 I don't know if you noticed, but all of the major or all of the desktops in Linux that I know of don't have telemetry on by default. That's very weird, almost as if in the Linux world everyone cares about privacy and this sort of stuff, you know, so when you actually build something in the Linux world, it won't have telemetry out of the box in theory, or at least it's open source and you can take it off. That's the theory. And as far as I know, that's also the practice for all of the Linux desktops. So this also comes with an however, however, for whether actually designs those desktops because sure, you do care about the privacy of your users, but that also means that when you design something, you are not able in any way to actually know what they want, what they use their settings like, what they use the desktop like, you don't know anything about your users and sure, you can go to forums and ask people that's happening for sure. I myself go to the KDE forum a lot to check what people think of KDE Plasma. However, that's a very skewed perception. It's not statistically accurate, obviously. So something that I saw a lot is the major desktops in Linux actually trying to understand their user base better. And that's very natural. And I think it's very interesting how different desktops approach this, trying to be as, you know, respecting of the user's privacy as possible, because obviously that's a common goal for everyone involved. So let's start off with elementary. And elementary did last year something very interesting. And I even did a post about that to actually get people in a UI study, which was actually like a form that you could go and compile. And I compiled it. I made my girlfriend compile it. And based on that, they actually tried to understand some aspect of how their users use computers in general and regardless of the desktop, if I understood that correctly. And that was actually very interesting. And of course, they also published the results publicly so we can actually go through them. And with all of the questions and there are stats. And it is stuff like how often do you click tap an app in an app launcher to open your frequently used apps versus, I don't know, the doc or the search in your system, this kind of stuff. And that was rather interesting. Of course, you still have a very strong bias in your selection because people who actually participate in the studies are not representative of your users. Probably the users that want to keep the default are those who don't care about much about this kind of thing. So they're probably underrepresented. If you do this kind of studies, probably you're over-representing whoever customizes their desktops as an example. But nonetheless, this is interesting. However, there's only so much that you can actually ask in Google form as an example. And stuff like, I don't know, what is your system version or how often do you open this application or what extension do you use in GNOME? That's not going to be very easy to actually measure for you. There's this other survey, which is still a survey, which was rather interesting. And this one also focuses a lot on what extensions to the GNOME shell were used. This is older, 2017. And one of the results as an example is that Dash2Doc is actually really popular. And I think I agree with the explanation that the GNOME team gave, which is Dash2Doc's popularity doesn't necessarily mean that we should ship that extension by default, but rather it shows us the users would like the doc to be visible at all times. So maybe that year was to have an option for that. But it didn't happen as far as I know, but Inubuntu as an example, which is who actually revealed the results that does happen out of the box. So it does make sense. This is still a survey. So let's actually go with something different, which GNOME just did, very recent this one. And this is GNOME info collect. And how this works is it's not more a survey, which also probably takes more time compared to this GNOME info collect. And it's actually a program that you install and you launch. And this program takes all of the interesting data in your system. Of course, nothing that could actually expose you, not your personal data, but the version of the system. There's a whole list in here, this kind of stuff, whether we have a flatback installed as an example. That's very useful. I would like to know that when they end up with this one. And so you install it, you download it, you actually see what data will be sent. And you give an OK, and it will be sent. And of course, they have an entire section about data privacy, and they express that a salted hash of the machine ID and username used. So they, as far as I understood it, won't have your username at all. And the data will be anonymized when it gets to the server. So very interesting. This is another approach. And of course, a bit of a flaw on this one is that it's slightly weirder compared to the other option. I mean, if you do a survey, you know what you're getting into. In this case, you have to download a program, sometimes manually, if it's not packaged, and run it. That's an interesting approach for sure. But maybe for the users, it's slightly weirder. And it might provide with even more biased results, because maybe not everybody is really willing to go ahead and install a program and run it, whereas a survey, maybe it's more likely for a user to have that. But surely this actually allows to grab some info that wouldn't be able to grab with a survey. Now, let's actually get to KD Plasma or KD in general. And I think my personal opinion that KD does have the most advanced system compared to all other desktops, which is K-user feedback. Now, by the way, I saw some users saying some very wrong stuff about K-user feedback, such as I'll go quickly through them, that it collects data when it's turned off. And that simply falls. I can do another read about that. Or the fact that it's mandatory to use KD Plasma, which is false. Or I don't know that it shouldn't be even installed, because an update might actually turn it on a hundred a hood without you noticing. And that's also impossible. So I could talk about that later. But those kind of privacy issues are not actually there. And let's actually talk about the interesting part of K-user feedback rather than the false myths around it. So this is very similar to GNOME InfoCollect with some major differences. The first one is that it's usually, usually installed by default when you actually get your Plasma desktop, but it's turned off. So it doesn't collect nor send any data by default. And as I said, an update could not turn it on. That's not how KD works. So what you can do, and it's very simple, you open up settings, and you have here, send user feedback, you push it up to the maxed, you apply. And you can actually see all of the data that will be sent. And you can actually customize how much of the data you would like to be sent. Again, always in this list. So this means that compared to before, you don't have to install a third party program to actually be able to send your system data. And you can actually decide how much of your data you want to send. So this also, and this is another very pro, whereas GNOME InfoCollect only works one time, the K-user feedback, as far as I understood it, again, is actually over time. So you do it one time, and you know that you can turn it off later, if you want. Nothing keeps you from having it enabled. But if it's enabled over time, it should keep on sending KDE the system data. And this actually allows us to have some interesting information, such as actually checking whether people are updating to the latest version of KDE Plasma or KDE. There's another very cool thing is that this one does not stop to, as it says again in the pop-up, to yes, this is no data. But if I do like this, this doesn't stop to actually KDE Plasma, but it can also be implemented by third party KDE applications. So as an example, if you have KDE installed, KDE does support K-user feedback, meaning that the KDE developer can actually do a graph and they have done it in the past and published them about whether people are updating to the latest version of KDE, which is pretty cool and useful. Now, it does have some issues, such as the most important things that we would like to know about, that we would probably ask in a survey, are not available here for reasons such as, I don't know, whether you're using, I think, single-click versus double-click is not included, and that's something we're very interested in. So that's a bit of a pity. Also, it's very interesting that in theory, K-user feedback does support surveys as well. I've never actually seen that used. But again, there's also frequently asked questions. Why does the telemetry data contain no unique user machine ID? And so, again, this doesn't send over username or anything. It is everything is anonymized. So it is as privacy-respecting as all of the other instruments. I think that in this case, KDE Plasma does have the most advanced tool to actually get data. But sometimes I would like to see some service as well from KDE Plasma because not everything can be grabbed by K-user feedback. However, sadly, we still have some bias in who actually enables this K-user feedback because if you do enable K-user feedback, again, it's more likely for you to be somebody who customizes their systems. So there's still a bit of bias, but it's still a very interesting data. So through a cap, I think it's clear that, yes, you can design a desktop without knowing anything about your users. But you do want, if you want to make a better and better product, add some ideas on what version your user is using as an example. That's something that's useful to know. And how they use your computer, what settings do they have enabled. Of course, we can't have such telemetry by default out of the box because that's very against privacy, focused on how KDE GNOME everybody is. But via surveys or programs to install that send you data, or an API which can be actually used by all KDE programs and that can be turned on, obtained by the users, we're actually able to gather something over the users that will actually help us make KDE Plasma, GNOME, elementary, whatever better. So I think that's interesting. Do you have any better ways to approach this? I think that a discussion about this could be useful. Oh, by the way, today an episode of the Linux daily podcast, the first episode was published and it talks about, as an example, the fact that KDE now has half of a desktop environment inside of it. So you can check it out on a lot of platforms going from Spotify to Apple Music. So also link in the description if you're interested, description, whatever.