Pinned post

It's time!

- 🇮🇹 geek in his mid-thirties, based in 🇳🇱

- 🎓 M.Sc in computer engineering.

- My current job is about fixing and automating global supply chains, one line of code at the time, but I have worked in a wide range of industries over the past (nearly) two decades.

- My hobbies often involve automating everything around me.

- :linux: user since 2001. My experience as a Linux admin started back in a time when I used run my IRC and Apache servers on a repurposed Pentium 1 under my bed, and it still took about 10 💾 to install a full Slackware system.

- :arch: Linux and rolling release enthusiast.

- 🛠 Creator and main developer of (platypush.tech), an open-source (mainly :python: and :vue:), general-purpose platform/framework to automate everything - from smart devices, to cloud services, to robots, to DevOps operations, to everything in between. With hundreds of available integrations, you can think of it as IFTTT+Tasker+SmartThings on steroids, scriptable, and runnable on almost any device. Or maybe like HomeAssistant's lighter brother.

- Admin of social.platypush.tech, a Mastodon instance where I may talk a lot about Platypush, automation, programming, electronics and maths. I tend to write a lot, so if you're looking for an instance with a 10,000 characters per toot limit...

- Looking for relays with instances dedicated to similar topics. My dream would be to build an experience, when it comes to , that is akin to curated lists, where admins can create curated federated experiences for the users on their platforms, rather than the open-to-everything overwhelming stream of toots on the federated timeline that most of the relays provide nowadays.

- 🤖 Machine-learning enthusiast. I have published a book on it link.springer.com/book/10.1007, with simple computer vision exercises that can be run on a , and I did some academic research back in time where neural networks were still a green field fabiomanganiello.com/#research, and I never stop learning new stuff.

- 🧪️ Physics, chemistry, biology, maths and astronomy enthusiast.

- 🎵 Music addict, 🎸and 🎹 player, and occasional composer/producer You can find some of my music here open.spotify.com/artist/5H6BJf and here soundcloud.com/blacklight01

- I may often write about random politics/economics/philosophy. I may sometimes be very passionate on topics such as open-source, open data, open protocols, tolerance and social inequalities. I mostly belong to the progressive/social-democratic field. You are welcome to try and change my mind, as long as you do it in a civilized and data-driven way.

- 🏄 and 🛹️ rider. And, as a good Dutch resident, 🚲 enthusiast.

- 👪 Full-time father.

"After we reminded GitHub of (a) the pending questions that we'd waited a year for them to answer and (b) of their refusal to join public discussion on the topic, they responded a week later, saying they would not join any public nor private discussion on this matter because “a broader conversation about the ethics of AI-assisted software seemed unlikely to alter your stance”".

To me, this is the biggest reason why must die in a huge ball of fire and leave no traces behind.

They make big profits with FLOSS software, included the code that I've developed and hosted on their platform over the past >10 years.

They don't share any of those profits back with me and other FLOSS developers.

They claim to be on the developers' side, but they offer no chances of appeal nor legal support to developers when a DMCA takedown requests comes in and developers lose months or years of work, often without explanation.

They claim to be the largest FLOSS platform on the planet, but their own source code is not open, and they already refused multiple times to open it up.

They trained their co-pilot with my code and tons of other FLOSS code, when they could have simply used their own codebase, or that of Office/Windows, if they're really in good faith and into the idea of giving back to the world.

And, when we ask them to sit and talk together because we have legitimate concerns (especially around the amount of open-source licenses that the co-pilot violates, since the underlying code isn't Github's intellectual property and licenses must be respected), they behave like jerks and say "well, our conversation won't change your mind anyway [about the fact that we're filthy jerks], so it doesn't even make sense to have it".

Change my mind ABOUT WHAT? It's my f*cking code, intellectual property and licenses that you're talking about, and I'm entitled to decide how you use it, or at least opt out of your programs if I don't like them, you filthy pieces of shit!

If I had ripped your own source code and gave it away for free, with no references nor credits about the original authors, and you asked me for explanations, would I be able to get away with "my explanations aren't going to change your mind anyway, so I don't even want to talk to you"? Or would you unleash your whole army of lawyers against me? So just because individual FLOSS developers can't rely on an army of lawyers like you do we're expected to be your unpaid slaves, you scum of the earth?

These disgusting parasites are still bent on their parent company's embrace-extend-exterminate strategy, nothing has changed since Ballmer's times. But keep in mind that, unlike with Windows, they are NOT the authors of the content on their platform. WE are the authors. If we all pull our code out of Github and move it to Gitlab, Gitee or other solutions, they'll be left with nothing. Let's make sure that this is indeed the case and Github joins the cemetery of evil projects with no added value.

sfconservancy.org/blog/2022/ju

I have been testing for the past few days three replacements for 's notifications: , and 's Unified Push project. A few observations after a bit of tinkering:

1. The idea behind is amazing. An open protocol to share push notifications over any asynchronous channel (websocket, Redis, MQTT etc.) is what open-source apps have needed for years. Sure, there will always be those who say "push notifications are a distraction, and I'm happy to ditch them". But individual choices/behaviors shouldn't shape the development of a technology - especially when people want a genuine open alternative to something that they like/need to use.

2. UnifiedPush support from individual apps is still scarce. So far I've only found the NextCloud app itself (which only supports UP-NextPush), and . Support on has allegedly been implemented in the latest release, but I haven't yet managed to make it work. Let's roll up our sleeves and make sure that more and more of the apps that we like support open notification services!

3. The notification providers' client apps themselves are still quite buggy, and documentation still very sparse. I have used UP-Example from F-Droid to test the UP services. Only ntfy managed to deliver notifications end-to-end to my devices. Gotify reported an "unknown error" without many details from the logs. UP-NextPush is still very unstable both on the client and server side and I couldn't manage to deliver any notifications.

4. The protocol (and the apps that implement it) needs to slowly be extended to cover as many as possible of the features that have been implemented in the past decade. Action buttons, icons from URLs, custom background images, updates to existing notifications etc.: a couple of these features have been (partly) implemented by 1-2 providers, but we need open standards (especially for action buttons and gestures) if we want to ensure inter-compatibility.

So mostly relies on comments in the code to figure out what a certain block of code is about.

So the best way to prevent copilot from using your code is to avoid comments.

Who would have known that the best advice to protect software licenses would have been the worst possible advice when it comes to code maintainability?

Vatican praises U.S. court abortion decision

inkl.com/a/WAkprXHygzy

Heaven forbid a woman’s right to choose should limit the supply of altar boys for Catholic bishops to bugger.

#catholicChurch #vatican #pope #catholicism #religion #fundamentalism #childAbuse #abortion #humanRights

People value being able to log in to a platform, connect with who they want to connect with, consume the media they enjoy, find the information that’s relevant and then be gently nudged off the platform in a way that fits their time management goals. Social media could have a healthy, meaningful place in people’s lives. But that’s just not the way it’s being designed right now.

Custom lists (prompting users to organize their contacts in lists, and then consume content from those lists instead of the algorithmically curated main timeline) have a huge value - something I already realized when started to move too far from a chronological timeline and started inserting trending tweets and other shit.

Spotting conflict and prompting users to resolve their discrepancies in private instead of the public timelines can also make social media much less toxic, since 1-1 interactions are more personal and less likely to be as abrasive as public conflicts where each of the parties feels the urge to "defend" their side in front of others.

We also need to clearly condemn the social media that are exploiting and abusing our psychological reward mechanisms (the "keep digging this endless bucket and you'll find what you're looking for") to create addiction cycles similar to those of created by substance abuse and gambling. Social media should prompt users to set the amount of daily time that they expect to spend on them, and then gently nudge them off if they spend too long, instead of rewarding them when they spend all of their time doom scrolling.

Most of the people feel a sense of disgust and revulsion when they realize that they've spent too much time on social media instead of doing something productive in real life. We don't want to use social media that make us feel bad by design and steal our time from family, friends and work. If those social media don't change their approach, then their usage needs to be discouraged just like we discourage tobacco usage.

scientificamerican.com/article

is basically selling code that other people wrote, without providing them with any compensation nor credits, and disregarding any possible license incompatibilities.

And will probably get away with it because the whole thing is "AI-curated", so no human purposefully violated any intellectual property.

What leaves me speechless is that Github is making so much profit out of other people's open-source code, and they also claim that in a couple of years "there will be no more code sitting on local machines", but they are very, very wary from releasing THEIR own source code (like Gitlab and Gitea did) so people can run their own instances.

I'm still on Github because that's where all the devs and projects are. But all of my projects are now on a self-hosted Gitlab instance, mirrored to Github (and I'm even considering dropping that mirror). If you don't want Github to get rich from your code while giving you nothing in return, you should probably do the same.

A well-run business can’t afford to switch to a new approach—one that ultimately will replace its current business model—until it is too late

spectrum.ieee.org/deep-learnin

Here's a computer problem, 9/10 developers can't solve it.

you have two computers,

One you are able to upgrade any of the parts or even use an entire server room full of computers as good as money can buy

The other is a computer-like device picked randomly from anywhere on the planet statistically it's going to be a low end android phone

You have to display an image on the 2nd device

where do you choose to do the bulk of the work required to render the page?

Worst timing for to go down, after doing a factory reset of my old PixelC last night and trying to reinstall the apps.

I hope the folks at @fdroidorg
can fix it soon 🤞

It's interesting to see so many discussions around , sentient AI and so on.

It's just a shame that, beyond the two partisan camps ("it's human!"/"it's not human!"), there isn't much room to ask a simple question: if an AI had ever to become sentient, how would we know that?

We're way beyond the Turing test here. You can now entertain a long conversation with an AI without even realizing that you're not talking to a human.

So how do we know if the AI is actually coming up with independent thoughts or just mimicking and elaborating on pieces of the huge datasets it's been trained on?

How do we know if those thoughts were genuine or just words that statistically maximized a sophisticated cost function?

And how do we even define "independent" and "genuine" thoughts? Humans say and do a lot of things just to maximize social acceptance or personal gain: does it mean that our thoughts and actions are sometimes "less" human?

We are discussing a lot about human intelligence and sentient models, but we haven't even come up with a shared definition of what makes something "sentient" or "conscious". How can we even get to agree on something that we haven't even agreed on how to measure?

K-9 has been acquired by and will soon become mobile, and I have a mixed feeling about it.

I have recently migrated to K-9 as an Android mail client after the drama that recently happened with .

Being an independent developer of an open-source mail client for mobile isn't easy. The developer of FairEmail has recently been quite open about the levels of stress he has been going through - also because of Google's harassment always looking for ways to take down his app.

So the shielding of Mozilla can definitely help and ensure that an email app keeps getting developed and supported. Sure, it's also an implicit admission that they've screwed up so much their mobile app strategy that they had to pick an open-source app built by a random guy and ask him if that could become the new Thunderbird mobile.

But I can't help with the feeling that independent developers who build stuff that big tech mafia dislikes can only keep doing so if they are protected by another company or by a foundation.

blog.thunderbird.net/2022/06/f

Accurate temperature sensors, non-linear avalanche effects on diodes, and even just recording the background radiation, could all make for good true random number generators.

But they will never be as cool as a project that generates random numbers through bananas (well, technically through the random decay process of potassium atoms in bananas, but you get the point). And it even comes with a 3D printed banana holder! 🍌

hackster.io/news/the-just-bana

Here we go with one more reason to drop .

If you have a decently sized hard drive and a spare laptop/desktop, you have no reason not to run .

You don't have to pay anyone for storing your stuff.

You decide how big your storage is.

You decide what to upload to it.

You won't have tons of trackers recording your activities while working on the Dropbox/Google Drive web page.

And no faceless company can ever take your storage or your account down for violating some ToS.

I hope I won't have to wait longer for new Rick & Morty episodes just because of Dropbox wiped the creators' account...

mastodon.social/@hn_discussion

As expected, Google and Amazon have already started their lobbying efforts against the "self-preferencing" antitrust bill.

The bill says something that is obvious to everyone, except for these companies: if you have a dominant position in a field, you can't leverage you dominance to give other products in other fields an unfair advantage, because that harms fair competition.

For example, if you are Google and you have a dominant position on search engines, you can't boost your links to Google shopping or travels above those of your competitors.

If you are Amazon and you have a dominant position in online shopping platforms, you can't boost your own products above others in the marketplace.

Antitrust legislation in the US is stuck at the early 20th century, Europe had to pick the slack way too many times, and the bill discussed by the Congress is really a first basic step to guarantee better competitiveness, it's not even the endpoint. So let's make sure that the efforts of Google and Amazon are doomed to fail. If you're American, write to your Congress representatives and explain them why they shouldn't listen to Google's and Amazon's arguments.

on.ft.com/3Q2drhi

There are only two hard things in computer science: caching, and naming things (and off-by-one errors, as someone would say).

This article makes at least one of these things a little bit less hard.

thenewstack.io/best-practices-

#Belastingdienst #DutchTaxOffice #DutchGreediness 

They have already breached the EU law with an illegal tax on savings.

They have already received a huge backlash (and somehow survived it) because they tried to squeeze childcare money out of poor families that were actually entitled to the benefits.

They already have an open investigation about institutional racism and harassment of ethnic minorities.

But nothing can stop the greedy Dutch tax office from being an institution populated by racist, inhuman and greedy jerks with the emotional intelligence of a jellyfish. On the contrary, they seem to enjoy their reputation of absolute jerks, as the local population doesn't even seem that much bothered.

The last case is that of a low-income woman who was entitled to an inheritance that she never received (that's because the legal executor of her father's will gambled all the inheritance away), and yet the tax office blindly kept claiming its inheritance tax share.

This country deserves a better system than this. Today's Belastingdienst office is a sad institution whose only purpose is to confirm all the stereotypes about the Dutch greediness.

dutchnews.nl/news/2022/06/tax-

is a quite interesting category of attacks: just annoy the user with popups that ask for admin privileges until the user clicks Yes.

About mobile

Yes, I do obviously hate Google from the bottom of my heart. And I do also hate it that all the notifications from and to any Android device go through their Firebase engine. So the backlash again them (as well as 's choice of flagging Firebase usage as an anti-feature) is more than welcome.

But I think that things are getting too much out of hand. Nowadays apps like K-9 Mail and Tusky almost brag about their missing support for push notifications, as if it was a feature itself rather than a missing feature. And I've seen people write things like "oh, notifications are an annoying distraction anyway, good that we don't have them", or "thanks for staying pure".

I totally disagree with this stance. The ability to get nudged when we receive an sms, an email, or somebody mentions us on a chat or a social platform, is one of the core features of any mobile platform. Sure, one should have the ability to mute or disable specific notifications if they don't want to be bothered, but not providing the feature at all means that we (as a community) are providing an incomplete mobile experience, and our apps are unlikely to get traction with anyone who isn't willing to trade usability for privacy.

So I have recently set up my servers running , and 's unified push. I have some automation routines that poll whatever I want to be notified about (emails, Mastodon activity, new content by a specific RSS feed or social profile...), and deliver them as notifications over these channels so my mobile can be up-to-date with everything.

Let's not forget that asynchronous push notifications aren't rocket science. All it takes for them to work is either a websocket or an HTTP stream endpoint - and both the technologies have been around for about two decades. And we've also got a standard () to ensure compatibility.

So I'd like to invite more developers to build apps that come with UnifiedPush support. Yes, you guys deserve to be praised for saying no to Google and Firebase, but if you don't provide any alternative then you're just shipping a half-baked product. I can patch it for now by running a polling logic on another server and delivering messages over ntfy, but people who just want to be notified when they receive an email or get mentioned on Mastodon shouldn't run their own server-side logic to get things to work.

Show older
Mastodon

A platform to talk about automation, open-source, software architecture, data science, science and tech.