Fabio Manganiello

:platypush: Creator and #developer @ platypush.tech
:booking: Senior #software engineer @ booking.com
⚙ #Automation addict
🤖 #AI builder
:linux: #Linux user since 2001
🔓 #FOSS contributor
:arch: Prone to unsolicited "btw I use #Arch" statements
🏡 #SelfHost all #tech!
🔬 Open #science and open #data advocate
🎶 #Music geek
🎸 #Guitarist + occasional composer
🛹️ #Skater
🏄 #Surfer
👪 #Dad of a small geek
🇮🇹 ⇒ 🇳🇱

Public Key

npub13uunvh7djw9ep54nswkuxlneyee7ehcpc7e53t68krykrdeg6j4qrdpvgs

NIP-05 Address

fabio@manganiello-social.mostr.pub

Profile Code

nprofile1qqsg7wfktlxe8zus62ec8twr0eujvulvmuqu0v6g4armpjtpku5df2spzemhxue69uhhyetvv9ujumn0wd68ytnzv9hxg9trrwv

Publishing to

relay.nostr.band

Author Public Key

npub13uunvh7djw9ep54nswkuxlneyee7ehcpc7e53t68krykrdeg6j4qrdpvgs

Show more details

Published at

2025-03-18T06:40:01Z

Kind type

0 Metadata

Event JSON

{ "id": "2a74bd657e969123e12c1a1bfb8c5e38a1f938e5cedb5c08ad7b386f3753ff4a", "pubkey": "8f39365fcd938b90d2b383adc37e792673ecdf01c7b348af47b0c961b728d4aa", "created_at": 1742280001, "kind": 0, "tags": [ [ "emoji", "arch", "https://manganiello.social/emoji/custom/arch.png" ], [ "emoji", "booking", "https://manganiello.social/emoji/custom/booking.png" ], [ "emoji", "git", "https://manganiello.social/emoji/custom/git.png" ], [ "emoji", "github", "https://manganiello.social/emoji/custom/github.png" ], [ "emoji", "linux", "https://manganiello.social/emoji/custom/linux.png" ], [ "emoji", "matrix", "https://manganiello.social/emoji/custom/matrix.png" ], [ "emoji", "platypush", "https://manganiello.social/emoji/custom/platypush.png" ], [ "proxy", "https://manganiello.social/users/fabio", "activitypub" ] ], "content": "{\"name\":\"Fabio Manganiello\",\"about\":\":platypush: Creator and #developer @ platypush.tech\\n:booking: Senior #software engineer @ booking.com\\n⚙ #Automation addict\\n🤖 #AI builder\\n:linux: #Linux user since 2001\\n🔓 #FOSS contributor\\n:arch: Prone to unsolicited \\\"btw I use #Arch\\\" statements\\n🏡 #SelfHost all #tech!\\n🔬 Open #science and open #data advocate\\n🎶 #Music geek\\n🎸 #Guitarist + occasional composer\\n🛹️ #Skater\\n🏄 #Surfer\\n👪 #Dad of a small geek\\n🇮🇹 ⇒ 🇳🇱\",\"picture\":\"https://s3.nl-ams.scw.cloud/manganiello-social-media/82d5ffba-50f5-4f6a-a778-6a92245627b4/profile_pic.jpg\",\"banner\":\"https://s3.nl-ams.scw.cloud/manganiello-social-media/f03cc3ce-7224-49b3-b708-efbc2e94cddc/16b40f2138025168.jpg\",\"nip05\":\"fabio@manganiello-social.mostr.pub\",\"fields\":[[\"🏠 Home\",\"https://fabiomanganiello.com\"],[\":git: Git\",\"https://git.fabiomanganiello.com\"],[\"📹 Videos\",\"@fabio@video.fabiomanganiello.com\"],[\"💬 XMPP\",\"fabio@manganiello.tech\"],[\"🔐 PGP\",\"https://keyoxide.org/6E3F54DA3F4286D20AB4009ED90FBA7F76362774\"],[\"✉ Email\",\"fabio@manganiello.email\"],[\":github: Github\",\"https://github.com/blacklight\"],[\":matrix: Matrix\",\"@blacklight:matrix.platypush.tech\"],[\"📝 Blog\",\"@fabio@manganiello.blog\"],[\"📷 Photos\",\"@fabio@manganiello.photo\"],[\"🎧 Music\",\"@fabio@music.fabiomanganiello.com\"]]}", "sig": "4e5585240b8800b20b008bec84d5738f825e8e36eacb53f8979d9e2e1c36d94d50ea5286375e2d7fbfe9cb1d6fd147e04145a22e0e6b6c1d6ebc2f6baaaa9a66" }

Last Notes

2025-03-27 11:12:24

As the maintainer of a huge (>150K LOC) #Python codebase like Platypush, whose development spans over more than a decade, I agree and disagree with the points made by the author.
I agree that Python’s very tolerant and flexible typing system makes it way too easy to shoot your leg.
And I also agree that meta-programming, albeit powerful, makes it hard to understand what you’re actually handling (and IDEs also get easily confused by it).
But I also think that using typing annotations whenever possible addresses these issues.
Sure, Python’s typing system doesn’t run at compile-time (there’s no compile-time anyway). So in theory it’s nothing more than nice guidelines that developers can happily ignore and still ship broken stuff to prod. But I also believe that it’s a civic duty to program in Python with editors that have Black or any other LINTer embedded, and treat any type warning that they raise as you would treat a compiler error in another language. And it’s also a good practice to add pre-commit hooks or LINT tests on your CI/CD that would prevent you from merging into main if there’s any issue.
If you set up your environment like this, and you use strong typing annotations in most of the cases, then it’s not very different from using Java or another other strongly typed language.
That being said, I still like to have the ability to do duck-typing or meta-programming sometimes.
After working on huge Python codebases for many years I’ve come to the conclusion that you don’t really need such dynamic typing cowboy styles that often. On average it’s probably going to be beneficial in <5% of your codebase. But having that flexibility of saying “hey, what I’m passing to this deserialize method can be nearly anything - bytes, strings, dicts, collections, already deserialized object etc. - just don’t fuss about it and give me back something already normalized”, rather than declaring 10 different overloaded methods with different signatures, is a big gain compared to, say, Java or C#.
I wish these languages had implemented something that resembles Python’s Union type - it’s a good trade-off between strictness and flexibility whenever you have some function that can accept a bunch of stuff, and you want to enumerate what all that stuff is.
https://dustri.org/b/friends-dont-let-friends-write-production-software-in-python.html

2024-10-28 09:55:27

Today I’ve learned that there’s a manufacturer of Linux laptops that indefinitely stops shipping its products to entire geographical regions whenever they have an argument with somebody from that region (Colorado and Maryland are already on the list).
And for as much as I agree in principle with some of their arguments, and for as much as I agree with them with the issue of poor documentation of open-source projects, I’m also appalled from childish reactions like these:
System76’s Principal Engineer decided to chime in and make a fool of himself, so we banned the entire state of Colorado for life.
I’ll remind myself never to get a laptop from #Malibal. A company that may stop supporting or shipping items to all of their customers in a whole region whenever they have an argument with someone isn’t a company that can be trusted with anything, no matter how principled and well-intentioned.
It’s a shame because it seems like they build good Linux laptops that are 100% open, but someone who’s so unprofessional and undependable doesn’t deserve to stay in business for long.
p.s. To all their potential customers in the Netherlands: I’m sorry if this post ends up with them banning shipments here.

2024-10-23 06:42:48

- reply

@npub15zu…s7lx I guess that UK interference in US politics is ok only if it's about Nigel Farage teaching the orange scum how to complete a sentence.

2024-10-17 22:31:53

Compared to America and China, #Europe doesn’t invest much in technology.
At some point a couple of years ago Macron saw the point of a common European investment fund in technology to avoid ending up as technologic subjects of either side of the Pacific Ocean, but then he came up with the wrong solution - we don’t need “European champions”, as he called them, large enough to compete with the American giants, but a healthy, sustainable and competitive environment for innovation to sprawl. Pick the willing and not the winners.
In absence of a wider strategy to fund European technology, the EU has however, since 2020, allocated a small fund that goes to small European-based open projects that need financial backing to lift off the ground, or to keep the lights on.
The #NGI (Next Generation Internet) doesn’t even get a lot of funding. The previous cluster allocated about €27 million. Peanuts compared to the $8.5 billion that Biden poured into Intel with the CHIPS act, or to the hybrid ownership scheme common for Chinese tech giants. A medium-sized company burns that much money in a single week. However, it’s been enough to fund many European projects that would have begged for Github sponsors otherwise. The list of projects sponsored by NGI includes accessibility and graphics input support for KDE Plasma Wayland, decentralized storage solutions over BitTorrent (Scion), a privacy-preserving infrastructure for electronic payments (GNU Taler), mobile operating systems (PostmarketOS), and much more. All projects that provide open solutions good enough to compete with those provided by non-EU businesses that get much more financial backing, for a tiny fraction of the price tag.
Keeping the funds for NGI would have thus been a no-brainer. Tech talent in Europe apparently can do great things even with little money, imagine what they could do with more money - or, better, with a proper EU-funded investment bank dedicated to fund open tech solutions. Something like what the SVB was for the Silicon Valley, but without the whole profitability, financial mismanagement and high-risk bets factors that took it down. Or something like KfW, Cassa Depositi e Prestiti or EIB are for Germany, Italy and the EU respectively, but with a strong focus on funding a sustainable technological infrastructure on open protocols and open software.
Unfortunately, there’s no mention of NGI in the working draft of the Horizon Europe programs for 2025.
Not sure if it’s a glitch, if it’s an intentional omission or if the EU has other plans to fund open software. But it’d be nice to have some answers.
https://pad.public.cat/lettre-NCP-NGI#

2024-10-13 08:43:59

On one hand, I’ve always been skeptical of professors pushing for insane reading schedules. If you expect your students to read Pride and Prejudice today, Crime and Punishment the week after and War and Peace by the end of the quarter, then you have unrealistic expectations. Besides showing no respect for the free time of your students. Most of the students will end up skimming through pages just to meet the deadline, and they won’t have time to absorb the meaning of such important novels. Assign less books and give students more time to read them, otherwise your students won’t learn anything more about the book that they can’t read from the first paragraph of the Wikipedia page.
On the other hand, the volume of reading assigned by some professors seems to be the last of the problems. The biggest problem that also starts affecting top-level colleges apparently isn’t that students struggle to read War and Peace in a month, but that they also struggle to read a Shakespeare’s sonnet without getting distracted halfway. And high school has simply given up trying to keep students focused and read, so when then land in college most of them have literally read zero books in their lives - and their writing skills are also quite primitive and childish as a result.
This is a big problem. We’ve forged a generation too easily distracted to function properly, and with too many instant gratifications preventing focus time. And technologies like ChatGPT that amplify the illusion of knowledge at your fingertips are doomed to make things even worse. And that also applies to the kids who eventually end up in Yale or Cambridge. We have to undo this mess before it’s too late. If the solution is not to give a phone to your kid until he/she is of age, then it’s time to go for it.
https://www.theatlantic.com/magazine/archive/2024/11/the-elite-college-students-who-cant-read-books/679945/

2024-10-11 07:44:51

The #Nobel prize in #physics went to the seminal work on neural networks.
The Nobel prize in #chemistry went to the application of AI to predict proteins folding.
I guess that the Fields medal may go to some kid who asked ChatGPT to prove Riemann’s hypothesis.
We as computer scientists may not have our own Nobel prize category, but we’re apparently learning how to take over the others.
https://s3.nl-ams.scw.cloud/manganiello-social-media/accd16e4aeec6451666283ad6be312f27e60d8b8a47681be9dc323d803f4db0e.jpg

2024-10-05 14:32:55

The banality of evil always manifests through the unnecessary harassment of the weakest in society.
Imagine crossing the desert, being harassed and beaten up by thughs from the Sahel all the way to Libya, being detained for an undefined period of time in a cell in the lybian desert, crossing the Mediterranean on board of inflatables that are barely supposed to take you on a guided coastal tour, bouncing from a EU country to another to submit your asylum request, finally landing in the Netherlands, and being welcomed by a big sign saying "We are working hard to send you back home".
A quarter of the Dutch population actually had the guts to vote for such revolting human beings at last year's elections. #PVV
https://www.dutchnews.nl/2024/10/fabers-go-home-sign-would-be-bullying-refugees-says-bbb-boss/

2024-10-02 12:08:30

- reply

@npub1k06…zyu9 capitalism doesn’t necessarily have to get to the extremes where all basic necessities are privatized and prices free to fluctuate up to a point where basic needs become a luxury for increasingly large segments of the population.
I mean, sure that’s the natural outcome when you let capitalism free of mutating until it devours everything around it, but it doesn’t have to be like that. And I’d argue that it wasn’t like that in Europe until the past decade or so.
In the Netherlands at least it’s been since Rutte and his ultraliberal gang that checks and balances, public participation in the economy and safety nets have been gradually eroded to a point where I see late capitalism issues here (public infrastructure falling apart, rampant homelessness, families that can’t afford to pay for public transport, food, healthcare or childcare etc.) that were completely alien to me as a European until a few years ago, and I had only seen in the States before.

2024-10-02 08:34:16

What happens when you let prices of services float in a free market fashion?
That those who have interest in prices going up will inevitably push the prices up, until common people can no longer afford those services.
It’s been happening in the Netherlands for housing (median income can’t afford a house in any of the major cities, as most of housing supply in Amsterdam and The Hague is controlled by private businesses that have interest in pushing prices up), it already happened with childcare (the scandal that brought down the previous Rutte government), it’s happening with healthcare (the cartel of insurances increases prices every year in a synchronized way at rates that are much higher than the average salary increase), and it’s now happening for public transport too.
The “americanization” of the Dutch economy is doing permanent damage to the fabric of our society.
When you have people in a supposedly advanced country who can no longer afford housing, nor childcare, nor healthcare, nor transport, all while inflation remains higher than the EU average and many can’t even afford food, all while you get more and more aggressive homeless people on the street who have been kicked out from the economy, you have the perfect recipe for political grudge, polarization, extremism and chaos.
And we know already how it ends up. When you start those races to the bottom, it’s rarely the progressive left that wants to boost equal opportunities for everyone that gets a shot at running things. It’s usually the ugliest identity politics that wins on the promise of giving you the things you need by taking them away from someone else - me-first, me-vs-them, urban-vs-rural, local-vs-migrant etc.
The triumph of a populist imbecile like Wilders at the past elections is probably just the beginning of a long-standing trend.
https://www.dutchnews.nl/2024/10/middle-incomes-also-struggling-to-pay-for-transport/

2024-09-20 13:06:24

Springfield pets unite.
We have a duty to keep laughing about this guy and expose how ridiculously incompetent and stupid he is.
Laughter and ridicule are the best weapons against naricissists with hyper-inflated egos.
https://s3.nl-ams.scw.cloud/manganiello-social-media/8778357e3b627d335778e1c977b4745da0b42baca5e8d2e51cf72eb6522fdd54.mp4

2024-09-20 10:34:40

A word or two about #Palestine
https://s3.nl-ams.scw.cloud/manganiello-social-media/a766ed59f4127d491f85f4a5a590d583e971017ba95b04476b4e80554160da6a.mp4

2024-09-16 08:22:42

Summarizing technological enshittification in a single picture.
https://s3.nl-ams.scw.cloud/manganiello-social-media/3c97195847d8f05f7cfa072dcde55923756c6dcfdc8cb5366140c756e1908604.png

2024-09-11 07:51:07

A country that keeps shutting down the Internet for everyone every time a major protest occurs can never expect to become competitive, modern, nor it should be considered a reliable business partner.
https://www.dw.com/en/indian-state-shuts-down-manipur-internet-and-imposes-curfew/a-70182884

2024-08-08 10:14:43

- reply

@npub1c8p…ce39 things like expectations of what tables/constraints exists and if exists/not exists clauses between versions are still up to the developer.

2024-06-21 23:36:54

It happened to everyone.
Your municipality inadvertedly the fiber cable that serves your block.
Your son watches Netflix while your wife is on Zoom and you’re gaming and the modem starts choking.
Your federated Mastodon server running in your cabinet starts saturating your upload rate (ok, maybe this is a bit less common).
For all the times when you wondered “can’t my modem just restart by itself when the connection drops?”, there’s now a simple Platypush script.
https://notes.fabiomanganiello.com/s/YL0Vfze8Q#

2024-06-12 10:06:44

Two Dutch cities lead the 2024 sustainability index - and the whole top 10 is composed only of European cities.
Amsterdam 🇳🇱
Rotterdam 🇳🇱
Copenhagen 🇩🇰
Frankfurt 🇩🇪
Munich 🇩🇪
Oslo 🇳🇴
Hamburg 🇩🇪
Berlin 🇩🇪
Warsaw 🇵🇱
London 🏴󠁧󠁢󠁥󠁮󠁧󠁿
https://dutchreview.com/news/rotterdam-amsterdam-most-sustainable-cities-in-the-world/

2024-05-28 21:48:45

The batteries. The batteries for the sensors that measure the uplift of the soil under the center of the most dangerous active supervolcano in the world. How can someone on earth even think of stealing them?
I wanted to be in the room where this conversation happened. “Hey, I’m bored tonight, shall we go and steal the lithium batteries that power the Campi Flegrei monitoring network?”
https://www.isnews.it/2024/05/28/terremoti-nei-campi-flegrei-rubate-le-batterie-e-vandalizzati-i-sensori-ingv/

2024-05-24 10:20:27

Those who kill a man are accountable for murder.
Those who burn books are accountable for murdering the whole mankind.
https://s3.nl-ams.scw.cloud/manganiello-social-media/4bc62591a93d676c54cef03d3a6e0ec63f43296f48023135068b37aaaf8dda8a.jpg

2024-05-22 07:12:56

90% of today’s wars and territorial disputes can be traced back to “a British guy with fat mustaches decided that he knew better how to draw a border on a map”.

2024-05-14 09:34:36

#American #capitalism is that game about open and frictionless markets that everyone can play, as long as America always wins.
https://www.theguardian.com/business/article/2024/may/14/joe-biden-tariff-chinese-made-electric-vehicles

2024-05-09 18:06:05

- reply

@npub1kyz…psa4 according to the article the change was announced this week, but it doesn’t mention when it’ll be rolled out.

2024-05-04 10:39:43

I’m sick of the intentional meddling of the concept of “safety” and “comfort”.
Everybody in a healthy democracy has the right to feel safe. But people do not have a right to feel comfortable in their ideas. Especially in universities. Universities are places meant to challenge ideas. If you don’t feel comfortable about it, then you don’t probably just don’t belong to a university.
https://www.theguardian.com/us-news/article/2024/may/03/college-gaza-protests-antisemitism

2024-04-30 01:12:41

#H5N1 is adapting to spread between mammals - cows to cats in this case.
https://arstechnica.com/science/2024/04/concerning-spread-of-bird-flu-from-cows-to-cats-suspected-in-texas/

2024-04-23 07:50:31

Annual recurring revenue, rent-seeking behaviour and everything-as-a-subscription will one day permeate so much of our economic system that even drinking water from a public tap will require you to download an app and activate a subscription.
That day is today.
This degeneration of capitalism with zero added value for the customer must be ended at all costs.
#note16m5…tnfp

2024-04-01 16:22:16

- reply

@npub19j7…e8r4 yeah I thought of it, but Vivaldi is way too blown and opinionated for what I need.

2024-03-23 22:47:38

In today’s episode of “#Javascript never returns you what you’d expect from it”: numeric sorting.
❯ [1, 11, 4, 3, 2].sort()
Array(5) [ 1, 11, 2, 3, 4 ]
@npub1lhn…s0mj

2024-03-23 08:48:11

No one would trust an arsonist promising to light fewer fires, and no one should trust big oil and gas climate pledges and plans.
https://www.theguardian.com/us-news/2024/mar/22/oil-companies-emissions-goals-report

2024-03-23 00:37:42

I wish text/markdown or application/rtf where part of the RFC 5322 (email format).
Everybody hates HTML emails, for a good variety of reasons. But text emails with a text width of 80 characters feel so much 1980s.
Why not support Markdown/RTF as native formats, and let the client render them however the user pleases using any converters (which are usually much cheaper than a browser’s rendering engine)?

2024-03-21 12:05:20

- reply

@npub15nw…s0l7
You can not take an existing market and apply different rules to it and then predict some future.
I took the current operating costs, the total number of users, and came up with the average operating cost per user as a benchmark of the minimum price that should be asked. This is a quite standard way to identify the lower bar for an ask price, as the minimum price that you should charge to at least keep the boat balanced. Sure, not the most accurate, but a good one to at least get an idea of the order of magnitude. If you have better ideas, you’re welcome to share them.
it’s very poetic of you saying “free as in beer is never sustainable” on a free as in beer/freedom platform
I run my own instance on my own server in my own closet in my own house. And I’ve made contributions both to the software and to the protocol. I can talk about sustainability because my digital presence doesn’t weigh on anybody’s shoulders but mine - and I’ve come to the conclusion that I should host most of my services exactly to avoid the “free beer” entitlement trap. If I didn’t have time nor skills to do this myself, I’d probably contribute financially to offset the costs of running somebody else’s instance. But many don’t have this level of sensitivity. Many, both here and on Facebook, take their free beer for granted. And this is the mindset that I’d like to challenge.

2024-03-21 11:43:27

- reply

@npub15nw…s0l7 My hunch is that it's an estimate of the average cost of a user on the platform - approximately the total cost of running the platform (engineers, moderators, accountants, lawyers, servers, facilities, support staff...) divided by the number of users on the platform.
Even in that case, it's way off the target. Facebook's operating expenses accounted to about $80B last year (https://tradingeconomics.com/fb:usoperating-expenses). And the platform has about 3B users. That would make the actual per head contribution closer to $26/year, if they wanted to reach revenue parity (and I'm pretty sure that many would be horrified even by that price tag). But a cost of $1000/year actually makes sense for smaller platforms - on one extreme of the spectrum you have single user Fediverse instances, where all the costs are spread over a single person, and on the other extreme you have Facebook, where the costs are spread over a huge user base and economies of scale definitely help.

2024-03-20 23:19:33

- reply

@npub1klj…36ce @npub1lhn…s0mj I wouldn’t use ChatGPT for any problem that I could solve by searching on StackOverflow/Reddit - that’s where most of its training content came from anyway.
But I definitely find AI assistants helpful. I used TabNine even on vim for many years - it’s a mostly local AI assistant that has pioneered the field before everybody jumped in the same pond. And Copilot…I hate what Github did, training such a big model without asking for anybody’s permission, with no concerns for the licenses of the underlying code, and when you already host 99% of the FOSS code out there, was nothing but a dick move. But when I use it (and I tend to avoid using it when possible), it really boosts my productivity by at least 70%.
Even in hard problems with innovative solutions, there’s always in my experience some boilerplate involved. The struct/class/map that you have to fill from a JSON response, the unit test that needs to check what your code does when you pass any of the values in a big enum, the CI/CD pipeline, the Helm chart, the Gradle configuration…
This is all stuff where these assistants actually do very well. I wouldn’t let them touch my code that adjusts the thrusts of the motors of my drone depending on the accelerometer readings even with a 10ft stick, because that’s the kind of stuff that requires knowledge of a bunch of fluidodynamics equation as well as real-time programming principles - and if it something goes wrong the drone may go off the tangent hit somebody’s face. But generating a template for my API definitions, filling in the blanks in a schema/adapter definition or initializing the dependencies in my Gradle or package.json? Definitely! That’s not the kind of stuff that excites me about engineering anyway, those are more like the bureaucratic papers that I need to fill up in order to do my job.
I actually see a lot of potential for AI assistants on the augmentation. Just like LINTers, and IDEs with runtime errors/warnings checks before them, and compilers with error checks before them, boosted the productivity of developers in the previous decades.
But asking ChatGPT to whip up a concurrent, production-ready application and deploying it without engineering oversight (i.e. the replacement part), I don’t see that happening any time soon. At least for backend developers though, since our field usually requires a lot of application-specific problem solving skills. For frontend devs and designers, where the same design patterns and best practices are basically applied to all the UIs out there (and these models are actually great in inferring knowledge from common patterns), I’m not so sure…

2024-03-20 20:04:50

- reply

@npub176r…y7cj I agree, and things are even worse when it comes to corporate compliance. I wish I could just dist-upgrade to 24.04 next month. From what I’ve seen on the Palo Alto/GlobalProtect forums, they only released their client for 22.04 somewhere around last autumn. So it’s likely that we’ll have to wait 1-2 more years before the new LTS also becomes compliant™. I don’t even want to imagine what it’s going to be like to run a distro with software that may be ~3-4 years old by then (so much for “we want to increase your security”). By then I’ll probably have compiled most of the software from sources, including the libc, and my system may look more like Gentoo than Ubuntu.

2024-03-20 07:03:22

Banning China from accessing advanced silicon meant only one thing: that China is now at the frontline of innovation on an open architecture like #RiscV, while America is throwing its money at a dead body like #Intel.
https://go.theregister.com/feed/www.theregister.com/2024/03/20/alibaba_c930_riscv/

2024-03-19 22:30:02

21st-century colonialism unfolding in its most shameless shape.
https://www.theguardian.com/us-news/2024/mar/19/jared-kushner-gaza-waterfront-property-israel-negev

2024-03-19 11:14:22

- reply

@npub1tl0…q252 @npub1qew…jery that could be interesting - unlike a VM the published serial number may probably be the same if the right volumes are mounted. But it’d probably be cumbersome to say the least to reroute all the host connectivity through a container that runs GlobalProtect, while making GlobalProtect believe that what it sees is actually the full image of the OS. But it could be an interested weekend project…

2024-03-19 08:48:21

- reply

@npub1qew…jery well the only effect achieved by this forced migration in my company so far has been several Linux users giving in their Dell laptop and asking for a MacBook 😆
And a part of me keep believing that this was the end goal all along…

2024-03-19 08:21:02

“We like #Ubuntu LTS / #Debian stable because they make your system stable, predictable and easy to maintain, while #Arch and friends make it too easy to break”.
If I had a penny every time I heard this argument, I’d probably be a millionaire by now.
In the meantime, I’ve been forced to migrate from Arch to Ubuntu LTS on my work laptop because the Palo Alto endpoint security crapware (calling it a piece of software would be an offense to anyone who takes software development seriously) is only compatible with Ubuntu LTS. And, without that endpoint security crapware, you are not compliant™.
My Arch system worked flawlessly. Some stuff installed via pacman, some installed via AUR, one system upgrade a week, and everything worked well together.
A few months into my adventure with Ubuntu LTS, and I feel already like it’s a castle of matchsticks just one step away from crumbling.
The packages on apt are ancient because the distro is ancient (almost two years old). Want to get reasonably recent versions of Neovim, Node.js, Docker, Firefox or Wayland? Well, are you ok with just forgetting about it?
So I’ve already ended up with a mix of apt, PPAs, backports, Snaps, and just compiling a lot of software from source because (like in the case of Neovim, foot or i3) all the solutions currently supported on Ubuntu LTS are either outdated, broken or incompatible with the rest of the system.
To be clear, I can’t recall the last time that I had to build some popular piece of software from sources on Arch because I was provided with no alternative.
Do people really use this crap and convince themselves that it’s easier than Arch?

2024-03-16 17:12:21

We can discuss for a while about the headscarf and its historical and religious meaning when it comes to women's rights and subjugation.
But "my body my choice" should really means what it says, whether it's abortion or hijab.
https://phys.org/news/2024-03-france-abortion-rights-hijab-highlight.html

2024-03-15 23:23:51

If you are a public elected official, can you block people who criticize you on social media?
https://www.npr.org/2024/03/15/1238907249/supreme-court-social-media

2024-03-15 13:04:39

EU approach to tech #regulation: “We don’t like what you do, it harms consumers, here are some laws that you need to comply to in order to address our concerns”.
US approach to tech regulation: “We don’t like what you do, it harms our undisputed supremacy, either you sell your business to us or we’ll ban it out of our country”.
#TikTok
https://www.theguardian.com/us-news/2024/mar/14/mnuchin-investors-buy-tiktok

2024-03-13 15:57:57

- reply

@npub1qew…jery @npub1lhn…s0mj Linus Torvalds entered the chat 😆

2024-03-13 15:28:52

Time to change this #xkcd I guess. In 2024, the #1 excuse for slacking off is no longer “it’s compiling” - it’s “the PR is still waiting for approval”. Closely followed by “the pipeline is running”, I guess.
https://xkcd.com/303/

2024-03-13 14:39:10

If your software needs customer/dev/user support, it means that it was either poorly developed or poorly documented, and you’re charging your users for your shortcuts and bad decisions.
Change my mind.
@npub1lhn…s0mj

2024-03-03 14:34:24

- reply

@npub1car…g4j0 the compliance requirements are actually quite straightforward: whoever is a gatekeeper according to the DMA (and that btw may soon include may employer too, so it’s nothing personal against Apple) have to open up their platforms in order to prevent monopoly consolidation.
It definitely applies to Android too - which already supports 3rd-party stores and PWAs installed through other browsers btw.
KaiOS or PostmarketOS don’t have to comply because those definitely don’t meet the bar for being considered gatekeepers, but they may have to in a hypothetical future where they grow their market share to reach that bar.
Compliance is a consequence of success.
I like to rephrase that as “with great power comes great responsibility”.
Btw, as I was writing this post I just noticed that Apple just published a list of requirements for “Web browser entitlement” on iOS - which involves allowing users to set your browser as the default on iOS.
There’s a very lengthy process involved apparently, and I’m really wondering how much of it can be actually enforced and how much of it exists just to create enough friction to discourage major adoption.
Anyway, it’s at least a step forward compared to their previous stances.

2024-03-02 23:47:36

- reply

@npub1car…g4j0 I understand quite well what’s required to do the job right: Web engines not baked into the OS.
Apple’s arguments are actually quite similar to those made by Microsoft 15-20 years ago or so, when Mario Monti was the EU antitrust commissioner and Windows was forced to unbundle the Media Player from the OS and then provide the browser choice panel. And the scrutiny continues to this day - it’s recently been forced to unbundle Teams from Office suites.
In both cases, counterarguments go against the lines of “bundling these services/engines within the OS provides greater security/scrutiny/privacy/blabla, and unbundling them and opening up is too operationally expensive / poses security risks / blabla”.
Bullshit. A big smoking pile of bullshit.
Bundling within the OS what’s not supposed to be part of the OS only benefits the OS maker by increasing entry barriers towards potential competitors who don’t sell a pre-installed OS to customers, while denying consumers a choice.
I know exactly what Apple is supposed to do to comply with the Web engine constraint: open up iOS APIs (permissions, sandboxing etc.) to alternative Web engines, without assuming that there’s always a specific version of WebKit on the other side. Then other Web engines can simply tap into the OS APIs and have the same level of security and controls of WebKit.
And I understand exactly why it’s dragging its feet so much. It’s only about defending its privileges over the platform. Sure, technically it’s not the kind of thing we expect Apple to do right in a couple of days (unlike flipping a switch and killing the PWA background worker on iOS). But Apple had an official 6 months notice before the DMA kicked in, it already knew even before then that it was likely to be labelled as a “gatekeeper”, and it already clashed with the EU in the past over Web engines on iOS.
Had Apple been genuinely interested in solving this problem the right way, it would already had plenty of resources and time to do so. But Apple doesn’t want to solve this problem, because it greatly benefits from this problem. Thus the PWA move has been malignant compliance to get concessions out of the deal.

2024-03-02 14:51:19

I wish #Python developers didn’t write their libraries with asyncio.get_event_loop() and loop.add_signal_handler everywhere like it’s JavaScript.
I know that managing the lifecycle of your own loops can be a hassle in Python (and I wish that the asyncio API did a better job at it), but writing Python libraries in the async JS flavour is basically a guarantee of being thread-unsafe and thread-incompatible.
add_signal_handler uses the set_wakeup_fd C API under the hood, which only works if the file descriptors are initialized in the interpreter’s main thread.
I’m now working on rewriting the #Telegram integration in #Platypush (it’s been broken since the Telegram bot API migrated to the new asyncio paradigm). Since the new code uses get_event_loop and add_signal_handler everywhere, I have to come up with a complex architecture that involves at least a separate process for the Telegram listener (with its own event loop which uses the bot API), another thread as a bridge to run commands on the loop using run_coroutine_threadsafe, and at least two inter-process queues to send commands and receive responses.
To all the Python library developers out there: I know that it’s tempting to use the asyncio API like it’s JavaScript, and forget that your code may run in multi-threaded contexts. But please, try and go the extra step and manage the lifecycle of your own loops. Prefer new_event_loop+set_event_loop over get_event_loop. Avoid add_signal_handler if possible, instead opt for asyncio.Event with an option to specify a compatible event class (like threading.Event or multiprocess.Event), so the library user can adapt the synchronization mechanism to their code’s concurrency model rather than the other way around.
Many users of your libraries who want to run them in a separate thread will very very grateful for helping you keep their code simpler.
We may still have to use the *_coroutine_threadsafe APIs as a proxy, but at least we won’t have to come up with multi-process architectures and communication protocols to accommodate your library’s constraints.

2024-02-28 15:36:29

Four more #Python packages connected to North Korean hackers.
They contain a test.py file that decodes an intermediary DLL which generates a payload (disguised as IconCache.db) which in turn connects to a command-and-control server.
Affected packages:
pycryptoenv – 743 downloads
pycryptoconf – 1344 downloads
quasarlib – 778 downloads
swapmempool – 392 downloads
https://www.bleepingcomputer.com/news/security/japan-warns-of-malicious-pypi-packages-created-by-north-korean-hackers/

2024-02-16 15:52:49

- reply

@npub1z49…xzcg @npub1l3g…vu48 @npub100c…gkwp you may want to give SurfingKeys a try. Granted, it takes quite some fiddling to configure properly, but there are many good out-of-the-box configurations on the web too. I've been using it for years and I've even forgotten what the default browser bar is. In exchange, I've got a scriptable floating omnibar that I can tweak to show me whatever I want depending on the keybinding I want.

2024-02-08 22:01:20

- reply

@npub1vlw…xkpy We learned how to use a hammer, now we're entitled to assume that the whole world is made of nails.

2024-02-08 11:53:50

- reply

@npub1g3u…4u2g Circular dependencies are allowed only when they are allowed.

2024-02-08 10:27:47

244 partners, seriously?
I just scroll the news on a website, and each single tap and scroll will be shared with 244 other data brokers?
And then these motherfuckers dare to complain when people like me start using NoScript, PiHole, AdNauseam and uBlock to protect my privacy?
We're resorting to extreme solutions because THEY have completely lost their moral compass!
https://s3.nl-ams.scw.cloud/manganiello-social-media/a999950d-1156-449a-90fe-ce5212b9233a/Screenshot_20240208-112207.jpg

2024-02-07 22:54:37

- reply

@npub1kpx…uak4 numen is quite cool, I didn't know of it before! But it comes with a couple of cons after some testing:
1. It takes forever to compile.
2. It seems to be intended as a stand-alone executable that transcribes the speech, while my use-case is more that of a programmable voice assistant flow (with hotword detection, speech detection that goes on only on hotword, and programmable events that I can hook to). Maybe it's worth looking a bit into the vosk API.
3. At least the small-en-us model seems to struggle a bit with my non-US accent (and apparently even with that of a native speaker), but maybe it can be solved by just using a larger model I guess.

2024-02-07 20:59:58

- reply

@npub1urf…n7l3 thanks for the response. I wasn't aware of the context (i.e. that picamera was maintained by a 3rd-party developer who dropped the support). All I saw was that with the Bullseye release picamera was suddenly out with no alternatives.
Maybe outsourcing the development of the software the the camera wasn't a wise decision in the first place?
I guess that I'll do a bit more testing/reporting/contributing to get picamera2 to properly work on Arch and Ubuntu too then.

2024-02-06 22:55:41

I think that @npub15nn…sdr0 has got a #PiCamera problem.
PiCameras are amazing pieces of hardware with a software that could probably be managed better.
The original PiCamera libraries haven’t seen a commit in 4 years, as they were deprecated with the Bullseye release. I can still run some code that uses that library on some old RPis just because I’m keeping them on some ancient version of Raspbian. Otherwise, raspistill and friends break in many possible ways both on both Bullseye and Bookworm, and the package doesn’t even install on Arch ARM because raspberrypi-firmware is now gone.
A couple of years down the line, and its replacement, PiCamera2, is still in beta. It can be installed through a relatively smooth process only on the last two versions of RPi OS via apt, and they are otherwise very hard to get installed on any other distro - on Arch it depends on packages that officially aren’t even available for ARM (like python-av), when installed via pip it tries to build the world even if some packages (like numpy) are already installed on the system, and I didn’t manage to get it to run on Ubuntu because of permission issues.
It’s really a pity because a vibrant ecosystem of camera apps and scripts had been written using the old version of PiCamera, which could do a lot of things with very low entry barriers. Then a sudden deprecation was announced without a viable alternative, and a couple of years down the line that alternative isn’t quite stable yet. I’ve eventually resorted to leverage the native v4l2 integration over ffmpeg as a cross-platform workaround, but that moves most of the burden to the CPU and I’m not really leveraging this hardware at its best, plus it probably raises the technical bar for a lot of hobbyist makers.
Why was something so important to many users deprecated without any stable alternative on the horizon?

2024-01-31 10:36:10

"What happens if someone in the EU runs into a bug that isn’t happening in other browsers? How do we troubleshoot it?"
Just test your stuff in multiple browsers to ensure cross-compatibility, maybe? It's not like there's 34 web engines out there. There's only 3. You lazy arses.
Apple has spoiled a whole generation of developers who have now become nothing more than IDE-jockey who oppose any decision that would break Apple's immoral closed garden with "but then we have to test our things on more things!"
They currently build their thing only once, test it only on one type of device, only with one type of web engine, and that's it.
It's time for them to finally learn that testing your software against multiple devices and browsers isn't the exception, it's the norm.
https://www.theregister.com/2024/01/31/web_developers_worry_apple_ios/

2024-01-30 13:32:16

- reply

@npub19lt…qfjk I will definitely give this a try, and probably also contribute to it.
For context, I’ve reduced the usage but I’ve basically never stopped using Foursquare (just in the form of Swarm app now), especially when travelling. You still get a lot of valuable tips in the app after checking into a new restaurant or hotel!
I have replaced it in the meantime with a Tasker script that reads my GPS location every 10 minutes, forwards it to my MQTT broker, and then all of my Platypush automation magic happens to store it on the db together with an estimate of the address and nearest POIs. It leverages the Google Maps APIs for now, but migrating to OSM is something that has been on my backlog for a while.
It works great, and I can even run aggregate queries on the db like “show me the countries where I’ve spent most of my time in the past 3 years”. But it lacks the “social” part of it (location sharing with my circle if I want to), and this project definitely seems to fill the gap.

2024-01-29 12:52:19

- reply

@npub1q37…k2ta @npub1kw7…mwjd I thought that the problem of visibility and discoverability had already been solved two decades ago with syndacation.
If I bump on an article I like on somebody's website, and I like other articles on that website too, I instinctively look for a feed to subscribe to.
If there is, then I add it to the monitored feeds on my Miniflux instance. And I can even set rules on new feed items to automatically tag them, mark them as read or notify me through other means.
If the author doesn't write anything for a while, they'll still be there on my feed list. As soon as they publish something new, I'll see it regardless of any algorithmic black boxes.
Algorithmic timelines and the silent damnatio memoriae placed on RSS/Atom feeds have created a false problem of "discoverability and visibility" that didn't exist before.

2024-01-29 12:28:34

- reply

@npub1hyy…vcf2 @npub1r9j…4h54 I think that there alternatives to public libraries in the digital world too.
First, the job of the Internet Archive is enormous and incredibly understated. Not only it barely receives any funding from governments (those who fund the preservation of physical books and cultural heritage should also fund the preservation of digital media, but that's not happening), but their activities are increasingly seen by Google and friends under a hostile lens.
Countless hostile actions are taken almost on a weekly basis against youtube-dl, yt-dlp, and other tools whose job is simply to download a media file from a URL. Instead of being properly funded, archivists have to constantly operate on the edge of illegality and fears that the tools that they use to scrape the Web are just one DMCA takeover or API change away from being pointless.
Not only there are no financial incentives to ensure that the job of archiving the Web can proceed unimpeded, but the laws of online content are also extremely skewed towards the platforms (and, in part, a tiny minority of high-profile creators) rather than the users - and even less towards our sons and grandsons who may be around when Google and its servers may no longer be around.
Second, if you already know that publishing your content to some closed platforms comes with many risks when it comes to accessibility, preservation and trustworthiness of the platform itself, then we shouldn't upload content to that platform in the first place. The "I can't migrate" excuse only really applies to those who already have tons of followers on their YouTube channel - and, as underlined by this study, that may be <1% than the total number of people who uploaded at least a video to YouTube. Even without running our own servers, we can just leverage services like PeerTube that rely on open protocols and decentralized architectures to distribute the media.

2024-01-28 23:39:27

So many interesting insights in this research.
First, the technique used by these researchers to find out the actual number of videos hosted on #YouTube is definitely unorthodox (and inefficient), but it worked. Since Google won’t provide these numbers, and relying on traditional crawling techniques is likely to bring to the surface only videos that somebody has already interacted with, researchers have run an algorithm on a bunch of supercomupters that simply brute forced all the possible combinations of YouTube ID strings, and kept track of the requests that didn’t end up with a 404.
Second, even a conservative estimate of the number of videos on the platform is massive. 14 billion. Or nearly two videos for each human alive. With an unfathomably long tail.
To dig more in detail, videos with 10,000 or more views account for nearly 94% of the site’s traffic overall, but less than 4% of total uploads - a quite extreme version of the 80-20 rule. About 5% of videos have no views at all, almost 75% have no comments, and even more have no likes.
This sheds an interesting light on what YouTube actually is. Not a product that should be monetized at all costs, but a collective memory of basically all the media content that the human race has created in the past two decades. It’s vital infrastructure that should require no entry barriers, and it should be treated as such.
Most of the minutes of videos stored on YouTube’s servers aren’t from MrBeast, Veritasium or Tom Scott. They are from church services, weddings, condo-board meetings, graduation ceremonies, school lectures, and all other things that humans record and want to save on a permanent storage - for themselves, their families, their co-workers, their friends or their classmates. With absolutely no intention of monetization, wider reach, or whatever stinky corporate metrics YouTube PMs are obsessed with.
I was at a ceremony for my wife’s grandma a while ago. A guy took a high resolution video of the whole thing. It lasted about 2h and weighted 40GB. Of course, everybody from the family immediately wanted to see it, but a 40GB video isn’t exactly the kind of thing you share on the family’s Whatsapp group. Since I was involved, of course the video eventually ended up on my Nextcloud and Jellyfin servers, and from there the link could be shared with everyone. But what would have anyone else done, if they didn’t have a spare Nextcloud server with a couple of TBs attached in their closet, or didn’t want to upgrade thier Google One plan just to store a single video file on their Google Drive? Simple: upload it to YouTube.
And this is exactly how most of the people use this platform. As infrastructure to store their memories and share them, mostly within their own social circle. Not as a curated place where content is produced mostly by a very small (but very profitable) niche of creators who do that as a job. It was supposed to be the TV-killer, but it’s gradually become just another TV service - with a few channels getting most of the views, and advertisements everywhere.
When you store most of the media content that our whole species created in the past two decades, you have a strong duty of making it accessible to everyone, all the time, with the smallest amount of disruption. And that’s exactly the opposite of what Google has been doing lately.
Another example of how YouTube is used: I just installed a new washing machine in my house today. It came with a minimal cryptic manual, and full graphical instructions in English were only available by scanning a QR code which (of course) redirects you to a YouTube video.
Nowadays, YouTube allows you to watch videos only if you’re logged in. And it’ll keep throwing ads at you during the whole process.
How is this compatible with content such as “base instructions to install household appliances”? Luckily I have tons of way to bypass YouTube’s dumb checks and immediately consume their content everywhere I want. But what if it was up to the average Joe? They’ll need to have a Google account to sign it to YouTube, or they won’t be able to get the full installation instructions for the product they have purchased. And they’ll probably get some annoying ads while following those instructions (maybe while holding some piece of the appliance in uncomfortable positions and waiting for the next step). Because the PMs at YouTube think that all people use their platform for entertainment purposes, and thus they won’t be bothered by a small ad while they watch an entertaining video on their couch. And they don’t even realize that they are just focusing on a tiny subset of the use-cases, while breaking the experience for everybody else.
Going forward, we should all consider not publishing to YouTube anymore. Unless we are professional creators with some actual following there. Because YouTube is increasingly been shaped only around this subset of users, because they are the easiest to monetize and those who bring most of the views, and the company honestly doesn’t give a fuck about everybody else unless they bring views (and thus more eyeballs that spend time watching ads).
It should never be used for storing things to be shared only with a small circle, and even less as a permanent storage of your memories.
Google can’t be trusted in anything, and yet we’ve donated them all of our creations of the past 20 years, thinking that they’ll take care of them forever. Until a few years ago, it would have been unthinkable to invite some friends over to watch a VHS/DVD of your wedding, and getting interrupted by ads every couple of minutes, or getting all the comments and reactions of all of your friends profiled by a 3rd-party for advertisement purposes. How did we let a single corporation change our habits and expectations so deeply, and mostly for the worse?
https://www.theatlantic.com/technology/archive/2024/01/how-many-videos-youtube-research/677250/

2024-01-28 22:26:47

- reply

@npub1pss…6x53 yes, if you tune - Xmx/-Xms and friends properly. And if every corner of your code knows how to deal with complex hierarchies of objects, knowing that a forgotten reference is forever.
In practice, I run several Java applications on my prod servers (Elastic, Kafka, Keycloak...), and even with very low workloads they end up taking between 500M-1G of memory - or whatever you allocate for -Xmx.
If a language's memory performance requires both developers to know how garbage collection works and how they should handle references (because, unlike Rust, the compiler won't tell you anything if you leak references across contexts), and sysadmins to tweak heap allocation options based on heuristic estimates of how much memory the application requires, then that language has a long way to go towards sustainable memory management.

2024-01-28 19:21:40

One day I'll probably understand how a #Kafka broker sitting basically idle and with no traffic on a server I own eventually ends up eating ~1GB of RAM.
#Java and Java-based languages are everywhere, and just because we've decided to just pack more RAM in our machines every couple of years we've forgotten just how messy and wasteful they are when it comes to memory usage.

2024-01-27 09:11:41

- reply

@npub1ztc…n459 @npub1xzm…eykm If you prefer a very closed and controlled environment, then just keep using the App store and funnel all of your transactions through them. I don't think anything will change for people who are still keen on this use case / business model.
But it's very dangerous to assume thay millions (maybe billions?) of iOS users around the world all fall into this bucket. Give people freedom, and see what they'll do with it.
Of course, on the other end of the spectrum you have people like me, who despise Apple, its philosophy, its business model and everything it has built in the past two decades, and have never purchased a single product or service from them.
But between the Apple fan boy and the FOSS purist who wouldn't touch them even with a 10 feet pole, there are many shades of gray. And you can never know what they really want until you give them the freedom of choosing it.
Apple so far has strongly opposed giving its users any form of freedom, leveraging the excuse of "everyone who uses Apple products likes our monopoly and our closed environment", and I'm happy to see this finally challenged and validated by the market.