untested.sonnet.io
After 6 years and 2 million words of daily writing I feel like I've learned enough to make Ensō simpler and more accessible.
Related thread: https://news.ycombinator.com/item?id=38025073
Brain dumping also works the same way. You write whatever you have in your mind, without even correcting spelling errors. It really brings out things you don't know they are there and bothering you or taking space.
You should at least try once. Takes an hour or so.
I also use a similar method for drafting my blog posts if I have the idea, but can't bring out the rest of the text.
to me the ability to manipulate thought fragments is the killer feature of writing. If I want to get stuff out without needing to see it immediately, i just record my voice talking
I argue that brain mechanisms are different for speaking, typing and writing.
Speaking with intent is very powerful, indeed because hearing what you say is very different than an inner dialogue, but from my experience writing and typing has a greater power for dislodging what's in the nooks and crannies of your mind.
I can speak my mind, but that path is generally more filtered and shallower than sitting and writing what I have in my mind. Typing is a bit more inferior, but comparable.
The trick in writing/typing is it allows "blurts". You dump the initial 10-20 items at a quick pace. Then you start to slow down. This is where magic happens, because while writing/typing you experience small "a-ha" moments and continue writing things triggered by the initial "burst". This small blurts which drive the rest for an hour or so, and you feel tired but relaxed, because now you have a blueprint of what "keeps you up" is in front of you.
This is not equally possible with talking + auto-transcribing, because brain needs a different pace for that dumping process to work.
Are you suggesting that this experience of yours applies to everyone or that it applies to every context (alone vs not alone)?
Nothing stops you from voice record 10-20 items and stopping or pausing the recording. The idea that talking plus transcribing does not let you do what you do when typing seems odd and likely not based on actual facts. Stephen Hawkings did not seem to struggle to do highly intellectual work without typing
I'm VERY conservative with adding new UI elements, especially those introducing new possible sources of distractions, so I might hide it behind a bunch of menus. That said, I've spent ages yak shaving / working on those problems already :)
When I’m at home, I do most of my writing now with voice input. Would somebody please invent a sound cancellation device that will enable me to talk to my devices in coffee shops and on public transportation without being heard by others?
I think it's funny that it's very similar to ensō in many ways, but also the complete opposite: ensō is calm, mindful, soothing. MDWa is hectic, terrifying, sadistic. Funny how a tiny difference produces products that look almost the same, and feel completely different.
huge props to rafal for creating ensō, personally really love it
So glad to come across it again!
At least my https://meat-gpt.sonnet.io gets indexed well, including 100s of AI websites who webscraped it and hallucinated product descriptions.
0: https://storyempire.com/2025/04/28/nanowrimo-closing-what-we...
I made something similar inspired by this few times in the past.
I think this is already quite perfect, ambient music I can provide myself.
While I did thought of new features, they are really not needed. I especially like coffee shop mode. I often feel self conscious about things I am writing, so hiding text is fantastic.
Apparently it’s an app. I suppose that blog post is only for people who already know what the app is?
cat > ~/my-notes.txt <<EOF
Type here anything you want.
To save and finish, type:
EOF
Nevertheless looking forward to following this project!
Sometimes providing TLDR means you are providing a way for people to instantly ignore you without further thought. Maybe there is a desire to engage people who think.
Thanks for the feedback, most of my usual readers know about Ensō and it seems I forgot to leave my little bubble when writing this post!
If you want it changed further, you're welcome to email us at hn@ycombinator.com.
On one hand it’s hard to disagree with the statement “it should be clear what your product does in a single glance”. In fact there’s a whole meta that’s been developed around this, with well established common wisdom on how to structure your landing page to quickly frame value to prospective users, call them to action, etc.
On the other hand it’s kind of fun to stumble upon something and feel like you missed the beginning of the conversation, and to figure things out piece by piece based on context.
I was also confused when loading this website but it led me to trying the app and it was kind of fun.
Sometimes the most optimized, clearest path isn’t necessarily the preferable one.
Untested is my playground/a place where I "work with the garage door up", so generally I allow myself for more flexibility, especially since this post is more of a devlog entry than a one-glance product page (that would be https://enso.sonnet.io). That said, I had a break from writing, so ended up putting too much content in one place, which made it harder to edit.
What's going to happen in the next few months is this: I'll post more dev/design-log style posts on untested.sonnet.io, then extract some of this information into the product page.
I'm glad that you had some fun with the app!
Not to be confused with Ensso, maker of fountain pens <https://www.ensso.com/>.
What? Damn.
The hard thing, I find, is structuring text so that each paragraph has a purpose in relation to the others. I was once taught this in school, but I haven't kept up with my practice.
So, maybe a tool that takes previous paragraphs and--contrariwise to letting them recede into obscurity--shoves them repeatedly in my face?
Anyway, very elegant and pleasant. Like a foggy quayside cafe.
Also, you might like The Fieldstone Method (Weinberg).
PS. Andy Matuschak's notes: http://notes.andymatuschak.org have some good tips on a similar subject. (My "digital garden" is more of a choose your own adventure book, I'm not married to a single methodology, but I appreciate much of their work)
If so, I recommend looking at Barbara Minto's Pyramid Principle.
https://www.amazon.com/Pyramid-Principle-Logic-Writing-Think...
We don't know what it is based on the description, so even the simplest "Try xyz" or even some goal would help us discover what it is.
I think this is also a problem of framing.
I've collected a lot of high quality feedback over the years*, plus have defined user personas/problem areas (examples: writers, developers, neurospicy folk, people working on their mental health through journalling/expressive writing, YouTubers, video essay creators, ...).
Over the next few weeks/months I will continue writing/thinking about those on untested.sonnet.io (working with the garage door up, so to speak).
Then, once I come up with more terse/clear ways of expressing this -- I'll put it on the product page (https://enso.sonnet.io)
* thanks to relying on an email link over analytics in the app
Enso it pretty overloaded as name for tech things.
Good to know.
But when I clicked around I found what the app was and I liked it. Here the cuteness was charming. Great work!