HNHacker News
TopNewBestAskShowJobs

KolenCh

350 karma · joined September 18, 2021

submissionscomments
KolenCh··on One Handed Keyboard
Interesting. I worry about its ergonomics though as RSI might develops over time after long term usage of that design.

I've been designing my own one-handed keyboard for 3 years. My main problem is that both of my wrist suffers from RSI and either wrist can ocassionally acts up with different levels of pain. (I also have shoulder problem.) They can become practically disabled temporarily for a few weeks, or just quite painful for me to avoid using it. So my desiderata are a bit different from permanently one-handed people.

Interestingly my right wrist is acting up in the last couple weeks so I've been going through a iterative redesign phase recently. I probably will write up a blog post in the future when I have the final design, I'm going through it briefly below:

Desiderata: (first three are directly from the temporarily and random disabled hand criteria)

- primary used for two hands - each hand should be able to single-handedly control the computer - skill transfer from two hand to one hand: since one hand use is oacassional, retraining time should be minimal - based on the research in http://www.tandfonline.com/doi/abs/10.1207/s15327051hci1101_... which eventually becomes a producten https://matias.ca/halfkeyboard/ , the concept of a mirror key becomes a requirement: with the hold of a mirror key, the key at the mirror image position is active. An implementation detail is that the mirror key is a dual function key: on tap it is space, on hold it is mirror. I've implemented other possibility but find that the design in this research is better than others I come up with. - symmetric keyboard would facilitate this, where many split keyboards already is. - I must be able to buy them off the shelf. I do not have the skills to design it from scratch, nor do I afford to put more strain to my hand to assemble it from parts. - ergonomic must be one the of the primary goal of the keyboard, to minimize RSI. Speed is not important at all for example. - from my empircal experience, split keyboard, espeicially true split keyboard would encourage a better wrist and shoulder ergonomics. Hence I require split keyboards.

Based on these criteria, I bought ZSA Moonlander (QMK based) personally and Kinesis Advantage 360 Pro (ZMK based) for work.

The mirror key based design is currently at

    - Moonlander https://configure.zsa.io/moonlander/layouts/QwA3z/latest/0
    - Adv360 Pro: https://github.com/ickc/Adv360-Pro-ZMK/tree/dev
The key concept are that the mirror key can be implemented as a layer, and shift also functionally acts like a mirror. Thumb cluster are then dual function, where on hold a key could be the mirror key (via layer), another key could be the shift key. And since mirror+shift is needed, you either hold both (which is a bit less pleasant for the thumb), or have another layer serves as the shift-mirror key. Over there, every key is implemented as holding shift+key at mirrored position.

The ZSA training site is useful to iterate this design process: after each iteration I'd train per single hand and see if it works. For example in my earlier design I mainly focused on testing single left hand use and later found it doesn't quite work for single right hand.

Finally, macOS sticky modifier is used to hold modifier with a single hand. I.e. Ctrl+Opt+A becomes Ctrl+Opt, release, A. This is because OSM in QMK cannot handles one-shot of multiple modifiers well. Without doing fancy thing, you need to do Ctrl, release, Opt, release, A.

Same design working across Moonlander and Adv360 is important. Layout differences is not that much thankfully, but firmware difference can be a pain.

Lastly, I recently bought a Silakka54 for the ocassions where the setup hassle of either is too high. Basically either lap use or going to a meeting. I think my current layout design is adaptable to it but I'll see.

KolenCh··on Linear algebra explains why some words are effectively untranslatable
Big claim but not much substance. They should try to really understand linear algebra first, and also linguistics a bit. Semantic domain (from linguistics) is a better way to describe it, where using sets (from math) might better convey what they want to say.
KolenCh··on Physicists Take the Imaginary Numbers Out of Quantum Mechanics
This is referring to the fact that overall phase is not real (no observable difference) but relative phase has. The word “except” is not downplaying its importance, but to emphasize the fact that overall phase isn’t physical.
KolenCh··on Why formalize mathematics – more than catching errors
Are you sure you are really talking about Physics? Are you talking about actual research in physics, or physicists applying their way of thinking in other things?
KolenCh··on Windows Subsystem for FreeBSD
But Apple enforces code-signing, and that prevents them to ship those binaries under GPLv3 in the OS. The users can always compile them on their own (or via their favorite package manager) but Apple just can't ship it (without changing how they operate fundamentally.)
KolenCh··on Windows Subsystem for FreeBSD
That’s related to anti-tivolization in GPLv3 and basically Apple is forced to stop shipping things that’s updated to GPLv3. That’s not just about scared, assuming being scared means they are irrational and they could have adopted it if they want. Legally they cannot ship it unless they are changing their business model.
KolenCh··on Models Can Prompt Now
My wife is doing a freelance project where basically humans are evaluating how good an LLM-generated prompt is, and if there’s any problem to be fixed. Some times the humans are amazed and can’t believe they are written by LLMs (other times they are flabbergasted by how bad it is, of course.)

So definitely it has been improved in this area, intentionally.

KolenCh··on U.S. abandons hunt for signal of cosmic inflation
It really hurts to see this happening. CMB-S4, stands for stage 4, is a huge collaborative effort that units many CMB scientists and experiments, in a sense that stage 1 to 3 experiments are “converging” towards this.

To put that in context, CMB-S4 being a DOE project has many side effects to other CMB experiments in a lot of ways. For example, a few years ago the LiteBIRD mission led by JAXA from Japan has been gaining momentum, and many international CMB scientists got involved. European scientists got funded by ESA, and US scientists were expecting NASA to fund them too. In the end they denied the proposal (despite generally positive impressions when discussing with various people there), partly because they think the satellite based CMB experiment LiteBIRD has significant overlap with the goals of the ground based CMB experiment CMB-S4, deeming it unnecessary to support LiteBIRD.

And then CMB scientists used to get generous access to NERSC, a top 10 HPC system in the world. But as CMB-S4 becomes a DOE project, NERSC being DOE funded also, it becomes a bit of conflict there, in the sense that they feel they need to prioritize access for CMB-S4 to guarantee its success. There are many other factors in play but in the end it becomes much more difficult to even get access to the system, not to mention having any sizable allocation.

All these might not be so bad as CMB-S4 is supposed to be our endgame. It would benefits the CMB community as a whole so much. But now? It’s game over.

It also hurts particle physics progress as well. Long story short, the CMB B-mode holds a promising sensitivity to inflationary models, that its discovery may finally makes inflation falsifiable. At the very least, it involves an energy scale so high that no experiments on Earth can reach, and therefore is a good complement, to high energy physics experiments such as LHC.

There is a reason it is deemed so important in both the decadal survey and the Particle Physics Project Prioritization Panel.

Project like this has under heavy scrutiny from both the scientific community and the funding parties around feasibility, risks, costs, etc. It wasn’t a a light decision to have had happened in the beginning. Why would the end come so abruptly without explanation?

KolenCh··on Tell HN: I Lost Joy of Programming
May I suggest stop eating cats and the joy will come back.
KolenCh··on How I set up new MacBooks
I have custom scripts to bootstrap my macOS as many others here, and I can testify that migrating to nix-darwin greatly simplified my scripts.

P.S. I don’t use home manager though as I also need to bootstrap systems without nix.

KolenCh··on How I set up new MacBooks
Interesting! Do you know how to convert com.local.KeyRemapping.plist to the JSON format you used for remapping keys?
KolenCh··on Show HN: Infat – Declaritive application assocation manager for macOS
I see that you mentioned it is inspired by duti. I think may be adding a comparison on how it solves what it couldn’t would be beneficial.

From quick skim and memory it seems a notable one is url scheme association.

One problem I had with duti is to associate extensions automatically from a pre defined “table”. Automating that involves shell scripting[^1]. Also associating extension to softwares that hasn’t been installed. Another pain point is there’s two ways to associate and depending on different cases one is better than another.

[1]: https://github.com/ickc/bootstrapping-os-environments/blob/m...

KolenCh··on Sapphire: Rust based package manager for macOS
To those who downvote, care to explain? Homebrew community is notorious in how they treat users on their GitHub Issues and I wonder why anyone is downvoting without showing how that situation have changed.
KolenCh··on Sapphire: Rust based package manager for macOS
I think you mean “the hell I got from the team”.

I wonder if anything changed substantially over the years to make you say that?

KolenCh··on Meta antitrust trial kicks off in federal court
From an actual email from Mark Zuckerberg on February 28th 2012:

> Integrating their products with ours to improve the service is also a factor. But in reality, we already know these companies’ social mechanics, and we will integrate them over the next 12–24 months anyway. The integration plan involves building their mechanics into our products rather than directly integrating their products, if that makes sense, by a combination of these two things: neutralizing a potential competitor, integrating their products with ours to improve the service.

> One way of looking at this is that what we’re really buying is time. Even if some new competitor springs up buying Instagram, Path, Foursquare, et cetera, now will give us a year or more to integrate their dynamics before anyone can get close to their scale again. Within that time, if we incorporate the social mechanics they are using, those new products won’t get much traction because we will already have their mechanics deployed at scale.

https://www.acquired.fm/episodes/meta

KolenCh··on Tell HN: Announcing tomhow as a public moderator
I thought it is an April fool joke about having Tom Holland as moderator.

My bad.

KolenCh··on Ask HN: Any jobs that don't force you to always be advancing career wise?
ie is abbreviated Latin and usually written as i.e. I know that mobile and auto correct can make it difficult, but IE must be deliberate upper cased. It is the worst because no matter how ancient it is, it still stands for internet explorer, not id est.
KolenCh··on Stoicism's appeal to the rich and powerful (2019)
There are two kinds of Christians, the Bible studying kind, and the one who hearsay. Philosophical arguments like this are handled extensively in the field of (Christian) theology. There are many things many Christians claim nowadays are not biblical. Imagine “physicists” start claiming things that no falsifiable, tested theories in physics attested to. Now do you call them physicists? If there are enough majority of them is doing it, do you say “physicists“ (as a generic term) now believes in so and so? This the kind of situation that’s happening in Christianity. Because, the core believe is anyone can be a Christian (subjected to minimal check such as the basis of their salvation, some also include baptism), unlike in physics you need to be trained and qualified. I’m not saying that’s right or wrong. But I’m pointing out the situation, so that one can understand, for outsiders one would say “Christian believe this or do that”, and for the “insiders” who really study the Bible and trained in theology and know differences between denominations for example, would be completely puzzled and say it is not.

I guess the former focuses on Christian as a social construct and includes whoever claim to be Christian or even influenced by Christian thinking. The latter focuses on Christian as what it is ought to be and knows that all are sinners, including Christians, and I’d fall short of the standards the Bible uphold. Among these Christians, they know that the majority of Christians has gone astray.

KolenCh··on Ghostty 1.0
Search through the whole documentation, not just current page.
KolenCh··on Ghostty 1.0
Looks great. Switching from wezterm and kitty before that, and the native feel on macOS is indeed better than those emulated tabs. Also it is really quite zero config where to me it seems only one line of color theme is needed, together with setting TERM for ssh. (Default font happens to be mine.) Bundling so much color theme felt a bit waste of space, but it is nice that I don’t need to define them myself which also makes switching theme tedious. I also love to see it lands at 1.0 where features, documentation, and promotional materials seem polished already.

Chances are, it will become my default terminal in 2025!

The only nitpicking is the doc does not seem to have a search function. It would be great if I can search color or ssh and find the relevant page address that immediately, without navigating the tree or relying on external search engines.

KolenCh··on Python packaging: why we can't have nice things, part 1 – The Old Refrain
Interesting. I’m looking forward to your future posts.

One question I have in mind when dealing with PyPI vs conda is that why wouldn’t conda already solved all the problems with PyPI. My own answers to that includes 1) PyPI is still the de facto place to claim a namespace and it would be foolish to only release it in conda and not PyPI, and 2) there are many more packages available in PyPI and not conda, and the effort to just make all PyPI packages I used available in PyPI is too laborious.

Another issue is that many packages available on PyPI are not packaged well, from version constraints to setup.py madness to licensing issues. People managing the conda-forge channel is more gate keeping about those issues. I wonder even if the tooling around PyPI is improved, the community culture is much more difficult to be changed.

KolenCh··on Python packaging: why we can't have nice things, part 1 – The Old Refrain
There’s a chain of stuffs happening regarding packaging and the first step in the chain using PyPI limits what one can do to fix it. Conda starts with a different design in how to build a package and also how to bundle it. It specifically handles compiled code much better, because in that non pure Python case, compiler is needed at some point, and that is going to cause problems if you have no control over it. Conda essentially achieve reproducibility, with the necessary compiler considered as part of the dependencies. While in a certain sense of reproducibility can be achieved with PyPI packages, my understanding is that, eg with manylinux, it takes a hope for the best approach and can break in exotic cases, which should never happen with conda packages.

> The genesis of Conda came after Guido van Rossum was invited to speak at the inaugural PyData meetup in 2012; in a Q&A on the subject of packaging difficulties, he told us that when it comes to packaging, "it really sounds like your needs are so unusual compared to the larger Python community that you're just better off building your own"

From https://jakevdp.github.io/blog/2016/08/25/conda-myths-and-mi...

KolenCh··on Python packaging: why we can't have nice things, part 1 – The Old Refrain
Ok, thank you. It’s a pity for me as I think having a resource that explain the big picture of both would be great. But I don’t have enough of a big picture to write about it yet.

The thing about conda is that it is not just a tool, it is a completely different ecosystem to handle what you describe in your post. You touched on the topic of building & packaging, distribution, environment management, etc. And all these are different in conda land. To me the answer to the problems you mentioned in the post is conda. But I think it is commonly misunderstood and some people hate to use it. Not to mention needing to maintain publishing in both PyPI and a conda channel is laborious.

KolenCh··on Python packaging: why we can't have nice things, part 1 – The Old Refrain
Will your series at some point incorporate conda in the picture?
KolenCh··on Home Manager: Dotfiles Management
How home manager users handle the case of setting up the home directory on a system they have no control over, and therefore nix is not usually available?
KolenCh··on Singlefile: A web extension to save a complete web page into a single HTML file
But they are talking about gzipping base64 encoded data though.
KolenCh··on Conda: A package management disaster?
As far as the solver is concerned, there should be no difference as it has been streamed. But I personally can’t see a reason to go back, as mamba is supposed to be a drop in replacement of conda. I default to use mamba and switch to conda only when necessary. There are some cases mamba can’t handle correctly, such as the case where you want to roll back to an earlier revision: https://github.com/mamba-org/mamba/issues/803
KolenCh··on Why are UK electricity bills so expensive?
Good point. Not to mention the circulation will be all wrong had the radiator been cool instead of warm. (The primary means of heating by radiator actually comes from convection rather than radiation.)
KolenCh··on Conda: A package management disaster?
In this sense I personally prefer pixi because of this. It is pixi like but resolves using conda channels like conda, and similar to conda it supports PyPI packages via uv.

With a background in scientific computing where many of the dependencies I managed are compiled, conda packages gives me much more control.

P.S. I’d like to point out to others to differentiate between package index and package managers. PyPI is an index (that hosts packages in a predefined format) while pip, poetry, uv are package managers that resolve and build your environments using the index.

Similarly but a bit more confusingly, conda can be understood as the index, hosted by anaconda but can also be hosted elsewhere, with different “channels” (kinda like a GitHub organization) where conda-forge is a popular one built by communities. Conda is also a reference implementation of a package manager that uses anaconda channels to resolve. Mamba is an independent, performant, drop in replacement of conda. And pixi is a different one with a different interface by the author of mamba.

Even more confusingly, there are distributions. Distributions come with a set of predefined packages together with the package manager such that you just start running things immediately (sort of like a TeXLive distribution in relation to the package manager tlmgr.) there are anaconda distributions (if you installed anaconda instead of installing conda, that’s what you get), but also Intels distribution for Python, mini forge, mambaforge, etc.

KolenCh··on Conda: A package management disaster?
The situation has since changed as the solver is rewritten. (Upstreamed from the work done in mamba.) I encourage you to try again.
← PreviousPage 2 of 13Next →