I'd be curious to hear more. I presume you're using some kind of semantic search? Any other kinds of semantic technology? What kinds of insights into your journaling has it offered you/users?
The work thus far has mostly been on a seamless mobile UI, & to make engaging with the LLM like less of a chat and more like browsing an interesting dynamic wiki about your own life. But also, yes, a highlight for me re: having your journal in context can be to search and find commonalities, patterns.
I would say the insights frontier models have given me, at a high level, match some of those offered by professionals in a theraputic context--which is one reason I'd be curious to make an affordable/accessible app. Although I tread lightly into depersonalizing such a human area with techno utopian naivete...
I suppose "the Internet will interpret censorship as damage and route around it" was utopian naivete that got us pretty far, and "the LLM understands me and can help me understand myself" might be naivete it's useful to disprove/explore the truth of. I think utopianism is important even if it's Sisiphean, it's good to have a north star even if it can't be reached.
Fear that postmodernism "infects" the humanities has always read as condescension to me: "this idea is too dangerous and 'normal' people won't be able to see the ways it will be abused, so we'd better reject it."
At some point (not sure if it is still true) Jai had or was going to have language-level implementation of relative pointers, i.e. a 16 or 32 bit relative offset from “this field’s memory address.”
> It makes me sad because while they have awesome professional accomplishments, they would have been awesome mothers and parents as well.
I just wanted to point out politely how strange this language reads to me. It's ultimately the decision of each person whether to have kids or not. This being "sad" for them is an overreach, morally, IMO--and part of the fight feminism led was to carve out a space for women in society not to be mothers, if that is their choice.
I have many female friends in their mid-30s and early 40s, and the ones without kids are almost always sad/regretful about it themselves. Some won't admit it, but some will, especially in more intimate settings. I've had one cry on my shoulder - she is very successful professionally, but feels she missed out on the opportunity for motherhood.
Whether it was her "choice" is hard to say. Very few people get the chance to engineer their lives the way they want. Most people instead play the cards they are dealt. And now that there are many more different types of possible cards in the deck for women, they are coming to the understanding that it is possible to be dealt a bad hand, and not realize it until it is too late.
The issue is that there's no objectivity here. Are these women actually sad because they truly wanted to experience motherhood, or is it more because of pressure (even unconscious pressure) from family and friends to have kids? Or just the built-in feelings due to upbringing that having kids is "just what you're supposed to do".
I wonder, though, if these same women did have kids, and sacrificed their career for it, would they look back and have regrets in that regard, too.
So maybe it's just... given a multitude of options where we can't choose them all, perhaps humans will just naturally have some regrets around the path(s) not taken?
As counterpoint, I know women in their 40s and older who are very pleased with their decision to not have kids.
I didn't reference sadness as a moral projection but rather out of empathy as some of these friends will share fears, doubts, and regrets from time to time. If a woman (or man) doesn't want to be a parent, then more power to them and we should all support that.
I think the idea of sadness is that choice were not made with relatively few externally imposed restrictions from a place of independence. Rather the decisions were taken from a set of choices made in a context of a society that imposes restrictions than necessary based on somewhat arbitrary traditions, or genderism, or biased, or at least questionable economic restrictions.
Making a choice properly requires having knowledge about the particular options. Fertility education particularly in regard to age is not part of the sexual education curriculum in the UK [1] nor in international guidance documents [2,3] (just Ctrl-F "fertility"). And a recent poll in the US showed that 77% of women did not properly understand the relationship between age and fertility [4]. At some point this stops being choices made with appropriate knowledge and starts being a massive and tragic policy failure.
The statement you reference is "77% of women do not know that when a woman is 35+, her age is a better indicator of her fertility than her overall health." As a women in her 30s, I'll say that every damn woman in America has heard of a biological clock by her 20s, and is quite aware, and many are getting calls from Auntie every two weeks about whether she's getting on it yet.
Looks a bit like Odin (https://odin.handmade.network/) --a language explicitly designed to be small and simple I've been enjoying learning and toying around with.
Actually, you can use DevTools to inspect DevTools. Hit option-command-J to bring up devtools, make sure it's in "window mode," i.e., in a separate window from the browser itself, and then hit option-command-J again to DevTools that DevTools.
You can actually just keep recursing and inspect the previous inspector ad nauseam...
Using React every day has me sold on functional reactive programming for UI. But there's definitely some friction at the boundary between the nicely declarative React tree and the rest of the world with all its mutable state.
I'd love to experiment with with React and Elm (or maybe Ur/web?) together.
You could also be interested in PureScript[1] then; and its React bindings[2]. It is a statically typed language which compiles to JS. It has row polymorphism and effects (yay!).
This is the stack I'm planning on attemtping to use for my next project. The purescript-react docs could use some work, and I think the best example is the one in the pull request "Another example", but it seems like purescript + react is a really nice combo.
Have they been working on compilation speed? About a year ago I was trying some WebGL "bindings" that had definitions for all the constants and functions, and the game I was hacking on would take 20 seconds to compile on a fairly recent Macbook Air.
It has improved in recent releases, but I think they are focusing mostly on incremental compilation via the -watch flag. My projects still take 5-7 seconds to build from scratch on my MacBook Air.
You probably know this, but just for others' information: Type inference is expensive, as it requires global information about the project. As a result, compiling one TypeScript file in a project results in the compiler recompiling all of the referenced modules and definitions to re-build up that inference information from scratch. Using the -watch flag keeps the compiler alive with that information in-tact in memory, enabling incremental recompilation.
No, it's still rather slow. We've "fixed" by doing conditional compilation (using an md5 of the files). A full compile of our project takes a few minutes.
