I agree with this. We don’t know even know what’s going on with the developers on their plans in Anki for everyone in the foreseeable future.
I applaud for proposing this idea and good job on trying. I know this idea can actually impact for beginners, but I prefer the add on for now or trying to convince people to approach Anki like the same as Arch Linux (DIY type of operating system)
And I see that this is going with the environment of politics as well, agreeing and disagreeing with each other instead of helping each other out (I mean, it’s what humans are I guess)
In my opinion you already have that skill. The Python used primarily in Anki is the easiest programming language in the world to learn and Anki is the most powerful learning tool in the world, plus you already do advanced calculations and logical explanations and are working diligently to improve Anki, so there is no reason why you cannot learn it.
To be honest, if you can read the manual and have long-term use with Anki, as long as you think the mentality of “not broken, don’t fix it.” You’re on set. And I even think Anki is getting simpler too, even with FSRS, and the manual is just a way to make your use go smoother.
I believe that Anki is a “do-it-yourself” program for flashcards that will depend on people-to-people. There’s a lot of info and videos online to do just that
My beginner side
Last year during school, I just want an alternative to Quizlet to do all my work and searched “alternatives to Quizlet.” Anki popped up, installed it, search for an addon to import Quizlet decks (I’ve even searched “how to install addon anki” and voila, my school work is saved for the day. By the way, I didn’t read the manual back then.
Most people don’t. It’s a very well known phenomenon in the technical writing community. There was this research paper called “Life is too short to RTFM” that actually shows asking people to read a manual contributes to negative emotional experiences. (Fun fact: more educated people are much less likely to read a manual).
This has made ideas like topic-based writing and “Every page is page one” very popular. You write the manual to be referenced and not to be read.
I plan to develop such a fork in the future, mentioned a bit in this thread.
I think another way to make Anki easier to use for beginners would be to simplify the Anki GUI to such levels that no instructions or manuals are needed. e.g. if I make FSRS the default algorithm and incorporate an add-on it could have two buttons so I don’t need to explain about the buttons. Similarly many non-essential features can be disabled or hidden, such as deck options, browser, editor, etc. plus I can incorporate all the standard and popular add-ons from the beginning.
But I’m thinking of this fork as just advanced and personal add-ons, so I’m not thinking of a generic fork such as the one discussed in this thread. e.g. if I introduce Anki I need to explain how to download Anki, how to configure it, how to install recommended decks and add-ons, if I create a fork of Anki I can package them in one click and distribute them to new learners.
I’ve said this before: having two separate versions of Anki will achieve anything but decreasing confusion. Do you really think that saying “Here’s the official Anki and here’s another Anki!” will make people less confused?
I think another way to make Anki easier to use for beginners would be to simplify the Anki GUI
many non-essential features can be disabled or hidden, such as deck options, browser, editor
But advanced users don’t want simplification and they don’t want their favorite features to be hidden. If only there was a way to have a toggle that enables/disables simplified UI…yep, we’re back to two layouts ¯\(ツ)/¯
The reason I don’t like two layouts is because instead of solving a usability problem it’ll promote a “We don’t know how to make it more intuitive, let’s just shove it in advanced tier” mindset. Take the discussion with “Learn ahead limit”, if we had two layouts there would’ve been more unwillingness to discuss that issue.
Maybe your thought of fork and my thought of fork are for completely different purposes, here’s what I mean by fork:
Before:
me: “I developed a new game add-on! (youtube)"
Learner: “What is this? How do I use this?"
me: “The Anki is an open source flashcard learning app, these are the recommended shared decks, and these are the recommended add-ons, this is an add-on, so you can install it and use it. Recommended settings include these…"
After:
me: “I developed a new game add-on! (youtube)"
learner: “What is this? How do I use this?"
me: “plz download this: (fork url)”
Or I can put this on my Anki wiki:
“Hello welcome to the my Anki wiki! I’m going to explain in detail how to install Anki, how to configure it, and what shared decks and add-ons I recommend! If you don’t want to read them you can download this packaged version I made: (fork url)”
In short such a fork is intended for me to promote myself and distribute to new learners, so I do not plan to distribute them to advanced users (they can download the add-ons), nor do I intend to incorporate them into the official Anki to reduce confusion for the average Anki user.
In my opinion if we simplify Anki for beginners we don’t need the latest Anki features either, thus I think it is not necessary to incorporate such a feature into the latest Anki. If users download the simplified fork and then install the official Anki, it will work the same as the “enables/disables simplified UI”. This should simplify Anki’s code and documentation and make it easier to maintain.
Expertium probably thought it will be something like Telegram versus Telegram X with seperate icons and such and it’ll be distributed widely. I was actually thinking the same thing you were. It’s a good idea actually.
I think AnkiHub can do it too. But they are fine with the current way it seems.
Off-topic. But this is interesting, I thought I wasn’t the only one who’s seeing the phenomenon of people not actually reading the manual on products they use.
Do you have recommendations that delve on this? I wanna learn more about this
@dae, just to clarify, I’m not trying to force this on you or argue that you should implement this into Anki core ASAP. Just providing it as an example
We will continue to automatically subscribe all Ankihub users to the deck and perhaps in a few months we can share data regarding how useful it is. One thing we plan to due is to collect review data for this deck (e.g., are people actually completing the deck) and collect survey feedback about the experience. At that point, we could revisit the convo and discuss if Anki core should adopt it based on the likelihood of it being a significantly positive contribution improving the UX and learning curve of getting started with Anki. Currently, 40K people have subscribed to the deck but we don’t have data on whether people actually installed it and reviewed the cards. “subscribing,” simply means that created an AnkiHub account and were automatically subscribed. It’s possible some of these users didn’t even do the next steps of installing Anki and providing AnkiHub login credentials.
However, I do think the tutorials we are making with https://www.iorad.com/ could be useful and I’m happy to discuss how we might be able to make make them more discoverable to all Anki users and more useful to the community. They are free and also embeddable.
Yep, the big change the more difficult to develop. e.g. I developed an add-on for highlighting by Anking request earlier and if possible there was a request to incorporate it into Anki for desktop, but when I actually developed it I found it was not very suitable because of its unstable working. Such results are very common in development and cannot be known until the actual development is done.
So if we try to develop the toggle system or other feature as you suggest, I think it is reasonable to develop a small add-on or small fork first and see if it works effectively, if it is popular or effective it would be a good candidate for PR.
I mean how much can you delve on something like this? People not only don’t read product documentation people also generally dont read on the web. See Myth #1: People read on the web - UX Myths.
I have heard about something called information foraging theory that tries to elaborate on this behaviour. But it’s quite simple to understand. People live busy lives, so we can not afford to spend a lot of time on everything. People read if something actually interests them. Otherwise, they are satisfied with a superficial understanding of it.
There is also an expectation of usability in products. If someone fails to understand something, they play around with it to solve it themselves instead of referencing something like a manual.
If you are interested in technical writing though, read IBM’s handbook about developing quality information. Quite a good read on topic based writing.
I’m not asking for an apology, I’m asking you to be more considerate moving forward.
Different people have different opinions about what should and shouldn’t be included in Anki, and the order that things should be done in. We are not always going to see eye to eye on what should come next, and I understand that may be frustrating sometimes. But your response to this has to be to mock me and the decisions I’ve made, on multiple occasions.
It’s clear you’re really enthusiastic about spaced repetition, and I have seen you make valuable contributions to Anki. I have been trying to look past your disparaging remarks and focus on all the good. But I’m human, and I find it unpleasant to see comments like that. If you keep it up, I imagine it will get to the point where I no longer want to interact with you.
Having data to guide decisions is always good.
I do see the value in step-by-step tutorials, and am open to discussing this too. In the short term, we could potentially link to guides from the relevant section of the manual for example.
I’ll copy what I said to David in a private conversation (with a few minor edits to make some wording more clear):
Beginner-friendly UI Token: Dae makes a Github issue called “Two layouts” or “Onboarding deck” or “Interactive tutorial” or anything that sounds vaguely similar and can be considered at least remotely UI-related. Adequate: Dae is either personally working on one of those things or is cooperating with someone who’s good at UI stuff.
“Hard” button misuse Token: Dae doesn’t openly say that users should just read the manual. He makes a promise that he will actually, really, in real life work on making button usage more clear (not via the manual). Adequate: Dae is implementing any suggestion proposed in the “How to prevent users from misusing Hard?” topic and replies, or a an idea that wasn’t proposed before but is also good.
Not UI, but something that has been requested for almost a year by many users - automatic optimization (AO) of FSRS parameters Token: Dae gives a deadline that is more specific than “in the future” and less than “in 10 years” for when the new sync protocol (or whatever is needed to address potential sync issues with AO) will be implemented. Adequate: There is a new sync protocol/new whatever thingy that is a prerequisite for AO, and now AO is possible.
Ideally, of course, it would be great to reach “Adequate” level on all 3. Right now we are below the “Token” level on all 3.
I’m not demanding that all of that must be implemented immediately, I’m not insane. I’m saying that right now these issues are basically swept under the rug, and I don’t see any signs that any of these issues will ever be addressed in any way, shape or form.
Can we get to at least “Token” level on at least one out of these three, please?
