Of course, back when I was starting with Japanese, my teacher insisted that we use spaces until we knew enough kanji to make our writing legible without them.
Vicious :)
Rather, that you can look at something like "食べる林檎" and, using just orthography, find the word boundary "食べる 林檎".
I threw katakana in there just because a word written using katakana instead of kanji does not actually present a problem for this.
Words written in hiragana, however, do present a problem for this naive approach of finding word boundaries, but do not pose a problem unless you have a long string of them.
Ask even a native, and they'll tell you that characters are a big part of the problem (though less of a problem in Chinese than in Japanese).
Another confusing item in learning the language is that so many people are named after things, so sometimes it's not obvious in a sentence that you're talking about a person rather than an object and you have to reparse the sentence a few times. Obviously the language works because there's a giant country using it without people constantly having to second-guess each other, but as a learner, it's one of my pain points.
I use Anki flashcards to review vocabulary; the Chinese Support plugin makes it easy to create cards with colored Pinyin Ruby, pronunciation and translation. I read just about anything for practice; lack of space has never really been a problem. When I find a sentence I can't tokenize, I'll have to look it up anyway. (I use the Pleco dictionary for Android.)
BTW: I saw you complaining on your blog that 望 is being displayed as 亡月王 instead of 亡匀王. Since I'm probably never going to email you about that, I'm tacking it on the end here. It's the font reflecting different handwriting preferences [1] that Unicode decided not to encode. I used to have the opposite problem and had to install a new font to have it look the way you don't want it to look. I suggest finding a font you like instead of putting up with the font you currently use.
[1] https://en.wiktionary.org/wiki/%E6%9C%9B#Alternative_forms
I guess you have a really good memory to remember words just from flashcards. I tried that, but could never remember the words after class. Learning words is much easier for me if they're in context.
https://pingtype.github.io/docs/blog.html#vocabulary
In the end, I discovered the Pingfang font, and now I load that every time you visit the Pingtype website. It slows down the page, but it looks better and has fewer glyph problems.
But thats kind of the advantage too, there is no preset standard grammatical way of looking at anything, you simply take it all in and infer everything on context. No learning verb conjugations and endless rules like in some languages (I'm looking at you, French)
People are saying Chinese is harder to learn than other languages, and it's true, but mostly only for the first few years, after that, it's way easier to learn than a new romance language for a native speaker of a romance language
I don't speak Chinese, so I don't have a personal intution about this, but in this case, I think you're conflating as many as three different linguistic issues.
* Word boundaries not being represented in orthography means that there is more ambiguity about whether something should be regarded a compound word, a fixed phrase, or a single word. Speakers of languages like English may disagree about items like "mail box" vs. "mailbox" or "zu Hause" vs. "zuhause" ('at home' in German). In Chinese orthography there does not necessarily need to be an answer to these questions and people don't necessarily need to take a position on them.
* "No grammar" might be an intuition that people form about Chinese, but it's not true at all. Chinese doesn't mark a large number of grammatical categories that other languages do, but it marks some that other languages don't, such as classifiers or measure words used with numbers.
https://digchinese.com/measure-words
Also, Chinese word order is not free and forms a part of the grammar. So 我爱你 'I love you' is a different sentence from 你爱我 'you love me', and that distinction is a grammatical distinction. In Latin, "ego amo te" and "te amo ego" have the same meaning! (The inverse is any permutation of the words "tu me amas".) In comparison with other languages, this is one way in which Chinese speakers don't just "take it all in and infer everything on context", because they get to use word order to derive the meaning, which would be a far less useful signal in Latin or Russian.
* Chinese pretty much never inflects words, so there aren't tables of inflections to learn as there would be in an inflected language. But conceivably an inflected language speaker might feel that Chinese is being verbose (!) by requiring the expression of subject pronouns in order to indicate grammatical subject. For example, a speaker of inflected pro-drop languages (where subject pronouns are optional) like Latin or Spanish might find it weird that you have to use two "words" 你有 to express what to them is just a "simple" word like "habes" or "tienes" 'you have', that inherently reveals the identity of the subject in a concise and intuitive way. :-) Or for tense, speakers of these languages might find it strangely verbose to have to throw in 以前 'formerly' or 曾经 'once' or something, instead of a concise infix like -ba- (amabas, in both languages, instead of something like 你以前爱 which can be regarded as three distinct words ... or maybe, as you said, not always analyzed that way by some speakers!).
DeFrancis' counterexample for the myth that each Chinese character is a word is "coral", 珊瑚 (I hope). According to the dictionary he quoted, the two characters, individually, mean "coral" but they're never used separately.
Unfortunately for me, I'm a programmer, not an English teacher, so I spend all my time reading instead of speaking. So I have to learn it the hard way.
But:
* where is the code? i couldn't find it. Is it a private repository? Ability for others to contribute might go a long way.
* Presentation matters, a lot. The interface looks more like a 90ies windows program than a modern web-based tool. Working on the design and UX a bit could go a long way.
* a short explanation of what it is and does and how to use it right on the start page, which can be toggled on and off, might help too.
- How should I improve the interface? Specifically. I mean something like "orange buttons" not "use React".
- I used to have an explanation text that was in the translator input areas when the page loaded. Soon I realised that it was quite annoying, because I had to manually clear the inputs before I could enter my own text to translate. Toggling on and off would require a cookie, and then I would need a cookie warning (which is another barrier to the user).
I, too, went to Taiwan when I began learning Chinese. It was very interesting, but ultimately a waste of time for anything except visiting the lovely museums, developing an early (in my Chinese linguistic education) feel for the transition between traditional and simplified Chinese characters, and the political reality of Taiwan's steadfast refusal to accept communist pinyin - a clearly and embarrassingly superior romanization system to anything Taiwan's academics could ever muster.
What I later learned (on the mainland) was that speaking to a wide variety of people with a wide variety of accents about a wide variety of every day matters and things of interest was the more efficient way to learn, but also learn to understand and be understood both verbally and in writing. Of the 1.4 billion people in mainland China, over sixteen years I've met perhaps ten who gave a damn about the bible, and nine out of ten were clearly and literally impoverished.[0]
I would wager that this is the source of your issues: frankly, nobody cares. You are learning a vocabulary without practical community or utility. It may be very important to you, and perhaps Taiwan has some happening churches, but in the context of Chinese language it's totally irrelevant to almost everyone.
What Chinese and Japanese need are spaces.
I struggled for 2 years to study Chinese here in Taiwan. I tried many ways (see the blog), and wasn't picking it up.
Then I wrote Pingtype, which puts spaces between the words, colours for tones, and a literal translation. I started reading the Bible every day using it, and also listening to music and reading the lyrics. Since then I've learned so much more.
https://pingtype.github.io
The problem is not the language. It's the education materials. Now I think that traditional characters are better, because I can rebuild a character from typing only the parts I know into Pingtype's keyboard and searching for a suitable match. I had to build my own tools, but I wish these could be popularised to help the thousands of foreigners who give up learning Chinese.
Support for Taiwanese Hokkien is also included in Pingtype, thanks to the Taigi dictionary for POJ transcription. See the Advanced -> Regional section. It is very different to Mandarin, but learning Chinese is a necessary stepping stone. I'm working on a lyrics video of some popular songs by Fire Ex as some educational materials.
Please give me feedback on Pingtype; only a few hundred people have visited the page and it's kind of depressing. I don't know how to do marketing, so it's not catching on. Even if the tone colours and pinyin don't become widely used, I really wish that spaces between words could be introduced to this beautiful language.