1,431 karma · joined December 16, 2009
Contact: gergely@imreh.net
Based at http://gergely.imreh.net
Taipei Hackerspace co-founder https://taipeihack.org
[ my public key: https://keybase.io/imrehg; my proof: https://keybase.io/imrehg/sigs/Vqc30FUDkAzeR6b4jTTQown3Qjned_Xju7DqvvT6DfU ]
Definitely thinking about categorization differently (folders which are similar to this, tags, organic order by linking; bookmarks, notes, all are in the same boat as well).
Nice to see the Rust angle here, coming from Python I'm also trying to pick up that a lot more these days, and it's just quite a different mindset that needs more anchoring in real world use, like here, to stick.
Tooting my own (failed) horn, a similar, just one-challenge writeup from last year: https://gergely.imreh.net/blog/2022/09/programming-challenge...
Except last time I've checked, even the English version needed a Windows system with Chinese characters to run properly (not even any of the Traditional Chinese, say Unicode, but some specific one). Also, there are printed paper bits to file sometimes in the end.
Nonetheless it's still good to have something "blessed", and it's also evolving.
Funny part, now living in Taiwan, when the forms ask me for my "English Name", I never know it means "First Last" (which would be the opposite order for me of the original form), or merely "Name in Latin characters".
Names are hard.
I think it's useful to add as a note how to add credentials when running inside Docker since that mode is included right there in the readme (ie. adding a volume with the credentials).
As an idea, for the "Adding string descriptions" section, having stuff like dbt docs (or even just the manifest that generates) could supercharge the context.
Very interesting in general, definitely trying it out for our database!
I was looking at the project to spin it up with a BigQuery connection, but I couldn't find an example in the docs (expected it on this page: https://dataherald.readthedocs.io/en/latest/api.database.htm...), nor can I find it in the repo. Any hints on this? Otherwise I'm fine going source code diving. :)
On the other hand, looking at MS Team & Zoom desktop apps (in MacOS), there are some options that suggest that _some_ sort of USB hacking would be possible.
Teams > Settings > Devices > Sync Device Buttons; and Zoom > Preferences > Audio > Sync buttons on headset seem to be the key to enable that, but it's unclear just now what sort of device are they looking for.
I'm guessing it can be the "General Desktop Page 0x01" / "Call State Management Control 0x13" kind of device, as briefly mentioned in the blogpost, with the Call Mute Toggle 0xE1. It's a different kind of value (changes on a signal edge) rather than absolute like the button done here, and would need a bit more experimentation to send the right kind of signal. My initial checks are inconclusive.
Looking at MS Presenter+, it seems like the button behaviours can be customized by software, so I wonder how much of its behaviour is hardware devined. Also, it's Bluetooth, so likely the control mechanism is different from this USB-based control.
Thanks for the question, though, the rabbit hole goes deeper!
> 𝗍𝗋𝗎𝖾+2:𝖨𝗇𝗍 means “𝗍𝗋𝗎𝖾+2 has type 𝖨𝗇𝗍”, which makes even less sense, as the expression 𝗍𝗋𝗎𝖾+2 is nonsense and does not have any type at all.
Well, Python called where `True + 2` is actually an int and is equal to 3. :) This doesn't say anything about whether this should be the case or not, just whether it is.
Mentioned in the "What" section is the matching of the stated goal of the PR with the reality: so rare to find anyone who puts any effort into writing good PR messages (not to mention good commit messages, but that's for another time). Linking to tickets, explaining the rationale, most other context is just skipped because being assumed (or assumed to be communicated through some other channels). Less than ideal.
In general the "what you put in is what you get out" applies very well both for writing good changes and PRs as well as good reviews. Applying similar principles as the author listed, I have a lot of good feedback from people that the comments in reviews helped them improve and occasions for rare actionable feedback on technical know-how.
I frequently have the feeling as we well that I could get so much done if I didn't have to work -- this is not necessarily _correct_ (most likely it isn't), but aligns well with the case made here, so there _might_ be something to it to explore how to get more free (creative time), almost at all cost.
While checking past issues on the web, it's a bummer that they didn't work out how to typeset the actual Mathematica commands in the articles as text rather than the current image format. This prevents copy-pasting code, which can spend dozens of lines, making it harder to try them out. That's at least not the case in the Wolfram Comunity pages - I'm guessing those are just notebooks at their core, so much easier to do.
Originally I thought the keyboard printing would be a sticking point for them not to ship there, but now that I just tried to customise a Framework 16, they do have Traditional Chinese keyboard (the usual English + 3 Chinese writing system on each key), and that's great. The 13 doesn't have it (yet?) though.
Super fascinating to read the user experiences so far, really excited and cheering them on!
Langchain has a lot of surface area and it is still on 0.0.x, so also much advised to pin the versions hard.
Or am I missing something?
> [NOT AFFILIATED W/ THREADS BY INSTAGRAM] All in-one team communication platform
I bet they had a bump due to the name clash, and loads of support request too!
One thing that makes me wonder all the time are CI/CD runs that commit back to the repository. That feels weird for some reason I cannot articulate very well yet, but makes me think there should be a better way. Especially if the results are artefacts like that. Any experience by others to weigh in?
Cheers for highlighting it!
Cheers for the inspiration, an that now I crave scones...
The discussion of the reported issue on the OpenZFS GitHub is very educational, both on the maintainers side (useful, sensible), and the other commenters syde (kinda "thermonuclear")[1]
In the meantime I just downgraded the kernel as well, fortunately still had it in the pacman cache; then pinned it for now.
Sent an email to to the patch creator / signer-off too, to feed back on that "this is not expected to be disruptive to existing users." part of the commit (as in [0]).
[0]: https://lore.kernel.org/all/20221107170747.276910-1-broonie@...
(Used to run various geek events in Taipei, it has been a while since doing it though, so timely to get back into it!)