Hi logophiles! I am a total word nerd. Over the last six years--mostly accidently--I ended up creating a bunch of vocabulary learning materials and spent way too much time thinking deeply about how we learn new vocab and how to teach it. My story: basically, via word of mouth, people with kids taking the SSAT and the SAT kept asking me for my materials which I continually iterated on as I got feedback. It wasn't my day job, lol, it wasn't even a side hustle.... just an obsession :) As I shared my "system", I kept dreaming of even better ways to make vocab learning effective, easy, and fun.

Some interesting things about learning vocab. The "keyword method" is extremely effective. (The keyword method is associating a target word with a similar-sounding word (the "keyword") and then creating a vivid mental image connecting the keyword with the target word.) [Ávila & Sadoski, 1996; Shapiro & Waters, 2005]. Further, connecting the new word and its meaning to your own personal experience is much effective than rote memorization. ("...engaging in deeper semantic processing and relating information to personal experiences can activate distinct neural circuits compared to those involved in rote memorization." [Andreasen, O'Leary, Cizadlo, Arndt, Rezai, Watkins, Ponto, & Hichwa, 1995]).

There are a lot of other cool things I discovered on my (research-obsessive) path to make learning vocab radically easier. A core driver for me has always been thinking about the epistemology of word-learning. What does it mean to "know" a word? "Knowing" a definition is different from truly knowing a word, where you can deploy it effortlessly when the context is right. That led to endless rabbit holes of learning about polysemy, colocations, and a whole lot more.

The first day I saw Dalle my jaw dropped. This was it! This was the missing piece for learning vocab 2x, 3x, 4x(?) more efficiently than has ever been possible. The image generation AI tools can make a custom image that packs in your own favorite keyword mnemonic and your own personal story into a cool image. Whoa! Because what has been my total obsession could finally be created in the real world, I teamed up with two good friends with the technical chops to build what had been percolating in my brain for six years. We've built a beta version over the last four months and it is ready to test!

I love Tildes, and I don't want to self-promote, so I am not going to drop the app name / website, but I am here with an ask. We want feedback! We want to make this the dream app for anyone who is serious about growing their (English) vocab. We want you in our beta test group.

The commitment I'm asking of our beta testers is a bit onerous. I want to hop on a zoom call with you while you use the app for an hour or so and have you tell me what you love and hate. I want to ask you a bunch of questions about what you want to see in your dream learning app. Then I want to give you the app for a month a two; hopefully you'll use it and learn a bunch of words; then I want to hop on a 20 minute call with you and get your hot take on the whole thing.

It is such an intense passion project for me; I want to make the app just rock-your-world-awesome. That's why I want to do live user interviews. (Which is a little out of the ordinary for sure.) And I can't do that without talking to real people who care about growing their vocab who are willing to hang out with me for an hour or two. :)

As a thanks for your help, when we go to the paid version, you'll get three months free, and a massive lifetime "friends and family" discount. But more than that, you'll really impact what we build next, and how we can make it better. While it's maybe a little idealistic, or might even sound silly to some, I feel like better vocab = better communication = better relationships. So I am all-in, fervently devoted, and hopeful that you'll come along for the ride and help me make it epic.

Who is it for? Studying for standardized tests? Oh yeah. This will help a lot. Want to raise the ceiling on what you can read. Let's go! Want to improve your English skills? This is for you. Love words. Yep! I'd love to meet you! Basically, if you love words, and/or have something coming up that requires that you know more words, I really hope you'll be part of our test!

More interesting stuff about vocabulary:
--Average high school graduate has a vocab size of 16,000 words
--Average college grad, 20,000 words
--Average PhD. 28,000 words

Tildes is a very smart and well read group, so I' bet the average vocab size around here is 25,000 to 35,000. Want to know your (approximate) vocab size. One of the best (easy and fast) tests is here:
https://preply.com/en/learn/english/test-your-vocab
(I have nothing to do with that site or company, and do not endorse them. It's just that their vocab size estimator is really well done.)

Want to be a beta tester, or just talk more about vocab, shoot me a message!
pandacat@onmail.com

I'm looking for a French grammar checker. I think I'm in that intermediate-level plateau where I just need to keep talking / chatting in French but I want to eventually get to a point where I have correct grammar, maybe even some suggestions for idioms.

Some info for my use-case:

• I don't expect to go past 100 "consultations" a month.
• Would be nice if there was an extension that helps for email / Messenger / Telegram / WhatsApp.
• Would be nice if it did help with idioms.

I did my homework and found out that:

• Grammarly does offer this but only in English.
• Language Tool exists but it's 20 euros monthly or 60 euros per year, which are both steep prices for just trying it out.
• Asking ChatGPT works most of the time, but it's a bit annoying to load up that website every time and ask. I'm open to coding something based on the API if that would be the most cost-effective option.

Thanks in advance for all your suggestions!

The New Yorker has a provocative article asking the question "Can We Talk To Whales?" It boils down to utilizing language learning models to process a dataset of sperm whale clicks, their codas, and crossing one's fingers to see if "ClickGPT" can produce actual sperm whale language.

Which makes me wonder if a language learning model been given a library of Chinese sounds and ideograms, without context, then communicated in workable Chinese?

Using a language learning model to learn to speak to whales is an interesting idea, but I'm thinking any LLM assigned the task will wind up chunking out a word salad or something akin to Prisencolinensinainciusol. I'd like to learn more.