4 ms·
Apart from all issues mentioned so far (all valid (edit: utunga got there while I was still writing)), there's also a problem where you can't apply good UX in s
by viraptor 5y ago
Apart from all issues mentioned so far (all valid (edit: utunga got there while I was still writing)), there's also a problem where you can't apply good UX in small patches here and there. You can fix some terrible experience where it's actually a problem, but to really improve UX in an app, you need to touch almost everything at the same time, which means a lot of time spent on the task and required commitment to the idea from the leaders.
That's really hard to organise in opensource world. And when you try to introduce leadership which can do it and try to collect information, you get the issues Audacity ran into. Sure, Tentacrul can lead the UX effort and will make Audacity much better, but the community impact was pretty negative.
The only opensource project I really remember pulling off a well organised UX update is Blender. And even that was after years of people screaming "don't touch right-click-to-select, we're used to it and newbies need to learn".
- hrnnnnnn 5y agoWhat happened with the Audacity community when Tantacrul started? edit: found the thing where people got upset about telemetry - https://github.com/audacity/audacity/discussions/889 https://github.com/audacity/audacity/discussions/889
- viraptor 5y agoThe telemetry was one thing, but there was also a few people with "what do you mean muse owns this project now? I'm forking rather than accepting forced leaders".
- pdkl95 5y agoA good solution to this is the traditional[1][2] UNIX approach of separating mechanism and UI. First develop the core functionality without a UI as a library (or daemon or whatever). The actual UI is then developed as a front-end to the library. This allows multiple front-ends to coexist so a new UI can be developed without disturbing the people that like the old version. Too many projects tightly couple their UI and mechanism, which always leads to problems. [1] http://www.catb.org/~esr/writings/taoup/html/ch01s06.html#id2877777 http://www.catb.org/~esr/writings/taoup/html/ch01s06.html#id... [2] http://www.catb.org/~esr/writings/taoup/html/ch04s04.html http://www.catb.org/~esr/writings/taoup/html/ch04s04.html
- bruce343434 5y agoInteresting way to develop a program while keeping ui seperated, would have never thought of that approach. But in my experience, developing the UI alongside also leads to insights into your product you hadn't foreseen.
- pdkl95 5y agoSure! Developing both the backend and an initial UI front end in parallel is a good idea. The point is to separate different aspects of the program into manageable, modular sections. Separating the mechanism and UI is just an another way to practice information hiding and modular programming.
- Nextgrid 5y agoI agree with your first point but I disagree that telemetry is required to improve UX. You can "collect information" in a respectful way by asking users instead of stalking them (and letting a malicious third-party such as Google Analytics do so as well).