xk3's recent activity

  1. Comment on What programming/technical projects have you been working on? in ~comp

    xk3
    Link
    Two projects this week have provided moderate enjoyment: One is a geospatial search tool. The bones are finished but could use a lot of tweaking to figure out the right sorting. Two of the cool...

    Two projects this week have provided moderate enjoyment:

    One is a geospatial search tool. The bones are finished but could use a lot of tweaking to figure out the right sorting. Two of the cool things are the ability to search places directly from PBF files (without indexing first). It's slower but needs a lot less space than Bleve. The second cool thing is a similar idea but pmtiles.

    PBF is the most complete data to seach because it's the full OpenStreetMaps database and still it is surprisingly fast for small areas!

    PMTiles are a much smaller subset of data but they still scale up to larger areas because you can use their built-in spatial index.

    https://github.com/chapmanjacobd/poisearch

    The other thing is a tool to make global satellite image pmtiles maps from the ESA's preaggregated quarterly mosaics. The code is much smaller but almost equally satisfying:

    https://github.com/chapmanjacobd/satmaps/

    1 vote

  2. Comment on Shawn Phillips - Rumplestiltskin's Resolve (1976) in ~music

    xk3
    Link
    interestingly... I think they misspelled Rumplestiltskin's name... but I'm sure it counts. I'm sure they would have pronounced it perfectly well ergo they probably could get the baby back, etc....

    interestingly... I think they misspelled Rumplestiltskin's name... but I'm sure it counts. I'm sure they would have pronounced it perfectly well ergo they probably could get the baby back, etc. Probably couldn't be bothered to go to a library and loan a dictionary as they were too busy making music!

    1 vote

  4. Comment on George Benson - I Always Knew I Had It In Me [Version 2] (1977) in ~music

    xk3
    Link
    Version 1 is also pretty good (and completely different)! But listening to Version 2 first seems to lend to a more powerful experience

    Version 1 is also pretty good (and completely different)! But listening to Version 2 first seems to lend to a more powerful experience

  8. Comment on What programming/technical projects have you been working on? in ~comp

    xk3
    Link Parent
    It should be similar to wireguard. That is, it depends how you set it up. If you want to map a specific port to the open internet it's a similar situation either way. The difference is just...

    It should be similar to wireguard. That is, it depends how you set it up. If you want to map a specific port to the open internet it's a similar situation either way.

    The difference is just connecting between your server and your home computer.

    Tailscale does have some products (some are free) for exposing to open internet without your own server https://tailscale.com/docs/features/tailscale-funnel --but you don't need to use them. You can use Tailscale like wireguard because it is running wireguard under the hood but with hole punching and other features which make it easier and more robust for many scenarios like not having a static IP address.

    You could just do that final part of VPS to public port via reverse proxy like Caddy, HAProxy, Nginx. I think... maybe it is slightly more complicated:

    https://old.reddit.com/r/teamspeak3/comments/fm39qx/nginx_reverse_proxy_for_teamspeak/

    https://old.reddit.com/r/Tailscale/comments/1cskoya/port_forwarding_public_vps_local_machine/

    Well for low latency stuff you probably want as few programs streaming the data so it sounds like wg-easy will be a better fit

    1 vote

  9. Comment on What programming/technical projects have you been working on? in ~comp

  10. Comment on What programming/technical projects have you been working on? in ~comp

    xk3
    (edited )
    Link Parent
    I tried it out for a few months and to be honest... not too impressed. Something like qbittorrent-nox (that is, no X11, no GUI) or transmission-cli is a lot more responsive and scalable. I've...

    rtorrent

    I tried it out for a few months and to be honest... not too impressed. Something like qbittorrent-nox (that is, no X11, no GUI) or transmission-cli is a lot more responsive and scalable.

    I've slowly built up a collection of qbit scripts that make managing thousands of torrents across different servers a breeze (...once I remember the script name!):

    library torrents-info does a lot of heavy lifting and it's my main interface into qbit:

    parallel server.ssh {} lb torrents $argv ::: (servers.ssh.connectable $servers)

    This uses GNU Parallel, fish functions, lb torrents, also see allocate_torrents

    nb. if you use these you'll need rb_libtorrent-python3 on Fedora, in Arch the correct python bindings are already included in the main libtorrent-rasterbar package (other distros might be even different but most likely installing via package manager will be very easy once you locate the correct one)

    There's also these scripts which I mostly only use in an automated way (if you search the script name in my computer repo you might find a reference in daily.fish or weekly.fish, etc):

    https://github.com/chapmanjacobd/computer/blob/main/bin/ (see the qbt_* and torrent_* scripts)

    Let me know if you see anything confusing, or if you have a specific problem, and I can try to explain how I would deal with it.

    There are a lot of other qbittorrent CLI options too so you aren't limited to my scripts... I just have less experience with them--but if you don't like what I have, I'd still encourage a deeper look at qbittorrent!

    2 votes

  11. Comment on Yonatan Gat - American Quartet (2022) in ~music

    xk3
    Link

    Playing with Curt, Yonatan and Greg is a dream come true. This is the dream team group to go on this journey of learning, memorizing, and then eventually deconstructing a string quartet to be played at very loud volumes. Although we only performed the piece live together a handful of times each time was different from the last and was better and better until finally we got to record it. Recording took place in a day. We setup all together in room and played each movement once or twice. Very little overdubs and editing occurred on this record as we wanted to capture the piece as the band intended, raw and live. Mixed beautifully by Greg Saunier.

    Mikey Coltun

    Brahms would instruct those brave enough to seek his compositional advice that no matter how long or sectional a piece of music it must be “singable” from beginning to end. Dvorak was one of those seekers and my feeling is that Yonatan in turn absorbed the lesson by memorizing the entire melodic thread to Dvorak’s American Quartet before contacting Mikey (who then brought me in and, finally, Greg) about filling in the accompaniment— at times doing more— always arranging our parts in a manner that suited the instrumentation of a surf rock quartet. After a few sessions in Yonatan’s Joralemon St attic studio it was too late to turn back from the task at hand. Once we had the notes in place it was Greg’s motivic and melodic approach to the drum part that added the final combustion. The arrangements were still somewhat fluid until the recording date— we were finding new ways for the drums to enter the counterpoint, for Mikey's bass or my organ to take those crucial melodic turns. I still find it so refreshing how we were able to interact with the string quartet score, hewing closely to it at times and at other times taking the opposite improvisational approach, which we really developed by performing the quartet at music festivals in NYC and Canada for a beautiful brief period of time. In those settings we had no choice but to make the music and the arrangement totally convincing.

    One thing that I would like for listeners to understand is that these are live studio recordings. Yes there are overdubs here and there to add certain colors but really what is being presented is faithful to how we played it that night in Big Indian, NY. Old school classical (before tape splicing or digital editing) and punk ethos therefore converge. Most early attempts to reconcile European classical to American/British rock music failed because of prog tendencies which rendered the musics to alike to one another and therefore effectively neutered each to each. I think like Dvorak before him Yonatan with this project has truly tapped into something of the American mythology where ambition is all, nothing is sacred, and competition is fierce. Like Doc Holliday taking in a Shakespeare production in Leadville, CO. Or the tragic career of Junius Brutus Booth. Knut Hamsun. Bob Marley. Mark Rylance. Those who got lost here for a while, but made their way back out.

    Curt Sydnor

    The promotion of favorite European composers to God status by 20th Century academia and the marketplace have not done those composers any favors. Treating their work as sacred objects to be revered, rather than living provocations to inspire play, has robbed them of any emotion other than gravitas and any mood other than somber. Sorry but Dvorak was a human being like the rest of us, capable of humor, capable of revolutionary spirit, capable of violence. For the four of us, playing the string quartet as we would play a rock song did not deface the piece, but revealed it to be something bigger than the stifling box into which it normally gets put. As we struggled to wrap our heads and bodies around it, the piece kept yielding new surprises and ever more complicated feelings. My bandmates surprised me not by straying from the piece but by plumbing its depths.

    Greg Saunier

  14. Comment on Suggest media in which the antagonist is an idea or an abstract concept rather than a person or intelligent entity in ~talk

  15. Comment on What programming/technical projects have you been working on? in ~comp

    xk3
    (edited )
    Link
    I made a small alternative to mergerfs because I have had traumatic experiences with FUSE in the past and just wanted something simpler: https://github.com/chapmanjacobd/nofs I think the most...

    I made a small alternative to mergerfs because I have had traumatic experiences with FUSE in the past and just wanted something simpler: https://github.com/chapmanjacobd/nofs

    I think the most interesting part is playing around with strict clippy settings for working with LLMs can be a mixed bag of usefulness. Some settings really do help and it is important to set "deny" so that the LLM doesn't make the temptation to write an #[allow(lint_name)], etc. But a lot of linting settings don't actually help much to guide LLMs to produce high quality code--it just forces them to make that low-quality code look nice :-)

    Another thing that I learned is that the CLI parser clap can silently (no build errors) overwrite variables if the positional arg and optional flags have the same name [eg. --paths and <paths> ...].

    I'm doing a similar thing now in golang via golangci-lint (this golden config was really helpful as there are way too many knobs to turn) and similar observations... cognitive complexity linters help a lot if you can strictly enforce them. Unfortunately, go doesn't have a "deny" setting but I've noticed that Qwen Coder doesn't want to lean on //nolint nearly as much as #[allow...] for whatever reason. So thankfully it's not a big problem!

    2 votes