@hasen6 thanks for the feedback! We had minutes and hours way early on, but it just seemed to cause reviews to pile up faster than anyone wanted, so we switched to days. To @mike-lima’s point - we wanted to keep the focus on getting lots of exposure rather than trying to really memorize or getting slowed down on any one sentence.
It depends on the context and how you learn best! For the level of granularity it sounds like you’re looking for Anki may be a better fit - we’re aiming for as simple as possible but still effective SRS. Anki has way more complexity, but also lets you get into minute and hour sized review intervals.
You shouldn’t need a workaround for this kind of software that is designed to make learning easy, that would be counterintuitive.
Ok please post me those forum threads, I’d like to read about that. Since you can set the review times yourself why would people complain about that? Seems somewhat obtuse.
Actually it doesn’t, that’s just how the mind works, SRS is well researched and documented. You say you had a problem with minutes and hours but if people can just set their own review times there’s obviously no problem?
I paid for the pro version for a month since I was limited to 30 sentences a day with the free version (a limitation no-one else seems to have had evident by the non pro members in the top 10) so I can’t get a refund. I don’t see why you just don’t give people the option since it’s already possible to edit the review times.
I appreciate your input and points about spaced repetition, but we’re simply not set up for minute and hour review intervals at the moment, and we have a number of other projects taking precedence for the immediate future. We’re happy to provide a refund if you’d like, just email support@clozemaster.com - we definitely don’t want Clozemaster to be a source of frustration.
All that said, if anyone else would find hour and/or minute intervals particularly useful, please be sure to let us know (@morbrorper I assume that’s still your preference based on your comment above ).
Well hours would be a huge improvement, there’s really no need for minutes and seconds or anything like that. Something from one hour and four hours would make a world of difference. A whole day for the first review is just way too long.
For durable learning, space out your review of the material over time; back-to-back repetitions are ineffective.
Figure 1 on Page 13 of the report shows that hasen6’s method is called the “massed practice” compared with the “spaced practice”. The difference of the two methods is that the massed practice is to review immediately after the initial study.
Kang 2016 report fits my own experience with Anki. I used to use Anki in the default setting (minutes or hours after the initial study). I practiced a new word as many times as I could finally type it in properly. This approach reinforced my short memory, but it actually hindered my long-term vocab acquisition. It’s like painting a wall over and over before getting dry. The surface always looks shiny, but the inside is very flimsy. Repainting in a short period merely slows down the reinforcement process.
Long time ago, I read another scientific report arguing that sleeping significantly affects on our memory system. The report recommended to review after sleeping rather than cramming everything in one night. So, I fully support the current Clozemaster’s interval setting for the initial review. I shortened my personal interval setting for the second and third reviews, though.
I find it’s very rare that I learn a new word only by going through the regular four reviews, unless it’s very easy or I was already familiar with it. My experience is that I need at least six or seven reviews before reaching 100% with some confidence. Even then, I will often fail the first recurring review, unless I set a ridiculously low 100% interval. In fact, a lot of my time with Clozemaster is spent manually setting and resetting the mastery level, and that doesn’t feel quite right.
I would really like Clozemaster to adapt to individual learning patterns, maybe by customizing the number of review steps, and making more use of the Easy, Normal and Hard buttons to modify the next review date at every stage of the review process.
The article is about spaced repetition which is EXACTLY the method I was talking about…??
So literally the definition of spaced repetition which was what I was advocating? And exactly what is used in Anki which you said didn’t work…??
This is the conclusion of the article, that it works and it’s exactly what is used within Anki:
" Spaced review or practice enhances diverse forms of learning, including memory, problem solving, and generalization to new situations. Spaced practice is a feasible and cost-effective way to improve the effectiveness and efficiency of learning, and has tremendous potential to improve educational outcomes."
The conclusion is that spaced repetition is great and that back to back repetitions are not, which neither I advocated nor does Anki use…so no idea why you even mentioned that.
It’s not just about repeating the card a few minutes after you encountered it, it’s about repeated study over a long period of time where the spaced intervals get longer and longer. You clearly don’t understand how SRS works or what it is. Anki famously uses SRS, it does not use back to back repetitions. I suggest you read up on SRS instead of just posting a random article without understanding the conclusion.
Exactly, you’ll need quite a few reviews to remember a word and way, way more if the spaced intervals are not optimal - as is the problem with clozemaster.
Yes it’s a shame it isn’t more customisable, especially with a pro account. I think it’s more suited to the more casual learner with reviews by the day, not for someone like me that would like more intense learning.
Fixed that Intense is playing 1000+ sentences per day. It really sounds like you’re looking for Anki and all the customization it offers. I’m curious - what’s keeping you playing Clozemaster vs switching over to Anki? (sometimes tone isn’t expressed well in writing - I really am curious, and it’s helpful to understand as we work to keep improving Clozemaster)
@hasen6
It’s so odd that nobody has clicked the hyperlink to Kang 2016 report so far, but you are arguing that your method is supported by the report… FYI: hyperlinks on Clozemaster forum usually display the number of clicks – not sure it’s a glitch or nor. I’m afraid you understand the definition on Figure 1 of Page 13 in a very different way from mine.
@morbrorper
You made a really good point! And I always admire your insights. But…
It totally depends on how many different sentences your course offers per cloze-word. I think 40 reviews (= 4 reviews x 10 sentences) are enough for one cloze-word while 7 reviews (= 7 reviews x 1 sentence) is still not enough. It’s seemingly nice to have Easy/Normal/Hard buttons as Anki does, but the problem here is that you don’t know how many new sentences of the same cloze-word are left at the time when you push the button.
By the way, I’m okay with the current features. I usually add some default sentences to my personal “extra review required” collections or “favorite” decks, and/or manually import different sentences with the same cloze-words to my personal collections. I also set the second and third review intervals shorter. These features are not perfect but manageable. In that sense, I agree with @mike-lima 's opinion.
I wish Clozemaster to prioritize different issues, one of which is to import extra sentences from more reliable sources than Tatoeba. I think reviewing one sentence many times in a short period is less effective than practicing a variety of sentences with the same cloze from different sources. I also believe that the source diversification is much more appealing to prospective users than customization of review intervals/times. @mike showed his interest in this matter in January this year. But no progress… Mike, is your team still interested in this idea? I totally understand that it’s time consuming to implement, so this is just a quick question, not a demand.
Thanks @mike for the courteous way you have answered all the questions/suggestions.
Having tried several sites, I can honestly say that ClozeM suits me perfectly and that what is available for the very reasonable fee is incredible. Also incredible is the way “Mike” responds to each of us, we actually have contact! No language site is perfect, why not mix n match if somewhere offers a method you prefer.
My Italian and confidence levels have vastly improved with ClozeM and for this I am truly thankful.
I literally posted a quote from the report in my post above…
It is not “my method”, I don’t have any method, the method I described in my initial posts above is SRS which is what the report is about and its conclusion is that it is good. I literally used the acronym SRS in my first post. The only thing I can be sure of is that you haven’t read it since you seem to believe that the conclusion of the article was that SRS is bad…
Well I said ‘more intense’ so it wouldn’t have any specific definition or number. As it is certainly less intense since the reviews are by the day, than they would be if the reviews were by the hours, that’s surely not in debate.
But yeah a 1000+ sentences a day would be what I would be looking to do but that’s not practical with the first review time being set at an entire day minimum.
Well nothing, I already stopped using it and sent an email to you for a refund this morning but didn’t hear anything back. I think it’s a good tool but it just runs a bit too slow for me with the reviews only once per day.
I wouldn’t rely on that count. I clicked through to the (most interesting) article twice at different times on different computers (I have three on at any given time, don’t ask) yesterday evening my time, many hours before you posted that.
(Edited Afterthought: Most of the time I right-click on links and select “Open in new tab” or “Open in new window” since (depending on how the HTML is set up) clicking on the link may take me away from the page that I’m looking at. It’s possible that those don’t count as “clicks”.)
I know that cloze words are Clozemaster’s thing, but I’ve long since stopped thinking of it in terms of individual words and more in terms of variations of words (verb conjugations (of which there are a LOT in Italian), verbs and related nouns, etc, etc.) , and the context of the words within sentences. Some are easier to commit to memory than others, some are “blind spots”. For myself, I’d be hard pressed to put a figure on it. because it will vary. This is why Collections are a great idea, though they could be even greater if (shameless plug for my feature request) the review time could be set by collection. That would allow people to shift their review time focus to better aim at the words that just won’t “stick” and let the ones that have slide further into the distance.
I suspect that such a thing may be dismissed in some quarters as a “counterintuitive workaround” despite the fact that that questions can be added to a personal collection with a single click these days, but there’s no pleasing everybody.
Yes I’ve no doubt it would suit certain people with the daily reviews, it depends on what your goals are, the time you have available etc. But to be honest if you like daily reviews, then set it to daily reviews, I really thing it should be an option, especially with the pro version. Having an option is not detrimental to anyone since it’s customisable to each person’s needs. Obviously it’s not currently possible so they would need to add that and it’s of course their choice if they don’t want to.
Sure, no problem, I’m new to Clozemaster and am still discovering its features, that’s why I found this thread interesting. I’ve already adjusted my settings as well.
For the level of granularity it sounds like you’re looking for Anki may be a better fit - we’re aiming for as simple as possible but still effective SRS. Anki has way more complexity, but also lets you get into minute and hour sized review intervals.
And I always admire your insights. But…