Changing wording assumes users read stuff, but we got many reports from users who simply think Anki is broken when they see the terminal window and do not proceed. A GUI is the first step to solving poor user experience here, and that’s thankfully being worked on currently.
@Shigeyuki As far as I’m aware the iPhone app is where any profit is made, which relies on the desktop app and cloud server to work. I should also mention I purposefully avoided using (or creating) any add-ons and have been very careful to avoid dependency hell, choosing boring technology less prone to break with future updates. I will look into previous builds if necessary, thanks for the suggestion.
The post was strongly worded, but let’s not get into any personal attacks. @dae and the team (as well as anyone who prefers the terminal setup) are quite rightly welcome to design Anki however they want, but it’s not currently to my tastes. A .dmg should be all that’s needed, and it seems to me best practice to follow what most Mac apps do: an (optional) silent background installer, a notification the app needs closing, then relaunch (or auto relaunch). That generally seems the agreed way to do things, but others have you download fresh and overwrite the previous app. I can’t think of any apps that do some fancy downloads within the app itself (other than VS Code). No consumer apps anyway.
@Danika_Dakika It was mainly for the reason I couldn’t edit this thread’s heading and the one I’ve recently posted better describes what I’m talking about.
Whatever way you cook the terminal or GUI you shouldn’t make the user think (book).That’s pretty much 101 for UX design and good design is silent (as in, users won’t notice when a design is just right, but will notice when it’s wrong). And yes @abdo it’s good to assume nobody will read anything properly, but it sounds like there’s a potential fix on its way which is good.
As for Tauri, another app I use has utilised that, but has some problems with the screen blacking out on Monterey and simply not working. Your mileage might vary.
Anyway, I made my point. Downloads should be quick and painless. There are a few design decisions in Anki that may mean I have to move away from it in future, but an easy experience for absolute beginners with zero experience in programming is what I’m generally aiming for. My stuff is still a ways off that goal also, it’s not always easy to achieve.
Thanks for the replies.
AnkiMobile(iOS) works standalone so it does not depend on Anki for desktop, many users in the community report already using only the iPad or iPhone.
Anki for desktop is primarily developed by the official Anki, but roughly 40% to 50% of its development comes from volunteers and third party contributors. It’s open source and users can use all features for free so no revenue is generated.
So IMO users can only make advanced requests like commercial apps when AnkiMobile users request features for AnkiMobile. Other programs are primarily developed by volunteers, so instead of complaints, contributions and volunteer work are generally what’s needed. (because we use them for free and no one is paying for them like free rider.)
So far a similar project by volunteers is voczi’s fork of Anki. This is an all-in-one so it’s probably similar to the old Anki.
- AnkiForums: Offline distribution of Anki
- Github: All in Anki
The GUI for the official Anki launcher is being developed by iamllama on this GitHub, I expect this will resolve the GUI issues.
- Github: feat: launcher gui #4390
Instead of complaints, contributions and volunteer work are generally what’s needed …
Look, it seems like you’ve done a lot of work on add-ons for Anki so I’ll respect that. But I’m sorry, you’re talking to someone who’s developed on top of the Anki ecosystem — for free — for almost 10 years, so I feel entitled to complain about things that break the onboarding process (as well as valid suggestions I’ve previously added). I’m not a Python developer so haven’t got the skills to contribute that way (and at my age, nor the time).
I understand open source is a difficult thing to do, but the team have to somehow find a way to support the whole ecosystem or they’ll be limited to (mostly young?) developers to do free work. This guy (interview) has many good talks about just that. Anki mobile is fine for very simple front/back cards or downloaded packs, but anything more (like the above link to my programming flashcards) depends on the desktop app. My “complaint” (request) isn’t advanced, it’s a fundamental one. I don’t seem to be the only one who thinks this way.
At this point, I think I’m done with this discussion and will thank everyone for the app and requests support over the past few years. You can choose to listen (or not) but I suppose I’ll have to rethink using Anki’s ecosystem if it continues to become more complicated for users.
It’s not a good use of my time (or yours) trying to convince people who seem to have other ideas.
It’s not clear what outcome you expected other than the response that it’s a known issue, and it is currently being worked on. You’ve gotten that.
@Danika_Dakika Fair. But. The outcome I’d suggest is more overall consideration towards ease-of-use and user experience for non-power users (beginners) throughout the Anki ecosystem and UI, regardless of who’s building what and whether it’s for profit or not. Appreciate it’s a known issue being worked on (which you’ve pointed out) and for the most part Anki has been great and useful, so I should make it clear that I’ve appreciated everyone’s hard work over the years.
However, it already takes a fair bit of training to onboard novice users (with the setup I need) and this decision in particular hasn’t filled me with confidence going forward (although this does look like an improvement). The comment @Shigeyuki made pissed me off, even if I understand his point of view, so I elaborated.
Perhaps it’s simply that my particular needs and the Anki app are starting to diverge. Anyway, that’s about it … I was trying to be helpful, but Danika’s curt response suggests it’s best to just wrap things up there.
I think such discussions are beneficial for development by volunteers but if you don’t like them it would be better to end our conversation here as you said. I already completely agree that the launcher is inconvenient and needs improvement.
I don’t believe the launcher/terminal window process was anyone’s ideal solution to the necessity that Anki move on from the existing installers. However, it worked (for many) for the July release, to get FSRS-6 out the door – and it even added an easier path to version updates (after this initial hurdle). There are plenty of users who are are not intimidated by the presence of a terminal window, and willing to read instructions (or ask for help) – they have handled it just fine.
It seems like everyone who can is chipping in with the expertise and time they have available, to improve the text-based launcher (for now) and move us forward to a new GUI-based launcher as soon as possible. I was brief with you because it didn’t seem like in-fighting, outrage, or catastrophizing was going to help much with that process.