I like the article. In fact just yesterday I quipped to someone about how the quality of AI output will be determined by the competence of its "operators".
I have always had a strong drive to produce awe-inspiring software. At every job I have had, I have strived for usability, aesthetics, performance and reliability. I have maintained million LoC codebases and never caused a major issue, while people around me kept creating more problems than they fixed. However I do not recall one single word of encouragement, let alone praise. I do recall being lectured for being fixated on perfectionism a few times.
I took me 15 years to realize that consumers themselves do not care about quality. Until they do, the minority of engineers who strive for it are gonna waste their efforts.
Yes software is complex. Yes, you cannot compare software engineering to mechanical, electrical engineering or architecture. But do we deserve the absolute shit show that every app, website and device has become?
I don’t think it’s true that “consumers don’t care about quality” but rather that their concern for quality doesn’t really manifest itself in those terms. Consumers care about critical tools being available when they need them and businesses often have a hard time situating feature requests in the broader desire for utility and stability (in part because these are things only noticed when things are really bad).
Part of my growth as a developer was connected with realizing that a lot of the issues with quality resulted from miscommunication between “the business” and engineers advocating for quality.
agree that we (users, humans, customers) all are desperately reaching for something steady, well designed, rugged.
something that people thought about for longer than whatever the deadline they had to work on it. something that radiates the care and human compassion they put into their work.
we all want and need this quality. it is harder to pull off these days for a lot of dumb reasons. i wish we all cared enough to not make it so mysterious, and that we could all rally around it, celebrate it, hold it to high regard.
> I took me 15 years to realize that consumers themselves do not care about quality. Until they do, the minority of engineers who strive for it are gonna waste their efforts.
N=1 but I would personally pay $15 for a higher quality garden hose valve when the competition is <$10, but I typically can't because I have no way of knowing which one of the products marketed at me is the higher quality one.
The worst case scenario for me is paying $15 for a bad quality product.
It's not that consumers don't want quality, it's that quality is hard to market.
Consumers don't care about your code, only what it does for them. If your crappy software provides its intended service quickly, accurately, and reliably enough, your customers consider it a win. Any further improvements on those axes are just gravy -- expensive gravy.
No. Consumers don't care about what they get. May be a small minority of tech savvy ones do but others don't or at least they don't know how to demand better software because software comes with no promises and guarantees.
Consumers certainly get annoyed if their software is hard to use, has errors, is slow or down. I've run many user tests with arbitrarily selected test subjects, and this part is universal. This is what good software development can fix, or ideally avoid in the first place.
One kinda heart breaking thing I observed was that especially older users wouldn't get mad at the software, but themselves. They thought everybody else is using this just fine, and they're somehow not smart enough. Motivated me to go that extra mile whenever I can.
What's nonsensical, here? How many of us can rid ourselves of the terrible Wifi router the ISP forced upon us? The sluggish operating system on the TV? The stupid navigation system installed by your car manufacturer.
A small minority can but in general the software you can use is determined by IT, procurement and leadership, and corporations controlling operating systems and hardware.
That is a very different argument from your original one. You are correct that consumers don't always have much choice about the quality of products, and they have to take something crappy or go without entirely. But that's not the same as "consumers don't care about what they get". They care, they simply are unable to get something better.
Could you expand on how you see that working? combining these extensions is what I’m guessing but I’d be super interested in hearing how it might work.
If all information for an asynchronous development process could be stored into git and distributed by it, this would be a very good idea for many projects.
The same way, one could add support for code review (aka PRs), todo list, custom entities that your workflow need (say, tracking documentation or custom requirement) ... It can also be entirely outside of the development process.
This would allow for really native linking between a tracked issue and corresponding commits/branches/tags, for modeling dependencies between issues as part of the git DAG,...
I am releasing JFK-TELL, a dataset I generated by extracting text from the scanned PDFs of the assassination records released until April 2025. The extraction was done with Google Gemini LLM API to generate Markdown text, using a very simple prompt. For detailed methodology, check out the Github repo at https://github.com/farhanhubble/jfk-tell
Who is independently error checking this? Surely given the depth of conspiracy theory, simply machine driven OCR with no independent validation is going to "feed the beast" more than it intends?
The intention is that it should be used for all kinds of research, including the efficacy of LLM OCRs. It shouldn't be too difficult to look at the original PDF to substantiate any "fact" anyone wants to quote from the dataset.
I'm more interested in seeing if people can find new insights, questions, or inconsistencies by reviewing the documents at scale, automatically.
Nehru place was almost 100% scammers than real geniuses but definitely the place if you wanted a custom computing rig. The food was good enough, like you said but the prospect of having a 2GB RAM module made you salivate even more!
The neighboring areas also had electronics importers that proved super helpful. You could find some of the most "lethal" tech with their help. I got myself some powerful FPGA kits that are normally not accessible outside defense, academia, and select licensed labs. I have great admiration for those folks who let me lay my hands over the most powerful technology of that time.
Lillian's writing is really crisp without compromising the technical details.
I learnt reinforcement learning from Udacity course. It was one of the finest on the internet. Whenever I couldn't find some thing in the course I went to this website. I created my own small introductory course for RL, a pure hands-on experience for learners.
In image processing at least, NN typically learn a Fourier or Wavelet representation in their first layers. Biggest benefit of applying a transformation beforehands is to reduce training time / obtain better generalization by "removing the dimension that doesn't matter".
E.g. in a suitable space, one coordinate could represent the rotation of an object. You could do the transform and discard this dimension if your NN should be rotating invariant.
In image processing I thought there was a whole host of specialized algorithms, such as edge detection, SCC, etc. that were run before the data was even fed into the ANN.
As a DIYer myself I am awed by how much Henrik has managed to flesh out in terms of complexity: signal processing, hardware, GPU acceleration, and algorithmic optimizations. I wish the HN community had some award for such incredible feats.
This is absolutely amazing work but I would not call him a hobbyist.
Again not diminishing the work but adjusting the expectations, this output is not from a curious guy in a shed. He is a trained Electrical Engineer working as a Senior Consultant/Scientist in RF design for a Finnish Engineering Consultancy.
This guy clearly has a breadth and depth of knowledge from his professional career which he brings with him into his out-of-office projects.
This is the way to go if you want get things running without wasting days. HA needs to be on on wired LAN for the best experience so one more reason to have it run on RPi and maybe have it sit next to your router.
It may not be Geospatial data at all and I'm not sure how much the users consented but the data collection strategy was well crafted. I remember recommending building a game to collect handwriting data from testers (about a thousand), to the research lab I worked for long time back.
I have always had a strong drive to produce awe-inspiring software. At every job I have had, I have strived for usability, aesthetics, performance and reliability. I have maintained million LoC codebases and never caused a major issue, while people around me kept creating more problems than they fixed. However I do not recall one single word of encouragement, let alone praise. I do recall being lectured for being fixated on perfectionism a few times.
I took me 15 years to realize that consumers themselves do not care about quality. Until they do, the minority of engineers who strive for it are gonna waste their efforts.
Yes software is complex. Yes, you cannot compare software engineering to mechanical, electrical engineering or architecture. But do we deserve the absolute shit show that every app, website and device has become?