kepano

There are dozens of options for this: Jekyll, Hugo, Quartz, Astro, Eleventy, etc.

When I think about Photoshop it is so tied in my mind to its history as an offshoot of ILM and the VFX industry https://en.wikipedia.org/wiki/Adobe_Photoshop#Early_history

ILM's famous t-rex scene from Jurassic Park contains very little text/dialog, but emotional, expressive, synthetic imagery: https://www.youtube.com/watch?v=Rc_i5TKdmhs

In this case the scene is not made up of "generative" images in the current definition of the term, but synthetic images generated from polygons, virtual lighting, etc. It seems that there could be artistic utility to manipulating text in a similar way.

Genuine question — do you think synthetic images pose less of a problem than synthetic text? If yes, why?

The repercussions of what the author summarizes as "could you colour-grade a book?" still feel wildly unknown to me, even after a couple years of thinking about it (see Photoshop for text [1][2]).

Partially it's because we're still wrapping our heads around what kind of experience this might enable. The tools still feel ahead of the medium. I think we're closer to Niépce than Muybridge.

In photography terms, we've just figured out how to capture photons on paper — and artists haven't figured out how to use that to make something interesting.

[1] https://news.ycombinator.com/item?id=33253606

[2] https://stephango.com/photoshop-for-text

It would be so cool to be able to save a delve exploration to a set of linked Markdown files that you can open in a local wiki app.

No need for a plugin, it's built into Obsidian, just turn on stacked tabs

https://help.obsidian.md/User+interface/Tabs#Stack+tab+group...

I preordered one already but will do!

I've been dreaming of a device like this to run Obsidian on. I made an e-ink theme for Obsidian[1] but the refresh rate and ghosting of e-ink makes it less appealing to work with. I am curious to see if this device will solve some of those issues.

[1] https://minimal.guide/features/eink

Yes, it seems uncontroversial to say you will eventually die if you don't eat food... And it seems uncontroversial to say that eating higher quality food will help you live longer. Quality food deserves your hard-earned cash too.

The relationship between health and software is less tenuous than it may seem. We spend so much of our time using software — it can have a significant effect on our health. See studies that explore social media use and its relation to depression, anxiety, suicidal thoughts, etc.

Big food and big software are similar in that they are cheaper to mass produce, more effective to advertise at scale. As such many people are trapped into assuming there are no alternatives. You may never develop an understanding of what it is you're being fed.

I don't think you will die if you buy cheap jam — at least not quickly. The choice people make to buy higher quality jam isn't a function of life or death. It seem related to a perception of value, which good software doesn't seem to evoke for the vast majority of people.

Converting time into money is easier for some people/professions than others. However converting less useful time into more useful time is a great thing for anyone. If an app gives you an extra 5 minutes every day, that's 30 hours per year that you can use towards something of your choice, whether it's sleep, exercise, recreation, work, etc.

Consider how much one hour of your time is worth — you are probably underestimating that number.

Assuming you haven't exited to a barter economy you must surely be buying food? Unlike food, software can create time. Time = potential earnings. Consider a piece of software that costs less than a sandwich but the leverage it creates frees up some amount of time that earns you more than the cost of a sandwich.

Some software has ongoing costs, such as support, servers, bug fixes, etc. But even in the hypothetical case that a piece of indie software has no ongoing costs, it seems fair that the creator should be at least somewhat rewarded for the value the software creates, regardless of how many copies are sold. Without this incentive we are stuck in a world where the only reason to create independent software is for the pure love of the craft, which is noble but only magnifies the advantages big tech has over indies.

Ah good catch. Yes the "D" key can be used to switch light/dark mode, but I didn't account for the bookmark shortcut. That should be fixed now. Thanks!

That's exactly my point. "File over app"[1] is just as relevant for businesses as it is for individuals — if you don't want your data to be used for training, then take sovereignty of it.

[1] https://stephango.com/file-over-app

In this case they mean leak into the global model — so no. You can have sovereignty of your data if you use an open protocol like IRC or Matrix, or a self-hosted tool like Zulip, Mattermost, Rocket Chat, etc

In summary, you must opt-out if you want to exclude your data from global models.

Incredibly confusing language since they also vaguely state that "data will not leak across workspaces".

Use tools that cannot leak data not "will not".

if your data is stored in a database that a company can freely read and access (i.e. not end-to-end encrypted), the company will eventually update their ToS so they can use your data for AI training — the incentives are too strong to resist

https://twitter.com/kepano/status/1688610782509211648

https://twitter.com/kepano/status/1682829662370557952

The project's readme about "How it's made"[1] is interesting. The content of the Markdown repo is enjoyable to explore as an Obsidian vault[2].

I’d like to say that I wrote some clever site scraper and then algorithmically generated everything but the real answer is that it was mostly done by hand with a template for the pattern structure, the help of Obsidian’s auto-complete combined with a keyboard macro to convert (and correct) the links, and many dozens of hours typing, copy-pasting, and tweaking. I reference these patterns directly in my Obsidian vault. I began in 2020 but I didn’t work on it in earnest until November of 2023 with a final push in April of 2024.

The “master” version is my Obsidian vault. I have a custom Fish function (apl-copy) using rsync to copy from my vault to the Markdown repo that contains the patterns (and list of patterns), the README, and the LICENSE as markdown files.

[1] https://patternlanguage.cc/README#how-its-made

[2] https://github.com/zenodotus280/apl-md

In searching for the primary sources for the quotes in this article I found that "Extreme Metaphors" (ISBN 978-0-00-745485-3) seems to be the best collection of J. G. Ballard interviews.

Quite a few are collected here:

https://www.jgballard.ca/media/interviews.html

One of the interviews referenced in the OP can be found here:

https://www.jgballard.ca/media/1987_november_i-d_magazine.ht...

This seems to be quoting from https://softpanorama.org/People/Knuth/index.shtml

> When I was at Stanford with the AI project [in the late 1960s] one of the things we used to do every Thanksgiving is have a computer programming contest with people on research projects in the Bay area. The prize I think was a turkey.

> [John] McCarthy used to make up the problems. The one year that Knuth entered this, he won both the fastest time getting the program running and he also won the fastest execution of the algorithm. He did it on the worst system with remote batch called the Wilbur system. And he basically beat the shit out of everyone.

> And they asked him, "How could you possibly do this?" And he answered,

> "When I learned to program, you were lucky if you got five minutes with the machine a day. If you wanted to get the program going, it just had to be written right. So people just learned to program like it was carving stone. You sort of have to sidle up to it. That's how I learned to program."

Obsidian runs on local Markdown files so there are many free options you can use for syncing and publishing notes.

I understand what you mean. This will be most problematic if you downgrade far in advance of the renewal date, especially if you are on the yearly cycle. We'll look into improving that process.

Downgrades are trickier for Obsidian than for cloud-based apps because Obsidian is end-to-end encrypted and stores files locally. We also have to deal with different client versions. Deleting files on the server could cause the Sync client to think these deletions should be applied locally. Therefore we put it in the hands of the user to set up the account so that it is ready for the downgrade.

But yes, this process could be smoothed out a bit. The growing pains of going from one to two plans :)

This clarification has been added to the spec:

> Nodes are placed in the array in ascending order by z-index. The first node in the array should be displayed below all other nodes, and the last node in the array should be displayed on top of all other nodes.

Converted the PDF to plain text for readability here:

https://gist.github.com/kepano/0153563e9cdaef3c9fb9cf997003a...

Chyrp was/is amazing. It's so great to see that this project is still alive. When I first set up a Chyrp site around 2010 I had intended to start blogging, but was never happy enough with any of my entries to make them public so I kept it private.

I used Chyrp for several years as a private journal, until I switched to TiddlyWiki and eventually building Obsidian. Now that I do have a public blog I often come back to some of those old drafts I wrote in my Chyrp days, and have polished and published quite a few of them.

I remember loving how low-friction Chyrp was, both to write and build themes/plugins for.

I can't believe it took this long for someone to post this haha

Whoa! Didn't expect this to bubble up to the top of HN. Some context about why we created JSON Canvas:

https://obsidian.md/blog/json-canvas/

We just released it today, so this is still a very nascent project. A little over a year ago we released Obsidian Canvas. The .canvas file format has felt stable enough to give it a name and resources that other apps can freely use. See the original Show HN:

https://news.ycombinator.com/item?id=34066824

The spec is conservative, and definitely does not support many features canvas apps will want to implement (yet).

The purpose of giving JSON Canvas a name and site is to encourage an interoperable ecosystem to grow around this format. We're definitely looking for feedback of all kinds!

It's great to see all the suggestions already shared in this thread because it starts to provide a roadmap for how this could become a more useful format for other apps.

I didn't mean untenable legally, but from an public/internal perception.

Now that Google, Meta, Mistral, etc, all have great open source models, it seems rather untenable for OpenAI to

1. keep "open" in the name

2. stay closed source

3. pretend to be a non-profit

at least one of those things must go, right?