Hacker News: datadeft

New comment by datadeft in "California farmers to destroy 420k peach trees following Del Monte bankruptcy"

datadeft — Tue, 05 May 2026 22:00:09 +0000

You are implying a centralized semi monopoly is the only way. If we had farmers to buyers direct distribution it would be much more resilient to this kind of problems.

New comment by datadeft in "Show HN: LinkedIn sucks, so I built a better one"

datadeft — Sun, 23 Mar 2025 19:58:43 +0000

Do you think that there is form of communication where you reach a higher understanding? Are we average Joes on HN?

New comment by datadeft in "NixOS and reproducible builds could have detected the xz backdoor"

datadeft — Sat, 22 Mar 2025 22:11:22 +0000

could have / should have => being smart retrospectively

New comment by datadeft in "Crabtime: Zig’s Comptime in Rust"

datadeft — Sat, 22 Mar 2025 19:12:02 +0000

> plenty of things Rust didn't solve

I agree on that too.

> Nor do I believe that Rust is the perfect tool for all tasks.

No disagrement here either.

I was just not really getting why you are attacking OP with assuming bad faith content from him immediatelly. Maybe I misread your comment.

New comment by datadeft in "Crabtime: Zig’s Comptime in Rust"

datadeft — Sat, 22 Mar 2025 10:01:12 +0000

I am sorry, but HN is probably the last place on Earth where you can wander off the mental plantation of single tech praiser people without much pushback. Perpetuating monocultural ideas and discouraging critical opinions is the way to create echo chambers that many online forums are poisoned with.

New comment by datadeft in "Apple shuffles AI executive ranks in bid to turn around Siri"

datadeft — Fri, 21 Mar 2025 09:22:49 +0000

True. I would love to have less features in MacOS actually. We are approaching Windows start menu level of crap being installed by default.

New comment by datadeft in "Moving away from US cloud services"

datadeft — Tue, 18 Mar 2025 08:51:05 +0000

My problem is that I do not want to replace one centralilzed service with another. I do not see any difference between the US and the EU (or Australia) in handling privacy. Most politicians are super keen on destroying privacy for people, for the "good cause". There are so many examples of this I lost count. We need strong encryption and true peer-to-peer networks where the connection is going through random routes (impossible to predict) and there is no government controll of any of the nodes it touches.

New comment by datadeft in "After 47 years, OpenVMS gets a package manager (VSP)"

datadeft — Tue, 18 Mar 2025 08:47:06 +0000

A lot. Problem is that sometimes the hardware is lost due to renovation to the building.

https://www.theregister.com/2001/04/12/missing_novell_server...

New comment by datadeft in "Rhombus Language"

datadeft — Tue, 18 Mar 2025 08:45:03 +0000

Is it possible to have Hindley–Milner type system for a LISP?

New comment by datadeft in "Zlib-rs is faster than C"

datadeft — Mon, 17 Mar 2025 13:07:38 +0000

True, however I only saw this happens to achieve max perf. I have very limited experience so this is confirmation bias from my end.

New comment by datadeft in "uv downloads overtake Poetry for Wagtail users"

datadeft — Mon, 17 Mar 2025 13:06:01 +0000

I see what you mean. You can use it with mise that has build support.

New comment by datadeft in "uv downloads overtake Poetry for Wagtail users"

datadeft — Mon, 17 Mar 2025 11:11:16 +0000

    uv build
    Building source distribution...
    running egg_info
    writing venv.egg-info/PKG-INFO
    Successfully built dist/venv-0.1.0.tar.gz
    Successfully built dist/venv-0.1.0-py3-none-any.whl

New comment by datadeft in "uv downloads overtake Poetry for Wagtail users"

datadeft — Mon, 17 Mar 2025 11:09:29 +0000

I have moved to uv few months back and never looked back. I use it with venv and it works very well. There is a new environment handling way with uv:

- uv init new-py-env

- cd new-py-env

- uv add jupyter

- uv build

These are executed super fast. Not sure if this could help your situation but it is worth to be aware of these.

New comment by datadeft in "Zlib-rs is faster than C"

datadeft — Sun, 16 Mar 2025 20:20:34 +0000

I thought that the point of Rust is to have safe {} blocks (implicit) as a default and unsafe {} when you need the absolute maximum performance available. You can audit those few lines of unsafe code very easily. With C everything is unsafe and you can just forget to call free() or call it twice and you are done.

New comment by datadeft in "DiceDB"

datadeft — Sun, 16 Mar 2025 15:06:34 +0000

Is this suffering from the same problems like Redis when trying to horizontally scale?

New comment by datadeft in "Cursor told me I should learn coding instead of asking it to generate it"

datadeft — Thu, 13 Mar 2025 10:07:29 +0000

The biggest problem what I have with using AI for software engineering is that it is absolutely amazing for generating the skeleton of your code, boilerplate really and it sucks for anything creative. I have tried to use the reasoning models as well but all of them give you subpar solutions when it comes to handling a creative challenge.

For example: what would be the best strategy to download 1000s of URLs using async in Rust. It gives you ok solutions but the final solution came from the Rust forum (the answer was written 1 year ago) which I assume made its way into the model.

There is also the verbosity problem. Calude without the concise flag on generates roughly 10x the required amount of code to solve a problem.

Maybe I am prompting incorrectly and somehow I could get the right answers from these models but at this stage I use these as a boilerplate generator and the actual creative problem solving remains on the human side.

New comment by datadeft in "New Zealand's $16B health dept managed finances with single Excel spreadsheet"

datadeft — Tue, 11 Mar 2025 10:02:44 +0000

The amount of issues I have seen in the last ~20 years caused by "off by one" type of errors in Excel is insane.

Few examples:

- incorrect schedule for the electricity grid for an entire country

- incorrect assessment of an airport use for an airline (causing few millions USD loss in revenue)

- incorrect financial position assessment for a mine (resulting incorrect deciosion to optimize the wrong business process, not sure about how much they lost)

Making illegal states unrepresentable is a concept that benefits programming langues and business processes alike.

New comment by datadeft in "Tailscale is pretty useful"

datadeft — Thu, 06 Mar 2025 10:27:53 +0000

So I have this small form box (it is hardly bigger then few mobile phones stacked on each other) and I run Tailscale on it as well (also as an exit node). I installed Tailscale on my MBP and mobile phones as well. Now it is forming a network where regardless of where I am in the world I can chose my home server as an exit node and also access the samba shares and the other devices (rpi3 and rpi4).

New comment by datadeft in "Apple M3 Ultra"

datadeft — Thu, 06 Mar 2025 00:01:54 +0000

Apple today announced M3 Ultra, the highest-performing chip it has ever created

I thought it was few weeks ago when M4 Max came by.

New comment by datadeft in "Tailscale is pretty useful"

datadeft — Wed, 05 Mar 2025 23:38:45 +0000

It is one of the tools I use as well and I pay for it. It makes life so much easier. At work we have to test a lot of country dependent settings and with TS and Mullvad is is very simple. I can also access my home network easily.