As many issues as we have, I’d rather have somebody who understands the consequences of a security breach having access to sensitive information than someone who does not.
Think what you will about who came before or after, but everyone involved here should have experience or training in how to handle and secure sensitive information.
Very neat and creative approach but I'm honestly conflicted whether the country/map metaphor is the best choice. In many cases the names are not that clear, so one has to zoom in to understand what they represent. It would perhaps be more interesting to do hierarchical clustering and show something like average connectiveness between the (super)clusters with lines, possibly with more descriptive/faithful LLM-generated labels for each cluster.
I was pleasantly surprised that it wasn’t a heavy line drawing creation. As someone who first did those in the 90’s and almost immediately learned their limits, I think this is nice because it doesn’t overclaim. It’s just a view, not a thesis.
I like diagrams where the axes mean something. Lines, shape, boxes/groups, distance, X vs Y, colour, thickness, texture, background, foreground. I also like simple. So often it’s lines to be fancy with no meaning. This one is just a pic, with some grouping, and it has personality. Yay?
I couldn't find a universal clustering algorithm yet: Frequently there is more than one way to group data that still makes sense, and as a result whichever final clustering option we choose - it will not be perfect.
Hm... unless maybe we do some sort of quantum clustering, which could be a fun project to explore!
It's a bit hazy now, but I remember trying hdbscan algorithm (hierarchical clustering), and on the graph of the GitHub size - I just couldn't fit it in memory.
I did end up using something similar to hierarchical clustering (mix of louvain/leiden/my own), and that's what we see in the final map.
Unitree is catching up to Boston Dynamics in locomotion, but has not surpassed them yet I think. Unitree's real advantage is the much lower cost of their hardware and their ability to mass produce their robots.
But IMO manipulation is harder than locomotion, both in hardware and software, and neither company is convincingly ahead there. I think the uncut laundry demo from Physical Intelligence a few days ago is better than anything shown by Unitree or BD for manipulation. https://www.physicalintelligence.company/blog/pi0#:~:text=Af....
I don't really want a robot wandering around doing laundry. I think what most people want is a box you dump clothes in and they come out folded, an extra machine next to the dryer. That would be a genuine time saver. I hate scaling Mount Foldmore.
I don't want more machines taking up space in my home. I don't want a bunch of special purpose "smart" devices with buggy software and dedicated apps requiring logins and firmware updates to plug security holes. I want one robot that can do it all. I'd get rid of my security system, cameras, smart thermostat, dishwasher, clothes washer, stand mixer, toaster, etc etc.
I put the laundry on this morning, came into the kitchen and got a bread mix going, made some toast while that was mixing. The dishwasher was just finishing. Having to wait for one machine to serially finish doing each job not quite as well as a dedicated machine, before doing the next job, doesn't sound ideal. Plus when the dishwasher breaks I can still have toast.
The robot can do the dishes and the laundry while you aren't even there. No need for that to happen during breakfast! It can start making breakfast before you even wake up, if you want.
Clearly it’s not easy to solve the folding machine problem because otherwise we would have one of those already. I absolutely wouldn’t mind a robot helper walking around the house and going what my maid currently does.
They put a lot of work into the control algorithms that allow the robots to move. From a control theory perspective, being hit is a disturbance that the control system is able to reject to maintain the intended pose/motion.
You have to make sure that robots know their place. Today you treat them too gently, tomorrow you have a robot uprising destroying the whole humanity on your hands.
Haven't the last century of sci-fi books and movies taught you this?
What? The humanoid robot in the video is walking on the most even surface possible. Boston Dynamics has a more impressive walking video from 11 years ago (obviously restricted by the hardware at the time): https://www.youtube.com/watch?v=SD6Okylclb8
Yes, human-looking gait is nice, but it isn't worth anything if it can't translate to real-world settings.
LMAO Boston Dynamics used to have a guy randomly coming into the frame and giving the robot a kick and show it could recover, Unitree goes all-out kung fu on them.
Something rubs me the wrong way when the dog is hit and kicked, and we don't have a dog.
I know it's a robot but the same could be demonstrated without clear violence.
what if the robot has little vials of acetone, IPA, etc. to wash off its lenses, while switching to mm-wave vision (not even dead reackoning) returns the favor to the offending human while cleaning its lens...
Unfortunately some humans are just violent in nature. Or perhaps most. When robots become as intelligent as humans, some people might enjoy abusing them like they once enjoyed abusing people of different skin color.
I guess it's the difference between an ensemble and a mixture of experts, i.e. aggregating outputs from (a) model(s) trained on the same data vs different data (GPT-4). Though GPT-4 presumably does not aggregate, but it routes.
Yeah, it's a riff on the old random factoid you used to see all the time in papers and books where someone would be talking about something like Spanish Rice and the editor would helpfully point out that in Spain they just call the dish Rice. Also, it's just fun to make puns sometimes.
A similar joke:
Batman and Robin are in hot pursuit of the Joker when they hop in the Batmobile. Unfortunately it fails to start so Batman tells Robin that he is going pull out the Batskates and continue to chase the Joker. In the meantime he tells Robin to check the battery. Robin agrees but after a moment pauses to ask "What's a 'Tery'?"
I think that would fall under the "must accept interference" portion of FCC Part 15, my phone communicating back to the tower might raise some alarms if I tried amplifying it at all though!
Having used Copilot for a year now I very much doubt this figure. In my experience it only works well in boilerplate kind of situations where most code is copy/paste work anyhow. As soon as the code gets a little complicated it stops working well. It has also gotten quite slow for me lately. So I doubt it increases my work efficiency by more than 5%, but I do like it for reducing strain on the hands. For that I find the price appropriate.
> In my experience it only works well in boilerplate kind of situations where most code is copy/paste work anyhow.
As a data point, this matches my personal observations. But reducing the time spent on that bolierplate plus not needing to search where the "copy" portion comes from, may justify $30/month (and probably much more than that). My 2c.
Is Paperless suitable for business use, say, for a smallish sized company with 25 employees and 1000 customers. I think in my EU country such systems need to fulfill certain requirements like versioning/tracking of changes.
