Sorry, but I’m not proving anyone’s point. There is a benefit to them using QT for the Win version in that it does form a solid core for Android, should KB decide there is a viable market. Even with cross-platform tools development and support is expensive and requires a solid user base prepared to pay.
As KB stated, the inbuilt development tools for Mac are (while not perfect) so good (and FREE) and easy to achieve great apps that using any so-called cross platform development app for Mac and iOS would result in a lesser app.
One of the reasons for no Android app ‘since years’ is that L&L clearly stated one would come (if one did) after Win V3 was completed. It’s now just ‘after’, so I’m guessing KB is considering if it’s worthwhile (or secretly the team are forging ahead and might surprise us). After the flack they received over delays in Win V3, I’m guessing if they proceed, the first we will know will be if there is a beta program.
Thank you! I didn’t take it as criticism at all - I was only joking about the rewrite.
I wonder if this is a Mac/Windows difference. I’ve never seen a cross-platform app that looks entirely at home on the Mac. There are some cross-platform apps that look and work great on the Mac, don’t get me wrong, but they’re still obviously built with non-native tools. They have the traffic light buttons in the toolbar and a lot of the general look, but their toolbars use different colours or aren’t quite the same height as other toolbars and things like that. All of which is absolutely fine for apps that want their own custom UI standardised across platforms with only minimal differences in where the “close/minimise” buttons are. But I like to work with a more platform-standard UI, which on the Mac means a UI that looks like Apple apps - and the Apple tools.
That said, you could prove me wrong and show me an app that I had assumed was native but turns out not to be, and it wouldn’t make an awful lot of difference. The truth is that for as long as I’m leading the design and development, I’m going to be using a Mac because I love the tools and programming languages (Objective-C and Swift) and work well with them. To move to a fully cross-platform toolkit for all operating systems, I would either have to retrain in tools and languages I’m not entirely enthusiastic about, or step back and oversee other developers take over. Right now, I’d honestly hate that. I love thinking about were to go next, coding up ideas and prototyping, throwing out ideas and refining them - and implementing them in the Mac toolkits. (One of my absolute favourite comments from last year’s survey was, “I don’t think Keith is fit to be the lead developer because of his personality.” Maybe they’re right, as it is to some extent my love of working with Mac tools that prevents us from moving to a fully cross-platform solution. It’s a conundrum, though, because if I gave that up I’d have less enthusiasm, and someone else would have to take over driving Scrivener forward. And I’d just end up writing my own new version anyway and launching a competitor product!)
With regard to other Windows tools (other than Qt), as I understand it - and you will know more than me on this, as I haven’t spoken to our Windows developers to check this so am going from hazy memory - one of the problems with Windows development has been the number of tools and languages available, with Microsoft introducing different ones and partly abandoning older ones. When I first had the idea for Scrivener, a few years before moving for Mac, I looked at developing it on Windows using C++ (and M… something? I can’t remember, but you’ll probably know). I didn’t get along with C++, so abandoned the idea until I moved to a Mac, but at the time MS seemed to be transitioning to C# and .Net, and since that there have been, what, two or three others? I know of other developers who have banged their heads trying to work out the best Windows dev approach. That’s why it seemed best to us to use a cross-platform tool like Qt for Windows (and hopefully one day Android) development, but to stick to the Mac native tools for macOS and iOS.
As for Android, I hope that won’t stay a missed opportunity forever. Our research certainly suggests that Android is far less profitable than other platforms - despite more people in the world having Android, iOS developers make more. But we’ve never been driven entirely by profit (which is actually an advantage of being profitable). We want to get Scrivener onto Android too, but we have a lot more ahead of us before that (things that will have a direct influence on the Android version).
I feel sorry for the grief the 1Password guys are getting. I understand their reasons and things must be very frustrating for them right now. I understand the concerns of users, though - I tried out out an app the other day that markets itself as a competitor to Scrivener and which is built on Electron, and its Chrome renderer background process was eating up anywhere between 10% and 40% of my M1 MacBook’s CPU after it had been left open in the background for a couple of hours. Ouch.
Anyway, I wrote far more than I intended there - back to the code!
I loved it! I’ve been sharing it with everybody; I guess I’m quite proud of my misanthropy. (The user went on to explain that I had said no to some of his feature suggestions, so I think he - or she - was just a bit peeved.)
There were some brilliant replies to the survey. Another of my favourites was this one-word reply to “What type of fiction do you write?”: “Smut.”
I have started drafting a blog post discussing the survey results, but I still want to read a few more replies - we had over two million words’ worth of extra comments from users, so it has been months of reading some here and there - who would have thunk that writers like to write?
I think you tried the same app as I did, with the same result
“What type of fiction do you write?”: “Smut.” Haha - that’s what my ex mother in-law said about two love scenes in my first book - Natanz. The rest of the book was aircraft flying and international intrigue, but less than half a page and it was ‘smut’.
I really must write some more and really wind her up!
I was the one who posted that thread (and kept bumping it ) that Keith replied to about sharing the results. I imagine reading through that many responses while continuing to work on the product must be difficult and I can’t imagine i’d make it through all of them without going insane so props to you there.
I must admit that I’ve been waiting for the “We’re bringing Linux support back!” announcement from the survey results but I think that is probably a pipe dream at this point because of how few of us use a Linux distro as a daily driver. I was thinking that using Qt as the framework for Windows may provide a launchpad for that though but I have very limited experience with Qt as I typically work with Java (Spark/Hadoop - Big data).
Anyways, still very much looking forward to your blog post Keith, whenever that may be
That’s part of it. The other part is that Linux support is inherently more expensive to provide for the majority of software vendors than supporting any other OS, simply because of the wide variety of distros, releases, package formats, library versions, etc. that one has to be on top of. A common mitigation strategy is to limit the supported Linux distros (only Ubuntu-dervied LTS releases, forex) but that usually just makes the people who fall outside of the support boundary even more upset and vocal about it.
Fortunately the situation is not as bad as it was several years ago and distributing an app on linux no longer requires recompiling for every distro out there. You can now distribute your application through flatpak, snap, or appimage. All of these methods bundle the dependencies with the application and take care of that headache as well. They are basically self contained application packages with all dependencies required. They can be installed, updated, and removed with a simple command - very convenient! This is, in my opinion, one of the reasons we now have more apps on Linux than we used to.
Unfortunately, the often unspoken part of the equation is money. Sweeping statement, but many Linux users are like Android (even more so?), reluctant to spend money. This can make the development and support cost of a niche Linux app prohibitive and I believe why the original Scrivener Linux support and development ceased.
Apart from wishful thinking, has anything changed? I do see a number using Win Scrivener via WINE, but it seems a very small number.
To be fair to potential Linux Scrivener users, I have seen enough people offering to buy a SfW license over the years that I don’t think that is as universal as you think, at least among the crowd that would use Scrivener.
Yes I know you mentioned support specifically but part of supporting a product on Linux (and you mentioned this in your original post) is around supporting different package formats (.deb, .rpm, etc…). Since this is no longer required because of the aforementioned methods of packaging an app, that part is less of an issue for support now than it used to be at least.
I currently use Scrivener on macOS and PopOS through Wine (I guess i’m the exception since I actually bought two licenses ). I do understand what you mean though - many who use Linux feel all their software should be top notch and completely free which is a real issue. I am an experienced software engineer myself so I don’t live in such a fantasy world.
Any more word on when the results of the Scrivener survey (plus action items and the like) will be shared? I certainly don’t envy whoever has to put all that together into a single post (KB?) but I feel like it will be really interesting at any rate
The truth is that for as long as I’m leading the design and development, I’m going to be using a Mac because I love the tools and programming languages (Objective-C and Swift) and work well with them. To move to a fully cross-platform toolkit for all operating systems, I would either have to retrain in tools and languages I’m not entirely enthusiastic about, or step back and oversee other developers take over. Right now, I’d honestly hate that. I love thinking about were to go next, coding up ideas and prototyping, throwing out ideas and refining them - and implementing them in the Mac toolkits. (One of my absolute favourite comments from last year’s survey was, “I don’t think Keith is fit to be the lead developer because of his personality.” 
Maybe they’re right, as it is to some extent my love of working with Mac tools that prevents us from moving to a fully cross-platform solution. It’s a conundrum, though, because if I gave that up I’d have less enthusiasm, and someone else would have to take over driving Scrivener forward. And I’d just end up writing my own new version anyway and launching a competitor product!)
) that Keith replied to about sharing the results. I imagine reading through that many responses while continuing to work on the product must be difficult and I can’t imagine i’d make it through all of them without going insane so props to you there.