Since it seems important, a few words of disclosure first: I’m a developer of commercial software based on .NET, using an older version of DXO and would love to see it on Linux.
As for the discussion;
The argument “which distro would you support” doesn’t really apply anymore than “which windows version would you support?” for the .net reasoning. There are at least 6 versions of just windows 10 and then myriads of different patch levels…what is done is that you support a specific .NET -version, and then possibly test on the windows versions you want to “support” (and find workarounds when MS has messed something up until they fix it or forever). That’s one of the main points of .net, you run the same clr on different architectures or OSes…the difference now is that they bring it to linux as well, but the idea stays the same, you don’t support a linux version or distro, you support a .net version, and possibly test a few to have a response when MS messes up.
Now, in practice MS doesn’t seem willing to support wpf on the Linux-variant of .net, so they are creating a divide between win-.net and linux-.net already even of the same version…maybe that will be fixed later on, but given that windows forms and wpf are more or less abandoned (but still “supported” =) ) and use a lot of native calls, they probably won’t fix it, so we’re probably out of luck as long as dxo uses wpf.
As for the (more limited than I think many believe) testing, documentation, deployment etc, I agree that Linux will be a hard businesscase, even if they would choose to only formally support say debian-based distros. (then the community would fix what’s needed on all other distros but that’s another discussion…)
But what could perhaps be done, depending on what path MS takes with .NET 5, is that if it comes “close enough” to support wpf somehow, or just a few components turn out to be problematic, perhaps an informal support (don’t say that it is supported anywhere) could be implemented by just making adjustments that would allow it to run under say linux.net 5 or perhaps more likely under wine, and I’d bet that although fewer linux users, if someone makes it run the first time and writes a guide, half of the potential users could make it work since as you say linux users while not paying as much are a lot more tech than the average win/mac user.
After all, 80% of the coding takes 20% of the time (and then there’s the test, documentation, marketing, sales, support etc costs that we skip this way since we don’t formally support it), so if that turns out enough for informal-wine-support or something, it may be still a businescase, at the same time as it puts development hours into moving to the more future-proof .NET 5. (But 4.x won’t be deprecated for years due to current legacy so maybe not as relevant as an argument)
Let’s be practical about this. DXO currently adds, at best, a few feature updates to the Windows and iOS versions each year. On top of that they are actively working on the Nik Collection and also at best have been able to add a few minor updates in each version. Even if DXO saw enough of a market potential for a third Linux version, which they probably don’t, they simply do not have the resources to even consider creating it, much less supporting it long term. They already have versions for iOS and Windows which covers 90%+ of all potential users. Considering any use of their very limited resources that does not work towards increasing market share on their existing platforms would be an extremely poor business decision. Any discussion on the subject is moot and probably a waste of time.
No Linux users I know would be willing to run anything Microsoft to get access to a photo editing program. You totally compromise the security of your box, half the point of Linux in the first place. What a mess Linux is with all these distributions and still difficult to configure preferences and software. It’s amazing that noone has been able to corral Linux into a viable desktop alternative considering how atrocious recent privacy violations are on both OS X and Windows. Certainly not a good place for commercial software to invest into development at this point.
No Linux users I know would be willing to run anything Microsoft to get access to a photo editing program. You totally compromise the security of your box, half the point of Linux in the first place.
I’m going to disagree with this. If you’re running Windows 10 in a virtual machine, it’s isolated from the rest of the host OS; the only file(s) being accessed by the VM are the virtual disk and/or shared files, if configured. Dual-booting would only compromise the security of the Windows partition / disk.
What a mess Linux is with all these distributions and still difficult to configure preferences and software.
When was the last time you tried Linux? I’m not trying to start a fight here, but at the same time spreading misinformation is not a great thing to be doing either.
It’s amazing that noone has been able to corral Linux into a viable desktop alternative considering how atrocious recent privacy violations are on both OS X and Windows.
Ubuntu, Linux Mint and Fedora would like a word with you on that. All are quite viable, actually.
Now, to swing back on topic, Photolab runs great on a Windows 10 VM. I share the photos folder to the VM, do all of my editing there and then any post processing is done with some scripts I wrote. I didn’t need to buy a separate Windows license, it activated on the VM just fine.
Disclaimer: I’m running on a machine with a core i7-7700k and 32GB of memory.
Yes, if you like half-functional OS which requires the command-line and you really, really enjoy fighting with drivers. I wish Linux wasn’t like this but it is. Switching to Linux would cost me 50% productivity for at least a year. As I’m the director of a software company that would be very, very expensive. If I were a twenty-two year old with no or few direct reports and a limited investment in my preferred OS (in this case Mac but it doesn’t really matter which it is), I’d definitely pay that price. Mac OS gets worse with every iteration so at some point, I hope Linux adoption will be smooth enough that the transition won’t be too painful even for me.
Put a viable hardware accelerated Linux on deck with good driver support and then talk about migrating expert end users to Linux (yes, newbies with very basic needs could get by once their specific driver issues are resolved).
Very cool that PhotoLab runs well in a Windows VM. That’s adapting to a real-world situation flexibly.
So, what exactly is your definition of half-functional? Everything I was doing in Windows, I’m doing over here in Linux. As far as fighting with drivers, you might want to take a step out of the early 2000s, because Linux has come quite a long way. Case in point, I’ve got an RTX2070 card in this rig. Care to take a guess as to how much time I spent ‘fighting’ with drivers? If you answer with: “long enough to check a checkbox, click apply and restart the window manager”, then you’re correct. Total time? Maybe 5 minutes. Oh, and hardware acceleration works great as well.
I’m not out to change anyone’s mind, however, I will correct people when they spread misinformation. Linux isn’t for everyone. It’s not going to be for everyone, and in fact I’m a large proponent of using whatever tool is going to make you the most productive. For some, that’s Windows or Mac, and for others, that’s Linux.
But, we’re steering way off-topic here and no one’s going to change how anyone feels about whatever OS, so I’ll just leave it there. In the future I suggest some research before making statements about an operating system you have limited experiences with.
I wasted two months of my life with a switching to Linux experiment about eight years ago (Linux Mint, Debian version) and have moved three programmers over to Linux for as long as two years to finally face them imploring me to be allowed to have Macs like everyone else in Foliovision.
Of course, I’m sure you’d like me to run this experiment every year, even every three months as new drivers and flavours and major releases come out. Sorry but no.
The latest hot compatibility issues with Linux I believe are battery performance and power management on laptops, which is poor and/or inconsistent. Well one of my important devices is portable, in fact have both a MBP and a MBA in regular use. Having to babysit and fight poor power management doesn’t sound like a good use of my weekend hours and I pay for my own work hours.
I firmly believe in the Linux mission and actively both support financially and with development hours FOSS every month. But until most of Linux is unified under a single flag with up to date drivers and power management and reliable and attractive utilities, I’ll reluctantly keep on with Mac OS.
If I had to choose at gunpoint Linux or Windows, it would be a difficult decision. I’d much rather deal with Linux (where there’s a chance of privacy) but it would be hard to cut myself off from so much pro level AV software. I’ve not enjoyed my time with the Linux equivalents.
Returning to the matter at hand:
Any mid-tier software publisher who thinks it’s a good idea to publish their commercial software on Linux has come away disappointed. I’d hate to see DxO waste man hours on another dead end project.
I’d just like Wine support. Maybe have a dev or two take a peek at running in Wine, and if it’s not a lot of work, get it working and provide instructions.
It’s simply not going to happen for the multitude of reasons mentioned in this thread and others. Are you planning on paying DXO for this casual request of yours for a day of a day two of their developers’ time? They are not just hanging around sitting on their hands. They’re running a formal business for profit and are not a bunch kids experimenting and trying things out on a whim.
I just want to confirm - Linux is not our priority - it will take a great amount of time and effort to organize the support of the new OS and if we would do that you could forget about new features for a long time. I do not think this is worth to be done now or in the near future.
Based on Svetlana’s response, apparently I do have an idea about what their priorities are.
On a separate note, how big do you think the PhotoLab development team is? DXO is a small company. You can probably count the entire number of developers on one hand, and likely you would still have one or more fingers left.
Mark
2 Likes
m-photo
( Marc (macOS Sonoma on MBP16" Intel))
55
Hi Phil,
Sorry I am not a Linux user and I have no idea how Wine works.
Maybe you can check with Wine community to help you and if you have questions contact DxO support in private to see if someone has time to help. But do not be upset if you get a negative answer.
I know Linux is not a priority right now but if it is a small step and someone wanna help you.
In the end that can help DxO too…
Thanks for clarifying this. There are no surprises at all in your response.
Some Linux fans here seem to think there is no negative impact to a small company like yours diverting limited resources to the development and support of core software onto a new platform. They don’t seem to understand that DXO is running a “for profit” business and that PhotoLab is not freeware.
Porting PhotoLab to Linux would be a poor use of those very limited resources. This is especially true because the potential for adding a large enough number of new PhotoLab users to its current user base, from the limited number of Linux users who would likely be willing to pay for it, would probably be insufficient and not cost effective. Linux users like to point out the large number of people using that platform these days, but that number is still just a blip compared to the combined user base of Windows and IOS. Perhaps at some point in the future things will change.
Great. Get a bunch of like-minded people together and get to work on WINE support. No one is stopping you. If at the end of making PhotoLab work on WINE, you find you have a small ask from DxO to improve compatibility, then have your best developer ask for the fix in very precise technical terms.
That’s how WINE compatibility is normally built, by people who want it, not by the original development team.
From what I understand, photolab is built on windows presentation foundation, a tech created by microsoft as “the” windows GUI framework for a few years a decade or so ago (or at least in their view). But the problem with it is that it has many native windows calls, so it seems wine doesn’t support any WPF apps at all.
MS has started porting and open sourcing a lot of .net libraries to “.net core”, which will be the base for .net 5, and has the possibility to build both windows and linux binaries, but even though support for WPF was added in it recently, that part (wpf) of core is just for windows due to the native calls, and doesn’t support linux. And it doesn’t seem like it will change on the MS side either in the near future (otherwise I’m guessing dxo will be forced to or choose to migrate to .net core sooner or later, but since just using the windowsonly wpf-variant means by far the least effort, that won’t change the compatibility outlook for linux)
I think it would be interesting to start a page like go fund me. Where you set a financial goal and see how much real interest there is. Cost of bringing someone new on for a fixed term contract to work for a set amount of time or something.