free form GUI panes wish


Photoshop in recent evolutions introduced a capacity to break up the components of the GUI so it can be arranged free-form. So applied to Scrivener; this could see the GUI editor, binder, inspector and toolbars as able to break apart and be arranged in a different configuration.
So for eg: toolbar placed along bottom of a pc screen, and editor in one monitor with binder and inspector in another. It would also allow for binder to be placed right side and inspector left.
In Photoshop it opens with the usual GUI but then this can either be locked together or unlocked and re-arranged.

It’s a great GUI design tweak enabling really focused work whilst addressing user ambidexterity-preferences / requirements.

The main issue with this idea is that Adobe actually went and reinvented the entire GUI, rather than use the native development tools for each operating system. On the one hand it means they can break all of the rules and do whatever they want, but on the other hand it means their software looks and acts like nothing else on the system.

The other problem with doing that is that you’re reinventing the entire GUI. :slight_smile: There is a reason why this isn’t typically done, and it has to do with how many years you want might not want to tack on to your project. In some cases that is definitely something worth considering. Graphic design software for example, sits in an arena where native GUIs just can’t do everything they need to do, or can’t visually get out of the way enough. Games are another area where typically everything is reinvented from scratch, although there often for more stylistic reasons (and if you think of how woeful even triple A game UIs can be in terms of how simple scrolling should work, you can maybe see how this isn’t easy to just do).

The question is whether a writing program should spend two or three years on that kind of project. In my humble opinion, writing software fits within the greater pantheon of software that should look and feel like 99% of the other programs out there. That does mean being limited to what these toolkits can do though.

I’m not sure i agree with your perspective. The move to apps functionable on small screen real-estate gadgets has caused a more flexible pair-downed approach to application GUI which has seen conventional static GUi builds broken up into constituate parts as part of making them available on mobile devices. From questions posed in the recent Scrivener survey, this is clearly a direction being considered. As software applications migrate to integration across multi-platform OS - surely innovation should be enabled to influence it’s presence on various platforms ?

The problem with the world at the moment is that coding can do a lot but there is an acute shortage of qualified coders globally. I would go so far as to state the Scrivener project reflects that reality. Maybe you should hire on short contracts emerging coders based in Taiwan, South Korea, Japan and California - they may be more can-do on seemingly difficult suggestions. :smiley:

I very much disagree with development practices that encourage “design UI once spew everywhere”. I’ve never seen a project like that I would actually care to use on a regular basis, and of those I have to use (like Resilio Sync’s awful UI), I just bite the bullet and suffer with it. Even Electron-based stuff, by and large, is something I give a pass to.

The vast majority of user interfaces should be hand-designed to each platform they are aimed at, with their unique advantages and disadvantages catered to, and using native widgets and local conventions. Then again I don’t know much about mobile at all. Honestly it surprises me to hear that a lot of developers port desktop GUIs into it, modular or not. That doesn’t jive with my experience, but I tend to see more of the opposite since all I use are computers—I see lazy stuff like modal “Edit” interfaces instead of Ctrl/Shift clicking coming from mobile, for example.

That all said, none of what you’re referring to has anything as sophisticated as user-configurable dock-able interface components, at least none that I’ve ever seen. Maybe Visual Studio does, but we’re talking about megacorporations again, in that case.

You seem to be conflating development practices, creating modular GUI in code for example, with user configurable interfaces. The latter is hugely more complicated than the former—and I don’t just mean from the development side of things, it also makes the software itself far more complicated to use. You can justify an advanced and fundamentally complicated UI in AfterEffects or Maya, but I think it’s worth noting that even Adobe stays away from that kind of modular UI in LightRoom, which is aimed not at professional career computer users, but photographers.

I don’t know about that, nor do I understand what you are getting at. Are you saying Keith isn’t qualified to make Scrivener, and so we should farm out work to cubical farms instead? Eh… :laughing: