TheDesk desktop environment | Hacker News Reader

TheDesk desktop environment

github.com

131 pointssmartmic87 comments

As someone who reads and writes documents all day, I just want a desktop that helps me visualise a stack of messy pages, books and things like calculators or clocks on a desktop. Doesn’t need to be VR, a top-down view would do.

There have been a few cheap looking, janky and slow demos in the ‘oughts and late 90’s, but I’m confident that personal computing is able to do this without significant performance loss, and it’s not such a massive leap from tiled window managers today.

I firmly believe that our current way of managing and interacting with information could benefit from an extra dimension: ragged corners, thickness, persistent placement on a desk, and other attributes that make a paper object much easier to identify, locate and differentiate in real life, compared to a bunch of documents named something something.pdf/docx

That model of visual-memory-centered skeumorphism died in the 90's. Because for a majority of people, they get no benefit from persistent visual placement of content, or visual differentiation of documents, and can't cope with the "desktop" in REAL life, much less the metaphor that was "every desktop UX between the Mother of All Demoes and the death of even trying that was the iPhone" now people interact with the world through vertically siloed apps which all have different management models for content, and make it easier to charge you feudal rent on what would otherwise be your own possessions.

And what especially galls me is that if you suggest this "skeumorphic" model to OSS devs, they act like you're crazy. Even if it's something simple like "the Indigo Magic Desktop's scalable line art icons that had animated states". Suggesting visual interfaces seems to grossly offend people who develop software.

>Because for a majority of people, they get no benefit from persistent visual placement of content, or visual differentiation of documents

You say that, but I don't think it's true. (I think you even argue against it later in the comment but I'm not sure??)

I agree.

I liked some things that looked more physical/tactile.

To be clear, I don't care if my calculator looks like a desk calculator as much as if clickable buttons show they are clickable. Underlining a link might be inelegant, but it is also functional and immediately knowable.

This is why I think people loved the concept of the "desktop" in Windows. Those that filled it with icons, in their own rows, columns, grids, corners, full of games, applications, documents. To me it is one of Windows's biggest successes. And yet some dismiss it as unecessary and even remove the feature altogether. I think it should be embraced even more, make the messy desktop a first class citizen for people to organize their thoughts, with ways to interact with it without having to minimize all windows. Windows 10's menu did this in a neat-enough way with its customizable menu, what happened to that?

When I moved to macOS back in 2011 (OS X back then), from Windows, I was shocked how much the OS pushes me to organise my digital world better than the mess I had in Windows. I never have piles of stuff in my digital life, it’s uncluttered now.

These days I use Linux with both swaywm on a laptop with Arch do-it-yourself Linux, and mainstream Gnome on a Fedora desktop (workstation edition) with almost no fiddling. Both desktop metaphors allow me to organise my digital world very well.

From that point of view the mess of icons on a Windows desktop looks horrible to me. I dread the very thought of using Windows ever again. I dread I’ll be given the locked Windows laptop and forced to use it.

I know it’s not that bad. Especially, if you just need to perform some tasks in some GUI apps, especially if those are Microsoft apps anyway. Still, I have the fear. My Linux familiarity created me a very vulnerable place of comfort and possibly an illusion the things would be that way forever (for me). Which can be true, actually.

I'm curious about why you're referring to the Windows desktop in particular. Don't pretty much all OSes do that?

Some desktops like GNOME remove the feature altogether.

And I will never forgive them for it. I aggressively uninstall all traces of ubuntu-desktop after a new build, and usually regret doing so by the end of the night.

I'm working on a modern 3d filesystem, ala fsn/fsv, (in godot 4.3), as part of a bigger project. I think the spatial opportunities will scratch that itch for some people.

I find it remarkable that there really doesn't seem like a lot of momentum to innovate the desktop. I'm dead sick of Windows and Mint is just a carbon copy.

I want several, hard, discrete desktops. I want one that I can toss a bunch of graphics files onto that has something kin to "Fences" so I don't have to navigate the file explorer. And with the flick of a button I can move to a different desktop that I've set up for work, one for play and etc... I think Ubuntu used to have something similar, though I wouldn't be surprised if it actually behaves much like Windows and is just a window manager.

I do a lot on the desktop to specifically avoid the file manager because file managers seem to be intrinsically unnavigable. And then there's inconsistencies between applications which is enough to press me ever closer to an aneurysm. Going between the Win10 file explorer and one from Win95 a la ImageJ is awful.

Shit, with a system like that I reckon you could almost get away without a file explorer.

windows 11 does have the option of discrete desktops. not sure if that is what you are referring to.

I'd say it has the opposite, the desktop itself is shared among several virtual environments. However, I believe W8 did have it.

You know how you swipe between pages on a phone desktop? Organize all the logically coherent things together for productivity and badda bing badda boom.

And Fences does that, mostly. It has "pages" and the "Fences" themselves. Implementation is... Passable. And unfortunately Stardock's reputation isn't stellar.

It'd be nice to see a native system for Windows or Linux that replicates it.

I'm on Windows 11, and when I click the "Task View" button which is usually right next to the windows button on the task bar, I get a layout of all my open windows on my current desktop. But there is a ribbon across the bottom of the screen that shows "Desktop 1" and "Desktop 2" and another with "Add Desktop".

I can have a dozen windows spread out all across my 3 monitors on desktop 1 and a different windows placed on Desktop 2. when I switch to desktop 2, it switches the whole view to the other windows.

And now, as I write this and test it out, I realize that you must be talking about the icons on the desktop rather than the open programs? That would make sense, because those icons do stay the same no matter which desktop.

I guess I just never thought about that because I almost never use desktop icons anymore. I just hit the windows key on the keyboard and start typing to find what I want. I actually rarely even see my actual desktop, because that would require me to close everything or minimize everything to see the icons and that is just an extra step.

I think this should be replicated per-project.

The ability to switch to a project, and see all my notes, documents, apps for only that project. In a way that doesn't collide/interfere.

maybe have ~/Desktop/project1, and all apps use that to store settings, notes, etc.

could be a group of desktops for a project.

maybe like virtual desktops for a project in columns, but each row is a different a separate project.

The idea would be close to macos with fast user switching, but one user per project.

A better way to frame it instead of one user per project, is to have one persona (casual me, working me, planning me) per workspace, each of which can only be used by one identity on the whole machine.

Multiple desktop shell sessions per user. Oh yeah, that would rock

lol, I wasn't framing it in terms of users, it was just that you could corrupt fast user switching, which exists, to be fast project switching (which needs to exist).

You just reminded me of "bumptop" - which I had totally forgotten about:

https://github.com/bumptop/BumpTop/wiki#background

Interesting to see it's up on github

I think this is one of the janky PoC’s I recalled from the ‘oughts. I don’t really like large touch surfaces (they remind me of broken interactive displays in museums), but I’m surprised the concept hasn’t evolved further.

Awesome. I've not tried this one. I've tried almost all the other ones though like Gnome, XFCE, Plasma, Fluxbox, the one that doesn't use a mouse much, Unity, FWVM/Crystal, and a bunch more that I can't even remember.

Have to admit that even though I liked XFCE the best from the Linux world, I still like MacOS the best because it is the most stable, the most consistent, and has almost no bugs, even if it has less features. Microsoft is too ugly to use at all.

What differentiates this from other desktop environments?

From what I saw in their video, there’s a top bar that on mouse-over expands and becomes a virtual desktop switcher with an arbitrary number of virtual desktops. Otherwise it appears to be a somewhat standard floating window setup. I didn’t see a menu, but my guess is that there is one or there’s some sort of rofi/dmenu work-alike or that it makes use of one of those.

Kinda made me dizzy - animating the whole panel with all that stuff (clock, etc) can't be any good for accessibility, but no easy task to do a whole DE regardless.

Oh man, you’re not kidding, watching those gifs made me reflexively long for the “reduce motion” settings toggle.

I really don’t like to put down the creative work of others, but yeah. The motion stuff made me realize just how great UIs like Win95 and early Mac System were. I love creativity and experimentation, but I dislike lots of motion, lots of animation, and especially stuff that moves under my cursor without me moving it.

It's a new one? I guess you are not the target audience of this Windows manager if you are not willing to give it a try and explore the features and the experience before rationalizing it's usefulness

While I'm sure -I'm- not the target market for a new Windows manager, it really is helpful to understand up front why a project exists.

Sometimes it's a different feature set. Sometimes it's the same (or reduced) feature set or an emphasis on performance. Sometimes it's the side effect of using a new language. Sometimes it's just an exercise in learning.

While these things are all completely valid (and more besides) the underlying reason is usually helpful when deciding if I am "willing to give it a try". Context matters, and it's helpful to understand if the developer's context is compatible with my context.

> It's a new one? I guess you are not the target audience of this Windows manager if you are not willing to give it a try and explore the features and the experience before rationalizing it's usefulness

To add some perspective, I am the target audience for a new Window Manager, but I'm still gonna need a goal or mission statement before I try it out.

No need to be fancy, just use one of the following stock phrases:

For people who like eye candy...

A clutter-free, spartan environment

Smart(er) tiling/stacking/workspace management

Lean and fast

Big with tons of value-added features

Highly customisable

"It Just Works(tm)"

A (simpler|more complex) alternative to (Gnome|Plasma|Windowmaker|Xfce)[1]

A (new|old|ancient) way to manage workflows

Designed for (programmer|designer|gamer|salesman) productivity

After all, creating a minimal Window Manager is about 2kLoC (in C). Someone who went to the effort to write maybe 20kLoC

a) Ain't stupid or incompetent in the least

and

b) Must have had a good reason for doing so!

Telling the world that reason can only help - the majority of us support new projects, we don't diss them for no reason (for that, we go to reddit :-))

[1] Yes, I know that some of those are desktop environments and some are window managers. No, I don't want to argue about the difference.

>Lean and fast

Moksha Desktop....

I think your assumptions around points A and B are both potentially mistaken.

Xmonad stayed under a thousand LoC for it's first few releases.

As for this DE, it's main feature seems to be having a topbar that expands itself on hover. But to me that seems like it's biggest issue.

The animation will always seem choppy. And you cannot click anything on the bar directly because everything moves around. And you will inevitably accidentally expand it. It's just bad UI...

> As for this DE, it's main feature seems to be having a topbar that expands itself on hover. But to me that seems like it's biggest issue.

> The animation will always seem choppy. And you cannot click anything on the bar directly because everything moves around. And you will inevitably accidentally expand it. It's just bad UI...

So?

I mean, I agree it's bad UI, but some people will absolutely love it. I mean if it was so universally regarded as a bad UI we wouldn't be seeing the same pattern on every third website, would we?

It's all different strokes for different folks. Doesn't appeal to me (or to you), but I think there's enough people who prefer prettiness over ease-of-use; after all, look how many people still love their Macbooks, even though it's still got pretty a poor UI.

A thousand lines of C is not equivalent to thousand lines of Haskell.

Well, I'd kinda like to have some idea of what I'm getting before I download it. Otherwise, why even bother?

You are asking users to commit a certain amount of time, effort and resources to testing out your software. The least you could do is set some expectation other than "LOL, n00 S0ftW4rez, yo. CZekkit, b1tches!".

Do you have any clue how many window managers are out there?

Just the ArchLinux wiki alone lists 66 window managers.[1]

Writing a prototype window manager takes maybe a day, it can be done under a 100 lines of code.[2] I am not going to test every new window manager.

[1]: https://wiki.archlinux.org/title/Window_manager#List_of_wind... [2]: https://github.com/mackstann/tinywm

I really wish we had a standard wayland "server" that we could write independent window managers for as easily as we can for Xorg.

Found more about it here: https://vicr123.com/projects/thedesk/

Not to be polemic but... Another floating windows manager ?

It's about time to came back to free tiling (like Emacs/EXWM where there is no rigid "tiling model"), since we no not need to keep moving windows, we do not need icons covered by any windows, overlapping windows and so on.

Most of the time we work on a single maximized application, otherwise we have some windows side-by-side to see both and so on. Use case for floating tend to be zero. Similar use case for most launchers and menus tend to be zero, most of the UIs have choose search&narrow over time, from "dashes" to Android preference passing though desktop quick-launchers witch are in the end a specialized CLI.

The widget model, the floating windows model is failed and proved to be even if too many still use it. The DocUIs are the past (SUNWs anyone? Postscript UIs?) and the present and the foreseeable future...

The DE as a tool proved to be next to useless, we use applications and need integration non desktop prove to been able to offer beyond very basic cut&paste and some protocols supported only by few apps and essentially irrelevant.

> Most of the time we work on a single maximized application, otherwise we have some windows side-by-side to see both and so on. Use case for floating tend to be zero.

Try telling [John Siracusa](https://atp.fm/96) that. Seriously, I'm with you, but I think there's a significant set of users who are not.

Well, I'm listening it now (so far no anti-tiling, so I can't directly answer for now) but I can share my personal journey, as nearly all I've started with floating windows, at a certain point in time I've realize that's not a nice thing, there was a path:

- first, no more icons

- then a quick launcher

- most of the time living in a CLI

- then a first test of tiling with i3 used just in mere tab-mode and I've felt the sense of being a bit lost for a moment BUT after anything seems smooth enough

- finally the enlightenment switching to Emacs/EXWM

I've experienced a similar path when I ditch my decades-old hyper-curated home taxonomy to org-attaching all my files accessing them via org-mode, org-roam managed, notes.

I've hesitate MUCH, I've tried for very long time with limited set of low importance files and so on. Evolution is scaring when it's a kind of revolution and you know it's not a game, it can be painful as well, but it's the reason we are not in caves anymore, so it's a thing humanity have embraced. Not all, of course, I bet a very small minority, but that minority is the one who steer the evolution, others follow.

Long story short if we advertise our findings, the reasons behind a specific move, and a little bit at a time we spread the idea, things will happen. Even if today many say we are at the end of the history the history never end. Those who think the contrary can talk about themselves, meaning they count not to change themselves, and generally are wrong even at that, we do change, we like it or not anyway...

> Most of the time we work on a single maximized application

I’m not sure who the “we” is here but I don’t. I don’t go to the extent of the sibling comments ATP link (worth a listen btw, toward the end of the episode from memory) but I have half a dozen windows open at any one time and it’s very rare I will make something full screen or just two side by side. Even if constrained to my laptop screen I will usually have 3 or 4 windows open and not as just an even split of the screen.

I have colleagues who work in Mac full screen mode and I get motion sick watching them move between (and forget the location of) different apps and windows.

Well, we meaning the most common kind of users, techies or newbies or luser all together: most part of humans today use a desktop like a browser bootloader, to a point someone have invented "the Chromebook" because of that, bringing also a limited kind of tab-based UI to the masses.

Allow me a question: when you read your emails how many windows you use for that? Personally I use one (notmuch-emacs) or two (notmuch+a note), when you buy some stock do you use some full screen software (no matter if local or web) to see market data or you have many other windows shown at the same time? These are just two example of common activities. Of course if you are a frontend programmer you probably have much more stuff shown all your work time, but I bet none of them specially overlapping. That's is the "we" I, a bit arrogantly, use above...

> I have colleagues who work in Mac full screen mode and I get motion sick watching them move between (and forget the location of) different apps and windows.

I use Emacs/EXWM and well, yes, often I forgot some buffers/X11 apps open, but who care? The point is that anything is at my fingertips. I can easily create windows layout, or save-and-restore them, I can easy flip windows (mode-line-other-buffer bound to a keystroke) is MUCH more efficient than moving the mouse around.

I long for the days of early MacOS X where there was no maximize and the green button just optimized the size of the window to its contents. The days of tear-off tools where I could pull toolbars and menus out into their own windows and just build a little workspace for the exact task I was working on.

The maximized single window of Windows stood in stark contrast to MacOS X’s floating windows. I came from Windows, and it took time for my brain to buy in but when it did I tell you, true productivity awaited in a way we are direly missing.

Nothing makes me sadder in life than seeing someone using an ultra wide with a maximized browser where the contents take the center 8th of the screen.

Did you try just seeing Plan 9 GUIs? They do more than what you ask and better, you do not need to "pull out toolbars", you simply create them typing functions names in a relevant place. Try Pharo and you'll see and even powerful model, the UI is just an element to makes the text active and user-programmable for anyone.

If you think about them you'll realize why the widget GUI model can't be an answer, it try to offer "some" flexibility, but can't be enough, trying to avoid the need of an acculturated user, well, the acculturate user is not an issue, it's a target. We MUST form them, because IT is the nervous system of our society. We should stop making Ford model workers as a kind of dumb-terminal meat-based robots, we must craft people, humans who bend their desktops to their needs.

You can't create anything like that up front, you need empowered end users. The current old mac model is rubbish today since we are in the era of WebApps, and we are there because we need DocUIs, and the current WebUIs are not flexible enough on the user side, and that's why we start again NotebookUI and alike, modern 2D CLIs, trying to deny we need end user programming while we slowly push it again. Old Mac like modern Gnome SHell want just to be at the center for narcissistic business purposes.

> Plan 9 GUIs

Mostly just Acme, though. The others are kind of mediocre and don't have this mechanism.

Oberon does implement something like this system-wide, and iirc that was the inspiration for Acme.

> Most of the time we work on a single maximized application, otherwise we have some windows side-by-side to see both and so on.

With a few caveats (some smaller floating windows, always on top, occasionally for reference or playing a video I'm half paying attention to) I'm with you there. Under windows FancyZones is my current tool of choice, effectively splitting my 32" screen into two (or occasionally three) by various proportions, I don't use Linux on the desktop ATM but last time I did tiling window managers were my preference.

Though given how I keep lots open from various trains of thought, particularly browser tabs, spread over a few virtual desktops, I'm starting to think nested virtual desktops might be a thing to look for and try…

While we are here: if anyone can recommend a good tabbed file manager for Windows I'd be happy. Dual-pane works for some things but not otherwise. I've tried a few over the years, some that wrap Explorer, some that are entirely their own thing, and they've all proven unstable, take an age to start or open new windows, or are very leggy in response to any input, or some mix of all three of those problems (yep, more buggy and unresponsive than explorer itself!).

Well, IMVHO you just look at Eclipse perspectives for desktop, meaning something to save and restore a set of apps in a relevant layout. Virtual desktops are the poor man's handcrafted solution for that. Try to visualize EXWM|a fullscreen Emacs, where you just call a function and get three windows, your code debug layout, than another for your mail reading routine and so on. Do you really need "virtual desktop visual"? You just call, search&narrow style or with some keybindings, a dedicated function. Personally I use my keyboard function keys for that, for instance F7 open my mail reading setup (notmuch + org-mode/org-capture notes), F10 open my Firefox and so on. Being in Emacs running "an app" (i.e. a window) or a set of apps is no difference, it's just a bit of elisp to be run. This allow MUCH more than simple virtual desktops with much less effort everyday.

Oh, sure you have to craft and evolve your desktop, but thanks to the user-programming concept Emacs have it's damn easy. It's hard only because X11 apps are not made that way, but try looking a Pharo (Smalltalk) demo and you'll understand.

About file management: I can't suggest a proper file manager but I can tell how I manage my files, as org-mode attachments accessed via search&narrow styles. I have notes about almost anything, let's say one about EDF (the utility provide me electricity) in the note I have all relevant contracts infos, invoices etc. Well, if I look at a specific invoice just hitting a key (org-roam-node-find) and typing EDF offer me the main note and many other "sub-notes", a simple title textual match. I can tab/enter to se the main note and find anything in it, or I can add some text like "november" to get results for all EDF stuff with november and add 22 to mach 22, they I get the EDF invoice for November 2022. Since it's free form text I can place it in other notes, also (limited, unfortunately) transcluding it elsewhere. For instance I have a note for my home, witch contains anything relevant, electricity contracts and bills as well. This way you made all your "views" without any GUI limit and without taking care of crafting an maintaining a file-and-folders taxonomy. File get attached, you can access then via text, you can craft org-ql queries to see different results, it's just a graph of textual nodes, fully searchable with ripgrep, files are just a backend storage stuff. Oh, and you can use it in Windows as well. It's a kind of Paperless-ngx without paperless rigid structure and limitations. Quick archiving of recurrent things, like an utility bill, can be automated via scripting.

I'm stoked for "mosaic tiling":

> a new window management mode which combines the best parts of tiling and floating https://blogs.gnome.org/tbernard/2023/07/26/rethinking-windo...

Honestly I'm not at all convinced, it seems to be another try to push childish stuff in modern systems... Yes, widgets based GUIs are designed with a specific aspect ratio and size in mind, DocUIs tend to be designed with an aspect ratio but scale well enough (think about web-apps, not just Emacs buffers). The lack of integration is only solvable with the classic model "the OS is an application, there is a single unique environment for anything", so all apps are written as part of the sole system/framework from the most lower level to the highest end-user programming.

Something perfectly possible, actually done in the past, with very nice results, see just for instance https://youtu.be/RQKlgza_HgE to not go too much back in history toward Xerox Smalltalk workstations. but also something commercial big player do dislike, since in this model you can only pay the programmer, not the program/service so no giants only competition...

Text and relative text rendering is the most flexible and integrable tool we have so far, and the web itself is a good proof and show as well the limits of trying not giving power to the end user.

On "modern desktops", honestly Ubuntu Unity was a thing because the side bar was just a quick launcher with some icons, something usual on a good place since we have larger and larger monitor with less and less vertical space, and the top bar was as small as fluxbox at al. Gnome SHell tried the narcissistic way, forcing people to see animations just for common windows switching witch surely have a kind of WOW effects for the kids, but far less interesting to work all day. Ubuntu also pushed the HUD, or a way to replace the apps menu with a CLI search&narrow style, witch was essentially the same principle than dash and quicklaunchers in general AND the Emacs M-x model. However in Emacs anyone can craft a function in a snap and run it through M-x, in modern desktops it's a hard, long thing with much boilerplate. Just to create a .desktop, witch is a kind of functions skeleton in the Emacs model, you need a dedicated file only for that, put in a specific place, than the code somewhere else etc. People do like seen beautiful gardens, but do want quick things, so the path of regulate the world in advance all modern system try is a failure in advance.

> The lack of integration is only solvable with the classic model "the OS is an application, there is a single unique environment for anything", so all apps are written as part of the sole system/framework from the most lower level to the highest end-user programming.

This is one reason why I wanted to design an operating system, although I also wanted to change many things from what other systems do, although there are also some similarities to features of other systems.

My ideas do involve better integrations between parts of the system (although different parts can still be individually reimplemented and replaced), and my ideas involve both low-level and high-level programming.

> Gnome SHell tried the narcissistic way, forcing people to see animations just for common windows switching witch surely have a kind of WOW effects for the kids, but far less interesting to work all day.

Sometimes animations may be helpful if objects are moving around on the screen and you want to easily see where they are moving to, but a lot of the animations are worthless and I would rather not see them. If an option is added to allow disabling all animations (and to adjust their speed), then that will be a good idea.

> However in Emacs anyone can craft a function in a snap and run it through M-x, in modern desktops it's a hard, long thing with much boilerplate.

It does look like a good idea, if anyone can craft a function and run it (which is also an idea of what I might intend in my operating system design). I don't use a desktop environment on my computer and do not use .desktop, although some programs do.

It is a interesting idea, although there may be some problems too. Maybe it will help, though.

Note that X window system already has the ability for programs to specify a minimim and maximum size hint, aspect ratio, and other things. This isn't new (maybe in Wayland it is new, but in X it isn't new since it is already there).

Anyone else remember seeing this on OSFT years ago? Kinda nostalgic

Don’t Linux people realize there are a limited number of people to work on these things in their spare time for free?

Perhaps the folks working on the project in what is presumably their spare time for free do know this?

The idea is that it's irresponsible to diffuse the limited amount of effort by launching your own desktop environment project rather than contributing your talents to one of KDE or GNOME.

I don't ever see myself using this DE, but the fact that it seems to be one guy's labor of love kind of makes me happy.

But presumably they aren’t interested in working on a part of a larger project that they have less creative and admin control over. This is a huge deal and precisely the answer to your question.

Something in the scale of a modern DE could become decrepit, outdated, and difficult to use real quick without major support.

There’s more DE’s than I could shake a stick at and like 5 of them are usable for a an average user.

Well, like it’s just your opinion, man… Opinions are like asses, everyone has their own.

As someone that uses the 5th most possible DE I appreciate the state of the landscape. But more to the point I think you’re not framing this correctly. I think this question is the best way to probe get to the heart of our disagreement:

How would you coerce or incentivize the developers of this DE to work on a more mature project?

> How would you coerce or incentivize the developers of this DE to work on a more mature project?

Being a cog in a wheel is always a sacrifice, but also more successful. It’s more likely to be usable by regular people and overall more likely to be used.

The biggest problem with desktop Linux is it’s a fragmentation of unpolished projects, many of which overlapping each other.

Do you say the same of all photoshop alternatives or all different AI companies doing the same stuff...or even Apple building their own OS instead of providing their computers with Windows?

>> How would you coerce or incentivize the developers of this DE to work on a more mature project?

> Being a cog in a wheel is always a sacrifice, but also more successful. It’s more likely to be usable by regular people and overall more likely to be used.

> The biggest problem with desktop Linux is it’s a fragmentation of unpolished projects, many of which overlapping each other.

Maybe you didn't realise this, but you didn't actually answer the question: exactly how would you convince the people having fun writing their own little project to switch to being an unpaid employee on someone else's project?

I want to hear the actual sentence you'd utter to convince someone to give up a hobby and become an unpaid employee.

I mean, what could I possibly say to you that would make you give up cycling/TV/reading/art/gaming/$HOBBY and spend that time as an unpaid employee for me?

When did anyone claim this was a hobbyist project? It looks pretty serious to me.

"Serious" and "hobbyist" are not antonyms.

An important distinction, I'm sure, but entirely beside the point. It's people scratching an itch in their spare time, so we can't expect them to do things the way we want. We can choose to ignore them at best.

That’s what I wanted to say as well. This might look and even be serious, but it’s still someone’s hobby project as well. They can do whatever they want.

> When did anyone claim this was a hobbyist project? It looks pretty serious to me.

You did, right here: https://news.ycombinator.com/item?id=40269321

>>>>> Don’t Linux people realize there are a limited number of people to work on these things in their spare time for free?

How were you expecting someone to interpret "things in their spare time"? What other meaning did you have in mind for that sentence?

Regardless of whether it is a hobby or not, and how that differs from your meaning of "spare time", I still want to know what it is I could say to you that would make you give up your spare time to do unpaid labour for me.

In context, I think this is a legitimate question - you've bemoaned the fact of people doing something "in their spare time" instead of being an unpaid employee, so the question

    "What can *I* say to *you* to make *you* spend *your* spare time being *my* unpaid employee?"

still stands

If nothing I could possibly say would convince you, that's okay too - just say so, like this "There's nothing you can say to make me your unpaid employee".

Even if this is a serious project, particular author of this project has a right to build something, and try to carve out an audience for it. Nothing wrong with that.

Have you considered that people do things for fun? They may not care about some abstract concept of "success"? Maybe writing their own DE is exactly the "success" they wanted.

Also, being a cog in a wheel does nothing but perpetuate the wheel. If the wheel has a flat tire, or the wheel isn't round, or the wheel is going in the wrong direction, then being a cog provides no benefit to anyone.

You completely avoided my question. How would you make, coerce, or incentivize them to work on a different project than the one they created or the one they wanted to work on initially?

I slowed down on linux as KDE slowly degraded. I love that someone is starting something new, maybe it’ll go somewhere.

> The idea is that it's irresponsible to diffuse the limited amount of effort by launching your own desktop environment project rather than contributing your talents to one of KDE or GNOME.

This is totally wrong.

The free software world is as great as it is precisely because everyone is free to start their own project without asking for permission.

And I don't think you meant it this way, but can we please not make people feel irresponsible for pursuing projects they find interesting? That's an easy way to kill a new developer's interest, motivation, and enthusiasm. And they're sure not going to pursue projects they don't find interesting.

If you have an idea for a project you want to work on: please go and do it! I don't care if there are already dozens of alternatives. Go forth and create. Maybe you'll make the new best version of the thing, maybe you won't. But you definitely won't if you don't try. And even if your version is no good, if you cultivate a habit of making things you want to make then you'll have the best chances, and at least in the end you'll have lived a life of creation rather than idle consumption.

> If you have an idea for a project you want to work on: please go and do it! I don't care if there are already dozens of alternatives. Go forth and create. Maybe you'll make the new best version of the thing, maybe you won't. But you definitely won't if you don't try. And even if your version is no good, if you cultivate a habit of making things you want to make then you'll have the best chances, and at least in the end you'll have lived a life of creation rather than idle consumption.

I agree. Furthermore: It might have some advantages compared with other programs even if it does not have all of the features or other advantages that the other one might have. Also, even if it is no good you or others might improve it later, anyways. There may also be reasons why one user or programmer might prefer to use one program than another, or even someone might use both if they really want to do.

> The idea is that it's irresponsible to diffuse the limited amount of effort by launching your own desktop environment project rather than contributing your talents to one of KDE or GNOME.

That's just a different way for some people to complain that someone, somewhere, is having fun that they don't approve of.

> I don't ever see myself using this DE, but the fact that it seems to be one guy's labor of love kind of makes me happy.

Yup. I heartily support these kinds of things. I remember how much flack Terry Davis used to get on the internet, but those flamers have long since been forgotten by history while both he and TempleOS have their own Wikipedia entry.

I still have "Use TempleOS daily for a month" on my bucket list :-/

> That's just a different way for some people to complain that someone, somewhere, is having fun that they don't approve of.

Yep, it ain't their business, period. It's like talking trash on people who like making pottery.

In a perfect world, perhaps, everybody would align behind a single effort, and they would all agree on a vision and be able to all coordinate and work together effectively and make use of all of the time that everyone has available in the most efficient way possible. And they would all have exactly the right skills to fill all of the gaps, and there would be no overlap.

But we do not live in that world. The idea that each person‘s efforts are interchangeable and can just be dropped into a project is mistaken.

There are many numbers between 1 and 100 my friend.

There are just as many between 0 and 1 my friender friend

Although this is true in a mathematical sense, it's also mathematically true that there are exactly 100 times as many numbers between 0 and 100 than between 0 and 1. Yay infinity!

Nothing scares me more than people supposing authoritarian efficiency is the dream state.

Open source is better because it's disjointed & chaotic. It's advantage is that it doesn't get stuck, is moving and changing. New ideas happen, are iterated upon. Different people try different things catering to different use cases/scenarios/users.

Open source covering so much terrain so many ways is its strength. It's an enduring base of antifragility, by not being one unsinkable vast idea.

The classic exploitation vs exploration trade-off.

working on a project like this seems like a good way to learn and perfect skills you wouldn't be able to if you were working on a pipeline of issues for an existing package.

Yep. That's why there are multiple distributions of the Linux Distribution Chooser.

https://distrochooser.de/

https://distrochooser.snehit.dev/

https://librehunt.org/

It's very important to regulate what people do in their free time, otherwise they might just pick and choose to do whatever they want.

Where would it end? Far better to have centrally-approved free time activities to ensure those limited people don't get to just enjoy themselves but are instead spending their limited free time wisely.

There are a lot more people out there than you realize. You might wanna double-check your Dunbar number ..

Why should "Linux people" realize that?

Looking at other comments here I feel like you’re making a lot of assumptions about this project. I’d suggest to zoom out a bit instead and find a different perspective.

I see a hobby project being built just for fun. Maybe author uses it daily themself, but they aren’t aiming it at a larger audience, that’s for sure.

Not everything has to be pragmatic by your standards and useful to millions of people.

Reply on news.ycombinator.com