Chills

Ironically, this thread has probably underscored why there is not a Linux version of Affinity apps.  One thing I learned is that you can't even count the low percentage of Linux users in the OS marketplace as the potential market that Serif would be exploiting. You have to discuss which distro(s) you target or whatever tools are there to enable people to get it to work on their favorite one.  SMH 
Whatever the opportunity is here, it is surely one geared toward a very passionate and opinionated niche in the market. My impression is that, in the creative arena, only high end CG production has embraced Linux in a serious way because of their pre-exising necessity of building custom tools and pipelines to accompliish their goals. Building that pipeline on Linux is less of a barrier to that crowd.  But I have not seen Serif target that segment very heartily. If so they would focus more on multiple color spaces, EXR, Cryptomatte, deep compositing, etc.  Their more mainstream graphics clientele which they target seems more likely to use a mainstream OS. 
My advice is that I would not hold my breath too long or expend too much energy toward this effort to get through to Serif with this esoteric request. Especially given how unresponsive Serif has been to requests such as a simple Blend tool in Designer. 

The market simply isn't big enough.   Linux is less than 4% of the market. Of that 4% how many would want affinity?  Estimates are less than 1% of the total market. Also, it is a Hell of a lot of work to support Linux.  Also, Most likely it will be ONE specific Linux (out of the 100's of distros out there) so you are down to 0,00x % of the market.
It is never going to happen.  

Not dishonest at all.  Because one of the prime targets would be Rocky Linux for the obvious reasons. 

The kernels ABI is very stable, it is one of the authors main rules, that you don't ever break user space. The OS which use Linux don't have a stable windows manager, this is really the issue, but then the windows management is not done by linux.

You could easily make the argument that... The stable ABI of the kernel is exactly why so many operating systems choose to use it.

It doesn't. And it's a feature, not a bug. There are workarounds suitable for commercial software including "Use containerized package formats", "Statically link everything" and "Target RHEL clones: 10 years support cycle is stable and predictable enough".

Linux has a very stable ABI.

I think this talk on "which distribution" to use explains exactly why some apps are not ported on Linux. Instead of having a global vision (out of the box), some of you have a very restricted vision, close of "my distrib is better than yours"... kinder garden, sorry.
Yes there are global containers that are working for most, if not all, of the distributions ; appimage is obviously one of them and more and more editors are using it, which is easier to use than wine... and we don't care if it is for purists or if it is not fully optimized... That is not the point.
The point is having Affinity apps running on Linux. If it is using appimage even not optimized, it is fine by me !
Purists are boring... 

You are going to hell if you say that around the True Believers. 🙂
Every couple of year Linux has an upswing from 2-4% of the market, and they think the revolution is "next year". 
It's been that way for over 35 years. 
That is true. However, don't wait for Affinity to do it. They won't.  The Linux community will have to do it for themselves. 

I think it is safe to say that is never going to happen.
What I think would be far more constructive is a discussion thread on how to get the Affinity softwares to run on Linux via VM/Wine/Bottles and what the native Linux alternatives there are.

So you are saying that SteamOS Holo holds 31%  of the 4% of the desktop users and they are all games players.
This means that the general desk top community of Linux users is about 3% of the overall market. 
 
 
This is true, but there are 100's of distros "based on" those three. If it was just those three, you would not need Flatpack, Snap and Appimage...  The reality is there isn't a single installer for Linux, and even the three or five mentioned don't cover all of them.   Commercial software developers need a single install system that covers the target as it does for Windows and iOS.Untill that happens you will all be asking in vain. 

Flatpack, AppImage and Snap all bloat the system so much that they are unbearable; I've just moved to mx linux, for my linux box, exactly because snap made it unusable; This is not really a solution, not to my mind anyhow. The multitude of operating systems that have Linux as their kernel, do not become one just because you enforce a static build policy with individual package specific randomly occurring updates. Its just makes an already overly complicated foundation, even more complicated; This is fine for lovers of complexity, there is no denying it! However, it does not make linux a desktop operating system.

Linux remains a superb kernel, I have yet to see anything that I would consider to be a linux desktop.

I'm going to use a Calendar as a way of showing a few useful features I found out - and some suggestions.
Firstly, dragging out a table is great! Select the Table icon in the tool bar, select the Snapping tool to get it to snap to the margins, and simply Drag Out the table - you will see it gives dimensions - rows and column counts too! Pretty cool!
There are only a few modifier keys available. A modifier key is when you are using the mouse to move/alter something and press a key on the keyboard to get it to behave a different way.
Shift = constrains proportions CMD = sizes from the centre point where you clicked on the screen What I think is missing is a modifier key to allow the amount of rows and columns  - simple arrow keys could add and remove rows and columns would be super useful!
 
 
For this calendar - I only want the First 3 rows to appear. Month - Days - Dates
I draw out the table and let go of the mouse and low and behold - I have 7 (that was lucky) columns and 3 rows, as I only dragged down that far.

The added feature that's pretty cool!
With the Table tool still selected - you can drag down the table from the tab as indicated. Or similarly - on the right hand side, grab the tab and drag to add more columns!
You can slide them up and down or left and rigth until you have the desired amount of columns or rows.

However, once you have made your decision - you cannot remove the rows/columns by dragging them back up. I presume this is to prevent deleting information accidentally, however, if the cells are empty I don't see why this can't be done
If you want to remove the column/row you can select it and right click and choose Delete Row
However if you have more than one row/column selected you cannot choose to Delete Rows!
        No option to Delete Rows 
To delete more than one Row/Column you have to use the Table menu and choose "Delete Rows" or "Delete Columns"

 
 
Onto the calendar!
Select the top row and right click and choose Merge Cells

 
View>Studio>Table (I believe this menu and sub menus are subject to change)
 

 
Select your table and add some "Insets" so the text doesn't butt up against the cell wall - this is optional and you can adjust to how you like. 
Very important to ensure you're on the CELLS portion - there is an option for the FRAME - which is above in the screenshot of the Table  Panel, this is relating to the Frame the table sits in and not the Table Properties.
That's a bit confusing becuase you'd expect the Table options to the forefront of the options, however, not a major issue! 

In fact it's all optional - I'm leaving mine plain because I want to show another feature.
Insert the Month top row - days on the 2nd row - and leave the 3rd row blank.
 
Select the entire 3rd row and go to View>Studio>Paragraph
   Expand the Bullets and Numbering List!      
 
 
Choose 1. 2. 3. 4.     Change the text section to read as. 
 
Note - when you edit this panel option - you need to hit Return - this shouldn't be the case - if you don't hit return your changes won't be implemented. 
Your table should now look like this!

 
Grag the TAB as shown before - to expand the table to add more rows. 

 
January 2019 starts on a Tuesday (any calendar will tell you that )
Cool feature! You can select cells noncontiguous - that means they don't have to be beside each other. 
To do this - select the A3 cell - then let go of the mouse to leave it highlighted. Press the CMD key (possibly CTRL on PC) which is the "modifier key" then select the remaining cells. 
From the List Type in the Paragraph Panel - you can then set this to NO LIST.

 
Your table will now automatically renumber to ensure all the days have the correct number assigned.

 
Drag the handle down on the side of the table -  to make it fill the page! 

 
Select the numbered rows - right click and select distribute rows evenly
 

 
Duplicate your page
 

 
Rename Month to February 
January ended on a Thursday - so February starts on a Friday and February 2018 has 28 days. 

 
 
Select the first 3 days - and set the Paragraph Numbering to No List

 
Voila! February is Done! 

 
Add your icons! 

 
 
Now you can repeat the steps to make the remaining calendar months! 

To create a calendar with only dates, you can find here an easy way to do it with listed paragraphs:
Or, if you want to personalize it with feasts days and so on, you could use Data Merge, following this topic:
You'll easily create the csv file in any spreadsheet with Auto Fill feature. 

The take home from this experience ought to be that each distro represents a different operating system. It is more realistic to wish that, a linux based operating system will be successful enough, such that all the software builders will release version of their software for 'The Linux Desktop'. Whereas  wishing that any software creator write a version of their software for every single operating system that chooses to use linux as its kernel, well... this is quite the ask. Would you expect it to run in automotive systems and the likes, that also use linux? On servers too?

I mentioned earlier on that GTK the GUI toolset that was used to make GNOME, was initially written to crate GIMP, the acronym means Gimp Tool Kit. So you see the scale of the dream that you have. Linux is really a superb platform upon which you can learn what an operating system is, and this is exactly why this thread persists.
[Cue Hotel California] ... set for endless repeat.

I gave a factual technical answer.   The answer was helpful in that it gave the reasons and timescales for the TPM hardware that the commercial/corporate IT departments wanted. Hence, the ISO/ IEC standard in 2009.   In my experience, having worked on both  ISO and ICE standards, the IEC standards tend to be industry lead and faster than ISO standards. This is why motherboards started to get TPM support over 15 years ago, it was not just Microsoft doing it.
Now most people don't change PC's every 2 years but most get changed after a decade because the software they use evolves. (it's far worse on Apple for having to upgrade hardware) So by the time Win11 came out all new PC's for the previous ten years would have had TPM And be Win11 compliant.    

So yes, my post was intended to be helpful to explain the timescales on the TPM and that it was industry lead. MS put it in because it was wanted by the majority of users (the cooperates) 
Your reply was emotive and lots of guessing and incorrect assumptions.
BTW my PC was spec'ed for video editing using BMD Resolve Studio. (there is a Linux version of Resolve)  It was also bough by my company to last for a decade (or more) hence the 128GB.  The minimum RAM for Resolve Studio is 32GB with 64GB recommended. So it made financial sense on the original build to go to the 128GB  which was the max for the motherboard.  The additional cost was negligible.   The NVIDIA 3080TI was "reasonable " but nowhere near the top spec for a video edit PC. the same with the CPU.   

So whilst my PC is overkill for Affinity and below par for serious gaming, it is the sweet spot for video editing with Resolve. Well, it was when it was built 3 years ago. now, it would have the next gen GPU and CPU's.  This is purely technical and economics because I am an Engineer and think like that.   
EDIT:- Just been told that my 1080 res screens will need upgrading for the latest version of Resolve V20, as it needs 1440 Res screens as a minimum.  Also Resolve is designed to worth with three screens.   So despite your obvious complexes it has nothing to do with "measuring contest of reproductive organs,"  but the requirements of the software my PC was spec'ed for.   The actual spec was worked out with a custom PC builder that does PC's for video editing, gamers etc  and the gamers have by  far the more powerful PCs

Yes, a fresh start would be ideal. 
 
In the meantime a Linux release would be more than welcomed by those of us who would like to break free of Mac and Windows.

To be fair: NO ONE understood the significance at the time.  It was only looking back with hindsight a few years alter that you could see how all the pieces moved. 
Fortunately, I can still remember the discussion over dinner as the same person was explaining why there is no such language as C/C++ 

You're still incorrect. UNIX is a trademarked lowest-common-denominator standard, not some magical inherently secure operating system which Linux is not.
https://www.opengroup.org/membership/forums/platform/unix
Aerospace and automotive embedded use-cases and related government agency approval peculiarities are not relevant for Desktop Linux discussion anyway.

Lot software Linux users, or at least some that I read seem to think getting software developers to do things SO Linux will grow as in what I quoted. Yes Linux needs to grow before serious software developers will start looking at it, but it is not the job of the software developers to grow the various flavors of Linux that are out there. 

I would love the entire Affinity suite on Linux, it would become the most (only?) powerful photo, vector graphics, and desktop publishing packages available on Linux.
However, in defence of Serif and Canva: one does not simply package software for 'Linux'.
Linus Torvalds on the subject at DebConf '14 (~10 years ago, so may not be as true now but explains some of the difficulties):
Linus explains he's not talking about X, Y, or Z packages improving over time... (some paraphrasing to make the below quote a bit more readable, some errors may be present)
I guess the somewhat uncomfortable truth about getting any of the Affinity suite on Linux right now is that it won't happen without a somewhat stable architecture/SDK, more-so than now according to Linus, admittedly 10 years ago now.
Compiling from scratch is out of the question because one presumes that Serif/Canva won't release the source code to the Affinity suite, nor should they be forced to.
Perhaps the tact Serif/Canva need to take is by contributing code/developer time/funding (as Serif/Canva do have commercial interests here) to existing systems, such as WINE or Proton, that can allow Windows binaries of the Affinity suite to run on Linux instead?
Serif certainly know the Affinity suite inside-out in terms of what ABI/API functionality is required to make each suite function, so investing time, money, or effort into WINE or Proton (preferably both) seems like a reasonable solution over re-packaging the entire suite just for one Linux distribution let alone several, should the commercial viability and business case exist.

 

But you certainly can't expect Serif and similar private companies to solve this problem, meaning that they invest huge amounts of money into product development and then "hope" that this money will be returned to them.

Yes, but that is only one of many OS's out there that fit the bill.  Linux doesn't but many certified UNIX and other POSIX OS do. 
For any Linux fans, I suggest you read the Tanenbaum-Torvold debate. 
I would think that as the world, at lest the hardware side of computing, has moved on 40+ years it would be worth a complete new ground up design (by experts)   someting with a similar message passing architecture like this  https://www.sciopta.com/safetykrn/ Though I would not suggest that OS but something new starting with today's CPUs for a General purpose computer. 

 
 

I do not. To be a certified UNIX, the system has to comply to certain POSIX APIs. That's it. The historical heritage and direct descent from the Bell Labs code base is not required. Linux distributions can be UNIX. It's a matter of money for certification.

It’s not the MacOS that I have a problem with really, it just that it out grows the computer. The lack of ability to upgrade or fix the hardware, Macs are built to last at least physically but the shelf life is decreasing constantly. 5-7 years and it won’t run the latest OS, and consequently up to date software; despite the fact that physically the machine is fine and only needs a couple of new components that Apple make impossible to upgrade. 
To summarise Apple make life very expensive, it even has a knock on effect to peripherals such as Wacom compatibility, software becomes obsolete add ons, docs cables the list is endless. 
I’m ranting now, sorry I’m just tired of the lack of thought and greed.
I left Adobe and moved to Affinity because of similar reasons. I currently feel the same way about Apple as I did about Adobe.

That is highly unlikely to happen and so CodeWeavers ought to make the first move. Why not head over CodeWeavers' forums and make that constructive suggestion of yours?