Pleco Software

Newsletter: New graded readers and more

by

Dear Friends of Pleco,

It’s been *way* too long since our last newsletter, but finally enough new product releases + requests-from-friends-to-tell-people-about-stuff have accumulated to send out a new one. (I really really really do hope to send these out more frequently from now on)

1) Graded Readers
2) Cross-Platform Policy Revamp
3) Outlier Chinese Character Masterclass
4) Research Study
5) 4.0 Update
6) Project Marzipan / macOS
7) App Store Reviews
8) Android Q

1) Graded Readers

In late 2017 we launched our first paid e-book add-ons, a selection of graded readers from Sinolingua and The Chairman’s Bao plus the amazing “Chinese History: A New Manual” by Endymion Wilkinson. They have proved to be quite popular, and while unfortunately there wasn’t another Giant Amazing Chinese History Manual for us to license, we did go ahead and license 50 more graded readers from Sinolingua, which are available now in the “Add-ons” screen.

35 of these are from their popular “Rainbow Bridge” series, fun little illustrated readers (most of them US$3) with a variety of old Chinese historical / folk stories, including full English translations. 20 of those 35 are specifically oriented towards beginning learners, with simple HSK1-level vocabulary, and the remaining 15 are scattered among various higher levels.

We’ve also got abridged versions of some classic works of Chinese literature, for more advanced (HSK5 or so) learners; the “Four Great Books” (Journey to the West, Outlaws of the Marsh, Dream of Red Mansions and Romance of the Three Kingdoms), Qian Zhongshu’s “Fortress Besieged”, and Ba Jin’s “Torrents Trilogy” (“Family”/”Spring”/”Autumn”). There are also some classic heroic tales (“The Generals of the Yang Family” e.g.) targeting more intermediate learners, and also some intermediate-learner-friendly collections of bite-sized short stories (~500 or so characters per story, but 20 stories per book).

All of these support both traditional and simplified characters, and we also updated our previous Graded Chinese Reader volumes to support traditional too (having previously been simplified-only).

2) Cross-Platform Policy Revamp

This is actually an old one from last summer, but since we rolled it out without much fanfare back then: I’m pleased to announce that due to an Apple policy change, Pleco licenses are now seamlessly cross-platform between iOS and Android. In other words, if you buy Pleco on one platform, you can simply copy down your Registration ID and enter it on a copy of Pleco on the other platform to activate your purchases, no need to register or transfer or whatever. This works in both directions.

On iOS, we also now have a free ‘integrate purchases’ add-on you can buy which will attach your non-App-Store purchases to your Apple ID, so that if you reinstall Pleco on another iOS device in the future your add-ons will automatically activate on that device just as if you originally bought them on iOS. So not only can you migrate your Android purchases to iPhone, but once you do, they can pretty much behave exactly like native iPhone purchases. (unfortunately, since Google Play does not currently support free in-app purchases we are unable to offer a similar feature on that)

Also, our online store at store.pleco.com now works for both iOS and Android add-ons. We also recently added Alipay and WeChat Pay support (along with Apple Pay and Android Pay), though support for those among users who are not Chinese citizens is unfortunately rather spotty/inconsistent at the moment.

3) Outlier Chinese Character Masterclass

Our friends at Outlier Linguistics – developers of the very popular Outlier Chinese Dictionary etymology dictionary add-ons – have launched an online Chinese character learning class, at:

Outlier Chinese Character Masterclass

You can use the discount code ‘Pleco20’ to get 20% off. (the website indicates that enrollment for the last session has ended, but you can sign up for the next one through the form there)

4) Research Study

Forwarding a request from Dr. Amanda Mason of Liverpool John Moores University. (the participant information sheet can be viewed here)

—
Dear Pleco User,

You are invited to participate in a survey which is part of a as part of a study we are conducting at Liverpool John Moores University,UK and Xi’an Jiaotong University, China.

By completing this questionnaire, you will provide valuable information on how you learn Chinese characters, your preferred approaches to learning and your motivation for learning Chinese. To take part, you must be 18 years of age or older and actively studying Chinese as a Second or Foreign Language. For further details on this study, please see the attached Participant Information Sheet.

The survey takes approximately 40 minutes to complete. By completing this survey, you are consenting to be part of this research study. Please note that:

* Responding to the survey is voluntary.
* The survey is anonymous and all answers will be treated with strict confidentiality.

Please click on the link below to access the survey:

https://ljmu.onlinesurveys.ac.uk/chinese-character-learning-final-jan-2019
—

5) 4.0 Update

So… yeah. This is taking rather longer than we had expected.

Most of the problem has centered around what people in the industry call ‘technical debt’; the accumulated cost of years of ports (across four different platforms!) / fixes / hacks without ever really going back in and re-doing everything in a proper, modern, streamlined way. We would start rewriting one part of Pleco for 4.0 and then inevitably find that we couldn’t really do that unless we also went and rewrote another part and so on until eventually we ended up rewriting most of the app from scratch.

On the plus side, the mostly-new app we’ve been building is *ridiculously* nice. Getting rid of that decade-and-half of baggage has enabled all sorts of wonderful things, some of which will only be hinted at in the initial release. (despite a great abundance of new features, we’ve also got a list of punted ones long enough to sustain many years of 4.x updates)

For example: we can now match long lists of words against other long lists of words about 3 orders of magnitude faster than we could before. The upshot of this is that if for example you come up with a list of 100,000 words you’d like to import into flashcards, you can open that in our new import screen and in real time (about a quarter of a second on an iPhone 6s) we can check that entire 100,000 list against your existing flashcards *and* all of your installed dictionaries and put up a report listing – in advance of the import – all of the duplicated / failed / ambiguously-mapped cards, so that you can go through the list and decide what to do with all of them before you run the actual import.

It’s also a lot easier for us to maintain and build on in the future, partly because instead of tacking on system after system to handle different things – our current app has no less than *seven* totally different search indexes (designed / tacked on at different times in the past 15 years) depending on what kind of content you’re searching for and where you’re searching for it – we’ve consolidated everything around a single, simplified engine and then attached various modular rules / behaviors to it. So despite shrinking the size of our search engine code by about 75%, you can now build a Pleco user dictionary even for a language we don’t support; for example, you can make a database for Japanese that not only seamlessly juggles kanji/kana/romaji but even lets you mix them in non-standard ways (e.g. a Japanese equivalent to our mixed pinyin/character search where you enter some characters in a search string in kanji and some in kana despite the fact that they would normally be written as kanji).

We’ve also applied a similar simple-system-built-up-from-rules philosophy to our other great algorithmic rat’s-nest, flashcard repetition; we’re planning to ship 4.0 not only with a new, simplified SRS algorithm of our own design (one which moves away from SRS orthodoxy a bit by taking a relaxed approach to long-term reviews and only rigorously scheduling short-term ones) but also two alternate profiles that more-or-less perfectly mimic the SRS behavior of 1) our current flashcard system and 2) Anki (right down to esoteric Anki behaviors like studying ‘learning’ cards with less than one day interval before ‘review’ cards but ‘learning’ cards with a greater than one day interval after them). And it’s all unbelievably customizable – can define custom score fields, perform arithmetic on them after each card answer, and use them as factors in adjusting card intervals or in choosing / prioriting cards – so if you aren’t happy with any of the built-in options, you can build your own from scratch; a review of the past decade of Extremely Complicated Flashcard Algorithm Suggestion emails suggests that nearly all of them can be implemented in our new system. (basically anything that doesn’t require lots of really advanced math / linear algebra / etc, which we actually do hope to support in a later 4.x update [via open-source libraries and/or letting you bring in custom JavaScript] but couldn’t manage for this release)

Progress-wise, the ‘core’ of all of this is pretty well designed / implemented / working well, it’s just really ugly and makes our current settings screen look like a one-button flashlight app by comparison 🙂 So while the ‘hard part’ tech-wise is done, there’s still a little ways to go to make it something non-technical people can use. As to the question of release dates… we’ve missed like half a dozen internal deadlines already, so I’m loath to start discussing them publicly now, but while I can’t promise we’ll actually meet this goal, I would say that we’re highly motivated to have a beta out by next fall because we expect to spend a good part of next fall/winter focusing on the next item in this email:

6) Project Marzipan / macOS

Last summer at their annual developer conference, Apple announced something very exciting called Project Marzipan. Put simply, it’s a framework to let developers easily port iOS apps to macOS. Apple themselves already used it to port four iOS apps to Mac for last fall’s macOS “Mojave” release, and it’s expected that they’ll open it up to third-party developers at their next developer conference this June.

While we don’t know exactly what requirements/restrictions might apply to this, we are optimistic that it will give us a way to finally respond to a decade-and-a-half of feature requests and offer our first official desktop app on Mac. We have already begun preparing for this business-wise (for example, by ensuring that all of our dictionary licenses included the right to distribute a desktop version / dropping any that couldn’t be negotiated to add one) and are hopeful that we’ll be able to turn around a Mac version of Pleco fairly quickly once this project becomes official.

Also, without knowing details or requirements it’s tough for us to say what this will look like on the user side, but our *intent* is that your existing Pleco add-on purchases (at least of content) will also work on that Mac version; once you buy a book, it’s yours, and we’re not going to hit you up to buy it again. However, it’s conceivable that Apple or somebody else might put up a roadblock to prevent that, so we can’t actually promise that your Pleco add-on purchases will migrate over to our Mac app until we actually release it.

(also, while it’s not on our immediate radar, if we ever did a web or Windows port all bets are off – unlike with Mac, those ports would require an *enormous* amount of new programming work that would have to be paid for somehow, and we’re skeptical whether they could bring us enough new customers to pay for that work just from those sales alone, so it might well be that you would have to purchase all of your Pleco add-ons again [albeit perhaps with a sizable discount] in order to use them on Windows)

7) App Store Reviews

This is a pretty minor one, but if you use + like our app on iOS and you haven’t left us a review in the App Store in a couple of years, would you please consider doing so? You don’t actually have to leave any text, all we care about is the star rating.

Apple have finally dropped their policy of resetting apps’ review counts on every update, but they didn’t grandfather in reviews made before that change, so the upshot is that other apps that either a) were launched after that change or b) aggressively nag their users to leave reviews now have way more reviews than we do, which makes our app seem weirdly unpopular when reviewing search results. It would be great if Apple listed install numbers on their app store listings like Google does (“Installs: 1,000,000+”), but sadly they don’t, and so the count of star ratings is actually the closest thing the App Store has to a popularity gauge.

I’m a big believer in writing apps the way I want other people to write apps, which is why we don’t charge subscription fees or upgrade fees or nag people about reviews or send out annoying push notifications hawking new products or any of that other irritating-but-surprisingly-lucrative stuff other developers do, but I also don’t want our refusal to play this stupid please-review-our-app game to cause us to actually lose customers, so it’s a delicate balance to maintain.

8) Android Q

Google just released the first beta of this year’s major Android update, “Android Q”, and unfortunately there are a couple of changes in it that could impact some popular Pleco features:

a) Background apps will no longer have access to the clipboard, thus completely breaking our “Clipboard Monitor” feature; and

b) Apps that overlay the screen (as our “Screen Reader/OCR” function does) will be severely discouraged; they’ll be totally barred from overlaying the screen on low-end phones running “Android Go”, and on other Android Q phones, you’ll have to re-grant permission for our app to overlay the screen every time you reboot, or, if you install our app from outside of Google Play, every single time you toggle Screen Reader on.

It also seems highly plausible that they’ll block screen overlay apps altogether in future releases. We do at least have a partial workaround for that, in that there’s another category of overlay – accessibility overlays for visual assistive screen reading – that they aren’t proposing to ban/restrict and which we could use for Screen Reader instead of the type of screen overlay we use currently. (they did make some moves towards banning those last year, but thankfully, enough people spoke up that they backed down from that; if they did ever ban them it would probably only apply to Google Play downloads and not to sideloaded apps)

Nevertheless, it seems clear that Google are moving in the direction of making Android a lot less useful for Pleco customers, so along with warning you all about this, I also wanted to ask that if you care about these changes, please contact Google and let them know that these features – overlaying the screen and monitoring the clipboard in the background – are important to you and that you don’t want them to be taken away in Android Q. Some ways to contact Google:

It might also help to contact the manufacturer of your Android device, since they can also apply pressure on Google and – if that fails – can potentially also intervene to reverse some of these changes on their end in their custom versions of Android Q.

Thanks for all of your continuing support / word-of-mouth / feedback / Angry Rants About How Our Flashcard Algorithm Doesn’t Apply Logarithms Correctly and we look forward to bringing you lots more exciting Pleco stuff soon,

Michael Love
Founder & President