@jni@fosstodon.org
@jni@fosstodon.org avatar

jni

@jni@fosstodon.org

Core dev and co-founder napari.org
Core dev scikit-image.org
Co-author elegant-scipy.org
Research Fellow https://www.monash.edu/discovery-institute

Blog https://ilovesymposia.com

This profile is from a federated server and may be incomplete. Browse more on the original instance.

simon, to random
@simon@simonwillison.net avatar

I put together some detailed notes showing how I use Claude and ChatGPT as part of my daily workflow - in this case describing how I used them for a 6 minute side quest to create myself a GeoJSON map of the boundary of the Adirondack Park in upstate New York
https://simonwillison.net/2024/Mar/22/claude-and-chatgpt-case-study/

jni,
@jni@fosstodon.org avatar

@simon “I would miss them terribly if they were no longer available to me.” This seems like a good reason to focus on open source, local models, despite their limitations.

(But, as a side note, congrats on llm — the ergonomics are wonderful.)

RL_Dane, to KDE
@RL_Dane@fosstodon.org avatar

My music stopped playing, the phone rang, I answered it.
After speaking for a few minutes, I hung up and the music started up again.

It took me a second to realize that the music wasn't coming from my phone, it was coming from my computer.

is groovy, y'all. It pauses your music when a phone call comes in. (MPRIS-compatible player required, of course, but even mpv on the command line can do that)

jni,
@jni@fosstodon.org avatar

@RL_Dane @a this works really well with a New Zealand accent. 😃

vicgrinberg, to random
@vicgrinberg@mastodon.social avatar

What's your favorite reviews of "Three body problem"? Especially ones that focus on discussing the physics - both in terms of depiction of physics and physicists and in terms of how correct the books are?

Preferably of the Netflix series, but I'll also accept those for the books if especially good on discussing the physics.

Please link to them! Boosts welcome :)

#3BodyProblem

jni,
@jni@fosstodon.org avatar

@vicgrinberg @muellerwhh did you read the later books? I found each a slog (and the sexism in book 2 in particular is 🤢), but the core idea of book 2 is (I think) genius and original, and book 3 is just chock full of outstanding ideas that changed the way I look at the sky. My physics is only from popular physics books but I think I have the basics. (I also 🙄 at the quantum entanglement comms.)

neuralreckoning, to random
@neuralreckoning@neuromatch.social avatar

Early prototype of Mastodon thread viewer:

https://thesamovar.github.io/masto-thread-view/test.html

Just paste the URL of the thread into the box at the top and hit the "linear thread view" button below and it will give you a view of the thread with hierarchical replies sorted by how many engagements they got (reposts + favourites + replies).

It's very early days so it doesn't yet show any images, the design is not ideal, not optimised for mobile, etc. But I already find this useful for getting a feel of big threads.

My aim here is to give people a better way to navigate overwhelmingly large threads and to allow for a sort of archive of interesting threads. If we want to make Mastodon into a viable option for having scientific debates (e.g. alternative to peer review), we need some way to make them more accessible to outsiders and to surface the most interesting and relevant content.

So I'm particularly interested in hearing suggestions for features or other ideas on how to display threads in the context of long lasting discussions with some permanence to them.

At the moment it's just a very simple idea, but I have other ideas for how to display threads that are a bit wackier and I'll add these as extra buttons as and when I work on this. I'm also going to see how feasible it is to make this into a bookmarklet so you can just hit the 'render thread' bookmark in your browser and open a tab with this. Should be straightforward.

If you're interested, please feel free to post suggestions and issues either here or on github: https://github.com/thesamovar/masto-thread-view

May be of interest to @NicoleCRust @jonny

jni,
@jni@fosstodon.org avatar

@jonny @neuralreckoning I’d prefer if most work was client-side or upstream. Per-instance features are kind of annoying, eg coming across mathstodon posts where LaTeX won’t render unless I view them on mathstodon directly. The more the instances diverge, the harder it is for new folks to decide what to join.

Obviously, instances are a great place to experiment. But I just don’t want it to be the endgame…

albertcardona, to vim
@albertcardona@mathstodon.xyz avatar

It's 2024 and most text editors don't have a column-select mode. How can anyone live without it.

jni,
@jni@fosstodon.org avatar

@albertcardona @jzsimon @weberam2 Ah you are talking word processors not text editors. ;)

Interestingly I don’t think I’ve ever felt the urge to do this for prose or tabular forms…

jni, to random
@jni@fosstodon.org avatar

“I haven’t been bitten by Google killing an app or service since Google Reader, because I never again trusted them.”

👆 this.

(Also note: I am aware of Gruber’s terrible takes on the EU, and Threads, and a bunch of other stuff. But on this he is spot on. The Google Graveyard is so damaging to Google’s brand, it’s astonishing to me that (a) anyone still uses Google products, and (b) the execs still kill stuff rather than keeping it on life support just for reputation.)
https://mastodon.social/@daringfireball/112204358739653554

vicgrinberg, to Astro
@vicgrinberg@mastodon.social avatar

I just love this image - it highlights why we need all the different telescopes: each of them looks at the same object in different ways. And only when working together a complete image emerges.

Here, 's wide field is combined with 's zoom-in and sharpest IR image we ever obtained, allowing us to study how radiation interacts with interstellar matter.

More here and in the linked articles: ▶️ https://www.esa.int/ESA_Multimedia/Images/2024/04/Webb_captures_iconic_Horsehead_Nebula_in_unprecedented_detail

jni,
@jni@fosstodon.org avatar

@vicgrinberg @napari thanks! yeah re:raw data I presume there are Python libraries to get NumPy arrays from the raw files, and hopefully not too hard to use as a domain novice! 😅😬😬😂

jni,
@jni@fosstodon.org avatar

@vicgrinberg Absolutely gorgeous.

Do you know how one can get their hands on the raw, full-res images? I'd love to overlay them all in @napari as a demo... 🙏

hynek, to random
@hynek@mastodon.social avatar

WAHHHHHHH

jni,
@jni@fosstodon.org avatar

@hynek did you post this as a warning so others on that flight can make alternate arrangements? 😂

amcasari, to opensource
@amcasari@hachyderm.io avatar

I have a talk looking for a home.

HMU if you have CFPs still open and want to hear a nerd question their entire professional data + software reality.

jni,
@jni@fosstodon.org avatar

@amcasari @janeadams I only just heard about @bangbangcon. Could be a fit?

jni, to random
@jni@fosstodon.org avatar

It’s 2023 and Apple still hasn’t surpassed the 2002 iMac G4 design (the desk lamp). (And the commercial is still one of their best too!) https://www.theverge.com/23808948/imac-twenty-five-visual-history-apple-design-models

jni, to random
@jni@fosstodon.org avatar

Is there a comprehensive archive, with references, of Elsevier's many sins against scientific progress? @albertcardona @brembs The lead authors of a paper I played a small role in want to submit to Cell 🤢 and I would like to dissuade them.

Follow-up Q: I have a vague vibe that, although the entire traditional publishing system needs to die in a fire, NPG are not quite as scummy as Elsie. Is that vibe justified or not really?

jni,
@jni@fosstodon.org avatar

@jonny @albertcardona @brembs Indeed the basic “all ‘legacy’ journals must die” premise is assumed correct in the question and includes NPG, but I am trying to be realistic about how much I can get in this battle — the authors list is overwhelmingly traditionalist and wary even of preprints. 😞 I'll push for eLife but even that is a stretch. Is it worth steering from Cell to Nature & co? I think so, but I'm worried it's just exposure bias. Profit margins at least seem lower at NPG. (26 vs 38%)

jni, to random
@jni@fosstodon.org avatar
jni, to scipy2023
@jni@fosstodon.org avatar

Cool, my @SciPyConf @scipy2023 talk on @napari is up! 🥳 With great thanks to the organisers. 🙏❤️ Check it out here:

https://youtu.be/s362FTRzdg0

As a reminder, the repo and slides can be found here:

https://github.com/jni/napari-talk-scipy-2023

though you'll need to adapt it to use your own data to run much of it.

Finally, check out all the other amazing talks from the conf! Really looking forward to catching up with the tracks I missed (bc conflicts or frantically preparing for my own talk 😂)

https://www.youtube.com/@SciPy-Conf/videos

jni, to SciPy2024
@jni@fosstodon.org avatar

Ok so @SciPyConf @SciPy2024 call for proposals is open (🥳) at https://cfp.scipy.org/2024/cfp! (See also https://www.scipy2024.scipy.org/#CFP for more info.) This year we have a data viz and image analysis track I'm co-chairing and WE NEED YOU 🫵!

This also means I must finally share the 3D scan of The Claw from last year's @scipy2023 lightning talks! You can download it in DICOM or ome-@zarr format from FigShare at https://dx.doi.org/10.26180/25264588! 😂🦀

image/png

jni, to random
@jni@fosstodon.org avatar

@protonmail wow I have just logged in to my account after briefly playing with it back in 2015, and you guys have come a long way. 😍

I have a couple of Qs:

  1. Is snooze coming to the mobile apps in the near future? I tend to snooze more when on mobile, not less.
  2. Is there an API one could hook into to snooze emails? I basically want to snooze all emails in my inbox to random times in the next three months, so I don't get them all at once again.

Anyway, congrats on all the progress! ❤️🚀

jni, to random
@jni@fosstodon.org avatar

Every now and then a kind soul upvotes an old answer of mine on SO, and it turns out to be a fun read years later. 😃

https://stackoverflow.com/q/62135639/224254

"For some reason, my mcp object didn't have an offsets attribute […] but searching the source code shows me that offsets is defined with the skimage.graph._mcp.make_offsets function. So I did a bad thing and imported from that private module, so I could claim what was rightfully mine — the offsets list […]"

Can we please start writing papers like this?

jni, to random
@jni@fosstodon.org avatar

I love the “reveal” at 2:25. 😍
https://fosstodon.org/@napari/112256746427753857

jni, to random
@jni@fosstodon.org avatar

Reminder to turn on full text search for your posts in Preferences > Public Profile > Privacy and Reach > (Search section) > Include public posts in search results! This goes a long way towards making Mastodon interactions less ephemeral.

Screenshot of mastodon preferences UI showing "Public Profile" and "Privacy and reach" selected.

jni, to random
@jni@fosstodon.org avatar

@simon ok I'm finally trying out your llm library and it's neat. I am intentionally limiting myself to open source and local models. Could you write a post about these? Specifically:

  • strengths/weaknesses of each model?
  • strategies for getting around context length limits? I'm currently interested in text summarisation and immediately hit this with mistral-7b-instruct-v0*
  • yes I did ask mistral itself to suggest strategies but they were distressingly manual and brittle/non-generalisable. 😂
jni, to random
@jni@fosstodon.org avatar

The fun thing is that this Sydney Brenner quote is from 2014, and things have gotten even worse in the past decade. 🙃
https://qoto.org/@enroweb/112150171468524351

jni, to scipy2023
@jni@fosstodon.org avatar

@SciPyConf @scipy2023 sprints day 2: hooking up fastplotlib canvas to @napari ViewerModel (+Dims, LayerList, and Layer, but that came basically free 😍) with @caitlinllewis and @kushal = prototype browser/notebook front-end for napari. Probably most elated I’ve ever felt in a sprint. 😭😍😍😍

image/jpeg

jni, to random
@jni@fosstodon.org avatar

Any updates on the new open source licensing terms that Prusa was exploring in

https://blog.prusa3d.com/the-state-of-open-source-in-3d-printing-in-2023_76659/

?

If I want to buy an open source, forever repairable, hackable 3D printer today, is Prusa the best choice or…?

CC @stargirl

jni, to random
@jni@fosstodon.org avatar

Hey advocates! Ever had the feeling that you wanted to download all of your open science / research software thoughts into a collaborator/grad student's brain, but your regular coffee catchups aren't cutting it??? Just point them to this new from the awesome @madicken!

https://urssi.us/blog/2024/05/20/applications-now-open-for-a-summer-school-in-open-science--research-software-engineering/

I have seen Madicken teach and chair sessions and it's going to be awesome I want to goooooo! 😭😭😭😭

  • All
  • Subscribed
  • Moderated
  • Favorites
  • JUstTest
  • GTA5RPClips
  • DreamBathrooms
  • InstantRegret
  • magazineikmin
  • osvaldo12
  • Youngstown
  • ngwrru68w68
  • slotface
  • everett
  • rosin
  • thenastyranch
  • kavyap
  • tacticalgear
  • megavids
  • modclub
  • normalnudes
  • cubers
  • ethstaker
  • mdbf
  • Durango
  • khanakhh
  • tester
  • provamag3
  • cisconetworking
  • Leos
  • anitta
  • lostlight
  • All magazines