thrwawy283

I think we're getting faster at iterating on bringing up dev environments.

Silverblue is my favorite, but it's becoming common for me to develop everything within a docker image. As quickly as we're committing to a project, we're updating the env and rebuilding that image at the same time? I'm new to this.

I have a friend who's really big into k8s and ansible. Right now it feels like I'm toying around in 1 pod. He can bring up a set of services around the thing he's developing in a few minutes. I want his power. :x

This is something I'd be comfortable with maintaining, but I couldn't leave with my dad. (I work on the road for months, and it's unpredictable when I come back. Longest I was out was 2 years..)

The GPU is important.. afaik the keyboard and trackpad aren't working yet in Asahi Linux.

I don't expect Apple to support the Linux community. It feels like this is trending in one direction. It felt like things stopped being "favorable" to us when they stopped supporting OpenGL and made no effort with Vulkan. The touchbar and some wifi chipsets were poorly supported for years before M1 debuted.

What has been implemented in Asahi is impressive, but it's not ready to be a daily driver. I hope this becomes my mobile device of choice someday. I'd use MacOS for work-work, and I would choose Linux every time for fun-work. So tired of everything having telemetry and vendor lock-in and basic pieces of software moving to subscription models.

Apple laptops became the dev machines of choice because they embraced the OSS community in pretty big ways. Right now the water feels tepid.

1) We are now 2 generations in without full Linux support for the hardware.

2) I used my Powerbook G4 for 12 years; this is mostly because I was a kid with no money. When I got something else, my sister used it for another 3 years. With thermals as they are in this device, instead of selling it after 4-5 years I'd rather keep it for one-off projects as a server like a Mac Mini. I know laptops aren't designed for server work, but I love that it's a server with a builtin terminal. Also a device I'd use for hiking, because I could charge it from backpack solar.

3) The worst experience I've had was telling my dad MacOS Catalina couldn't be installed on his $3.5k iMac.

As you said in your other comment, M3 will be 3nm TSMC? Maybe Linux will look good on Apple Silicon then.

I want it, but I'm not getting it until all the hardware works under Linux.

MacOS is great, but someday they will stop updating MacOS for this. I see it as a device with incredible longevity because of the fanless thermals and how it sips energy. I could see myself using it 10 years later. I want to run MacOS or Fedora Silverblue, with Silverblue being my true love. Immutable OS images <3

HAH, downvoted in seconds. Would not be surprised if there are Apple employees brigading this.

That's an important distinction, but I think it's getting more common to post for positions you have filled so there are resumes available if someone takes flight. Resumes don't mean someone is available either.

This is the most tactless thing they could have done to bring 2A proponents to the table to discuss responsible legislation. *puts on tinfoil hat*

I'm optimistic growing remote work means extending the working life of an SWE (counter to ageism).

I'm not saying it's a good idea to give everyone a gun, but I do like the argument that the disadvantaged have the same opportunity to pull a trigger. I hope humanity learns wisdom as quickly as we innovate. We've made it this far..

While likely not your situation, a free and open-source app doesn't guarantee it will safeguard your data :x

I'd like to think these ideas have been around and well-known for decades, and we're starting to see the hardware rise up to support them. If you told someone in 2001 that Linux would be rewritten in say, Java, with a garbage collector it would have been dismissed as starkly inefficient. Nowadays we have hardware that makes certain safeties negligible on performance (memory tagging, bounds checking, etc). Nowadays we can make an OS out of Java and it will stand just fine. Or Rust. Faster hardware means we run slower (and safer) software. Not everything gets isolated into an ASIC, of course.

That + the Temporary Containers extension can make every new tab a new container by default. I also use Container Proxy so I can route traffic from each tab through a different proxy if needed (mitmproxy). I've wanted to go to Chrome but Chrome has nothing like per-tab sessions/isolation. I looked at first party isolation but it's vague and doesn't seem like what Firefox provides.

Firefox is the only browser that can do this. Also Google's UI decisions are just unilateral and awful. At least with Firefox we can still tweak some of it.

Still coming to my own conclusion here, but I wouldn't dismiss "easy button" as marketing. We keep hoping for easy buttons and reasonable default settings in things like openssl or pgp. I do like organizations that understand an easy button is the safest default. Is that what we have here?

While likely not your situation, paying for an app doesn't guarantee it will safeguard your data :x

This is my own inexperience with OOP showing: I keep wondering if trait/protocol/interface programming has made inheritance entirely worthless. Love Rust. I've been working in Java for other reasons, but I've been noticing more and more Rust-like things in Java over the past year.

I think hardware attestation is the most dangerous thing in recent history. All they have to do is convince the public that certain programs are like owning a gun or a bazooka that can do real harm, and begin normalizing the idea that only applications signed by microsoft or the gov should be running on your computer for national security. We've already become friendly to companies owning devices on tablets/phones. Then extend that with attestation so virtually every website or service you interact with is confirming you're using unmodified software. I recognize this is tinfoil hat stuff, but the pieces are there to make things painful. At minimum it becomes inconvenient to use any software you like. At worst, the content we enjoy begins owning us. </tinfoil>

I think most of the country gets a bad rep for regulation and initiatives and laws that don't have positive effect. My parents are in the crowd that want to leave California because of supposed "over-regulation". I was telling my father this morning about the Songs-Beverly Act (California's Lemon Law) which essentially protects products you buy for 7 years, giving you the right to repair them. I just sued my employer for not allowing me 2nd meal breaks (every 5 hours), rest breaks, and for not giving me proper/reasonable notice when making schedule changes. These are all things I didn't have when I worked a federal job. Btw, it's hilarious that you're not entitled to (just 1!) meal break at the federal level. California has excess regulation, but I wish people weighed it against the positives instead of assuming it's all red tape.

I think California deserves more praise for how it protects people. Net Neutrality is effectively upheld by state-by-state regulation, because the federal gov couldn't secure this for us.

I try to humble myself there. While 32GB seems ridiculous for 2005, I've absolutely enjoyed having 64GB in 2022. Yes, apps have gotten more bloated. Yes there are many wasteful framework-built heavy apps. In many cases they are doing much more than in 2005.

I wonder if anyone is tracking the average amount of memory somewhere from 1980-forward. I'd take that value * 2 for a developer.

My understanding is that Intel boosts power so a single core can chew through a program quicker than being inherently energy efficient (performance-per-watt). Most of the time this benefits Intel, because a core will boost-up-then-down quicker over completion of the program than a more efficient AMD processor core.

(This is my layman understanding.)

I solidly believe AMD is the king for efficiency, but I wish I could find better benchmarks showing idle power use for AMD vs Intel (not peak power use). My understanding is that Intel has deeper power states its processors settle into.

I'm surprised someone doesn't artificially limit or undervolt their Intel proc to approach or surpass what we're seeing from AMD. Would it significantly lengthen the total execution time of the program? Would there still be a significant difference in "performance"?

Performance-per-watt is important, and the total amount of power to execute the program. Would this be in Kilowatt/hours?

I want to see how many Kilowatt/hours something like Cinebench consumes on similar AMD & Intel processors, so we can derive the "real computing efficiency".

I don't know why you're being downvoted. While I would commonly understand people want to avoid taxes, I think there's a real question in there:

Why don't people want to contribute back to an environment that enabled their success?

I have too many family members that will avoid acknowledging anything the government does for them.

"The social contract."

I just want to run a VM on an iPad. I don't consider the iPad or Apple to be the best device to protect my privacy, but it is the most comfortable tablet/portable experience. What I read last, was that Google is adding it's "Tensor" processors to its Nexus phones, and Android 13 will allow running full VMs as "apps" within Android. That's what we want to happen on the Apple side.

I don't like Android or iOS fully. I want the freedom of a full Linux distro. I think we're going to transition to that over many years. The next step is Linux (Android) running Linux in a VM. Once Apple sees that expect a competing technology on the iPad. Then we'll have real tablet development machines. It will take years more for Linux desktop environments to gain a tablet experience as comfortable as iOS. :\

I was mostly remarking on the lack of (significant) innovation. I'd say the best feature here has been airplay/miracast for drive-by screen sharing from my sister/mom/dad on their iPads or phones.

For what we pay, I wish they'd consider some of these niches rather than startup ads or an AI that looks at what you're viewing and suggests other channels mid-viewing. Creeeeeepy.

The thing that really bothers me are all these "AI" technologies in my new LG oled tv that don't actually get used. All of them to reduce "crushed" images, correct color, improve blurriness in motion scenes, etc are only used when not using HDMI/PC hookup. My feeling is they're used when viewing things streamed on Netflix/Hulu/HBOMax/etc. But I spend most of my time using the TV as a computer monitor. I'm in a niche group, but this was my only option for large-format oled.

I wish they did more with HDMI, even though HDMI is being phased out. I want to hook up my computer and have the computer gain an ethernet link from the TV, even though the TV is wirelessly connected to my router. The TV should be a "dock" that includes a 2nd monitor. I have a wireless controller reciever plugged into the USB port on the TV. I want that to give input to my computer from it. I also want my computer to charge while being hooked up to my TV. I think the only answer is this TV should have type-c and do all those things as a dock, but it's frustrating that we're 1 step behind.

I wish I could watch a program, while showing my computer hdmi input picture-in-picture. Hell this thing has 4 HDMI inputs on the back, let me do each input to a quarter of the screen. Another niche use..

Don't even get me started on the ads. Doesn't make sense to keep the TV on when my computer is locked for long periods of time, so I generally come back, turn on the TV, and unlock the computer. First thing I see is fucking ads. Takes 2 clicks on the remote to dismiss but it colors my experience that they're always pushing another $30 or $40/month service when I first see things onscreen. They know you can't get this quality elsewhere so they're happy to push you ads. And the telemetry, my god. It's my $3k TV!

Most TVs are still waiting to support the next hdmi standard so you can do 120fps & hdr simultaneously.

The 2 things that did impress me were this LG tv supporting both Miracast and Airplay with /no/ hoops to jump through. It just worked. I do wish I could "cast" things to the TV, and that's like pushing a link to it where the TV navigates to that stream and plays. No other device has to stream or push the video to the TV, it does it itself in the Chomecast paradigm. That would be nice.

</rambling>

You mentioned things I hadn't thought of. Google's Search accomplishes the goals of 10 years ago, but steps no further than that. It treats its power users like kids, and offers no complex filtering to do things like removing search results that require logins. Librarians love when you come to them to specifically refine your search. Google still has the most useful search, but they've taken away methods to get better results. I remember I was pretty upset when i couldn't search for images by exact dimensions anymore. Bing allows this.

Google's product direction has been inching backwards for a decade.

That is definitely close to what I'm thinking of. What I was trying to say is our 1st reach should be for a graph editor. Behind the scenes it may break down to a command line program, but what you'd deal with all the time is the graph editor that you can mouse around and zoom in/out on, and add/remove edges between nodes/packages. That would be a neat experience.

This is an aside, but I've been watching all the audio graph editor projects pop up and it's made me realize I want something like this for package management. A package maintainer should be able to specify something is a dev dependency, runtime dependency. I should be able to form a "user dependency" where I say Xwayland depends on sway - even though it doesn't. So when I remove sway it pulls out Xwayland with it. I should be able to remove dependencies within package descriptions, and see that one used to exist but has been overridden by me. I'd love to even see the history of how that graph has changed over time (step through package adds/removes). Further filter by package source to visualize the network clouds.

I want to visualize all of these dependencies as a graph, instead of trying to hamstring things together with CLI invocations. Editing the graph would invoke the package manager.

Now I'm annoyed that in 2022 I'm understanding Google+ better. I want to follow the /intersect/ of "Jane Smith" and "Catan". Both of these are topics. I might be broadly interested in "computers", and "board games", or interested in just my friend "Jane Smith". I likely don't want to know everything Jane is posting - or those posting things about her - but when I choose to Follow her, I should be prompted with a list of topics I've subscribed to that would narrow the posts I see which intersect with Jane. Also - G+ or Twitter - would be working to autotag posts into topics/subjects to help me identify how I want to intersect with those posts. (If the author themselves don't apply tags)

I can see why limiting what you're presented with never took off from a marketing point of view. I think this is rad though.

This is an interesting idea: I wish you could Follow a combination of a person and the #topic they post to. You could choose to follow "all of" a person, but more realistically the default behavior would be to attach to posts from a particular topic that person is generating. This would be similar to your example of playing a game of Catan with someone and enjoying just that limited bandwidth you have with them. Knowing and seeing all of who they are is not common in our business relationships, or friendships.. I think you're right, and I appreciate how you articulated it.

Someone below this mentioned this was how Google+ was intended to work.

It occurs to me you could do something like this by creating a dummy user and enabling a user service on login (systemd). I don't think this is the best method, because you can't predict all the ways to login without a shell. Just a neat idea for generalizing it.

Hmm. Sometimes I step back and I look at the version history in modern Office documents and it does feel convenient. I like the CRDT editing a lot. It still feels too loose and prone to issues, but I recognize its value to the average person.

Most of what bothers me is trying to view, read-only a previous version. Usually I have several Word or Excel documents up on my screen, and it's easy to begin modifying a past version that then becomes the latest. It's not reflex to open a previous version and untoggle autosave. It bites me all the time.. In Office's pursuit to help people avoid losing data, I wind up overwriting what I need to preserve. A lot..

I have not seen manual version names in Word doc history. Maybe I'm sheltered :s