shiny new code = new best case, they will recognise you as an alien from a different universe. Because as long as more code = more bugs at a constant bugs/LoC-ratio, more code is not an asset - under a failure metric it is a liability. You need to convince people why adding your code to the project is actually making life easier by avoiding future failure or improving recovery, or avoiding future irrelevance, because you are talking to professional pessimists.

You also don't want feature persons in an infrastructure environment, because they look at things from the wrong angle. So as long as your project is staffed sufficiently, you do not want these people on board. Educating them, and teaching them professional pessimism could be valuable if you needed them, but as long as you don't it's better to not integrate them. Just keep those that stay around anyway, because they have the right mindset that enables them to contribute even if they are invisible and nobody will ever thank them their work. Because that's what infrastructure is like.

There are many of these people around you. You never notice them. They are not only making your kernel and the libraries everybody links. They are also providing your internet, your power and gas, and they also edit your Wikipedia.

They don't want your contribution and they don't want you, because you are not helping. Unless you do, in which case you are like them, and fuck the community.

Post has attachment
How US politics work,
explained by the best conspiracy theorist ever in 1985.
Call me a cynic, but I think even 30 years later his views still hold true.

Post has attachment
The linear bearings for my next pet project have arrived.
This will become a low-cost CoreXY 3D printer. 
If my reasoning proves correct, it should be easy to scale, simple to build (no extrusion profiles), and quite quick.
3 Photos - View album

Post has shared content
Finally got around to redesigning my froggy doll - this the test print. He now has movable eyes and jointed fingers. He can also stand by himself.
Now to print him in better filament and write up assembly documentation. Then I'll publish him.
4 Photos - View album

I'm planning to build a CoreXY printer, and I'm having problems finding the 'perfect' low cost printable extruder. Can you find me help me find the one that matches the following criteria?
-Bowden: to keep the weight of the print head low
-Geared: to apply enough pressure for smaller nozzles even with small or TMC2100 powered steppers
-Ergonomic: quick filament release is a big wish
-Economical: 3d printable if possible, not too exotic parts, vanilla NEMA 17, ideally Open Source 

It's supposed to feed a E3D v6 1.75mm 24V hotend (which comes with 4mm PTFE tubing).
What would be your take?

Post has attachment
Is your 3D printer too loud?
I've replaced the stepper drivers of my Ultimaker Original 3D printer with TMC2100 SilentStepSticks, making it much more silent.
It's clearly one of the greatest upgrades you can give your printer, and they are not even expensive (at €10 per piece).

What's the difference? Hear for yourself:

Installing them took some time, but they've been working flawlessly  ever since (about two months now). I totally recommend them, check them out:
Wait while more posts are being loaded