rollcat

Augmented reality is what you already have right here, right now, on your wrist and in your ears.

In my book, AR is completely invisible, until you need it:

- A smart watch + strict notification filters (good idea either way), means you only see a notification when something worthy of your attention happens (you get a text, event reminder, etc). You can glance at the notification and decide if it requires your reaction, instead of reaching for your phone. (And likely, waste another 10min on it.)

- Wireless earbuds with turn-by-turn directions, especially for walking and public transport. Again, you don't need a screen, you can admire your surroundings or read a book.

- Pay for stuff. If you'd spend 10s on pulling a card out of your wallet, it saves you an hour of your life per year.

- Track your vitals. Overall not that important, until you notice and suspect something is wrong - you can compare things month over month, see trends, show it to your doctor, etc. I took a hard hit falling off a skateboard, my watch started a countdown to call for help - I stopped it, but I needed a minute to get up, so it was really reassuring to have this.

’Twould wax yet more marvellous to ye beholders.

> Societies should not require such things though. They need to have trustable institutions and intermediaries to function, in finance and many other areas.

...which is more or less the same conclusion that I've arrived at by the end.

Is there any modern programming environment / IDE / toolchain / etc that doesn't heavily rely on open source work? Even initially-proprietary solutions like .NET are mostly OSS nowadays.

First things first, I'm a crypto-sceptic - to put it in the mildest terms possible.

You're spot on with CPU usage. However: how would you design a RasPi-efficient, fault-tolerant, decentralised ledger with strict ordering and a transparency log?

Consider CAP. Existing banking systems choose partition tolerance (everyone does their own thing all the time basically), and eventual consistency via peering - which is why all settlements are delayed (in favour of fraud detection / mitigation), but you get huge transaction throughput, very high availability, and power efficiency. (Any existing inefficiencies can and should be optimised away, I guess we can blame complacency.)

The system works based on distributed (each bank) but centralised (customer->bank) authority, held up by regulations, capital, and identity verification.

Online authority works in practice - we collectively trust all the Googles, Apples, etc run our digital lives. Cryptocurrency enthusiasts trust the authors and contributors of the software, CPU/OS vendors, so it's not like we're anywhere near an absolute zero of authority.

Online identity verification objectively sucks, so that is out the window. I guess this could work by individual users delegating to a "host" node (which is what is already happening with managed wallets), and host nodes peering with each other based on mutual trust. Kinda like Mastodon, email, or even autonomous systems - the backbone of the Internet itself.

Just a brain dump.

Atlassian broke both Trello and Bitbucket for me, so at least, thanks for acquiring these browsers before I got a chance to get attached?

Anyone interested in trying NeXTStep, it's included on <https://infinitemac.org/>

Looks interesting, but my main obstacle right now is that the power switch is controlled via HomeKit - I could make my Mac an integral part of the setup (it's powered 24x7 after all), but that feels clunky. I think the way to go is to make a shortcut to gracefully shut down via Mac/iPhone. It's one tap, just requires some thinking to set up.

My original point stands, also per TFA - performance gains from SMT are questionable for certain workloads. Whether OpenBSD prioritises absolute performance is besides the point - they benchmark against their own goals, not someone else's achievements.

Lisp (and similarly, Forth) is closer to a mathematical construct than an actual language. We've had to wrap it in parentheses and named symbols to help make sense of it, to talk about it. Pairs, atoms, GCs, are all an implementation detail.

I'm surprised nobody has mentioned OpenBSD yet.

They've been advocating against SMT for a long while, citing security risks and inconsistent performance gains. I don't know which HW/CPU bug in the long series of rowhammer, meltdown, spectre, etc prompted the action, but they've completely disabled SMT in the default installation at some point.

The core idea by itself is fine: keep the ALUs busy. Maybe security-wise, the present trade-off is acceptable, if you can instruct the scheduler to put threads from the same security domain on the same physical core. (How to tell when two threads are not a threat to each other is left up as an exercise.)

> I had original plans to turn the lab off at night and then the phone could keep running from battery. But, the homelab became critical infra and has to be always on lol.

I've rebuilt my entire setup a while ago to "tolerate" power loss.

- Everything is sliced into three zones: "always-on", "desktop", and "homelab". The latter two through a smart plug, so that power can be cut.

- Always-on includes the SOHO router/AP, a RasPi4, and my Mac mini. The router's switch has only two ports populated: the Pi, and the Big Switch.

- Desktop is things like the screen, speakers, dock, wireless charger, camera (dummy battery), cute lamp, etc. All of this can tolerate power loss at any moment.

Homelab: this is the tricky part. There's a big (52 ports!) switch that "everything else" is plugged into, including my Mac (yay 10Gig Ethernet, not sure what for). A bunch of SBCs like Pis, NVidia Jetsons, x86's, etc. Nothing important, still figuring out what to do with all of that, so I kinda don't care if the power is cut.

The important bit is the NAS (RAIDZ 3x3TB). ZFS makes you feel like you're invincible, but I've still built something to keep it clean: 1. it's powered down until required; 2. the router has a cron job to WoL the NAS 5min before all backup jobs start; 3. the NAS has its own cron job that waits for the backups to complete, then waits for all SSH sessions to terminate, then shuts itself down. You can kill the script via SSH.

What I'm planning to do is to build a simple daemon for each box, that checks in once per minute or so, to ask if it's time to shut down. Once everyone is clear, cut the power. Somewhere far down my TODO list ;) until then, you can also use an iOS shortcut to SSH to each box before asking HomeKit to throw the switch, but storing passwords is fugly and handling separate pubkeys is too much bother. So I'm mostly happy with it as-is.

NAT is the hard problem. I'm tracking this one: https://github.com/tailscale/tailscale/issues/11563

You can also talk to your ISP (or their competition). My friend negotiated something like +5€/mo for a static IP. We can play Factorio together, yay!

Similar. I'm going to hold on to my iPhone SE2 (same year, size, mass, touch ID, etc) until it dies.

My 2002 TiBook's original battery is buried in sand, in a bucket. It should probably get recycled.

For simple live streaming, the standard setup is a mid-range/DSLR camera, a dummy battery, and an HDMI capture card. (I have a Sony ZV-1 and Camlink 4K.)

Asbestos wasn't understood well at first, and its continued use was due to its utility. Practically all countries around the world have successfully regulated/banned it. Removing asbestos is when it becomes most dangerous.

Online banking is dead simple to secure: require a TOTP token. It's something you have (like a debit card) + something you know (PIN/password). The poison being introduced is through a third party, and entirely unrelated to the original function.

You don't solve "abusive technology" by getting rid of the technology, you solve it by addressing the abuse.

Let's start with HW secure element and boot chain verification (IMHO the minimum bar that's met by e.g. a 2013 Thinkpad), then work thru the rest of GrapheneOS' checklist.

Also: Android. If I didn't need Android/iOS apps, I'd be using a Nokia 3210.

> the issue is with your bank then.

What next? Change jobs? Ride a bicycle? Move to a less oppressive country? Buy land and grow your own food? We're trapped into a vortex of *multi-trillion-dollar* global corporations wedging themselves into our lives and clawing away our freedoms, and we're supposed to *vote with our money*?

> the fact mobile hardware is restrictive is a problem caused by hardware manufacturers.

The manufacturers are complicit because the alternative is to be ousted. Every "alternative mobile OS" that tried to directly compete with Android/iOS got crushed.

Google now wants indisputable dictatorship over Android. Manufacturers can try creating alt-stores but you know exactly where it ends.

We have different definitions of an "escape hatch". A user is not an IT specialist. Ordinary people need unobstructed access to lifeboats.

Apple allows developers to self-sign a handful of apps (exclusively from source!) with short-lived certs - it's a complete PITA to maintain a simple app for personal use, and you still need an account. Google is heading in the same direction.

> Anyway, this whole library should have been a copy-pastable snippet for a dialog or toast

People under-value copy-pasting. I'd rather copy/vendor a thousand lines of code (with license+credit intact) than add it as a dependency.

I'm working on a side project, and needed a CPIO library for Go. CPIO is a fixed thing, a good implementation is "done". U-root[1] has a really decent implementation, so I've vendored 2500+ lines of code, as otherwise I'd have to (indirectly) depend on almost 700.000. Great value.

[1]: https://github.com/u-root/u-root

I won't leave my home to type a 20-digit IBAN into an ATM.

I won't travel to another city or country just to hand money in cash.

I won't travel to a branch to... I have never ever written or received a cheque, what the heck even is a cheque? A piece of paper someone can photo-copy?

I used to be able to do all of my banking from a web browser, from any browser/OS I liked. I've had a fob that displayed a 6-digit code rotating every 30s. This used to be simple and secure. What you propose is ludditism.

GrapheneOS developers are free to set their bar wherever they like it. It's an independent, non-profit foundation, driven by community contributions. They provide a web-based, hands-free installer. They offer their work for free, and owe nothing to anyone.

Personally, I wish there was an open/libre device on the market that GrapheneOS could target.

Viability of second-hand still drives market demand, as people have an incentive to buy devices that have resale value. The counter-argument is that otherwise this device will become e-waste. This is still a conundrum, but "don't give your money to Google" remains the active topic here, so...

> People expect products and services to have internet connected features. Accounts are what allow that to happen.

Why do I need an Apple/Google account to use my bank account? I can't even log in from a web browser anymore.

That's my entire original point.

A single, powerful box (or a couple, for redundancy) may still be the right choice, depending on your product / service. Renting is arguably the most approachable option: you're outsourcing the most tedious parts + you can upgrade to a newer generation whenever it becomes operationally viable. You can add bucket storage or CDN without dramatically altering your architecture.

Early Google rejected big iron and built fault tolerance on top of commodity hardware. WhatsApp used to run their global operation employing only 50 engineering staff. Facebook ran on Apache+PHP (they even served index.php as plain text on one occasion). You can build enormous value through simple means.

It's not the devices, or the operating systems. RMS didn't see TiVo coming, but TiVo was never the problem: by the time GPL3 was ready, the industry (e.g. AOSP) has mostly moved to MIT/BSD. In the end, none of this mattered.

The real problem is that @gmail.com or @icloud.com are now required to participate in society. I'm happy to use an iPhone, it's in my subjective opinion the best device on the market. My concern is that I need an iCloud account to talk to my bank. It's become nearly as powerful as my ID card.

I was one month into my first full-time job, when I've (unknowingly of his rank) challenged the CTO in a technical discussion - in a public email exchange. Regardless of the outcome - I've been treated like an equal. This one short exchange has influenced not only the rest of my career, but my entire worldview.

Apple: m68k -> PowerPC (32), OS 9 -> OS X, PowerPC (32, 64) -> x86 (32, 64) -> Arm. They've dragged giants like Adobe (kicking and screaming) thru most stages.

Windows NT has always been portable, but didn't provide any serious compat with Windows 4.x until 5.0. At that time, AMD released their 64-bit extension to x86. Intel wanted to build their own, Microsoft went "haha no". By that time they've been dictating the CPU architecture.

I guess at that point there was very little reason to switch. Intel's Core happened; Apple even went to Intel to ask for a CPU for what would become the iPhone - but Intel wasn't interested.

Perhaps I'm oversimplifying, but I think it's complacency. Apple remained agile.