GNU + Linux - The Future Part 3

“I’m interested in the future, because that is where I am going to spend the rest of my life”

Linux has a place in the enterprise server
Linux owns the cloud and the web
Linux own the IoT space (we didn’t talk about this one)
Linux runs at the heart of containers

One thing is for sure, we may not know what new technological needs will be 30 years down the road, but I think Linux will be at the heart of that new idea

Where I think Linux needs to work on is the Desktop

Microsoft has begun the move from the traditional desktop to the cloud with their Windows 365 Cloud PC.

Apple announced their XCode for Developers in the Cloud which offers a cloud based application in addition to the Desktop version of XCode.

The question is will Microsoft and Apple move their operating system entirely to the cloud

There is a need for a traditional desktop, not all applications are good candidates for use in a cloud only platform. The other thing is Linux is not controlled by a corporate (yet) that could mandate the move to the cloud, unless its one of the big 4 who does it.

If Windows and MacOS moved to the cloud, that would leave Linux as the last OS standing as a traditional desktop environment.

The application space has changed rapidly on linux already with the introduction and adoption of snaps and flatpaks. AppImages is there too but probably but the size of the applications seems to be holding it back for now.

Last is the Desktop Environments themselves, none of them offers a modern (I know GNOME thinks so) UI which a user coming from MacOS or Windows would find similar to the one they left.

New UI Kits like Flutter and MauiKit may offer the way to solve this problem as they get deployed to the mobile phone and tablet workspaces. Both of these technologies offer a way to easily move between screens and in the case of MauiKit, it offers a means to run on all major operating systems as well.

Chapters
00:00 - Intro
00:20 - GNU + Linux Future
00:32 - Quote
00:41 - Linux Server Success
01:20 - ChromeOS
02:12 - Windows 365 Cloud PC
03:35 - MacOS in the Cloud
06:13 - Linux Desktop in the Cloud
06:53 - Fees for Cloud Access
07:36 - The Year of the Linux Desktop
08:53 - Linux Apps
09:45 - Application spaces
10:25 - Flutter
10:59 - MauiKit
12:20 - Maui Shell (early look)
14:24 - Final Thoughts
16:00 - Wrapup

Attribution: Microsoft Windows 365 Cloud PC ad

Follow me:
Twitter @djware55

Licensed under Creative Commons: By Attribution 4.0 License

Licensed under Creative Commons: By Attribution 4.0 License

Werq by Kevin MacLeod

Industrial Cinematic by Kevin MacLeod

Music Used in this video
Licensed under Creative Commons: By Attribution 3.0 License

#linux #cloudos #ui

[zaharib]

You obviously are a life long student DJ Ware! Thanks for sharing your learnings with us! 🥰

[emvdl]

Thanks DJ, this one was really good 🤙

[dezmondwhitney]

Very interesting and thought provoking upload. Thanks D J.

[tonywise]

The future seems to be pushing it all towards a subscription model for everything. I.e. you rent the system so that the company concerned can leverage as much out of people as they can manage to! It is very pernicious and I for one fight against it in my own extremely small way.

[guilherme]

I remember a few years ago when a guy said: Let's stop worrying and put everything in the cloud!
To this day, this seems so wrong to me.

[pctlc]

G'day DJ, very interesting when you talk about the possibility of Linux being the last desktop standing, when you said that the first thing I thought of was Nvidia adding more support for Linux Lol, if you look around at social media today, there's not much video editing, it's quick self videos on Facebook, Tik Tok, Instagram Youtube shorts etc, granted some videos shared may still come from the desktop, I believe it's becoming less as we move forward. Being a Gnome user myself I think it resembles a lot of what most people use today, iPhone and Android interface however I suppose the expectation on a desktop may be different. Great video mate thanks for sharing!! plenty to think about.

[bradm]

Another great video. Wouldn't have guessed that's where you'd take this series, but definitely interesting. I think success on the desktop comes back to the killer app question. While enthusiasts enjoy operating systems for their own sake, average users don't: OSes are just a means to running applications.

I think what companies like Zorin and System76 are doing are interesting and certainly helping to advance the cause of desktop Linux. But, these efforts must extend beyond the desktop OS to the actual applications. And this is where the challenges with FOSS come into play: 1) FOSS really is, first and foremost, created by developers for themselves (and other developers) and 2) FOSS usage is not a transactional relationship.

For example, let's say I am an IT admin for a school and decide to replace Windows and Office with Zorin and LibreOffice. The Zorin folks help get me up running and things are working well. But, a few weeks into this and I start getting feedback from users looking for certain functionality that Office had. I do some investigating and confirm that LibreOffice doesn't have these features.

What do I do then? If I reach out to the LibreOffice team and request these features, what will happen? Will they just tell me to do a pull request and add them in myself? That's fundamentally the challenge with FOSS in these "production" desktop scenarios. Users can request features, support, etc., but because they haven't paid anything, there are no expectations between the user and developer. There was no mutual transaction of value and no threat of future lost value for developers. But what if I'm willing to pay? Will the devs work on my feature requests? Probably not. I can donate to the project, but developers don't want users influencing their development roadmap by paying for certain features.

Please note, I'm not suggesting this is a "problem" or something FOSS needs to fix. But, it requires a different paradigm. Back to my school IT admin scenario, if I really want this Linux-based technology stack to work, I need to hire developers who can add features, fix bugs, etc., because it's not guaranteed that the project team will do so.  

The "beauty" of FOSS is that the code is available for anyone to build upon, but unless I'm a developer myself, development costs must be factored into the overall cost of this solution for production scenarios. So, the solution isn't really free; rather, the money just isn't going to a commercial vendor.

[alexxx]

Apple will not move to the cloud willingly because they profit mostly from selling their brand consumer hardware, contrary to MS or Google.

But mass movement of general public computations to the cloud is the inevitable (dark) future, as the Internet availability and speed will constantly increase. Most people will choose the cloud computing for the convenience it offers, like: not having to maintain their own hardware and systems; inclusion of cloud-backed AI and automation in the package, etc. People will pay a hefty price for such convenience: not only paying for the cloud will be more costly than were owning your own systems, you will also become dependent on some 3rd party service and its rules, and privacy will generally become the thing of the past. Of course private workstations won't totally go away for special use cases, but they'll become rare and hugely expensive.

[taylor-worthington]

Super interesting stuff. It will be interesting to see how WASM develops (2.0 is now being conceived), because it seems like it could go close to even encroaching on desktop features if people wanted it to.

[pandasticus]

I don’t think Linux will ever be a dominant desktop OS. Not because it’s not good or anything, it’s obviously amazing, but purely because it’s not commercial. Linux is a broad concept with a lot of open source projects, and distros try to provide a good way of gathering those projects into a usable product. Now until something like Chrome OS happens and standardises a Linux system into something marketable and sellable with their own stuff and interfaces and a separate platform, then Linux will just be that, an amazing collection of free software, and that’s the spirit of FOSS really.

Secondly, I’m sceptical about cloud desktop computing. Surely Chrome OS has done it somewhat and it seems to be relatively popular but to me it only seems usable in enterprise settings, just like AWS. A good chunk of people will always need the power of a regular desktop, either for beefy software, for offline storage, for gaming, or in general for many places in the world where high speed internet just isn’t available. That’s why both MS and Apple are marketing these cloud offerings to developers.

[eugrus]

Any efforts of Canonical/RedHat on the VDI field will also make remote access technologies for our own desktop machines better, so it's a good thing overall.

[NikkiSatmaka]

What software do you use to create those slides?

[juhanleemet]

Distributed applications are hard, in spite of successes of client/server models of computing. In essence, with most services and some/many applications going to the cloud (somewhat like a transmogrification of the centralized mainframe of many decades ago) the personal workstation or desktop is again a kind of (super)smart terminal. If AI gets to be better, it might be able to compile an optimized resident display server, with optimized common functions (even for that individual user?), and a custom/optimized protocol. Perhaps instead of having standardized smart display protocol(s), one could have a multiplicity of protocols and resident functions to optimize individual performance.

[scottdrake]

For some reason I found this a bleak end to the series. It's clear to most of us ("us" being academics, workaday programmers, designers, etc.) that the current "cloud" vision is rather dystopian. We give up everything, and worse, its inertia is undeniable, because to the end user, it _looks_ like a great deal. Cheap hardware (and larger margins for manufacturers), the promise of no technical support needs (not that it's a promise that's going to be fulfilled; like everything else since 2000, it means no humans needed and the long tail of those in technical trouble, and those with accessibility needs, are discarded).

And with it, an inescapable walled garden, all personal data commoditized, security breaches "resolved" with a class-action and year of "credit report monitoring" (lol), and a prohibition on creativity (_real_ personal creativity, not just using Apple's or whoever's application to touch up photos or edit video, which means tools that someone employed by them has thought of, not you.) It's dark.

Not that it matters coming from a rando in YT comments, but my reasons for using GNU/Linux include the political and philosophical. Inconveniences are part of the support I give to FLOSS (and I contribute to and create FLOSS software, but it's fine if someone doesn't or can't.) Using GNU/Linux can be a sacrifice I make for a the future, and at the same time be an intrinsically amazing experience; ie., not mutually-exclusive. The calculus utterly changes if no one not employed by cloud OS providers can contribute to the future of computing any longer.

[CrustyAbsconder]

I think we will someday see GIMP 3.

I think we will someday see kernel 5.22.

I think some apps that are written in Rust will replace those written in C.

I think Fedora 37 will be pretty good.

I hope we will see new games written from scratch for Linux.

I think using Crostini on ChromeOS will get better.

I think the governments will ruin the internet.

I think we will have affordable s*x-robots running Linux in 2050.

I think I will someday play Freecell on an Apple M2 Max.

I think we will see motherboards with lots of PCIe 4.0/NVMe slots.

[lenwhatever]

Flying cars? no, but reading through the licenses in my car dash, I know all of them.... I know some of the developers and have even contributed code to one or two of them. Yeah, linux is in our cars, flying or otherwise.