Not OP, but maybe because it is a survey from a Linux group and discord has treated Linux like 2nd class citizens since 2015 and they don’t give a flying fuck about making the experience as good on Linux as windows. It is an afterthought.

And it is not like they did anything special at all this year to warrant a “of the year” award. Discord has been out for almost a decade. That is like saying windows is OS of the year when they have done almost nothing but bad decisions this year and the OS is already been out for a long time.

Nice try FBI.

In all seriousness, matrix encrypted chat would be better.

You pretty much have to get into self-hosting to truly have great replacements.

Google music - tidal, or qobuz are the same thing but not as bad as Spotify. Otherwise you can self host navidrome or jellyfin->finamp

RSS Feed: choose from like 10 Foss projects on fdroid

Google home: Home assistant is self hosted and like 10x better and can integrate with more things

PDF scanner: Microsoft Lens (not FOSS) but really really really good, doesn’t require a Microsoft account, doesn’t have extra permissions, and barely has telemetry. If you are OK with >10x slower, there is OSS Document scanner

VPN: ProtonVPN, Mullvad, maybe PrivateVPN. You have to pay for one. The app is wireguard or openvpn

Google photos: Immich is pretty much a perfect replacement

WhatsApp you literally can’t get away from. Signal is the easiest replacement, but i could only get mg core friends on it.There is no way 99.9% of people can get their friends and family to switch to matrix or XMPP.

GPU advances have also gone way way down.

For many years, YoY GPU increases lead to node shrinkages, and (if we simplify it to the top tier card) huge performance gains with slightly more power usage. The last 4-5 generations have seen the exact opposite: huge power increases closely scaling with performance increases. That is literally stagnation. Also they are literally reaching the limit of node shrinkage with current silicon technology which is leading to larger dies and way more heat to try to get even close to the same generational performance gain.

Luckily they found other uses for uses GPU acceleration. Just because there is an increase in demand for a new usecase does not, in any way, mean that the development of the device itself is still moving at the same pace.

That’s like saying that a leg of a chair is reaching new heights of technological advancement because they used the same chair leg to be the leg of a table also.

It is a similar story of memory. They are literally just packing more dies in a PCB or layering PCBs outside of a couple niche ultra-expensive processes made for data centers.

My original comment was also correct. There is a reason why >10 year old MCUs are still used in embedded devices today. That doesn’t mean that it can’t still be exciting finding new novel uses for the same technology.

Again, stagnation ≠ bad

The area that electronics technology has really progressed quite a bit is Signal Integrity and EMC. The things we know now and can measure now are pretty crazy and enable the ultra high frequency and high data rates that come out in the new standards.

This is not about pro gamer upgrades. This is about the electronics (silicon based) industry (I am an electronics engineer) as a whole

Double sided perfboards way more often than not don’t have plated vias. The vast majority are just etched from a 2 layer blank without any post processing.

You would have to solder specifically on both sides in order to increase the strength.

But yes pads lifting is almost exclusively from dumping way too much heat in it. Because there is no plated through hole, it also has ~1/4 of the thermal mass so it is easier to overheat and damage the adhesive.

It has absolutely stagnated. Earlier transistors were becoming literally twice as dense every 2 years. Clock speeds were doubling every few years.

Year 2000, first 1GHz, single core CPU was released by nvidia.

2010 the Intel core series came out. I7 4 cores clocked up to 3.33GHz. Now, 14 years later we have sometimes 5GHz (not even double) and just shove more cores in there.

What you said “it’s just that you don’t need to upgrade anymore” is quite literally stagnation. If it was a linear growth path from 1990 until now, every 3-5 years, your computer would be so obsolete that you couldn’t functionally run newer programs on them. Now computers can be completely functional and useful 8-10+ years later.

However. Stagnation isn’t bad at all. It always open source and community projects with fewer resources to catch up and prevents a shit ton of e-waste. The whole capitalistic growth growth growth at any cost is not ever sustainable. I think computers now, while less exciting have become much more versatile tools because of stagnation.

Well it is literally not going as fast.

The rate of “technology” (most people mean electronics) advancement was because there was a ton of really big innovations at in a small time: cheap PCBs, video games, internet, applicable fiber optics, wireless tech, bio-sensing, etc…

Now, all of the breakthrough inventions in electronics have been done (except chemical sensing without needing refillable buffer or reactive materials), Moore’s law is completely non-relevant, and there are a ton of very very small incremental updates.

Electronics advancements have largely stagnated. MCUs used 10 years ago are still viable today, which was absolutely not the case 10 years ago, as an example. Pretty much everything involving silicon is this way. Even quantum computing and supercooled computing advancements have slowed way down.

The open source software and hardware space has made giant leaps in the past 5 years as people now are trying to get out from the thumb of corporate profits. Smart homes have come a very long way in the last 5 years, but that is very niche. Sodium ion batteries also got released which will have a massive, mostly invisible effect in the next decade.

I always seed the big ones, but sadly, rolling release distros outside of Arch and smaller distros have abandoned torrents because they change snapshots too frequently and/or don’t have the user mass to support it.

Interesting, when I started using FreeCAD I had 0 training, and I have managed to make it work just fine. Was mostly frustrated by the topological naming problem, but that is fixed now.

You can do quite complex things with it.

You just have to put in a bit of effort and think in an additive sketch-extrude workflow. But yeah, not easy to transition from solidworks.

To be honest. I had a similar question for my girlfriend for drawing with krita. A drawing tablet + a traditional laptop is better for almost everyone except students who will be taking notes in class and people who have to be drawing in a chair or meeting room with no desk setup.

Otherwise a drawing tablet is more accurate, faster, and with better features than a 2-in-1. Much better sensitivity, generally better pressure and tilt functions, and a much better feel (more like paper)

You don’t even have to spring for a Wacom. They have been resting on their laurels for over a decade and have become completely uncompetitive in the past 5 years (kind of the Intel of drawing tablets).

An XPPen Deco Pro Gen II (as an example) has good ergonomics, rotary knobs for zooming, rotating, and scaling, and works over Bluetooth. Their Linux drivers (4.0.x) are pretty great at a fraction of the price of a Wacom or the price difference between a traditional laptop and a 2-in-1.

It ends up being way more ergonomic also to look at a screen and not having to hunch over a tablet. It just takes a week or so to get used to not looking at your hands.

I hope they fixed the broken shoulder button problem. That tiny piece of plastic should be replaced with spring steel for sure for such a high-ware part.

Had 2 break on me. Sent one back, fixed the 2nd via hot glue for a while and a 3d-print when I got a printer. 3rd one has still held up (5€ at the sellout along with the steam link)

Well the gadgetbridge comment is not really true.

It can’t pull activities from Strava, runkeeper, or similar or push to them. Syncing across services via APIs or Heath Connect stops it from really being a proprietary watch software replacement or a google fit replacement as they all do that and it is a core function because people want to use the apps they want to use. For example, lifting tracking from Progression or similar which other apps like gadgetbridge or strava just aren’t made for. It doesn’t have the functionality built in to cover every activity set.

GPX is extremely limited and is only GPS data, not good for fitness trackers as they track all sorts of activities. TCX or FIT are better, but of course managed by Garmin. There isn’t really an open alternative standard. I guess the closest we come is the Health Connect API which is completely local interoperability.

Just send the energy directly back to the power executives houses with a high power laser. They want the energy for free so badly to pad their profits and buy a 5th yacht, give it to them 😉

Only partially true. The solar panels almost all inject power back into the grid. Power companies started complaining about their profits when they had to actually pay the users for their power that they generated so now home power generating houses get paid pennies on the dollar for delivering power and reducing the power capacity needed by the power companies and of course the power companies didn’t lower prices at all, so they are just sucking up the difference in pure profit.

Ugh, I wish I was in a good place financially to donate to everything I use… There are so many great KDE projects, and non-kde FOSS projects too! One day…

Okular, KDE connect, plasma, krita, freecad, libre office, KiCAD, local send, antennapod, gadgetbridge, immich, paperless, bookstack, home assistant & esphome, leantime… The list goes on, but even if I only gave 5€ per month to each, that would be more than 75€ per month and that is way above my budget with our renovation costs.

Not even to mention the good YouTube creators too! Maybe if I get a bonus at the end of this year I can spare 150€ or so.

Depends on what your usecase is for what is “essential.”

I think keeping household documents, taxes, medical bills, etc… In a local only paperless-ngx instance is quite essential to the organization of a household where everything is searchable and able to be organized on multiple levels compared to a simple document folder on 1 computer.

Having a document or self-hosted wiki with an in - case - of - death document that gets backed up in an encrypted, but accessible by family place is probably the most “essential” thing.

Actually, the AI assistant fad isn’t all bad.

HomeAssistant has an open souce assistant pipeline that integrates into the most flexible smart home software around. It is completely local and doesn’t rely on the cloud at all. Essentially it could make Alexa’s and google homes (that literally spy on you and send key phrases back to your built data collection profile) obsolete. That is a way not to have to rely on corporate bullshit privacy invasion to have a good smart home.

Indeed transcribing and translating (and preserving dying languages and being able to re-teach them) are 2 of the best consumer uses for AI. Then there is accelerating disease and climate research.

If these were the use cases that were pushed instead of fucking conversational assistants, replacements for customer support that only direct to existing incomplete docs, taking away artists’ jobs, and creating 1984 “you can’t trust your own eyes and ears” in real time, then AI would actually be very worthwhile.

What?

He is saying that AI uses countries worth of energy by itself. Even a normal search query using AI uses orders of magnitude more energy than a traditional search query.

Literally tech companies have been buying or reserving entire power plants exclusively for training AI datasets. At least Microsoft reactivated an old nuclear plant instead of buying out coal plant energy shares.

And 90% of uses for AI are absolute dogshit corporate fluff or a shiny activity for 10 year olds to play with for 30 minutes.

There are legitimate uses like auto note taking, voice assistants, etc… But it is destroying the environment because corporations are shoving it into every possible thing they can, quadrupling the energy growth rate and straining our electrical grids and burning tons and tons more coal to do it.

Here in Belgium there used to be big government subsidies for solar panels 5-10 ago.

Now the same wattage battery + solar setup without any government subsidies is a good chunk cheaper than that time with the large subsidies.

Pretty cool and shows the power of government renewables subsidies. A huge percentage of houses in Belgium have solar panels now.(and electricity still costs 0.30€/kWh average because of fossil fuel energy lobbies)

Now that there is a local industry around it, most renovations and almost all new builds include them.

Steam is pretty much the only thing that stopped the Microsoft Store.

If that had happened, you probably couldn’t even run games anymore on windows unless they were installed through the Microsoft store. Mods would be dead, and we would be in the same, but worse situation.

Hell, maybe you couldn’t even buy games, but had to buy “game subscriptions” like game pass on xbox