@juanluisback Twitter archive

RT @timeyoutakeit: But does GitHub Copilot write your docs for you? 🤔

 Wed Jun 30 22:54:06 +0000 2021

Thoughts on trio vs asyncio? I found this by @vorpalsmith eye-opening https://vorpus.org/blog/notes-on-structured-concurrency-or-go-statement-considered-harmful/ and, after a few projects with asyncio, learned to hate unstructured concurrency.

 Thu Jul 01 08:08:32 +0000 2021

FLOSS people: "GitHub Copilot is bad! This is derivative works but IANAL! Copyleft everything!"
Also FLOSS people: "Wow look at that, issue forms, Codespaces, VS Code plugins, GH Actions, sign me up 😍"

 Thu Jul 01 08:27:03 +0000 2021

I'm not saying that this doesn't matter. I'm not even picking sides because I don't bother.

We have collectively surrendered to GitHub a long time ago. They have a monopoly, 99.99 % of people don't care, and the company will get away with anything they do.

 Thu Jul 01 08:27:04 +0000 2021

Sorry folks, but if we didn't want to give all the power to a single platform...

We should have looked more seriously at the alternatives instead of falling for every shiny new feature.

 Thu Jul 01 08:27:04 +0000 2021

Anyway, just in case it was not clear: I'm not trying to being super intellectual here, or point out "you're wrong, I'm right".

I'm just annoyed that everybody on my TL is debating about "copyleft and GH copilot" when most likely this will have ZERO practical consequences.

 Thu Jul 01 08:27:05 +0000 2021

Also: I am sure copyleft licenses are breached every single day, but the FSF and the "software freedom fighters" have been too busy justifying the abhorrent behavior of the biggest asshole in the history of computing, instead of doing any meaningful enforcement or research work.

 Thu Jul 01 08:27:05 +0000 2021

RT @willmcgugan: Feeling less positive about Github Copilot. It suggests stuff that looks correct on first glance, but is subtly wrong. It…

 Thu Jul 01 14:45:27 +0000 2021

RT @nicholdav: Please help us find mentors for #SciPy2021 sprints 🙏

Esp if you're a maintainer with beginner-friendly issues. Help us help…

 Thu Jul 01 18:28:25 +0000 2021

RT @choldgraf: Happy Saturday all ✨ enjoy a new version of Sphinx Copybutton, which updates the button UI/UX to take inspiration from GitHu…

 Sat Jul 03 18:50:19 +0000 2021

RT @bphogan: Well this blew up.

Good.

I have a SoundCloud and books and whatnot I could promote here. But instead I just want you to sp…

 Sat Jul 03 21:54:56 +0000 2021

Dear students, please don't do this:

true = False

 Sat Jul 03 22:57:30 +0000 2021

Disney predicted Artificial Intelligence in 1940

 Sun Jul 04 09:57:55 +0000 2021

 Sun Jul 04 10:02:39 +0000 2021

https://sourceforge.net/p/docutils/bugs/419/ 🙈

 Sun Jul 04 14:46:45 +0000 2021

Congrats Daniele, well deserved 👏🏼

 Mon Jul 05 15:00:26 +0000 2021

RT @SciPyConf: Don't miss a session -- subscribe to the #SciPy2021 Google calendar here http://ow.ly/qvId50FdPck

 Mon Jul 05 15:05:33 +0000 2021

RT @choldgraf: The Executable Books / Jupyter Book team had our first distributed team meeting last week! We hope to use these meetings for…

 Mon Jul 05 17:38:15 +0000 2021

Si buscas "COVID-19" en @Google todavía te muestra información desactualizada y engañosa, supuestamente de @WHO.

No, las gotículas que transmiten el COVID-19 no caen al suelo rápidamente. La principal forma de transmisión es por aerosoles en suspensión.

Enlace más abajo 👇🏼

 Mon Jul 05 17:55:59 +0000 2021

"Ten scientific reasons in support of airborne transmission of SARS-CoV-2" en @TheLancet https://doi.org/10.1016/S0140-6736(21)00869-2

 Mon Jul 05 17:56:00 +0000 2021

RT @TerraMeijar: My career took off once I started attending conferences. I remember saving for almost a year for one in particular where a…

 Mon Jul 05 17:58:58 +0000 2021

Well, if only push notifications on Calendar worked, or at least if there was someone triaging issues... https://github.com/nextcloud/calendar/issues/3000

 Mon Jul 05 20:07:23 +0000 2021

RT @david_perell: Read to collect the dots, write to connect them

 Mon Jul 05 21:31:14 +0000 2021

🤬 Now that they don't have "the former guy", they think their country is perfect.

 Tue Jul 06 06:40:53 +0000 2021

I suffered badly during the worst months of the pandemic seeing our national health system collapse. In Madrid, doctors and nurses are underpaid and have terrible working conditions.

I am glad the pandemic is almost over. But poor management and harmful politics isn't.

 Tue Jul 06 07:07:27 +0000 2021

Having my (first) COVID-19 #vaccine tomorrow, and thinking a lot about my friends from other countries less privileged than mine.

@EUCouncil @EU_Commission is still sabotaging the #TRIPS temporary waiver for vaccine patents at @wto, which is an absolute shame.

 Tue Jul 06 07:07:27 +0000 2021

I want to do more, but I'm powerless. I'm just a random guy. Institutions and companies now react to social media pressure by proclaiming small symbolic gestures, while people still suffer and often die. There's so much work to do, and yet we keep wasting our time and energy.

 Tue Jul 06 07:07:28 +0000 2021

Two months ago I decided to step back, cool down, and spend less time reading the news. My mental health has improved considerably.

I just wish they lifted the fucking COVID-19 patents. It's cruel. It's unfair. It's dystopian. It's madness.

 Tue Jul 06 07:07:28 +0000 2021

It was said that the pandemic would change us. New economies would flourish. We would take sustainability seriously.

But corporate greed is still the main driver of European and Spanish politics. There is no public sector anymore, all the work is being done by "the Big Four".

 Tue Jul 06 07:07:28 +0000 2021

Sending my best wishes especially to my dear friends in Argentina, Brazil, Colombia, Namibia, Nigeria. Keeping Palestine in my thoughts every day.

Stay safe everyone. This will be over soon.

Hugs.

 Tue Jul 06 07:07:29 +0000 2021

Indeed. It breaks my heart that, in the middle of this situation, there's also people refusing to take the vaccines. Happy that you're getting your shot soon too.

 Tue Jul 06 08:42:09 +0000 2021

RT @readthedocs: Learn more about Read the Docs directly from its team members: check out this summary of our appearances in the media duri…

 Tue Jul 06 16:10:46 +0000 2021

RT @ericholscher: We've been adding various guides to the RTD docs and working on an official Sphinx tutorial. We've worked with the contri…

 Tue Jul 06 17:45:42 +0000 2021

RT @reydelhumo: "Any news on this?" https://twitter.com/alexellisuk/status/1349445747868127237

 Wed Jul 07 06:57:51 +0000 2021

*Narrator: Juanlu wanted to read Lawful Good papers, but he only encountered Chaotic Evil ones* https://twitter.com/JBorrow/status/1412341091551039488

 Wed Jul 07 07:34:03 +0000 2021

I confess I muted dabeaz some time ago. I'm just tired of the endless negativity towards what Python is today.

 Wed Jul 07 09:07:33 +0000 2021

RT @ralfgommers: If you are stuck in a place where you need to deal with HPC clusters running something like CentOS 6 (past its EOL date) -…

 Wed Jul 07 12:47:15 +0000 2021

And if I wanted to do `from sympy import cos` instead? ;)

 Wed Jul 07 16:55:37 +0000 2021

RT @Mbussonn: Want to speedup time to first plot in @IPython_dev/@ProjectJupyter ? Tired of importing things by hand ? Have a look at how…

 Wed Jul 07 17:56:47 +0000 2021

¡Nos hemos vacunado el mismo día!

 Wed Jul 07 23:16:07 +0000 2021

RT @Senficon: #GitHubCopilot is not infringing your #copyright: The #FreeSoftware community should not adopt a maximalist copyright stance…

 Thu Jul 08 08:44:24 +0000 2021

"That is why it is so absurd when copyleft enthusiasts argue for an extension of copyright. Any extension of prohibition rights not only strengthens the enforcement of copyleft licences, but also the much more widespread copyright licences, which aim to achieve the [...]opposite" https://twitter.com/Senficon/status/1411997803098914819

 Thu Jul 08 08:53:16 +0000 2021

RT @lgworld: High time to talk about Audacity and privacy in a less heated way, eh?

https://librearts.org/2021/07/audacity-privacy/

 Thu Jul 08 08:58:06 +0000 2021

"Expecting people to not use any proprietary technology and judging people for not meeting this standard pushes people away from our movement. People who are coerced into using proprietary technology deserve our empathy and invitation into our movement, not condescension."

❤️

 Thu Jul 08 09:28:21 +0000 2021

https://cooperativetechnology.codeberg.page/

 Thu Jul 08 09:28:22 +0000 2021

A mí Moderna no me pegó tan duro pero estoy "hecho bolsa" un poco jeje, ¡ánimo que en dos días se pasa!

 Thu Jul 08 19:35:36 +0000 2021

RT @Senficon: Apple founder Steve Wozniak backs #RightToRepair movement: https://www.bbc.com/news/technology-57763037

 Thu Jul 08 20:46:34 +0000 2021

RT @readthedocs: In our freshly published July newsletter, you will find our company highlights, what new features we have shipped during t…

 Thu Jul 08 20:59:36 +0000 2021

Aquí lo que falla como siempre es la comprensión lectora. El ministro ha recomendado "comer menos" y algunos han escuchado "no comer". Con este nivel no se puede hablar de absolutamente nada.

 Thu Jul 08 21:46:15 +0000 2021

RT @jacobian: FWIW I'm neither "for" nor "against" Copilot -- and in fact think drawing battle lines that stark is a massive part of the pr…

 Thu Jul 08 21:50:38 +0000 2021

Looks like several people are attempting similar things at different places https://twitter.com/juanluisback/status/1413067423847178240

 Fri Jul 09 07:01:04 +0000 2021

RT @wuoulf: Here is a summary of the latest mamba release – let us know if package resolutions are improving for you! This might be the las…

 Fri Jul 09 13:34:23 +0000 2021

Thanks @tk0miya and the Sphinx development team! 🙌🏼 https://twitter.com/tk0miya/status/1414275310162485251

 Sun Jul 11 17:45:48 +0000 2021

I'll be all week at #SciPy2021. If you want to chat about scientific documentation, open-source, @readthedocs , or just say hi, reach out in Slack or here! 🚀

 Mon Jul 12 14:04:59 +0000 2021

Starting the tutorial on @ProjectJupyter widgets! 🎥

 Mon Jul 12 14:05:29 +0000 2021

RT @poliastro_py: Or Astrodynamics 🤓 https://twitter.com/juanluisback/status/1414586590329024512

 Mon Jul 12 14:13:43 +0000 2021

😉 https://github.com/scipy/scipy/issues/14390

 Mon Jul 12 14:14:07 +0000 2021

RT @numpy_team: Last year’s inaugural community survey provided invaluable feedback from the NumPy community. We'd love your input once aga…

 Mon Jul 12 17:08:43 +0000 2021

Unpopular opinion: I am not excited about billionaires going to (almost) space.

I would be excited about spacefaring nations agreeing on a peaceful (hence non-commercial), collective (hence non-competitive), sustainable (hence non-capitalist) use of space instead.

 Tue Jul 13 07:39:04 +0000 2021

RT @jtpio: JupyterLite: Jupyter ❤️ WebAssembly ❤️ Python

https://blog.jupyter.org/jupyterlite-jupyter-%EF%B8%8F-webassembly-%EF%B8%8F-python-f6e2e41ab3fa

 Tue Jul 13 09:46:30 +0000 2021

The most "popular" opinion, around me at least, seems to be that we *ought* to exploit space resources for commercial purposes, because not doing it is a barrier for our civilization, a wall to our "progress".

 Tue Jul 13 13:08:58 +0000 2021

Tsiolkovsky said that we would "leave the cradle" one day. But he didn't say that we have to put a McDonalds on the Moon.

 Tue Jul 13 13:10:12 +0000 2021

RT @AdamChainz: ✍️ “The Boolean Trap” is a programming anti-pattern where a boolean argument switches behaviour, leading to confusion. In t…

 Tue Jul 13 13:13:06 +0000 2021

RT @brunosan: ❤️SO excited about this. We are working really hard to put the best geospatial datasets, OSS and compute. This is a huge leap…

 Tue Jul 13 18:13:26 +0000 2021

I don't have any certifications, but I think they give you a clear goal and usually a deadline, which are great elements to get motivated. In other words: the usefulness of certifications is the preparation, rather than the credential itself, IMO.

 Tue Jul 13 18:31:40 +0000 2021

Not really. It's proof for yourself 🙂 And to brag on LinkedIn 😛

 Tue Jul 13 21:36:34 +0000 2021

I mean applying the ISS philosophy to *everything* that goes beyond geosynchronous orbit

 Wed Jul 14 06:58:07 +0000 2021

So much for Big Data and Artificial Intelligence if humans have to manually clean browsing history.

 Wed Jul 14 06:59:49 +0000 2021

On a related note, I lost count of how many people connect with me on LinkedIn or other platforms where my name is clearly visible and open with "Hola Jose Luis" 😒 https://twitter.com/freakboy3742/status/1414403854456684546

 Wed Jul 14 07:01:23 +0000 2021

¡Jaj! Voy a compartir esto en LinkedIn, "la red social de concreto" 🤣🤣

 Wed Jul 14 07:02:13 +0000 2021

Congrats! 🙌🏼

 Wed Jul 14 07:07:06 +0000 2021

Long story short: SciPy started depending on Pythran recently for parts of scipy.signal and scipy.optimize. Pythran transpiles Python to C++ ahead of time! You can read the wonderful announcement of the 1.7.0 release by @Tyler_Reddy here https://mail.python.org/pipermail/scipy-dev/2021-June/024896.html

 Wed Jul 14 08:11:56 +0000 2021

The tale of a circular build time dependency in SciPy https://github.com/scipy/scipy/issues/14335 I love this thread so much, for various reasons! Keep reading 🧵

 Wed Jul 14 08:11:56 +0000 2021

Pythran uses directed graphs https://en.wikipedia.org/wiki/Directed_graph to detect cycles in its transpilation process. Pythran 0.9.11 depended on NetworkX, a wonderful package for graph manipulation and network analysis.

However... and here comes the juicy part 🍿

 Wed Jul 14 08:11:57 +0000 2021

In particular, SciPy 1.7.0 depended on pythran==0.9.11. The reason is that, so far, Pythran was not following semantic versioning, and new features were often introduced in micro releases, as @ralfgommers pointed out here https://github.com/serge-sans-paille/pythran/issues/1838

 Wed Jul 14 08:11:57 +0000 2021

Therefore, the situation was that SciPy 1.7.0 depended on Pythran 0.9.11, which in turn depended on NetworkX, which in its newest version depended... back on SciPy.

Isn't it ironic that a network analysis library introduces a cycle in the dependency graph? 🤓

 Wed Jul 14 08:11:58 +0000 2021

NetworkX was a pure-Python library, and some parts of it, like their PageRank implementation, "were often so slow that it was painful to run", as @JarrodMillman puts here https://github.com/networkx/networkx/pull/4967#issuecomment-877338272

Therefore, NetworkX 2.6 switched to faster versions... by depending on SciPy! 💥

 Wed Jul 14 08:11:58 +0000 2021

However, a macOS user that wanted to build SciPy from conda surfaced this issue, which could also affect High Performance Computing users, and started a chain reaction ⚙️ https://github.com/scipy/scipy/issues/14335#issuecomment-876976102

 Wed Jul 14 08:11:59 +0000 2021

Luckily for everyone, this only affected a subset of users: those willing to build SciPy from source. As there are wheels for pip in many architectures and Python versions, and conda only installs binaries, the issue went unnoticed in the CI systems of SciPy. 🥷🏼

 Wed Jul 14 08:11:59 +0000 2021

And finally, Pythran decided to trim its dependency tree and be as slim as possible by vendoring a custom implementation of directed graphs, as well as removing some other ancillary deps https://github.com/serge-sans-paille/pythran/pull/1839 thanks Serge Sans Paille for the blazing fast response time 😍

 Wed Jul 14 08:12:00 +0000 2021

(By the way, thanks @pypi for improving the usability of the platform so much in recent years, we don't appreciate your efforts enough! 🙌🏼)

 Wed Jul 14 08:12:00 +0000 2021

At this point, the only options were to release SciPy 1.7.1 with a fix or yanking the problematic NetworkX releases. Luckily again, @JarrodMillman and the rest of the NetworkX team were super receptive, and NetworkX 2.6.0 and 2.6.1 were quickly yanked from @pypi 🏃🏼‍♂️

 Wed Jul 14 08:12:00 +0000 2021

Thought-provoking question by @ralfgommers : "Perhaps the better solution is to stop distributing sdists altogether." https://github.com/scipy/scipy/issues/14335#issuecomment-877061819

As I said some months ago, "I would love to move to a wheels-only world" https://twitter.com/juanluisback/status/1387700243202969603 Could we make sdists opt-in in pip?

 Wed Jul 14 08:12:01 +0000 2021

Anyway, despite the stressful timeline and the potentially disruptive consequences, everybody reacted very quickly, all emergency procedures worked, and the conversation stayed civil at all times. Kudos to everybody 👏🏼

To close this thread, I'd like to share two more things:

 Wed Jul 14 08:12:01 +0000 2021

We have seen this in the past (@pradyunsg and @brainwane have suffered this with pip *a lot*): are development releases worth it? They impose a significant maintenance burden, and apparently they don't get tested enough. Our automated systems should help us with this.

 Wed Jul 14 08:12:02 +0000 2021

And more food for thought: NetworkX "intentionally had release candidates up for longer than usual for this release hoping to gather some feedback on this particular change" https://github.com/scipy/scipy/issues/14335#issuecomment-877085893 and yet, this problem went unnoticed until the final release.

 Wed Jul 14 08:12:02 +0000 2021

If you have read this far, please leave a comment! I'd like to connect to people that also are passionate about weird things, like open-source scientific packaging 🥰

Pay the maintainers, be kind (all else is details, as @gvwilson said), and happy coding everyone!

 Wed Jul 14 08:12:03 +0000 2021

I think that's about the only advantage, but I concede it is quite needed, and also highly entertaining 😃 You put me in a difficult compromise, but if I had to pick between dinosaurs and Musk fanboys... I would stand with the fanboys. At least they bring joy and excitement.

 Wed Jul 14 09:18:06 +0000 2021

RT @pradyunsg: @juanluisback @brainwane One of the things with release candidates is that it doesn't matter how long you keep them up, if n…

 Wed Jul 14 09:27:01 +0000 2021

RT @esc___: @pradyunsg @juanluisback @brainwane At @numba_jit we have this issue too! To help us wit that, we also have https://t.co/jeAyxm…

 Wed Jul 14 10:36:30 +0000 2021

Absolutely, sdists are essential and they cover many use cases that binaries will never do.

However, sdists are opt-out at the moment (pip install --only-binary), and I think it would be cool to make them opt-in instead, when the ecosystem is ready (pip install --allow-sdists)

 Wed Jul 14 10:39:16 +0000 2021

RT @ballingt: @juanluisback @JarrodMillman I especially like @ralfgommers’ concise description of the issue for @JarrodMillman, this is the…

 Wed Jul 14 12:15:46 +0000 2021

RT @jcea: Great thread. My opinion about doing wheel only releases: 1. Only works if you deploy in a supported platform 2. “impossible to b…

 Wed Jul 14 12:16:18 +0000 2021

Totally agreed! I think doing wheel-only releases would be a big mistake. However, I wish pip can, at some point, ignore sdists _by default_, so that only power user ask for them. 99 % of the people just don't want to compile from source, they couldn't care less.

 Wed Jul 14 12:17:19 +0000 2021

Hi Greg, you sent me an email 9 years ago asking if I'd want to organize a SC workshop, and I felt I had won the Internet that day. You were an inspiration for me before that, and you still are. Please keep trying to change the world, we need you.

 Wed Jul 14 12:22:49 +0000 2021

RT @AstroBrigi: @juanluisback @pradyunsg @brainwane The solution is not to remove RC releases but to make sure they are tested as widely as…

 Wed Jul 14 12:59:41 +0000 2021

"This is the 20th SciPy" 😍 #SciPy2021

 Wed Jul 14 14:01:55 +0000 2021

"The tests should become operation procedures, and conversely, you shouldn't use anything during operations that you didn't test" #SciPy2021

 Wed Jul 14 14:23:03 +0000 2021

RT @SciPyConf: Talks are underway at #SciPy2021! Thanks everyone for being here, we hope you have a wonderful time, learn something, and ma…

 Wed Jul 14 14:27:42 +0000 2021

Enlace roto :/

 Wed Jul 14 14:40:22 +0000 2021

RT @jkru: This is awesome (and you can see the crowd go wild via reactjis). the F' framework is open source, so you can actually go look at…

 Wed Jul 14 14:43:58 +0000 2021

RT @HEPfeickert: 100% in agreement with @NASAJPL's Michael Starch's #SciPy2021 keynote takeaway! Plan for reuse when you build software and…

 Wed Jul 14 14:50:33 +0000 2021

"Scientific Python Ecosystem: Planning for the next decade" https://scientific-python.org/ @stefanvdwalt announces the initiative at #SciPy2021

 Wed Jul 14 15:09:47 +0000 2021

RT @amcasari: 📣 hey #OpenSource community managers, project directors + budget mavens

✨ 2021 is >50% complete

📅 do you know the financial…

 Wed Jul 14 16:06:18 +0000 2021

RT @readthedocs: Do you want to use Jupyter notebooks in your Sphinx documentation and don't know where to start? Are you looking for ways…

 Wed Jul 14 16:56:48 +0000 2021

RT @mightaswellcode: in a few hours, @shr3ya_malv1ya & i will host a BoF session on Python in Aerospace and Astronomy at @SciPyConf.

we go…

 Wed Jul 14 19:04:46 +0000 2021

RT @ericholscher: Thanks again to @cziscience who has provided the grant that made this work possible. We're excited to be deeply engaging…

 Wed Jul 14 19:05:29 +0000 2021

RT @lukedones: @juanluisback reveals vigorous inner solar system trade network with @poliastro_py at #SciPy2021

 Wed Jul 14 20:38:33 +0000 2021

RT @dillonniederhut: In case you want to plan your own flight to Mars, you can use poliastro to solve for your astrodynamics

- @juanluisb…

 Wed Jul 14 20:38:35 +0000 2021

Interesting! I didn't readily see that in the jupyter-sphinx docs but I don't have prior experience with it. If you have some pointers I'd be keen to have a look :)

 Wed Jul 14 20:46:38 +0000 2021

RT @WillingCarol: Thanks @ericholscher @readthedocs @cziscience! Hey @SciPyConf folks, you may be interested in this guide. Great content a…

 Wed Jul 14 22:08:32 +0000 2021

Quite productive networking session about documentation at #SciPy2021. We talked about Sphinx, mentioned https://diataxis.fr/, discussed what makes writing documentation difficult, and more. Thanks everyone, and see you tomorrow!

 Wed Jul 14 23:13:32 +0000 2021

RT @BMtHart: Just uninstalled Matlab and installed PyTorch to do some deep learning.

 Thu Jul 15 10:42:15 +0000 2021

RT @ralfgommers: Had a great time at #SciPy2021 yesterday, best virtual conference experience I've had so far!

Gave a talk together with…

 Fri Jul 16 05:33:53 +0000 2021

RT @martinRenou: Want to make your custom #JupyterLite deployment? Thanks to @jtpio, it's as simple as using the http://github.com/jupyterlite/demo re…

 Fri Jul 16 05:37:02 +0000 2021

A kernel update bricked mine in December, it doesn't boot anymore and I'm trying to find a spare weekend to fix it... So, it's now a paperweight :/

 Fri Jul 16 05:49:00 +0000 2021

For the uninitiated, what is the quote? :)

 Fri Jul 16 06:04:02 +0000 2021

More on that: what if we reflect on the things listed there as "advantages of MATLAB"?

- "complex system to install libraries": conda!
- "3-D interactive graphics with transparency": Plotly, PyVista, others
- "powerful debugging tools": Admittedly depends on the IDE

👇🏼

 Fri Jul 16 06:09:55 +0000 2021

- "image processing toolbox": what is missing in scikit-image and Pillow?
- "signal processing toolbox": what is missing in scipy.signal?
- "optimization toolbox": what is missing in scipy.optimize?
- "bioinformatics toolbox": ?
- "virtual reality toolbox": ?

👇🏼

 Fri Jul 16 06:11:45 +0000 2021

- "the MATLAB compiler": hmm maybe PyInstaller comes close, but it's difficult to get right, and it's not exactly AOT compiling. To my knowledge there are no good AOT Python compilers yet, and probably there won't be for a long time, if ever.

👇🏼

 Fri Jul 16 06:13:02 +0000 2021

It's easy to scoff at these rants and cherry-pick silly things (like 0-based indexing) but they have a core of truth. Listing these alleged weaknesses and working on them will make the ecosystem stronger, and allow Python to break into places where "open tools" are not a priority

 Fri Jul 16 06:14:34 +0000 2021

Once a year, the Scientific Python community picks on a "MATLAB vs Python" poorly-written rant.

And once a year we need to remember that MATLAB is a powerful ecosystem that we still have a lot to learn from.

(My responses are in the thread) https://twitter.com/mrocklin/status/1415675325904670720

 Fri Jul 16 06:16:26 +0000 2021

Bookmarking this 🙃 Thanks!

 Fri Jul 16 06:19:33 +0000 2021

Despite my best GitHub search-fu abilities, I can't seem to find it 🙃

 Fri Jul 16 06:21:28 +0000 2021

Super exceited about @fperez_org keynote at #SciPy2021 😍

 Fri Jul 16 14:07:14 +0000 2021

RT @HEPfeickert: It is nothing short of amazing to think that IPython started with @fperez_org thinking of trying an "afternoon hack". :) T…

 Fri Jul 16 14:25:54 +0000 2021

RT @fperez_org: Thank you so much to the #SciPy2021 organizers and this entire community, that has made my entire career possible, from ear…

 Fri Jul 16 17:54:47 +0000 2021

RT @pyblogsal: After a lot of work, we have finished the implementation of PEP 657 🚀🎉. In Python 3.11🐍, tracebacks will annotate where exac…

 Fri Jul 16 19:38:10 +0000 2021

This is incredibly insightful, it will take me weeks or months to fully process it... Thanks a lot for writing it!

 Fri Jul 16 20:52:00 +0000 2021

- Scientific code was written in MATLAB,
- Professional code was written in Java,
- Fast code was written in C++,
- Fastest code was written in Fortran

Robert Kern at #SciPy2021

 Fri Jul 16 20:56:41 +0000 2021

Tomorrow I will be leading a documentation sprint at #SciPy2021 - if you want to write docs to your project or someone else's, ask for advice with Sphinx, reST, MyST, or just chat about scientific docs, join us! https://www.scipy2021.scipy.org/sprints

 Fri Jul 16 22:03:20 +0000 2021

RT @HEPfeickert: That's a wrap for the conference agenda for #SciPy2021, so a huge THANK YOU to all the @SciPyConf organizers who made it h…

 Sat Jul 17 08:07:03 +0000 2021

So far so good! Lending a hand to the sktime folks from @turinginst 🙌🏼 and getting some fixes in @readthedocs as well 💪🏼 Lunch break now!

 Sat Jul 17 12:31:47 +0000 2021

Wrapping up, might (or might not) connect tomorrow afternoon for more sprints. Have a great weekend everyone!

 Sat Jul 17 16:06:30 +0000 2021

RT @NumFOCUS: We are proud to introduce our newest addition to our fiscally sponsored projects: GDAL 🎊🎉

Learn more about GDAL and how to s…

 Sun Jul 18 09:40:25 +0000 2021

👋🏼 from the human behind @poliastro_py 🙃

 Sun Jul 18 17:58:08 +0000 2021

RT @AlexReibman: This one quick hack has saved me *countless* hours fixing graphs with Jupyter/Python

Explanation + details below: https:/…

 Sun Jul 18 18:50:06 +0000 2021

@PyLatam ¿Cuándo notifican a les ponentes seleccionades? 🙈

 Sun Jul 18 19:11:31 +0000 2021

¡Hola! ¿Cuándo avisan a les ponentes seleccionades? 🙈

 Sun Jul 18 19:12:24 +0000 2021

@vantpc Hola, ¿vuestro email de soporte técnico es sat@vantpc.es ? Es que... hace un mes y medio que me ignoráis completamente y ya empieza a ser molesto. Mi último correo es sobre un problema bastante llamativo de construcción.

 Tue Jul 20 07:53:42 +0000 2021

A kernel update broke poweroff and had to rollback. I can't control the fan speed and my laptop overheats (and I've been said this has no solution). Sometimes my desktop randomly freezes.

Look folks, Linux on the desktop is better than years ago. But don't pretend it's *easy*. https://twitter.com/nedbat/status/1417155521493274630

 Tue Jul 20 07:58:23 +0000 2021

Me acaban de llamar por teléfono, ¡muchas gracias!

 Tue Jul 20 08:23:18 +0000 2021

cc @civio 🤔

 Tue Jul 20 12:08:28 +0000 2021

Puf, me encanta cómo suenan. Deberían grabar cuatro o cinco canciones y subirlas a https://bandcamp.com/, yo compraría con gusto. Mientras tanto ojalá me los encuentre por Lavapiés algún día. Muchas gracias por el reportaje ❤️

 Tue Jul 20 12:18:29 +0000 2021

Welp https://forums.linuxmint.com/viewtopic.php?f=18&t=353361

 Tue Jul 20 13:25:14 +0000 2021

Hot take: More funding going to open source technologies is a good thing, not a bad thing

 Tue Jul 20 22:12:27 +0000 2021

I know who they are! 😂

 Tue Jul 20 22:42:32 +0000 2021

RT @paimadhu: Hoarding vaccines is a moral failure, but allowing Covid vaccines to expire is a whole another level of moral bankruptcy

 Wed Jul 21 09:25:10 +0000 2021

RT @ddbeck: Using the second person (you, your) in your documentation continues to be good and proper https://ddbeck.com/second-person-is-ok/

 Wed Jul 21 09:31:58 +0000 2021

Do you happen to have any pointers at hand about this?

 Wed Jul 21 10:56:56 +0000 2021

Thanks!

 Wed Jul 21 11:11:33 +0000 2021

On @readthedocs we are now a team of 6 distributed across 5 different timezones!

 Wed Jul 21 11:47:19 +0000 2021

6+2 @ethicaladsio on 2 different timezones, to my count :)

 Wed Jul 21 11:57:56 +0000 2021

RT @matrixdotorg: Incredibly exciting to get official confirmation that Germany's health service (@gematik1) has standardised on Matrix for…

 Wed Jul 21 13:58:42 +0000 2021

Sorry but this is FUD. There have been numerous standardization PEPs (Python Enhancement Proposals) on recent years, related to library development, dependency management, and the like. pip is *way* better, virtualenv is more maintained, there are alternatives like Poetry.

 Wed Jul 21 14:51:17 +0000 2021

Congrats! 🥳 May I ask a sneak peek into what tools will you use for the talk? 😇

 Thu Jul 22 08:58:56 +0000 2021

RT @europython: Are you into data science? Then we have something special for you:

The EuroPython 2021 Data Science Mini-Conference, embe…

 Thu Jul 22 09:14:01 +0000 2021

RT @KevinLeichtman: Are research journals outdated?

After obtaining my PhD, I’m not tempted to go through peer review process to push out…

 Thu Jul 22 09:16:37 +0000 2021

Loosely inspired by the Douglas Adams quote:

“I love deadlines. I love the whooshing noise they make as they go by.”

😂

Really, it's the only method that has worked for me over the years!

 Thu Jul 22 11:50:32 +0000 2021

RT @BrookeMacnamara: Likewise, Nobel laureates were more likely to have been educated or worked in fields other than the field for which th…

 Thu Jul 22 11:51:10 +0000 2021

That bio made me cackle 😂

 Thu Jul 22 11:58:44 +0000 2021

I'm feeling that pain now 😭

 Thu Jul 22 14:30:25 +0000 2021

First time I saw it was in an @Sydonahi presentation at Python for Astronomy at Flatiron

 Thu Jul 22 21:32:47 +0000 2021

Hah, true!

 Fri Jul 23 07:04:21 +0000 2021

"Speaking of GIL, can you please check the GIL documentation and let me know if this is a thing that we need in Python."

LOL

https://twitter.com/theJohnHerrick/status/1418260277217296387?s=20

 Fri Jul 23 11:36:49 +0000 2021

RT @pwang: @anacondainc Python continues to dominate and be a popular language, particularly among younger generations. 88% of students sur…

 Fri Jul 23 16:24:34 +0000 2021

RT @stsewd: New post! Tips to secure your development environment from your computer to your terminal.

https://stsewd.dev/posts/securing-your-dev-environment/

 Sun Jul 25 07:17:50 +0000 2021

I learned a thing or two from this, thanks a lot!

 Sun Jul 25 07:17:59 +0000 2021

RT @ralfgommers: First major milestone reached: SciPy now builds with @mesonbuild on Linux (and it's⚡️fast)!

Blog post about the journey s…

 Sun Jul 25 19:19:17 +0000 2021

This is a-m-a-z-i-n-g, thanks for sharing the detailed writeup as well! Excited about SciPy leading the way for other smaller projects with compiled extensions to switch to Meson 👏🏼

 Sun Jul 25 19:20:47 +0000 2021

RT @petersuber: Major step forward from @ERC_Research: When researchers list publications in grant applications, they may include "relevant…

 Mon Jul 26 10:49:22 +0000 2021

Little project I hacked over the weekend to learn some modern JavaScript and @fastapi : an online conversor from reStructuredText and MyST (idea by @ericholscher , design by Ana Costa) https://github.com/astrojuanlu/mystyc

What do folks think?

 Mon Jul 26 11:51:41 +0000 2021

Hah, thanks! I was so absorbed with the conversation that I forgot about it 😅 You can try it live here:

https://mystyc.herokuapp.com/

 Mon Jul 26 13:36:55 +0000 2021

"can't get rsync to exclude directory" yahoo answers

 Tue Jul 27 10:35:37 +0000 2021

Generative art? https://p5js.org/ Live coding? https://sonic-pi.net/ Scientific visualization accidentally going beautiful? https://twitter.com/accidental__aRt

 Tue Jul 27 13:14:05 +0000 2021

The first time I saw a live coding performance it was actually at @PyConNA and it was AMAZING 😍

 Tue Jul 27 13:56:51 +0000 2021

RT @sktime_toolbox: You are invited! Join sktime doc #sprint on August 16-19 and contribute to a growing Python package for machine learnin…

 Tue Jul 27 15:29:54 +0000 2021

Remind me again how snarky tweets not accompanied by a reproducer or a bug report make open source move forward 🙄

 Tue Jul 27 19:17:32 +0000 2021

Well, you didn't explain if it was a problem with setuptools, pip, requirements.txt, http://setup.py, PyPI...

 Tue Jul 27 19:29:58 +0000 2021

Ah, I understand your POV now, thanks (and sorry for my unfriendly tweets). I think @kushaldas PEP 582 https://www.python.org/dev/peps/pep-0582/ might alleviate that pain, but in any case one would need a separate installation step. I see how AOT compilation wins here.

 Tue Jul 27 19:37:32 +0000 2021

RT @ethicaladsio: We scaled our #Django powered ad server to 100 write requests per second using off-the-shelf tools while keeping our clou…

 Tue Jul 27 20:45:14 +0000 2021

RT @natfriedman: We’ve just added built-in citation support to GitHub so researchers and scientists can more easily receive acknowledgments…

 Tue Jul 27 22:09:09 +0000 2021