HN user
To me this looks like a cool demo product. Yet, the problem it's solving could be equally solved by a well integrated all-in-one business suite.
I don't run a small business myself, but I assume the scope of administrative tasks in such company is well defined and understood.
Not my app. Saw it online and found it funny. I'm #27
In the recent Dwarkesh Podcast episode Jensen Huang (Nvidia) said that virtually nobody but Anthropic uses TPUs. How does that add up?
Wookash Podcast
That's completely relatable, and also a major point in my original argument. Using heavily abstracted frameworks will automatically cap you performance wise. The only way out is to not use a framework or one that's known to be lightweight. In backend or tooling like with the JS compiler from OP, one tends to not use heavy frameworks in the first place.
Nobody expects you to know that, but I'm curious to hear how do you know it for backend code but not frontend code. Have any examples?
It's not too diffcut in the browser either. Consider how often you're making copies of your data and try to reduce it. For example:
- for loops over map/filter
- maps over objects
- .sort() over .toSorted()
- mutable over immutable data
- inline over callbacks
- function over const = () => {}
Pretty much, as if you wrote in ES3 (instead of ES5/6)
It always comes as a surprise to me how the same group of people who go out of their way to shave off the last milliseconds or microseconds in their tooling care so little about the performance of the code they ship to browsers.
Not to discredit OP's work of course.
They use Claude to skip the typing, not the thinking. They're 10x faster than two years ago.
I'm not sure a 10x increase in typing speed makes you a 10x developer.
I've raised this exact point to many team leads throughout my career.
Yet, they unanimously said, they are interested or need to know the progress.
I can't say if thats what they have to report to their managers, but I assume it's something you won't be able to fix from bottom-up.
For the handful of regularly visited websites, I wouldn't mind.
When I connect my server over SSH, I don't have to rotate anything, yet my connection is always secure.
I manually approve the authenticity of the server on the first connection.
From then, the only time I'd be prompted again would be, if either the server changed or if there's a risk of MITM.
Why can't we have this for the web?
Perhaps our industry should adopt a different approach, that fills in the gap between those.
- You host open-source software on your own hardware.
- You pay a company for setup and maintenance by the hour.
I saw from the SSD was around 800 MB/s (which doesn’t really make sense as that should give execution speeds at 40+ seconds, but computers are magical so who knows what is going on).
If anyone knows what’s actually going on, please do tell.
LLM-generated code should not be reviewed by others if the responsible engineer has not themselves reviewed it.
To extend that: If the LLM is the author and the responsible engineer is the genuine first reviewer, do you need a second engineer at all?
Typically in my experience one review is enough.
That would be shared ownership again.
The fundamental problem here is shared memory / shared ownership.
If you assign exclusive ownership of all accounting data to a single thread and use CSP to communicate transfers, all of these made up problems go away.
Your reasoning seems counter intuitive as back in 2012 Facebook rewrote their HTML5 based app to native iOS code, optimized for performance, and knowingly took the feature parity hit.
https://engineering.fb.com/2014/10/31/ios/making-news-feed-n...
Reminds me of Casey Muratori‘s talk on Conway‘s Law: „I always know what I am thinking…“
The @breakpoint built-in
Inserting the literal one byte instruction (on x86) - INT 3 - is the least a compiler should be able to do.
To add to that, a lot of business context is stuck in people‘s heads. To reach the level of a human engineer, the coding agent would have to autonomously reach out and ask them directed questions.
While this is great for end users, this just shows again what kind of monopoly Google has over the web by owning Chrome.
I work at a company that also happens to run a CDN and the sheer amount of layers Google forces everyone to put onto their stack, which was a very simple text based protocol, is mind boggling.
First there was simple TCP+HTTP. Then HTTPS came around, adding a lot of CPU load onto servers. Then they invented SPDY which became HTTP2, because websites exploded in asset use (mostly JS). Then they reinvented the layer 4 with QUIC (in-house first), which resulted in HTTP3. Now this.
Each of them adding more complexity and data framing onto, what used to be a simple message/file exchange protocol.
And you can not opt out, because customers put their websites into a website checker and want to see all green traffic lights.
At computerenhance.com[0] Casey Muratori shows that memory mapped files actually perform worse at sequential reads, which is the common case for file access.
That’s because the CPU won’t prefetch data as effectively and has to rely on page faults to know what to read next. With regular, sequential file reads, the CPU can be much smarter and prefetch the next page while the program is consuming the previous one.
Remember that code that clearly solves just those problems is in no way a hack.
I think you can’t stress this point enough. In my experience anything that is not implemented by any norm or „clean“ or in an unusual way is considered a hack. Even if it perfectly solves the problem with the least amount of cruft to it. That makes me sad.
they definitely improves average programmer efficiency
Do we really need more efficient average programmers? Are we in a shortage of average software?
This sounds suspiciously similar to CSV.
Apps should be required to ask for permission to access specific domains. Similar to the tracking protection, Apple introduced a while ago.
Not sure how this could work for browsers, but the other 99% of apps I have on my phone should work fine with just a single permitted domain.
This has been something I’ve been curious about for a long time, so thanks for the write up.
When it comes to system design, we often ignore what modern hardware is capable of and ignore fundamentals like fsyncs/s or more generally IOPS.
So instead of assuming that we potentially need a horizontally scalable system, we should instead find hardware that matches our needs.
Any thoughts on Mullvad Browser?
Thanks, this exactly my point.