atrilumen

I look forward to when variable fonts work well on all systems. (I really like rsms.me/inter)

Meantime, rocking the system font.

https://tachyons.io/docs/typography/font-family/system-sans-...

https://css-tricks.com/snippets/css/system-font-stack

And don't miss "Turning the database inside out with Apache Samza"

<3 https://www.youtube.com/watch?v=fU9hR3kiOK0

Thanks!

Any way to bypass the paywall? I really wanna read this one

Thanks, Rabid. I know I'm not being clear.

The goal is to present a magic software interface.

I want to say, "forget about apps. What do you want to do? How can I help you?" ...and be an interface to the people who can make these things happen, whether it's location-based services, "business logic", or integrations with other platforms.

I want to offer a service like Magic, for concierge, but include software development as one of those services. Lazily, on demand.

Presenting everything in threads, including graphical elements, is what I see enabling this wizard of oz approach to creating new software on demand.

It's also how I imagine eliminating the clutter of 'traditional' software. Instead of throwing a bunch of buttons, links, and menus at people and hoping they can navigate it... Slater just tells you what to do next, and never shows more than a small handful of choices at a time... and always gives you a way to ask for something different. And if it's something unexpected, it gets handled by a human, and that's how we learn what to build. If a human isn't available immediately, it will ask for permission to holler back later (push notifications, email, etc).

I don't think words do it justice. I just have to show it. FML.

What up, Clay?

I'm a software developer with a weird background: autodidactic since early childhood; always working obsessively on a personal project; freelancing since about 2005 as a last-ditch solution to financial distress.

For the past several years I've been working on a novel computing interface, and approach to software creation. I've failed to launch or even complete a satisfactory demo because without funding I just can't seem to finish anything; every road looks too long at some point, and I reboot. Or I take on some paying work, and when I come back to it my thinking or dependencies have changed, and I reboot. The result is many broken projects and no demo / MVP.

I'm obsessive and highly resistant to compromise, and I need help managing time and money. I need a strong business partner who can understand my vision and help me to stay on track, and make things happen while my head is down (or in the clouds).

I need help getting funding. I need help scoping an MVP and getting it in front of early customers.

But what is it? See the deck linked in my profile, and please ask questions. I can't seem to get people to see it without very lengthy discussion. Even then, I know they aren't seeing what I see. I can't express how frustrated I am.

I can't stop. I feel like, because I see it, it's my responsibility to the world. I imagine all the people that it will help, and know that if I fail I will have let them down.

I'm doing unrelated work now to feed myself, and making very slow progress. Please, somebody unshackle me so I can do this!

Brilliant. I want one.

I've been using a Pixelbook for the past year or so.

I do like the look and feel of the hardware: nice screen, decent keyboard and touch pad, and a thin metal body which feels really solid. The silicone palm rests start to yellow a bit, but really increase comfort. Overall, I'm happy with the hardware.

The software is an interesting concept, and actually the reason I bought it. I like the idea of a minimal, verified Linux that only runs Chrome, and (WIP) Linux containers for everything else.

I've had many issues that disrupted my work, and required switching to different release channels. Obviously this is on me for choosing to use the "Beta" container support feature, but I am still annoyed by it often because somehow bugs make their way through all those channels: Canary -> Dev -> Beta -> and even down to Stable. Several times I was forced to move to a higher channel to escape a serious / disruptive issue, only to encounter different issues. And then moving back down again requires a "powerwash", wiping all your data. So I feel like I'm just being chased around by issues that should have been resolved in a higher channel.

As a result, there tends to be nothing on this machine that isn't in the cloud somewhere. I suppose that's a good thing.

Trust is another issue, entirely. I'm looking for something else. I may install an alt OS like Gallium.

I couldn't read the first sentence because I wanted to zoom in a little.

I'm not ready to show it yet (#FML), but my interface renders everything in threads, including graphical controls, which yield messages.

Nothing happens outside the context of a thread. So it leaves a log of everything that happened in that context. Graphical elements can go full-screen... but you don't lose connection to the thread that they appear in.

It will tile these threads across all your (funky, heterogeneous) screens.

I think this will help you to organize things, and homogenize and simplify your experience with those things across devices.

Imagine actually knowing where everything is, and being able to see who, when, and why anybody touched it.

And you're better off without learning those things? Man, if there's a problem, especially with me, I need to know it.

Also, you learn from missing deadlines. So having deadlines must be valuable even if you consistently miss them.

Good point. But I guess the lesson for me is to have deadlines in the first place. At least it gives you a target.

I think, "go! go! go!"... and then fuck around with small details until I go broke. I need way more awareness of the big picture.

It's so hard to balance. I need a partner, but I've really isolated myself at this point :(

Thanks for this.

I've been struggling to work full time on a project for a few years now. I think it's the most important thing that I can do with my life now, so I refuse to let go for any reason. (I'm a US citizen literally starving in Colombia.)

I can't get traction the way I've been going, so I have to make radical changes immediately. Thanks for the list.

    1. Tenacity (check!)
    2. Logging  (need a lot of improvement. where has all my time gone?!)
    3. Compounding (I fear I'll be contemplating this one all day now)
    4. Defending your time (prob need to back off on this one; too lonely)

I'm really loving Musk's rhyme: If the schedule is long, it's wrong; if it's tight, it's right. I've been moving way too slowly, spending too much time on small details.

I need to do my thing more openly instead of hiding out with my head down; maybe I would have never done that first project reboot, and I'd have customers by now!

A thing being impossible or highly cost prohibitive kind of hampers demand for it, wouldn't you think?

AR goggles for dogs, with remote-controlled training and control characters.

I think that a better approach to software is to only build what is absolutely necessary, lazily, according to real user stories fulfilled by customer service. Try to automate away all the customer service, but only in response to learning from it in the first place.

So I'm building a Wizard of OZ interface for creating custom Wizard of OZ interfaces.

I'm building (and betting everything on) 'an app' on Telegram, and this shit makes me nervous.

Telegram, please don't crash

I know it's off-topic, but fuck Medium. No, I don't want to join. From now on, I will not click links to Medium!

(I'm tired of deleting their stupid cookies just to read a goddamned blog post.)

This is why I want to design and develop software _lazily_, as a kind of perpetual wizard of oz experiment. I'm laying the foundations now, for a 'magic' software experience that learns to become whatever you need, lazily, driven behind the curtain by operators and agents, AI and human.

https://en.wikipedia.org/wiki/Wizard_of_Oz_experiment

I'm not sure why, but the word "enterprise" seems to trigger a knee-jerk tab-closing reaction for me.

I really enjoyed mine, keyboard included. I rocked it until it died and rarely needed more power for web dev.

I miss it! I don't think anything else competes with that form factor. I carried it in a little sling bag _just big enough for it_, and was content to keep it with me always. Like nothing. I tried maintaining that habit with my new Pixelbook and a backpack, but alas; that shit is too bulky for me now. Thanks, Apple. You ruined me there.

RIP 12" Macbook <3

Man, I keep thinking "just keep on chugging with Choo", but I'm really tempted by Svelte.

( https://choo.io )

Haha, it's a little skewed.

    section {
      transform: skew(-2deg);
    }

Seriously, hope someone is willing to tackle this hypothetical scenario:

Land owners are unaware of a buried cache of gold that was robbed from an armored car in the prohibition era...

And it has to be dug up.

The spot is marked with a stack of large rocks. The previous owners were ostensibly always deterred by the threat of blasting caps being buried on top of it.

What would you do about that? Anything?

Edit: Offer to split it with the land owners, and call the excavators? (and the bomb squad?)

For sure. This post freaked me out because my main machine now is a Pixelbook.

Cool, thanks. I'm tempted to make a Node binding.

Chromecast was Chrome OS, which is not Android. (And I doubt that's changed.)