HN user
Intel does not have their own version of this machine. This ASML machine is a component in a very large system of machines from many manufacturers that runs wafers automatically from machine to machine until a month later a chip comes out the other end. That system is “18A”. Intel designs, builds and operates that system. Intel does not make the machines, they buy them and figure out how to configure and operate them as a part of the system.
A black friend of mine did exactly this, asked for a permission to get a pen from his pocket. The cop laughed “sure” and the moment he put his hand inside his pocket they jumped him and arrested him.
It’s effectively the company saying that they believe the shareholders can get a better return by investing that money elsewhere. So when a company starts doing major buybacks it’s a signal that they have reached an inflection point.
You can thank Microsoft for that. Intel architects in fact did not want to waste area on an NPU. That caused Microsoft to launch their AI-whatever branded PCs with Qualcomm who were happy to throw in whatever Microsoft wanted to get to be the launch partner. After than Intel had to follow suit to make Microsoft happy.
we have to be honest that there is no unquestionable source of meaning in life
Think about what you would like to remain in the world after you are gone. Then think how you can connect with and advance those things, and act accordingly in your life. To me this has been a reliable way to find meaning in life. But obviously I don’t claim this is unquestionable or works for everyone.
In the automotive industry, pretty much the whole point of standards like cybersecurity (ISO21434) and functional safety (ISO26262) is to let the manufacturer claim in court that they followed “modern best practices” and therefore are not liable when something goes wrong.
It is really incredible how Apple has obviously broken and buggy UX in many primary use cases on their devices, and fail to fix it for generations.
The iPad is particularly bad in this respect. For a decade it would not support the most obvious use case for a device like this: Have it in portrait mode like a notebook, show a video or book app on the top half and notes app on the bottom half. A use case that was solved by the original Macintosh. The most infuriating thing was that you could split the vertical screen into two useless, thing vertical strips---a configuration I have never seen any use case for. Even today now that there is some more configurability and you can vaguely put two apps in this configuration, there is still massive wasted space on the sides and the apps overlap.
The Human Brain Project comes to mind.
Following the definition from the article, armed forces seems like a complicated system, not a complex one. There is a structured, repeatable solution for armed forces. It does not exhibit the hallmark characteristics of complex systems listed in the article like emergent behaviors.
The article is about persuading your peers and your management to get what you need, not about persuading your direct reports to do their tasks.
It was a life-defining piece of software for me too. As a teenager I found a server called “REALbasic Cafe” that inspired and helped me go from knowing next to nothing about programming to making my first money from shareware as a high school kid.
To this day I’m grateful I stumbled across the Hotline software and the server.
As someone frustrated in a team of 10+ that is struggling to ship even seemingly trivial things due to processes and overheads and inefficiencies, I would really appreciate some insights on how do you organize the work to allow a single developer to achieve this.
How do you communicate with the rest of the organization? What is the lifecycle and release process like? Do you write requirements and specs for others (like validation or integration) to base their work on? Basically, what does the day to day work look like?
They mentioned during the presentation that they need the new display controller on the M4 to drive the new dual panel display.
HotBits is my earliest memory from the Internet after getting access as a young kid around 1996. Back in those days computer magazines would print website reviews and links, and I found HotBits in one of those. It was fascinating to a young kid who was into computers and physics.
Over a decade later I read the Autodesk File and it was a major inspiration for founding my first startup.
He was an inspiration to generations of hackers.
Let me give a concrete example: I have had iPad Pro for years now. A primary use case for me is to watch a lecture and take notes, or read a book and take notes. The way this should obviously work is that I have my iPad in portrait orientation, split the screen vertically so that I have the video playing or the book open on the top half and notes on the bottom half. This was solved perfectly by the original Macintosh. It is still not solved by iOS (even though I've had tickets for years in Radar).
For some reason I cannot comprehend, I can split the screen horizontally, which results in two thin strips side-by-side that are useless for anything that I can think of. It is not possible to split it vertically so that I would have two reasonable aspect ratio apps on top of each other. Now they added a convoluted "multitasking" mechanism that kind of lets me solve this for reading, where I can have two 3/4 sized apps overlapping each other, but I still cannot just split the screen or have freely resizable apps (which, again, is a problem solved already in the original Macintosh).
This type of terrible UX has become endemic, and is even worse in non-Apple products. The root cause I believe is "authoritarian simplicity", where some UX designer or team thinks they know the best and force a single, over-simplified, over-specified solution on everyone.
I've always been confused why salaried workers think they only need to work 40 hours a week.
It's cultural. Living in Europe, I've always been confused how salaried workers would not have set hours per week in the US.
I'm literally selling my working hours to a company. Of course the contract is going to have to include exactly how many hours I'm selling and how much money I'm getting in return.
It is an employment contract. I'm not selling the completion of a specific task or a project (that would be a different type of contract and I would price it very differently), I'm selling specific hours my time.
In automotive, the key feature offered by QNX is safety certification. This does not (and cannot) offer it, so it is not a potential replacement.
But also before you consider healthcare, transportation, higher cost of living in some desirable areas, lack of time off, lack of any employee protections, dismal social safety nets if you fall on bad times. And for people who need that, no maternity leave, expensive childcare, etc.
Let me try to add some more concreteness to this. I work for a major US tech company in Western Europe in a "second class" tech hub (i.e., not London or Zurich, but with most of the major tech companies present). Total compensation is low six figures, which is over 1.5x what I was offered by large European companies.
On my payslip, my net is about 55% of the gross. I then pay 25% of the net as my half of the rent, and 20% for childcare (in theory you could get it much cheaper, but in practice there are no available spots). On average I've paid 10% of net out-of-pocket for extra healthcare costs not covered by the mandatory insurance (which costs 15% of gross).
Compared to my previous job, my gross is now 2x but net 1.4x. My lifestyle is not meaningfully different than when I was making half the gross, I just accumulate some modest savings instead of breaking even. I still have no chance to own a home anywhere near the city I live in. I don't know how I could "work my way to a great lifestyle" here. Maybe I'm just doing it wrong.
Overall, my peers in the US have a significantly better lifestyle than me. What I do get that my peers in the US don't (AFAIK) is 30 days of paid holidays. Also my wife can get a year of maternity leave with reasonable benefits.
The worker benefits are pretty theoretical, since in practice I have to work well above the tracked hours to be competitive on the career track. I can also still be laid off on reasonably short notice if the company decides to cut costs. I would get 60% of my salary for a year from the state as unemployment, but I would typically not get any severance from the company.
Physical gold is also tax free after a year in Germany.
The VPN exit point for the big tech company I work in Europe for seems to be in Poland for whatever reason. I had to switch to duckduckgo to not be served search results in Polish all the time.
I just completed a successful job search, including landing a great senior level IC role at a large tech company. I do not have LinkedIn (or GitHub etc.).
Based on what I've read and heard from hiring managers, I optimized two things: CV and video call setup. My CV was customized for every application, a single page dense with information but designed to be easy to skim and pattern match to the role requirements. For video calls, I had a high quality camera, microphone, lighting and background arrangement. There are tons of guides for YouTuber setups, which I followed.
I have no control experiment (it could be due to a hot job market or my experience profile) but I was honestly quite surprised how effective it all seemed.
This brought back memories. The first money I ever made from software as a teenager was through Kagi. It was some Mac utility shareware apps written in REALbasic around the year 2000.
By Litte's Law, the arrival rate to the waiting room (l) is equal to the number of customers in the waiting room (L) divided by the amount of time a customer spends in the waiting room (W), ie., l = L/W. If the arrival rate to the post-injection waiting room is less than the throughput of the vaccination stations, then removing (or reducing) the waiting time after an injection is going to improve performance of the system.
There are also legal and privacy concerns. I've worked on a few research papers where exactly one researcher had access to the data under a very strict NDA. And even they did not get full access to the raw data, only the ability to run vetted code against it and some subsets for development.
This is because the datasets were subscriber logs from mobile operators. They are both highly privacy sensitive and contain sensitive business knowledge. There is no way they will ever get published, even in some anonymized form.
Ultimately it always comes down to trust. You need to convince your peer reviewers to trust you that you have correctly done what you have claimed to have done. Of course, even when you publish datasets, you need to convince the peer reviewers to trust you that you didn't fake the data.
Actually it's 25. Around that time your neuroplasticity changes and it becomes a lot harder to learn new things. You still can, but it takes much more effort and deliberate protocols to follow.
Prof. Andrew Huberman from Stanford University School of Medicine has some great podcast episodes on this (among other interesting things) [0].
I would add a word of warning here. In the multiple research groups that I've worked in, the completion rate for industrial PhD students has been way less than 50%.
The industry part (whether 20% or 50%) easily ends up being close to 100% in practice. And many industrial PhD positions only last a few years, after which you get a normal full-time position and are expected to finish the PhD on your free time. Realistically, this makes it practically impossible to conduct the kind of research that you need to get a PhD from top programs.
I'm sorry the hear that, and I think it's mostly worng.
I think there is just a large variance between countries and/or fields. I've worked in CS/EE in three top EU universities, and in every case I've seen, PhD students working in a research group had a heavy load of both teaching and project work on top of their PhD research.
When I was a PhD candidate, I was teaching directly what I was doing research on to bachelor and master students. So did my professor.
The research absolutely impacts teaching to a great extent. We are only humans after all, and I don't have a separate brain for teaching and for research. If you separate your best researchers in research institutes, you're going to seriously limit the flow of knowledge to your bachelor and master students. And those students are your future researchers.
This is not true. The GI specifications are very strict on where and how the protected product is produced. With Parmigiano Reggiano, for example, the cows producing the milk must be within the geographic area, and even their fodder must primarily come from the region.
The problem is that because of the way the incentives are set up, it is not important to anyone involved.
What is important to the grad students is to produce research papers and to fulfill their mandatory obligations (teaching, project deliverables). And most grad students, even in CS, are not professional software developers anyway. Good luck convincing capable grad student candidates to join your group to do boring software maintenance for horrible pay and no job security.
What is important to the professors, who decide what the grad students will work on, is again to produce research papers, fulfill their mandatory obligations (teaching, project deliverables) and to continually file for grants. Spending grad student time on porting and maintaining libraries does not help with that. In the worst case your grad student is spending their time maintaining a tool that a competing group's grad students are using to churn out papers, beating you to publications and grants.
What is important for the funding agencies is flashy new research in the current hot topics. I never saw a funding agency that would even consider paying a grad student, let alone a full software engineer salary, to port an academic tool from Python2 to Python3 or do all the other maintenance you need to do on production codebases---nor do most universities even have salary classes and positions for that.
As a result, in the many years I spent in academia, I saw many important research tools rot (both software and large hardware testbeds). The solution is not grad students, but to have fully paid software engineer positions in academia. But realistically that is not going to happen.