I still can't forgive Apple to buy darksky and then killing it. It was so good and Apple Weather is still not even close. Why are these big companies buying small companies that have a great product and just let it die without using their tech?
Using LLMs for universal computing would be crazy inefficient. On top of that they aren't deterministic (at the moment) so building systems on top of that would be shaky
We have these at my company. They refuse to do any infrastructure work so you have to spoon feed the databases to them ready to go. It’s pretty annoying.
“ This way, the work done by a "low quality" software developer (this includes almost all of us at some point in time), is always taken into account by the process”
That’s a horrible take. There is no amount of reviews, guidelines and documentation that can compensate for low quality devs. You can’t throw garbage into the pipeline and then somehow process it to gold.
In general I agree but there is bit more complexity. I work in medical devices and there are plenty of situations where a certain output is ok in some circumstance but deadly in another. That makes a stopgap a little more tricky.
I agree with the previous poster that the feedback from the field is lacking a lot. A lot of doctors don’t report problems back because they are used to bad interfaces. And then the feedback gets filtered through several layers of sales reps and product management. So a lot of info gets lost and fixes that could be simple won’t get done.
In general when you work in medical you are so overwhelmed by documentation and regulation that there isn’t much time left to do proper engineering. The FDA mostly looks at documentation done right and less at product done right.
Agreed - This is essentially the corner stone of systems failure analysis - something I wish architects thought about more in the software space.
I'm a product manager for an old (and if I'm being honest somewhat crusty) system of software, the software is buggy, all of it is, but its also self healing and resilient, so while yes, it fails with somewhat alarming regularity, with very lots and lots concerning looking error messages in the logs, but it never causes an outage because it self heals.
Good systems design isn't making bug free software or a bug free system, but rather a system where a total outage requires N+1 (maybe even N+N) things to fail before the end user notices. Failures should be driven by at most, edge cases - basically where the system is being operated outside of its design parameters, and those parameters need to reflect the real world and be known by most stakeholders in the system.
My gripe with software engineers sometimes, they're often too divorced from real users and real use cases, and too devoted to the written spec over what their users actually need to do with the software - I've seen some very elegant (and on paper, well designed) systems fall apart because if simple things like intermittent packet jitter, or latency swings (say between 10ms and 70ms) - these are real world conditions, often encountered by real world systems, but these spec driven systems fall apart once confronted with reality.
> My gripe with software engineers sometimes, they're often too divorced from real users and real use cases
I agree. In my current job, no developer ever talks to users. Projects involve multiple layers of subcontractors, and cross-vendor APIs are part of the bid/spec. Each layer’s purchasing department is in perpetual Bidding Mode while dev teams are already in the middle of doing the actual work. So by default, everyone is Strongly Discouraged From Talking To $VENDOR | $SUBCONTRACTOR | $USER.
Completely safe as long as the block of metal was in place. So you couldn’t just prevent the machine from putting out that much energy, you had to prevent it from doing that without the block in place.
The earlier model that the 25 replaced was all mechanically interlocked. The belief was that software provided that same level of assurance. They performed manual testing but what they weren't able to do was reach a level of speed and fluency with the system to result in the failure modes which caused the issues. Lower hardware costs equals higher profit...
DOGE never tried to do good work. It was a hit and run job to satisfy Musk's ego and get rid of stuff he personally doesn't like. It didn't save much money and didn't create any efficiency.
Before Neuralink, there was no major investment into BCI tech as far as eye could see - because medicine is where innovation goes to die. We've gone from Utah arrays in 1990 to Utah arrays in 2020. All while computing and AI - the other key enablers of neural interfaces - advanced in leaps and bounds.
I have lots of ideas. Some are great, some are ordinary, and some turn out to be embarrassingly stupid.
So does everyone else who tries to create new things. Edison had dumb ideas, too, like his mining ideas. The Wrights also had dumb ideas like their persistence with wing warping, and the canard stabilizer.
The sub thing didn't hurt anyone, it was an emergency so he didn't have much time to think about it, so really it's uncharitable to slam him for trying to help.
Do you think his rockets are dumb ideas, too? Starlink? Tesla?
This discussion started with somebody complaining that people are dismissing Neuralink because it's Musk's company. My point is that his ideas are getting way more attention than other people's ideas, stupid or not.
And he is an attention whore who will go after people who are dismissing his ideas. The cave guys in Thailand had to waste precious time thinking about his submarine. If Musk had really been willing to help, he would have done testing in quiet and published things only when it was clear that it worked. But he is an attention whore because he knows it's good for business.
Same for DOGE. They could have done their work in quiet and with deliberation. Instead they fired quickly some random people whose work they didn't understand or like.
> his ideas are getting way more attention than other people's ideas, stupid or not.
His amazing track record with success means his ideas merit more attention than your ideas or mine.
> The cave guys in Thailand had to waste precious time
insulting Musk on CNN. They didn't have to do that. They could have simply said "no thank you, we'll handle it".
> But he is an attention whore because he knows it's good for business.
Yes, offering his company's considerable engineering talent and resources for free is pure evil. Sheesh.
> Instead they fired quickly
They didn't have the people to evaluate tens of thousands of individuals, nor did they have several years to do it in. The way they proceeded was the only practical way. It's the way all organizations above a certain size cut costs when hemorrhaging cash.
Vern Unsworth said that the submarine wouldn't work, Musk then called him a paedophile and tried to hire a private investigator to discredit him. Could Musk not have just accepted that yes, his idea was stupid, and not thrown false accusations with no basis at Unsworth?
According to research impaired structural brain development due to visual deprivation from birth is not fully reversible and limits functional recovery. So even if eye sight is fully restored, cortical function will not be able to fully take advantage of that.
Experiments and studies have shown that this might be due to the fact that the visual cortex will take over a similar role in blind people as it does for people with intact eye sight. The brain uses different sensory inputs in that case and the visual brain structure is not restored after eye sight recovery.
This is still an ongoing field of research of course, but so far congenital blindless seems to be incurable, regardless of whether the sensory apparatus could be restored or replaced. Note that this only means seeing like a non-blind person. Some limited visual perception is still possible, just not "normal" sight.
Do you have some links handy? I'd be very much interested in the description of experience from people that have gained sight after congenital blindness.
You might be interested in searching up "strong critical periods" in brain development - this phenomenon has been studied for decades and has interesting implications for lots of aspects of brain development: https://en.m.wikipedia.org/wiki/Critical_period
If you are young enough, yes. But after a while, the neuroplasticity is simply not enough. Seeing is a complex enough process, if you miss learning it in your childhood, the train is gone. This is a very common error people make, announcing implant technologies to grown blind people as if the cure was just around the corner. It isn't. You will never adapt to a point where the vision you just gained is actually useful. Imagine trying to learn to read print, at 30, with a pixelated implant? It is a naiv pixie dream of sighted people.
The absolute numbers are still astonishing and make it clear that without deep pockets you won’t get anywhere. Maybe you can make an argument that from a certain level on, money doesn’t help anymore. But the minimum amount is still very high.
Advertising is very expensive and political candidates have to pay retail rates. As the old adage goes, half of it is wasted but you don't know which half. However with political ads, all of the money spent on the losing candidate is wasted in a certain sense.
As media has fragmented, you really have to spend a lot to get in front of enough high propensity voters, and even more to turn out low propensity voters unless you're organically good at getting invited on to podcasts.
Consulting around elections is a huge business, and consultants that win races cost millions, but that can't be paid for by PACs at all because they work with the campaign directly.
Buy ad buyers and media agencies that only work in politics are definitely a thing[1] also.
reply