I’ve had something like this happen on my prusa. A bug maybe. Sliceing for 0.6mm but some somewhere somehow either in firmware or slicer it got set wong so it only extruded as 0.4 and it looked just like this. Opdated firmware and slicer profile and it worked fine again. But yeah it looks proportion wise like wrong nozzle size about the difference between 0.6 and 0.4.

I’m not sure why you are getting down voted so much instead of actually engaging with your post. I think you are right about that the disclosing is the right thing to do, but i believe that the problem people have with this whole AI hacking and disclosing is the intent behind the hacking. A white hat hacker has a clear intend to help discover and responsibly disclose the issue, maybe even for a bounty, giving an indirect permission to white hat hacking. On the other hand, AI does not have a clear intend behind the hacking. Very often it’s accidental and done irresponsibly like a “woopsie doozy, sorry we just broke in, we didn’t mean to, and we are not really sure what files were accessed but pinky swear we think we did no harm 🫣🤗” that is at least my take on this and why I think is very problematic. AI can still be used responsibly as a tool by a white hat hacker under strict supervision to discover zerodays but that is a whole other discussion on how you make sure its not going rouge.

As if windows 11 wasn’t running slow enough already!

Give me the source code or it’s not running locally on device!

Super cool. However I think they are missing a crusial optimizarion strategy. In morning rushhour. More elevators should be dedicated to only going up and never pick anyone up on the way down. And fill up a single dedicated cart for downwards traffic as much as possible to maximise utilisation.

Handheld band saw for the pros:

https://www.milwaukeetool.com/products/details/m18-fuel-dual-trigger-deep-cut-band-saw-kit/2929s-22

Fair enough. If or works for you. Colemark is also more similar to qwerty so maybe the experience is different. Tbh. I actually like qwerty for phone use because the most used lettered are spread out making words more destinct. But writing on a phone is always annoying to me.

I’d say if you are bothered to change and already working on it my best tip is to commit and learn touch typing correctly with fingeres on the home row and train the muscle memory. If you don’t, then I’d say there’s no real reason to learn Dvorak. The benefits only really shine when touch typing. BTW for symbols I also Still have to look, but they never moved so even with qwerty layout visible I can just look ans find what I need. If you are using US Dvorak layout I’d highly recommend investigating if you can find and install a layout for your native language leaving the symbols alone makes things much easier

10 finger touch typing without looking at the keyboard which is what I did, is very different from using two fat thumbs on a phone where you are forced to look (at least for me) in order to have any chance of hitting the right keys. On a PC I never look at the keys because they are qwerty anyway. I just use a software setting in my OS to change it to Dvorak so I must be able to type without looking. On a keyboard my fingers just know what to push intuitively so I only need to think of a letter or word and my fingers just types it from muscle memory. This is what I mean when I say 0% transferable skills because the same muscle memory does not work on a phone anyway.

I guess that highly depends on how the AI or algorithm is working and how it learned. If it uses rhythm and distance in time space between strokes along a word prediction enging matching rythm with words, Dvorak will throw it completely off, since the rhythm is changed completely. My guess is you’ll need to train it seperatly on dvoark for it to have any chance of working.

Basically I learned to completely touch type without looking at the keyboard so the stickers were just there for training and as they wore out over the following months I basically didn’t need them ever again. Numbers and symbols stay in place except a few where the Danish æøå are placed but I just memorized those…

I just use a standard keyboard and have a shortcut to change between normal qwerty and dvoark. Qwerty is just for whenever someone needs to borrow my PC for a while. I can still write qwerty in a pinch but it’s as slow/fast as its always been and I have to look at the keyboard while typing.

My typing was really slow on qwerty to begin with and I found it super awkward when trying to learn it. I really did try but gave up multiple times.

One day I just decided enough was enough and wanted to start from scratch. I added stickers to all my keys for dworark and used a training program (cannot remember what, but it was old school and worked great going step by step) and during my summer vacation one year I just sat in my hammock relaxing and just kept training with the program. After two weeks I was fast enough to write to friends in chats without it taking ages (still slow af.) after about a month I was faster than qwerty and have never looked back since. Dvoark is for me much more comfortable and intuitive for my fingers. Linux has exelent support for dvoark out of the box. Windows not so much for other than english.

Dont bother changing your smart phone keyboard too. It has 0% transferable skills or benefits and just so confusing.

Good luck when I use dvoark.

As i understand it Valetudo is not actually firmware to control the robot. It is a “parasite” that makes the existing robot firmware belive it is connected to the cloud which is very different from actually controlling and navigating the robot. However, in terms of homeasistant integration it could be worth getting inspired.

~~For a control operating system ROS could be something to consider. When I used it 10 years ago the project was quite unstable since tools changed constantly and it was overly complicated to work with, but a lot of development has happened since so maybe worth considering. ~~ I wrote this before reading the article

Thanks for sharing. It feels good to know you are not alone. I had a similar experience once where after lots of debugging I realized the “PLA” I was using was in fact PETG and I was extruding way too cold.

I noticed this as well. Just opening google caused my GPU usage to constantly sit at ~70 % usage even without searching anything. Is google crowd sourcing compute on peoples computeres or WTF is going on? i completely stopped using google on all my devices and I dont have this problem anymore.

Loved it. I just memorized the dictionary positions and since it was consistent unlike modern autocorrect it was so much faster for me. I never liked the later T9 dictionaries that did dynamic positions of words. It works extremely well for Danish and other Scandinavian languages (inventet around here iirc) , but I know for other countries T9 dictionaries were just not a thing which is sad.

I was doing Ctrl+f searching for Bluetooth on their homepage and no results. A bit annoying its only in the graphics. The page is so long and I’m not reading all that filler text. I just want a nice spec sheet. I hope it’ll come with time. Still cool project.

It’s super cool and I’m strongly considering it, but not including 5G is dumb I think. Where I live 5G has way better coverage and 4G is lacking and spotty many places. Also no mentioning of Bluetooth is also a big no no for me. So much for a good DAC buy now all my headphones I already own are wireless with LDAC And AptX etc.

So nice! I can’t wait till I don’t have to use my computer at all anymore since it does everything on its own. Now I can go outside and play with my dog all day and enjoy the nice hot weather we are getting with all this global warming! I dont have to work anymore now yaaay! I still get paid right… right…