partial_accumen

The dual boot is the default install. The installer is a single terminal command in OSX with the installer being the guided setup. The installer is right on the front page of the distro web site: https://asahilinux.org/

It is literally just:

curl https://alx.sh/ | sh

The biggest decisions you have to make are how you want to partition the SSD between OSX and Linux.

I’ve been installing Linux in various ways since the late 90s using Slackware, and the Asahi installation experience was the easiest and seamless installation of Linux I’ve ever experienced. It on only occurred to me later why the installer could be so good. Asahi only runs on M1/M2 hardware. The developers knew exactly what the hardware would be and could tailor the experience around it.

I wouldn’t really recommend Asahi if you only have 8GB of RAM and a 256GB SSD in your Mac. It will certainly run, but is cramped in daily use.

A quick question before research: is it fully working by now?

Is every hardware function in the laptop that works in OSX available in Asahi? No.

I’d like to have a functioning display

I think you’re asking about “DisplayPort Alt Mode” which is where you can plug a dongle into one of the USB-C ports and output the local GPU to DP or HDMI. The answer to that is “yes, depending on how adventurous you are”. There’s an experimental kernel that does support it today. I don’t think its in the main branch yet. I intentionally run version 43 (1 behind the current 44). However, I use a USB-C DisplayLink HDMI adapter for an external display and it does most of what I want right now without the experimental kernel. I do want “DisplayPort Alt Mode”, and will use it when its available though.

If you have an M1 or M2 Macbook Pro with HDMI port built-in, those work right now. The challenge being worked through is a display port that gets unplugged, which only happens on the USB-C port Display Port.

I’ve got an impression that Linux can work on Apple Silicon, if you’re ready to abandon some things here and there.

I wouldn’t use the word “abandon” but rather “wait for”. Power management efficiency doesn’t come close to native Apple OSX, but under Asahi it has enough battery for my needs. I only charge to 80% (supported natively in Fedora KDE) and get about 3 hours of runtime on battery for light to moderate use. I also read that this has improved a chunk in version 44, but again, I’m not running that version yet.

Another piece of hardware not supported on Asahi yet is the MLX engine. I’ve been experimenting with running local LLMs, and they do run under Asahi Linux, but the hardware includes MLX in OSX. There are some models specifically made to utilize MLX which result in significant performance improvements in inferencing speeds. The unified memory of the Macbooks means system RAM is available for LLM use, so I can run 16GB models while still having 8GB of RAM left over for other applications and OS functions on this Macbook Air. The RAM footprint for LLM works in both Asahi and OSX.

Keep in mind, this is a dual boot system. I still have OSX available if I need one of those Apple OSX specific function or extended battery life only one reboot away.

My daily driver is an M2 Macbook air running Asahi Linux. There are certainly some hardware parts I wished worked better right now, but its fully for my needs usable as is. Improvements are occurring regularly by the development team. Apple hardware really is solid, and I’m very happy that in the rare cases I do have to use a commercial OS (Netflix streaming for example), I don’t have to use Windows. Its a dual boot machine (Linux/OSX).

Overall I’m pretty happy with Linux on this M2. Theres a handful of us here on Lemmy running it. You can find us at [email protected]

I’ll just leave this here:

I don’t use Brave, but one correction. From the article: "Brave sells Origin to strip added features—a $60 one-time fee (free on Linux). "

Brave thinks its Windows and OSX users are suckers, not its Linux users apparently.

DenverCoder9 strikes again?

So he’s saying [paraphrased] “It wasn’t broken, but I decided to ‘fix’ it anyway”

Whats really funny is that the interface is closer to MS Office 2003 which was Office pre “ribbon” interface. So the current version of MS office is actually the weird one that changed.

I’m confused on one main point from the author. Are these created projects for work or just personal? This answer changes the dynamic of the entire article.

Except for the SaaS, almost none of this is useful and I don’t want to maintain any of it.

If they were personal projects, then there’s nothing wrong. They were useful for a moment, or were fun to build, and if they’ve exceeded their usefulness, get rid of them. We do this all the time with hobbies, so why would it be a sin with personal code? Nobody spends an hour finishing a crossword puzzle and says “well that was a waste of time”. We spend money on hobbies too, so if your hobby is coding and you want to spend money on an LLM subscription for your hobby, as long as you get value and enjoyment out of it, its fine.

However, if these were supposed to be commercial marketable products, and that business resources were used they yes, clearly there is a lack of planning and resource allocation. Spending time and money building something which has no use can can’t be maintained is a major business error.

It takes more than a name to deliver the result. MS Office365 also claims to be great for collaboration, and its a trainwreck to try to use it in that way.

You’re proving my point. They just started working on it 3 months ago. It will be a long time, if ever, before it is a Gsuite replacement. They even said they tried this a few years ago and abandoned it. There’s a risk they’d do it again.

So if there’s another open source effort specifically focused on an online product, its a good thing.

I like and use Libreoffice, but its a local application. This makes it fine for working by yourself and on your own hardware. Google Docs (Gstuite) is an online application with the best collaboration integration I’ve seen and works on nearly any device you can open a modern browser on. Yes, both products have a word processor and a spreadsheet, but they don’t serve all the same use cases.

If this new Euro-office can replicate the Gsuite offerings (that Libreoffice lacks), then Euro-office could be a great addition to the open source community.

Corporations to workers: “Use AI for everything you do!”

Workers use AI.

Corporations to workers: “Not like that!!”

You’re going to be even more angry to learn that your apartment neighbor is using the shared building power to run an industrial aluminum smelter on his balcony as his side hustle. It does explain why he’s posting all of those pallets of 6061 alloy bar stock and ingots for sale on Nextdoor and Facebook Marketplace though.

The actual paper (PDF) this is based on gives much better information than the article. From that we get some really key information:

To allow FROST to measure SSD contention, the victim must perform activities that result in storage accesses to the same disk as the file used for contention measurement

This can’t ready your SSD. It can only listen in on the conversation between your CPU and SSD when something else reads it or writes to it. The whole FROST approach has a number of clever tricks to generate reads from open applications though. Further, it requires the attacker’s code to be running in an active browser session.

Also, If you have two SSDs, and your browser is on one, this FROST approach can’t see anything written to or read from on the other SSD.

Lastly, there’s a mention in the paper about hardware based SSD encryption being vulnerable, there’s no mention of Software Whole Disk Encryption. Given how the researchers are using the SSD timing exploit, I would guess that a software (not hardware) whole disk encryption might be immune to this attack because the patterns of timings would be different with encrypted data being written to the SSD (instead of the data being encrypted by the SSD when written.

something something tAx MoNeY

Nice fake quote

Idk maybe tax the fucking billionaires? Maybe instead of useless foreign wars spend it on schools? Tax the NFL? I’m just spit balling.

Get out of here with that lazy answer. Saying [paraphrased] “Have someone else pay for it somehow” is the most throwaway answer to any cited problem. If you want to contribute to the conversation looking for a solution, then contribute. What you posted here isn’t a useful or meaningful contribution.

I’m not opposed to taxing fucking billionaires, but how are you proposing? Increase on capital gains taxes? Higher taxes on non-resident properties? Wealth tax on unrealized gains? That’s a great conversation to have, but its pretty far away from school spending which is what we’re talking about here.

I’m not opposed to lowering defense spending, but in the USA you might as well try to boil the ocean than try to reform the Military Industrial Complex today. Short of revolution, that’s not happening in our lifetime. If we have revolution, school taxation will be the last of our worries.

They seem to think that what I mean is that we shouldn’t use tax money at all. No.

Incorrect. That’s not what I think. I’m seeing if you’ve thought through the consequences of your proposed change and have a solution I can’t see.

I’m happy to pay for the same education for all children. A standardized education.

Okay, so right now many schools are paid for by additional local income/sale/property taxes (these would be over and above any state or federal tax money that all schools would get). Those local taxes are passed by vote of the people in those localities. Under the current system, if “City A” has voters choose to tax themselves at a higher rate for better schools, then they get better schools. If “City B” has voters shoot down their tax increases, their schools pay the price and decline.

I’m happy to pay for the same education for all children. A standardized education.

I’m not trying to strawman you here. I’m trying to understand your proposal. Are you proposing to take the additional tax money generated in City A that they put on themselves and give a portion of it to City B to create your standardized education?

Here is study referenced in case interested - https://www.sciencedirect.com/science/article/abs/pii/S0967070X2600051X

First, I want to thank you for taking the time to engage on this topic, and also finding this great research paper. There’s always room for criticism in a source, but overall this is a great analysis done by the paper’s authors within the scope they define. I only have access to the abstract through the link, but may see if I can find the full paper from my library.

Pros of the paper:

• the WTW (Well-to-Well) metric the authors use does encompass nearly all of the energy/emissions from using the stated fossil fuel for transpiration. This is a great way to have an apples-to-apples comparison of the various sources of energy with regard to their pollution cost in vehicles of various types which is our core question.
• For Diesel the authors factored in emissions including not only CO2, but also Particulate Matter and oxides of Nitrogen which is much worse from diesel emissions than other forms of petroleum compared to gasoline or methane combustion. I appreciate this level of detail from the authors. However, in just the abstract I can’t see the formula for how they define a percentage of pollution on these other to inputs.

Cons of the paper:

• the WTW (Well-to-Well) metric starts at the well. There’s no accounting for the exploration pollution associated with fossil fuels (or battery materials for that matter). As an example, the Deepwater Horizon oil disaster was the largest oil spill as well as the largest release of methane in history. Deepwater Horizon was not an active production well. It was an exploration well (which is the step before production). Therefore it wouldn’t be included in the WTW metric, yet represents a huge amount of pollution that would not exist if we weren’t using oil and methane.

Also, to our question about coal derived power for EVs, we may have enough information from the authors to extrapolate the coal figure. Since the paper includes detailed analysis of methane generated power, we can likely get the efficiency and emissions numbers for that power source. This will let us use the author’s methods for defining the percentage of efficiency for comparrison once we get the coal inputs. We can likely get the coal inputs from looking at an existing coal power plant and getting its ingested coal, CO2 cost for extraction of that amount of coal, then the published emissions numbers from the plant for the KWh of electrical energy generated over a set period.

Overall, this paper, and your read of it is a fantastic contribution to the conversation. Thank you!

Last I read the only time ICE beat EV in emissions was when electric came from coal fired plants.

I’d be interested in your source on that as it contradicts my understanding.

For the ICE vehicle did your source include all of the environmental costs associated with producing the gasoline or did it just consider the tailpipe emissions. Or worse, did it include all the Coal costs as a environmental burden on the EV, but exclude all of the petroleum value chain environmental burden (besides the tailipe emissions)?

And then the battery production/disposal of course…

I’d guess the entire environmental damage of petroleum exploration, extraction, refinement, distribution, and combustion is greater than the entire environmental damage from battery material exploration, extraction, refinement, manufacturing, and eventual battery disposal when talking about a single ICE car vs a single BEV.

All of the ICE vs BEV pollution metrics I have seen to-date include the environmental cost of the battery, but only include ICE tail pipe emissions and exclude the environmental cost of everything needed to bring petroleum to market.