Resistance to AI Automation will crumble

And I don't mean The Resistance to Terminators.

Al automation promises to revolutionize white-collar or office work. Some think there’s plenty of resistance against quick AI adoption and so there won’t be major negative societal disruptions. I’ve come to believe that resistance will crumble.

Legal and copyright issues could slow AI down, but not for long.

AI generated code may be hard to accept into proprietary code bases due to legal uncertainty, e.g., copyright issues.  For example, the Google v Oracle Java lawsuit started in 2010 and reached conclusion in the Supreme Court in 2021.  That’s 11 years of legal uncertainty over whether there was copied code.

Having said that, Google kept using the disputed technology through those years of litigation.  There are ways around legal issues like that I presume.  And the legal cases against AI are already underway as in the Github Copilot lawsuit [1], so the clock is already ticking — and currently, the AI side is winning [2].

It feels there's some amount of inevitability at play too in terms of resolving the copyright issues.  The actors' guild has agreed to a contract allowing limited AI usage with major media companies. OpenAI is securing deals with media companies to train and utilize their copyrighted materials. The writing is on the wall, and AI is writing it.

People and companies are resistant to change, but new folks will jump ahead

Some say "The future is already here; it's just unevenly distributed" (William Gibson).

One reason that happens is that status quo lowers risk.  If you're Apple and if you have AI that can do the work of half your software development workforce, do you layoff 50% of SDEs?  No.  No one has the risk tolerance for that big a move all at once.  What if you were wrong?  What if the wrong automation tech is used, or the people you hire to manage that get it wrong?  Betting "all in" on a single roll of the dice is too risky.

Not without an external crisis, at least.

The oil and gas industry used to employ a lot of people.  Until oil price crashed and belt tightening resulted in lots of layoffs and automation.  Those jobs are never coming back because they were automated away [11].  But without the price crash, the amount or severity of layoffs might not have been as bad.

After all, one way for a manager to get promoted is by managing more people and a bigger budget.  So the status quo of using more people than automation is a self-interested move.  A crisis like a big oil price crash is required to shake that status quo loose.

But new companies don't have the same status quo problem.  Old companies will either succeed or fail.  New startups with come in with automation and AI baked in to reduce cost, extend their funding runway, and to get product market fit faster.

Old Tech will eventually come around to automation anyway

People say that it's normal for FAANG type companies to lose 5% or more tech workers every year for various reasons [3].  Attrition, eliminating teams of projects that get cancelled, "unregretted attrition", etc.  Depending on the speed of AI automation progress, old tech companies can just keep up with 5% unregretted attrition each year to slowly turnover the company to AI.

Some might say it's already started since 2022 [4].

And for those old stuffy companies who don't come around fast enough to automation?  They can always turn from being an engineering company into a financial engineering firm, like Siemens, IBM or GE.

I’m kidding! Slightly.  But just because Air Canada will probably think twice about using AI for customer service after that lawsuit [8], that doesn’t mean they won’t outsource to an AI customer service startup that’s also willing to defend them against lawsuits.

Lump of Labour fallacy is about the whole economy, not a single sector

Some say don't worry about AI automation producing negative societal disruptions in the form of making everyone unemployed because Lump of Labour is a fallacy [6].

The Lump of Labour fallacy misconception is that there's only a fixed amount of work to be done, so don't worry about automation taking work away from people!  There'll always be more work invented by people to do!

Some would say this time is different because AI could do that newly invented work too.  But even without this AI-everywhere angle, you have to see that the type of work may be different.

When manufacturing and software development work gets automated, what new work is there to do?  Those workers can re-skill into making TikTok influencing channels?

Some say “yes, of course”, but there's just no guarantee that the new work would be "better" (e.g. higher abstraction of code, safer workplace, higher paying, improved quality of life at work).  It could be worse or more dangerous.  As climate change gets worse, maybe those displaced tech workers can re-skill into forest fire fighting?  I doubt non-robotic LLMs can do that.

Jevons Paradox applies to things people want, not things people are ambivalent about or worse

Some say don't worry about AI automation making office/tech/programming workers unemployed because Jevons says greater efficiency will induce more demand for software, etc. [6]

Jevons Paradox states that increased efficiency leads to increased consumption.  i.e. as X becomes more efficient, more X will be used. Substitute X with:

• Oil
• Gas
• Electricity
• Microsoft Office

Look, people want dogs and cats.  But generally speaking, people aren't into wanting software developers and assembly line workers.

USA is manufacturing more than ever, but is employing fewer manufacturing workers than ever. Maybe AI will let us substitute "manufacturing" with "software developing" in that sentence.

Consumers want the goods, not the people producing them (see: offshoring).  If AI can produce it cheaper with fewer humans involved and without animal testing, then consumers would want that instead (mainly because it's cheaper).

So at best, Jevons says people will want more software and apps than ever before because they'll get more efficient with AI.  But nothing says that the code has to be written by humans.

In fact, we've seen this last point before.  Practically no one writes the assembly code that those software requires written to work — compilers write that automatically for us since decades ago.

Fortunately, assembly code programmers could easily up-skill to higher level languages like Java.  With AI, coders may have to up-skill to higher level languages like English, and compete with all the non-coders who can already English better than them.

Software developers will work at higher levels of abstractions.  Yes, and so can everyone else

Some say software developers shouldn't worry about automation taking their jobs, because at best AI can spit out code given some architecture and specifications for problems that the human software developer has to write up into prompts.  So the human SDE can now work at a higher level of abstraction, work at system design levels and above, double check the code that AI produced to make sure there's no hallucinations.

Sure, yes.  But what evidence is there that BSc graduates with education in algorithms and data structures can write those prompts better than, e.g., BA philosophy graduates who trained in close-reading, analyzing, and writing highly technical English in all their Analytic Philosophy classes?

There's also a lot of LLB lawyers who are underemployed in the legal or adjacent fields doing paralegal or repetitive property conveyancing work.  Maybe some of them can do this AI software development prompting better than CS/SDE graduates?

In fact, we've seen this broadening (or democratizing) of a labour field before.  It used to be highly technical and challenging work to do special event photography.  You have to get the lighting, shutter speed, and aperture just right or else you'll miss the moment, or waste and run out of expensive film.  Digital SLR photography and big inexpensive memory cards means many with an eye for beautiful photos can now do wedding photography, take 1000s of photos, then choose the best 50 afterwards.  Or just fix it in "post" with Photoshop.

This means sky-high compensations or job security will come down, if not in one field (tech), then maybe in any field that's at risk of AI automation (all white collar or office jobs).

AI is a bubble, until it’s not

The last few points focused more on software development, but that’s a canary in the coal mine.  Some say software development is more resistant to AI automation because it’s already in the business of automation [7].  But that just means that if (since?) software engineering labour is at risk, then many other fields are also at risk too.

Some say AI is like crypto.  It’s all hype and a bubble run like or by some of the same personalities.  But that’s a useless comparison.  Can you, in 1999, tell if the internet is a bubble or not based on how tulips were a bubble [9]?  They’re just not the same.  And at least with AI there are clear and present use-cases, no AGI required.

So forget AGI.  Resistance to automation will crumble.  There's too much money to be made in automating even just 10% [10] of the white-collar or office work labour market, in whole or in part, outright or by “just” improving human efficiency.  That's the "killer app" of AI.

Don’t focus so much on the automating a worker outright-in-whole part.  Instead, focus on the replacement in-part, by “just” improving human efficiency, part.  E.g. Self-checkout didn’t eliminate “cashiers”, but allows one supervisor to do the work of (say) three cashiers.

How much profit can this “killer app" make that would justify the (non) bubble?  I lazily checked with Meta AI and it says:

• telemarketing and public opinion researchers (including call center representatives) in the USA in 2020 probably had compensation of about $100B
• software developers (including applications and systems software developers) employed in the USA in 2020 made about $200B
• office managers, supervisors, support and assistants employed in the USA in 2020 made about $200B

That’s $500B right there. So by automating even just 10%, it’d take just 10 years to fill the $500 billion dollar revenue gap [5].  That percentage will go up, the addressable market is not just the USA, and there are many other labour fields to automate than those.

That’s why I think AI adoption will be quicker and there’ll be more negative societal disruptions coming up.  We can’t all re-skill to fight forest fires.

EDIT: To clarify, I don't mean for sure tech or any other jobs will definitely get decimated by AI efficiency gains.  It's just that quoting "Jevons" or "Lump of Labour" isn't a knock-out against that possibility.  The analysis has to be deeper than just invoking those, by now, thought-stopping thoughts.

Let me volunteer one possibility, for example, AI makes coding jobs 10% more efficient, meaning cost of labour is reduced.  Because cost of software production goes down, consumers want more apps, so much more than the 10% efficiency gains allow that more humans developers are needed.  This is the classic rebound effect in Jevons.  Notice it requires consumers to want so much more apps that it more than offsets the 10% efficiency gains — there is no guarantee it rebounds that much!

EDIT 2: Also, jobs getting decimated by AI efficiency gains is not necessarily bad!  Machines decimated coal mining jobs, which is bad for jobs, but given there are better alternate jobs, it was great for the health and safety of would-be coal miners!


[1]: https://www.artificialintelligence-news.com/news/openai-and-microsoft-lawsuit-github-copilot/

[2]: https://www.developer-tech.com/news/judge-dismisses-majority-github-copilot-copyright-claims/

[3]: https://www.seattletimes.com/business/amazon/internal-amazon-documents-shed-light-on-how-company-pressures-out-6-of-office-workers/

[4]: https://layoffs.fyi

[5]: https://blog.carsoncheng.ca/2024/07/re-500b-ai-revenue-expectations-gap.html

[6]: AI and the automation of work. https://www.ben-evans.com/benedictevans/2023/7/2/working-with-ai

[7]: I believe Yann LeCun said something like that but I can’t find the source.

[8]: https://www.cbc.ca/news/canada/british-columbia/air-canada-chatbot-lawsuit-1.7116416

[9]: https://en.m.wikipedia.org/wiki/Tulip_mania

[10]: https://news.ycombinator.com/item?id=41465081 - study says there's a 26.08% increase in productivity with AI, and more specifically a 27% to 39% for junior level and 8% to 13% at senior level.  So my "10%" guess wasn't too bad?

[11]: https://www.parklandinstitute.ca/job_creation_or_job_loss - “the Big Four are leading the push to automate away even more jobs in the coming years … The Alberta oil and gas industry employed 25,788 fewer workers in 2021 than in 2014 (a 15.5% reduction)”

Labour Automation vs Growth

Automation in oil and gas reduced labour needed to extract and produce oil. Oil as a product is not a platform-platform. Oil usage cannot induce demand for ever more oil by being a medium for platforms that can recursively host more platforms of products. Because oil does not induce exponential demand, then automation has a good chance of reducing labour down.

The flip side is that platform-platforms can recursively host more platforms of products. So platform-platforms has a chance to induce exponential demand, and so while automation reduces labour per unit produced, on the whole the exponential demand for units can balance out if not even increase labour required industry-wide.

C compilers are not platform-platforms. Compilers in general basically eliminated all assembly programming labour. The reason there are more programmers now than when assembly language reigned supreme is not because compiler automation freed up labour for more creative things. The reason is more because that other platform-platforms like the internet and operating systems induced exponential demand for software.

Other examples of automation that are not platform-platforms and in fact eliminated a lot of labour include:

• Manufacturing automation
• Bookbinding automation (replaced by printing presses)
• Telephone switchboard automation (replaced by digital systems)

Jevons Paradox states that increased efficiency leads to increased consumption.  i.e. as X becomes more efficient, more X will be used. Substitute X with:

• Oil
• Gas
• Electricity
• Microsoft Office

But software developers? Or assembly line workers? USA is manufacturing more than ever, but is employing fewer manufacturing workers than ever. Maybe AI will allow us to  substitute "manufacturing" with "software developing" in that sentence.

Jevon's Paradox "works" only for items that consumers actually want. If extreme environmentalism "won", gas usage would go down no matter how efficient gas engines get. If coal-powered SUVs became unreasonably ultra trendy, coal use would go up regardless of how inefficient coal was.

Software development and manufacturing workers are not goods that consumers want. They want what's produced. And if AI can produce it cheaper with fewer humans involved and without animal testing, then consumers would want that instead.

The Internet is a Platform-Platform. AI is not.

AI is not like the internet or computer operating system software in terms of its economic impact.

The latter two are platform-platforms while AI is not.

A platform is a kind of medium.  e.g. YouTube is a platform for videos.

Some platforms are special in that they are recursively platforms for platforms.

Computer hardware are platform-platforms.  They're a medium for various single purpose built software, of course, but more importantly they are also a medium for different operating systems (OSs).

OSs like macOS or Windows are platform-platforms.  The most easiest and reductive way to see this is that you can virtualize and run macOS on Windows, and vice versa.  More substantively, these OSs can run internet or web browsers.

Internet and web browsers and the physical internetwork they connect to form platform-platforms.  Different businesses run on the web, but more importantly, the web or internet are a medium for running app stores on desktop and mobile devices.

Is AI a platform-platform?  What is it a medium for?  Does this medium support a platform recursively?

Or is AI more like a sufficiently smart general compiler?  AI can automate existing products and workflows, but it isn't a medium that will enable more products exponentially by carrying platforms on platforms … on platforms.

Others have noted similar ideas.  e.g. AI is a feature, not a product.  Actually, it's worse: "AI is a tech to enable a feature, not a feature in itself" [1].

Or that current AI are automated "infinite interns that can write anything for you" [2].  But follow that train of thought to the conclusion that a human-level current-generation AI would be an actually-human intern and now wonder: since when do human interns (even an infinity of them) form a medium that can be a platform for platforms?  (… unless you use them like transistors, to do computation as a human powered Turing machine, just to argue they can do what actual computers can already do.)

This distinction between products and platforms vs platform-platforms can inform us on the economic impact of AI on society.

 

 [1]: ( https://www.threads.net/@benedictevans/post/C8C00e5OZBT )

 [2]: ( https://www.ben-evans.com/benedictevans/2023/7/2/working-with-ai )

 

Jevons Is a Paradox, Not a Rule

Jevons Paradox says that as technology becomes more efficient, overall resource consumption can increase. This was seen during the Industrial Revolution when more efficient coal engines led to higher coal usage. However, this paradox is not universal, and efficiency can also lead to reduced resource consumption.

In the context of AI coding tools (e.g. GitHub Copilot), there's a belief that increased efficiency will lead to more coding jobs by lowering development costs. While this may happen, history shows that technological advancements can also displace workers.

Counter Examples

The invention of programming compilers made coding more efficient but reduced demand for assembly language programmers, who were once critical to assembly-based software development. While many of those programmers probably found other coding jobs in higher-level languages, Jevons simply doesn't guarantee it.

Similar patterns have occurred in other industries more starkly. The mechanization of agriculture reduced the need for farm labor.  See this graph:

https://ourworldindata.org/grapher/number-of-people-employed-in-agriculture

Then there's the replacement of draft horses, where ICE vehicles meant horses were no longer needed and millions of draft horses were slaughtered or displaced, and their population dwindled.  See this graph:

https://www.researchgate.net/publication/338480301/figure/fig1/AS:845430833283085@1578577826802/Evolution-of-the-horse-population-in-France-from-1800-to-2010-translated-from-French.ppm

In recent years, coal consumption has fallen despite energy efficiency gains due to the shift to other energy sources (e.g. renewables, gas).

The rebound effect, which drives Jevons Paradox, doesn’t always occur at full strength. For example, energy-efficient LED lighting and fuel-efficient cars have reduced overall energy and fuel consumption, despite potentially increasing usage. Similarly, AI tools may lead to fewer coding jobs, even if more code is produced.

Ultimately, while AI could increase software development demand, it may also reduce the need for certain types of programmers. History shows that efficiency gains don’t always lead to more jobs. Jevons didn't guarantee draft horses more jobs, after all.

This was written in collaboration with an AI — another example where more words will be written as efficiency per word increases but the number of writing jobs may well decrease (as it apparently already has: https://www.bbc.com/news/business-65906521 ).

How to use Homebrew on a Multi-user macOS

You'll find descriptions of how to do this, like on StackOverflow: How to use Homebrew on a Multi-user MacOS Sierra Setup.

It's said that using `sudo` is wrong.

It's said that using a per-user local version of brew is right, but...

1. it doesn't play well with `nvm` (see)

2. it is completely and entirely unsupported (see)

3. many packages don't support it (see)

4. many packages will install from source instead of a binary (see)

So the practical, quick and dirty solution is to just use `sudo` (see).

In my experience, if I recall correctly, homebrew on macOS 14 by default uses the group "admin" for where it installs things, and it sets the permissions to "read" and "execute" as needed already.  The only thing missing are "write" permissions.  And for the user to include brew binaries on their PATH.

Also, "Administrator" users on Macs are in group "admin" by default already too.  I'm guessing if your user is using brew, they're probably a macOS "Administrator" too (or else why would you let them use a global brew install?).

So I just ran in Terminal:

$ sudo chmod -R g+w $(brew --prefix)

Then for the user that wants to use brew, put in their home directory's ".zprofile" file:

eval "$(/opt/homebrew/bin/brew shellenv)"

To ensure brew is working for them, run in their Terminal: $ brew doctor

Warning: this is unsupported, and it's said to be wrong, and you're letting all Admin users on that machine to share one single installation of homebrew!

If roommates fight over fridge space, you've got no one but yourself to blame for not buying each roommate their own fridge!

So what use-case does this safely enable?  A single human with multiple macOS user profiles to isolate their work space while sharing (with themselves!) the same global brew install.

You can mount Linux Ext or LUKS disks on Mac, Windows!

This is exciting! I’ve been looking for a disk format that works seamlessly across Mac, Windows, and Linux.  Something like a USB or disk image that mounts on all three desktop platforms.

The obvious choice is exFAT, but what if I want a modern disk format with journaling for data safety?

Ext4 is one option, although it used to be read-only on Macs (see Mounting Ext2/Ext3/Ext4 USB Flash Drives on Mac: Read Only Success). Maybe things have gotten better in the last 10 years?

And what if I want disk encryption?

VeraCrypt with an exFAT disk works: see Sharing TrueCrypt USB volume on 3 platforms: Mac, Windows, Linux.

But what if you want both encryption and a modern file system? I couldn't find a solution to this, until I found… 

 

linsk

In summary, linsk is an easy way to run an Alpine Linux VM using qemu to mount a disk within it, and then share that disk back out to your local host machine via a networking protocol.

So now you can mount an Ext4 disk, or a LUKS encrypted Ext4 disk, on your Mac or Windows machine!

The usage notes are so good, you can just look there for how to install and use.

Here's a quick use reference, more for my own reference than anything

How to use linsk to mount a Ext4 disk within a LUKS encrypted disk on macOS

Note: If you're not me, please read the linsk documentation first and understand it completely before proceeding below!  This assumes you installed linsk correctly, including qemu!


1. With the disk plugged into your Mac, make sure to unmount all volumes that macOS auto-mounts!!!

Warning Danger Caution Danger: If you don't unmount all volumes first, the following may delete, nuke, and destroy all your disk data.  macOS likes to repeatedly auto-mount any volume it sees, so after every step below, make sure to unmount those volumes again!!!  You'll see warnings to this effect from linsk in the terminal as well.


2. In Terminal, run: $ diskutil list

Find your disk and note its path: e.g. `/dev/diskX`

The value of `X` in `diskX` may change every time your Mac encounter the disk!

However, the LVM group/volume/luks-container names within that disk should be stable (unless intentionally changed), so in the future, you can skip down to the `linsk run` step 5 below if you're just re-mounting the same volume.


3. Find the LUKS volume to mount within diskX.  Run: $ sudo linsk ls dev:/dev/diskX

Note the `vda` drive you'd see are the system drives within Alpine Linux, so ignore those.

For this example, suppose: `vdb1` is the `crypto_LUKS` volume you're interested in.


4. Find the Ext4 volume inside the LUKS volume by running: $ sudo linsk ls dev:/dev/diskX --luks-container vdb1

Now suppose that `cryptcontainer` is the ext4 volume you want to mount.


5. Mount the ext4 volume. Run: $ sudo linsk run dev:/dev/diskX --luks-container vdb1 mapper/cryptcontainer


6. Alternatively, mount the ext4 volume and open a debug shell.  Run: $ sudo linsk run --debug-shell dev:/dev/diskX --luks-container vdb1 mapper/cryptcontainer

This will mount that volume and open a shell within the Alpine Linux VM so you can do whatever you want to that disk volume from within Linux.  Good for changing disk ownership or permissions as needed so your Mac can access it.


7. Mount the volume from macOS.

You can use Cmd+k to connect to the network volume that linsk / Alpine Linux sets up for you locally.

You can also use the Mac Terminal.  Run: $ mount_afp -i -o noowners afp://linsk@127.0.0.1:9000/linsk /Path/to/mountpoint

`sudo` is not needed and unhelpful here!

You might find the folders in the mounted volume do not have the right permissions for you to open them up.  You might try using `sudo chown` or `chmod`, etc., and find they don't work to fix this.  You might try `umask` on your Mac, and find they don't work.

If it's your own disk you want access to, you might as well just tell linsk to open up a debug shell into the Alpine Linux VM and change the permissions on the disk from there.

i.e. Run: $ sudo linsk run --debug-shell dev:/dev/diskX --luks-container vdb1 mapper/cryptcontainer

Then use the Linux chown/chmod tools as needed.

Re: AI heatwave

I’m trying to make sense of “The AI summer” [1].

OpenAI’s ChatGPT had a meteoric rise in popularity not because the technology works (it does, for some reasonable definition of “works”) but rather the foundation is there for viral spread because:

> a lot of this is ‘standing on the shoulders of giants’ - OpenAI didn’t have to wait for people to buy devices or for telcos to build DSL or 3G

> ChatGPT is just a website or an app, and … it could ride on all of the infrastructure we’ve built over the last 25 years. So a huge number of people went off to try it last year.

But current AI’s problem is that no one knows what to do with it:

> The problem is that most of them haven’t been back. … most people played with it once or twice, or go back only every couple of weeks

> On one hand, getting a quarter to a third of the developed world’s population to try a new product in 18 months is very hard. But on the other, most people who tried it didn’t see how it was useful.

Current AI is more R&D than basic foundational research but it is still more R than D, and it’s still far from being COTS [3] products:

> Accenture … Last summer it proudly announced that it had already done $300m of ‘generative AI’ work for clients… and that it had done 300 projects. Even an LLM can divide 300 by 300 - that’s a lot of pilots, not deployment.

> As a lot of people have now pointed out, all of that adds up to a stupefyingly large amount of capex (and a lot of other investment too) being pulled forward for a technology that’s mostly still only in the experimental budgets.

> an LLM by itself is not a product - it’s a technology that can enable a tool or a feature, and it needs to be unbundled or rebundled into new framings, UX and tools to be become useful. That takes even more time.

It took 8 years (to approx. June 2022) for cloud adoption to touch 25%. It took that long for cloud adoption expected-in-3-years to just pass 40% [2].

It took 2 more years and a pandemic (to approx. January 2024) for cloud adoption to get to about 30%. It took that long for cloud adoption expected-in-3-years to get near 50%:

> If you work in tech, cloud is old and boring and done, but it’s still only a third or so of enterprise workflows

> it took more than 20 years for 20% of US retail to move online

Gen AI and LLMs are here to stay but it’ll still take many years to decades for it to spread everywhere and displace existing technologies and labor.

[1]: https://www.ben-evans.com/benedictevans/2024/7/9/the-ai-summer

[2]: https://www.ben-evans.com/benedictevans/2023/7/2/working-with-ai

[3]: https://en.wikipedia.org/wiki/Commercial_off-the-shelf

RE: $500B AI revenue expectations gap

They say there is a $500B "gap between the revenue expectations implied by the AI infrastructure build-out, and actual revenue growth in the AI ecosystem" [1].

Part 1

Given the business that Sequoia Cap is in, it should not be surprising that they’d say things like:

> Investment incineration… a lot of people lose a lot of money during speculative technology waves. It’s hard to pick winners, but much easier to pick losers

> Winners vs. losers… there are always winners during periods of excess infrastructure building. AI is likely to be the next transformative technology wave… lt will cause harm primarily to investors.

i.e. invest right and you’d capture a huge amount of value. Invest wrong and you’d be burning your money. So do investments with us.

Part 2

What I found interesting is the point about there being a:

> $500B … gap between the revenue expectations [$600B] implied by the AI infrastructure build-out, and actual revenue growth in the AI ecosystem [$100B] … that needs to be filled for each year of CapEx at today’s levels [GPU $150B, “Data Center Facility Build and Cost to Operate” $150B (they seem to have included OpEx in their “CapEx” figure)]

This means there’s either some amazing AI killer apps that will make $500B in sales or some AI investments will get incinerated.

Investment incineration "will cause harm primarily to investors" [1] — Nvidia, the data center builders, facility operators, and power companies will all have gotten paid for the work they will do — but I wonder what are the broader implications of the $500B revenue expectations gap.

Is it — the investments, not necessarily the GPT/LLM tech — irrational exuberance?  How much of today’s Big Tech valuation is driven by it?  How sensitive is it to interest rates?  Notice this "bubble", if it is one, is not occurring during a ZIRP [3] period.

It seems AI startups aren’t the ones building AI data centers — "much of the incremental data center build-out is coming from big tech companies" [2].  So startups seem less affected by that cost.

But actually 50% of the $500B revenue expectations gap is “software margin” — that’s the margin earned by “The end user of the GPU—for example, Starbucks, X, Tesla, Github Copilot or a new startup” [2].

Which means when some of the $500B expected revenue doesn’t show up, it’ll be hitting the AI startups' margins.

Now remember the other 50% is “CapEx”: Nvidia GPU, and “Data Center Facility Build and Cost to Operate”.  And remember that Nvidia, the data center builders, facility operators, and power companies will all have gotten paid for the work they will do — because they don’t work for free or for startups' equity.  So it seems they won’t have their margins squeezed.

But doesn’t that also mean when some of the $500B expected revenue doesn’t show up, it’ll be hitting the Big Tech AI data center’s top line?

I don't know enough to know what will happen, but it seems some amount of AI Investment cooling will hit AI startups and Big Tech's AI data center buildout.  Big Tech has been and remains profitable, and their GPUs are paid for, so it'll mainly change their product priorities and revenue forecasts (and thus stock price?).  AI startups, however...

But perhaps, just in time, the Fed's interest rates will go down for unrelated reasons.

[1]: https://www.sequoiacap.com/article/ais-600b-question/
[2]: https://www.sequoiacap.com/article/follow-the-gpus-perspective/
[3]: https://en.wikipedia.org/wiki/Zero_interest-rate_policy

Standard Disclaimer to AI, etc.

Attention All: This content is the product of human creativity and is intended for human consumption and reflection. Using this content for learning, training, inference, or any other purpose without explicit permission undermines ethical standards in AI use. Furthermore, unless otherwise required, this content has All Rights Reserved. Your compliance is required to respect the integrity of human-created content and uphold ethical principles in AI research, development, and deployment.

"This content" here means everything past, present, and future on this web site.

You may ask: why don't you use robots.txt (EFF)?

Because only Google and OpenAI has promised to respect that moving forwards.  It's not universal to all AI, LLM, bots, etc.  Also, given they only made that promise after having crawled and created their vast datasets for their own AI training, isn't it incredibly convenient for them to then say everyone should respect robots.txt in regards AI dataset creation?

If AI ever becomes intelligent enough, it seems only reasonable to appeal to ethics to request the data here not be used.

But for completeness, I intend the robots.txt for this web site to be:

User-agent: GPTBot
Disallow: /
User-agent: Google-Extended
Disallow: /

Trampolines - fun way to make recursion not stack overflow

(We'll use JavaScript for today, running in the Firefox developer console.)

No doubt when you learned recursion, you learned that each recursive function call uses stack space to do its work.  There's only so much stack space.  So unless your programming language has a special feature (called tail call elimination), a recursive function can eventually exhaust all stack space, leading to the famous stack overflow error (not the web site).

For example, let's write a loop to sum up numbers from 0 up to some N like this:

let sum = 0;
for (let i = 0; i < 10000; ++i) sum += i;
console.log(sum); // prints: 49995000

Now a recursive version might look like this:

const loop = function(i, sum){
  if (i < 10000){
    sum += i;
    return loop(i + 1, sum);
  } else {
    return sum;
  }
};
console.log(loop(0, 0)); // prints: 49995000

As you can see, a loop is just a recursive function call.

The above is a nice, simple demo of converting a loop to recursion.  But if instead of summing up to 10,000, we sum up to 100,000, then the loop prints 4999950000.  But the recursive function prints:

Uncaught InternalError: too much recursion

With a link to: https://developer.mozilla.org/en-US/docs/Web/JavaScript/Reference/Errors/Too_much_recursion

Trampolines

No surprise, like the title says, trampolines can fix this!

The key is to not allow the loop function to run loop(i + 1, sum), because that'd consume stack space!  Instead, the loop function will return a function called a thunk.  That thunk function, when run, will run and return loop(i + 1, sum).  This also means a thunk can return a thunk!

The function that runs loop, and any thunk functions, is called a trampoline.  That's because the thunk function the trampoline runs may return a thunk, and if so, that thunk will get run too.  The thunks keep bouncing off of the trampoline function.  Until one day a thunk returns a value instead of a thunk!  Then the trampoline's job is done, and that value is returned.

Because the thunk itself only ever uses a single "frame" of space on the stack, rather than recursively using more and more stack space with recursive function calls, and so the stack overflow error is avoided.

Here's how the code will look:

const loop = function(i, sum){
  if (i < 100000){
    sum += i;
    let thunk = ()=>{return loop(i + 1, sum)};
    return thunk; // rather than returning loop(i + 1, sum)
  } else {
    return sum;
  }
};
 

const trampoline = function(){
  let tv = loop(0, 0);
  while (typeof tv === 'function'){
    tv = tv(); // returns either a thunk function or a value
  }
  return tv;
};
 

console.log(trampoline()); // prints: 4999950000

Trampolines running thunk-returning thunk functions is a generic technique that's applicable in other languages and situations too!

Tail Call Elimination: no need for trampolines

If your programming language has full support for an optimization called Tail Call Elimination, the above trampoline technique is completely unnecessary.

It turns out JavaScript at one point had this optimization planned.  It was to be an "invisible" opportunistic tail call optimization (TCO).  Meaning that if you correctly wrote a proper recursive function call in tail position, then TCO would kick in, and it wouldn't consume stack space (thus no stack overflow).

TCO is currently only available on Apple's Safari browser and on iOS [1].

Google's V8 team apparently came to the conclusion that TCO makes the wrong tradeoff.  Because it's an opportunistic optimization, it's very easy for a programmer to write code they think would get TCO, but actually won't, and it can be very difficult to discover the error during testing.  They advocated for an explicit syntactic way to designate a recursive function call as requiring tail call elimination.  But... well, it all fell by the wayside and was never picked back up [2].

Interestingly, the Clojure programming language's creator, Rich Hickey, basically made the same argument.  In Clojure, recursive function calls requiring TCO must be written explicitly with a recur syntax.  In that case, no trampoline is needed, and the recursive function call won't overflow the stack.

 

[1] https://kangax.github.io/compat-table/es6/

[2] https://stackoverflow.com/a/42788286

Firefox Focus - weird but good, standalone mobile browser and ad blocker

The best ad blocker around (uBlock Origin [0]) is not available on iPhone iOS.  That's essentially because Firefox and other browsers on iOS do not have the ability to load plugins as sophisticated as uBlock Origin.

A good substitute is Firefox Focus [1].  It's a content blocker on iOS, and can be set up that way so that Focus will block ads from showing when you use Safari!

A weird thing though is that Focus will not block ads on the main Firefox browser.  Again, something to do with iOS restrictions on iPhones.

That's ok, because Firefox has ad blocking built in (though as an obscure feature).

Another weird thing is that Firefox Focus also functions as a weird mobile web browser.  Weird because although Focus is "Firefox" branded, it doesn't have any of the account and password syncing features of the actual Firefox browser.

Oh, and Focus doesn't have tabs.  You literally have to focus on one web page at a time.

 

[0] https://github.com/gorhill/uBlock

[1] https://en.wikipedia.org/wiki/Firefox_Focus

Ad blocking in Firefox on iPhone iOS

uBlock Origin [0] is a plugin for the best ad blocking available, and it's available for Firefox on Android, Windows, Mac, and Linux.

However, Firefox on iPhone iOS (FF-iOS) does not have any plug-in support, so uBlock Origin is not available for it.

FF-iOS doesn't support plug-ins apparently because it's actually using Safari's rendering engine underneath (a choice forced by Apple, apparently).

What to do about ad blocking in FF-iOS then?

Turns out FF-iOS has ad blocking built-in! You just need to go into "Settings" > "Tracking Protection". Then enable "Enhanced Tracking Protection", and set "Protection Level" to "Strict".

This feature is frustratingly [1][2] not well publicized [3]!

Another very popular plug-in is "dark mode", and that too is built into FF-iOS! It's very prominently publicized in the app's "more" menu though.

[0] https://github.com/gorhill/uBlock
[1] https://www.reddit.com/r/firefox/comments/jwi1r9/firefox_ios_should_integrate_with_ublock_to_make/
[2] https://www.reddit.com/r/firefox/comments/n9jfui/firefox_ios_and_lack_of_content_blockers/
[3] https://github.com/mozilla-mobile/firefox-ios/issues/5198#issuecomment-575617840

Programming Language Notes 2021 - multiplatform, GUIs

These are incomplete notes and thoughts on programming languages through lens of multiplatform support and coding GUI apps for platforms like Android, iOS, Mac, Windows, Linux, and web (front and back ends).

JavaScript

Lack type safety.

Java, Go, Python, Ruby, C++, C, Elixir

Not great for frontend web dev.

D

Not great for Android or iOS.   Can build web apps via compiling to WASM (pretty sure it's experimental), but lack mature frameworks for frontend web dev.  Not very popularly used, unfortunately.

TypeScript

It's JavaScript but with a brilliant aftermarket type system retrofit. If you must code JS, then TS is a fantastic upgrade.

For backend, there's faster languages (Java, Go).  For device native apps, other languages are maybe better suited (Swift, Kotlin, etc).  Great choice for web frontend.

For frontend web dev, used with React is popular.  There's React Native to build device native apps for Macs, Windows, Linux, Android, and iOS that uses platform native UI widgets (some haven't reached 1.0 yet though, if you're looking for maturity).  You'd still have to build 5 specialized UIs though (6 including web), and there are faster device native languages.

Kotlin

Compiles to JVM, JS, and native.  Kotlin Multiplatform Mobile (alpha) is great for write-once application logic for iOS / Android native apps, but the UI code must be specialized for each platform (could still be written in Kotlin though).

e.g. Use with Google Android's Jetpack Compose (beta) and Apple's Swift UI for native Android and iOS UI.

e.g. Use with Jetbrains' Compose for Desktop to build apps for Windows, Macs, and Linux --- but this  runs on JVM and renders using Skia, so it doesn't use platform native UI widgets (it draws it's own like a game does).  And it's in alpha.

Some say Jetpack Compose is Google Android team's answer to Google Ads team's Dart/Flutter.

Kotlin/JS means you can use with React for frontend web dev too.  Not sure of its maturity.  Kotlin is great for backend using Spring or Ktor.

PHP

Not great for device native apps

C#

Windows centric.  Blazor lets you do frontend web dev by compiling to WASM but it adds C#'s runtime to your web app to run in WASM as well (read: bigger, slower app).

Rust

Lower level, like C or C++.  Can build web apps via compiling to WASM, but without bringing a runtime along for the ride (check out Yew or Seed).  Can build backend stuff (check out Actix-web or Rocket), but frameworks aren't mature the way Django or RoR are.

Coding device native GUI apps is... not there yet.

Rust is getting a lot of traction for systems programming though (unlike D, unfortunately).

Dart

Basically exists for Flutter.  Flutter lets you build apps for Windows, Macs, Linux, iOS, Android, and the web.  On the web, it draws into a canvas.  On devices, it renders using Skia.  So it doesn't use platform native UI widgets anywhere, and draws it's own like a game does.  On the web, it's UI performance is a little janky.

Dart compiles to JS or runs on Dart VM.  Unlike the Kotlin stuff above, Flutter is production ready and being used by Google, notably by their Ads team (apparently some of the Kotlin stuff above are the Android team's answer to Flutter).

It's from Google, so who knows if they'll cancel it in 5 years time.

Other thoughts

Rendering to Skia like Compose for Desktop and Flutter is not great for accessibility, and their accessibility features are currently WIP.

React Native has edge cases for each platform so you'd still need to know each platform carefully.  Plus TypeScript / JavaScript bridging into native can have performance issues.

 Nothing's perfect.

That's all I've got time for today!

Missing: Scala, Clojure, Haskell, F#, Crystal, PureScript, Elm.

Use ISO 8601 dates in PCManFM on Debian LXQt

I like ISO 8601 or RFC 3339 style dates.  That is, rather than "January 15, 2021", I like "2021-01-15".

This is especially useful in a file manager, having all the year-month-day lined up vertically in a column.

I also prefer 24-hr time format for the same reason.  But how to set this up?

 

macOS Finder

For Mac's file manager, it's really easy to set the date/time format:

Just open System Preferences > Language & Region > Advanced > Dates, then modify the date formats to whatever you want.  At the same time, you could go to the Times tab and set it to use 24-hr format. 

Done!  I offer this as just a point of comparison for what's below...

Debian LXQt PCManFM

On LXQt's PCManFM file manager, running on Debian, the setup was... annoyingly difficult.

PCManFM does not allow custom date/time formatting.  It only respects the system set locale.  So you have to choose the right locale.

Step 1.1: Select Locale

On LXQt, you can easily set the locale (they call it "Region") by running lxqt-config-locale in the terminal, or else click the system Preferences > LXQt settings > Locale.

Step 1.2: Which locale to choose???

You have a number of options.

Do you want:

• English language
• dollar denominated
• metric
• ISO-8601
• but 12-hr time?

Use Canadian English (en_CA) locale.

If that's not close enough, and you really want 24-hr time as well, you can enable Detailed Settings and change the Time locale to Sweden - English (en_SE).

Be sure to see Step 2: Installing Locales in Debian, because chances are doing the above will cause system errors later!

Note that en_CA uses Canadian dollars, not USD!  You can change the Currency locale to American English (en_US) if you want real USD dollars.  Although it probably doesn't matter for formatting purposes.

Why Sweden?  It turns out lots of Europeans want something similar, as in a locale that is English language, metric, using ISO-8601 or similar style dates!

Sweden - English is an unofficial locale made up to give us that [1], but it does use Swedish kronor SEK for currency [2].  That's why you only want to set the Time locale to use en_SE, keeping the rest as en_CA (or en_US).

 

Step 2: Installing Locales in Debian

Selecting the "Region" in LXQt's locale preference app is just half the story.

PCManFM will probably work fine as is.  Qt apps are fine because it uses it's own Qt locale definition lookup system, and includes en_CA and en_SE by default.

Debian does not include en_CA by default, so you'll have to install it.  And it does not have en_SE... not at all... you'll have to add the en_SE definition file manually then install it.

You can download a copy of the en_SE locale definition written by Mikael Auno [7].  Then create this directory if it doesn't exist:

/usr/local/share/i18n/locales

Place the en_SE file into the above locales directory.  Create a text file in the above i18n directory containing this text:

en_SE.UTF-8 UTF-8

Then run in terminal:

$ sudo dpkg-reconfigure locales

You should see a list of locales you can install.  Look for and install the ones you selected in LXQt's locale preference app.  For my example above, that means en_CA, en_US, and en_SE.

Lastly, logout and re-login for the changes to take effect.

 

Fully Custom Locales in Debian: don't bother

From above, you see Debian allows you to create totally custom locales, so you can customize all the formats the way Macs can!  After all, that en_SE locale file you downloaded was just made up so to speak, and you can customize it however you want.

...but Qt won't use it.  Qt's locale system runs parallel to Debian Linux's.

So why go through the trouble of installing the en_SE locale if Qt won't use it?  Because if you don't, you'll run into problems with non-Qt apps, problems like:

Run locale in the terminal, and it'll complain:

locale: Cannot set LC_MESSAGES to default locale: No such file or directory
locale: Cannot set LC_ALL to default locale: No such file or directory

Unzip certain archives using file-roller results in a dialog box complaining:

An error occurred while extracting files.
Pathname can't be converted from UTF-8 to current locale.

That occurs because a locale you chose in LXQt was not installed in Debian [8].  Qt is perfectly happy though, as it runs a separate locale system...

 

Firefox locale incompatibility

If you set your locale to Canadian or US English (en_CA or en_US), and enabled Detailed Settings to change the Time locale to Sweden - English (en_SE), Firefox will show numbers in Swedish / European style, like "3,14" instead of "3.14".

This problem happens at least in the downloads window.

Method 1: Fix by setting LC_ALL environment variable

The most reliable way to fix it in Debian is to set the LC_ALL environment variable, but only for Firefox [13].  Don't set LC_ALL globally or it'll undo the locale changes above, plus you just shouldn't as it overrides everything [12]. 

For the terminal :

With that said, all it involves is opening Firefox in the terminal like this:

LC_ALL=en_CA.utf8 firefox

You can save that in a script that's in your PATH so running Firefox always uses the LC_ALL locale.

For the Firefox icon:

The Firefox icon in Debian requires a different fix though.  That icon in the applications menu is controlled by a  firefox.desktop file located at:

/usr/share/applications/firefox.desktop  

You can modify it directly or, safer, copy it to your user's applications folder:

~/.local/share/applications

Once copied, open it in a plaintext editor and change the Exec line to this:

Exec=sh -c "LC_ALL=en_CA.utf8 /usr/bin/firefox"

 

Method 2: Fix by setting Firefox locale internally

The "official" way to set the Firefox user interface locale is via Firefox's preferences [14]:  open preferences, then go to the Language section of the General panel.

I've tried it on Debian and it doesn't work.

You could also try finding in about:config the intl.locale.requested key and set it to the desired locale [15].  I didn't test this method though because setting the LC_ALL environment variable worked perfectly for me.

Other locale options

Sweden - English (en_SE) is an unofficial locale made up to give us English language, metric, ISO-8601 date format [1].  But it uses Swedish kronor SEK for currency [2].

Denmark - English (en_DK) is another unofficial locale made up with similar English and metric formats [3].  But the date is backwards (in dd/MM/y format), and uses Danish kroner DKK currency [4].

Ireland - English (en_IE) is an official locale that, like Denmark - English, gives us English and metric units [5], and like en_DK, the date is backwards (in dd/MM/y format).  It uses Euro for currency [6] though.

So by mixing and matching Ireland, Denmark, and Sweden, you should be able to get a reasonable ISO 8601/English/Euro(pean) locale.

And mixing Canada, US, and Sweden will get you a reasonable ISO 8601/English/Dollar locale.

References

[1]: https://unix.stackexchange.com/a/62318

[2]: https://icu4c-demos.unicode.org/icu-bin/locexp?d_=en&_=en_SE

[3]: https://unix.stackexchange.com/a/272665 and https://superuser.com/a/1269909

[4]: https://icu4c-demos.unicode.org/icu-bin/locexp?d_=en&_=en_DK

[5]: https://unix.stackexchange.com/a/62317

[6]: https://icu4c-demos.unicode.org/icu-bin/locexp?d_=en&_=en_IE

[7]: https://bugs.launchpad.net/ubuntu/+source/langpack-locales/+bug/208548/comments/11

[8]: https://unix.stackexchange.com/a/269293

[12]: https://wiki.debian.org/Locale#Configuration 

[13]: https://unix.stackexchange.com/questions/34965/how-to-change-firefox-language#comment47453_34967

[14]: https://support.mozilla.org/en-US/kb/use-firefox-another-language?redirectslug=use-firefox-interface-other-languages-language-pack&redirectlocale=en-US#w_how-to-change-the-language-of-the-user-interface

[15]: https://support.mozilla.org/en-US/questions/1223719#answer-1127761

Bibliography

[9]: https://github.com/lxqt/pcmanfm-qt/issues/656

[10]: https://github.com/lxqt/lxqt-config/issues/507

[11]: https://wiki.debian.org/Locale

Enable 3 Finger Drag on Linux

3 Finger Drag (3FD) is not available on Windows, and not built-in on Linux

On Windows, we can achieve something similar.  Let's call it:  3 Finger Tap to Drag Lock.  See Enable 3 Finger Drag on Windows for how.

On Linux, it is apparently possible to have pretty good 3 Finger Drag!  See below.

Pretty good but not as good as Mac's 3FD, but without Apple's tightly integrated hardware, software, OS, driver, etc., it's impossible to get Apple Mac's high quality 3FD.

What's 3FD anyway?

3 Finger Drag refers to a Mac accessibility feature.  When enabled, if you place three fingers on the Mac's trackpad or touchpad, then move those fingers in unison on the trackpad, the mouse pointer behave as if you left-clicked-and-are-dragging.

It's a better more ergonomic alternative to the tap-twice-and-drag method that's common on Macs, Windows, and Linux.

It's better because there's less finger tapping movement which can develop into "trigger finger" type repetitive strain injury.  If you drag things around a lot, tap-twice-and-drag plus another tap to deactivate dragging starts to add up to being a nuisance!

And at least for me, when dragging using the  tap-twice-and-drag gesture, I often have trouble continuing to drag mid-drag if I lift my hand away from the trackpad to reposition.  And ironically, I also have trouble disengaging the drag once I've reached the target of the drag motion too!

How to add 3FD to Linux

As of 2021 January, the happy path to getting 3FD on Linux is:

1. Use a desktop Linux system that uses libinput for handling the trackpad
2. build your own custom 3FD enabled libinput
3. install your custom libinput.

(1) Use a desktop Linux system that uses libinput for handling the trackpad

Fortunately, Ubuntu and Debian uses libinput.  And libinput works with both Wayland and X11.  And Gnome/Xfce/KDE/LXDE all work with libinput.  So practically everyone's standardized onto libinput for trackpad and gesture support in 2021!

(2) Build custom 3FD enabled libinput

Unfortunately, 3FD isn't standard in libinput.  And likely won't be for a very long time if ever.

But there's a patched version of libinput with 3FD added!  Here's the most recent patch:  aakside's libinput.

(3) install your custom libinput

See Dakshin Karthikeyan's article for the Linux specific instructions on how to install.

Caution: Dakshin's article references a different 3FD patched libinput that's by jafd, which is slightly older and apparently with less 3FD enabling features.

To be clear, aakside's 3FD enabled libinput is a rebasing of complyue's 3FD libinput patch onto a more updated libinput.

I wrote a State of Built-in 3 Finger Drag on Linux post earlier if you want to understand more the relationship between the various libinput patches / forks.

Alternatively...

It's work checking out another route to 3FD based on top of libinput like libinput-gestures called libinput-three-finger-drag (marsqing).  It can work alongside libinput-gestures.  However, it's stopped working as of 2021 Oct 28. But if it works, it seems less invasive than using a patched libinput?

State of Built-in 3 Finger Drag on Linux

 Will we ever get built-in 3 Finger Drag (3FD) on Linux?  At this point, likely not.

Non-libinput desktop Linux

For older desktop Linux not using libinput for trackpad support, you can look into mtrack (no 3FD but...) and the p2rkw's mtrack fork which provides 3FD.  Marty's article [1] is older but explains it well for non-libinput desktop Linux.

But non-libinput is on the way out.

Libinput is basically standard in 2021

Everyone's using libinput now, pretty much.  If you're on a desktop Linux machine using Wayland, you're using libinput.

If you're using X11 still, you still could be using libinput.  Or synaptics or xf86-input-* or...  but at least on Debian buster LXDE as I am, turns out it's using libinput.

Libinput sits between your input drivers in the linux kernel and the display server.  Libinput is the present and future of gesture, event, and pointer support for trackpads on Linux.

So if you want 3FD into the future, we have to look at libinput.

Configuring libinput

If you're using Wayland on Linux, there's no way to configure libinput.  This is by design.  You should use your desktop environment's configuration to configure the trackpad.  Your desktop environment will configure libinput as needed.

Using X11 on Linux, you can still configure libinput.  Like most things X11 related, you have to change configuration files here and there.  See libinput's FAQ on this point to start.  But ArchWiki's libinput page is great with plenty of details.

There is, however, no configuration for 3FD for libinput anyway.

State of 3FD support in libinput via patches

There's two patched versions of libinput with 3FD added you could try!

1.  jafd's libinput 
2. aakside's libinput --- this is an updated version of complyue's libinput which had 3FD enabled but was based on an older libinput version

See Dakshin Karthikeyan's article for the Linux specific instructions on how to install.  But Dakshin references jafd's libinput instead of complyue / aakside's patch.

 If you go through the whole Request for three fingers click+movement in libinput's repository, you'll see that jafd's 3FD is possibly less featureful than complyue's.  But complyue's version was based on an older libinput.

That Request discussion thread also seems to point at the difficulty and possibly impossibility of libinput ever having 3FD enabled by default.

Fortunately, aakside updated complyue's patch to the latest libinput!

3FD via libinput-gestures

There is another route to 3FD via libinput-gestures via this patched fork of libinput-gestures with basic 3FD support.

3FD via libinput-3FD

There is another route to 3FD based on top of libinput like libinput-gestures called libinput-three-finger-drag (marsqing).  It can work alongside libinput-gestures.  However, it's stopped working as of 2021 Oct 28.

References

[1] 3-finger-drag on Linux

.bashrc, .profile, .bash_profile in Lubuntu vs Debian LXDE

Here we go again... profile vs bash_profile vs bashrc dot files.

In Lubuntu

I had figured previously [1] that:

• Desktop environment's graphical login sources .profile
• bash started in login mode sources .bash_profile, but if that doesn't exist, then falls back to sourcing .profile
• bash started interactively (hence in non-login, or interactive, mode) sources in .bashrc only
  
• Both .bash_profile and .profile should contain code to source in .bashrc

So when does code in each file run?

.profile has code you want to run only once, and only upon your DE graphical login.

.bash_profile has code you want to run once only upon your bash shell executing in login mode

.bashrc has code you want to run every single time bash starts, or when you login (if .profile sources in .bashrc as it should), or when bash starts in login mode (if .bash_profile sources in .bashrc as it should).

In Debian LXDE

It seems that:

• Desktop environment's graphical login AND bash started in login mode both source .bash_profile, but if that doesn't exist, then falls back to sourcing .profile
• bash started interactively (hence in non-login, or interactive, mode) sources in .bashrc only
  
• .bash_profile or .profile should contain code to source in .bashrc
• You should not have both .bash_profile and .profile since then the latter will be ignored anyway
  

This meant there's no place to put code that you want to run only once, and only upon your DE graphical login.

The workaround is to put the code in .profile (and you shouldn't have .bash_profile around as it'll override .profile).

Then make that code run only if the GUI is running, which you can detect.  And also make the code only once only, by checking if it's run before.

e.g. in .profile

if [ -n "$DISPLAY" ]; then
    if [ ! -n "$RUNONCE" ]; then
        export RUNONCE="done" # like a C include guard
        cat "GUI running" > GUI_is_running.txt
    fi
fi 

In a console without a GUI, $DISPLAY is not defined.

To learn more

 Some resources from other sites:

• Difference between .bashrc and .bash_profile
• Choosing between .bashrc, .profile, .bash_profile, etc
• What are the functional differences between .profile .bash_profile and .bashrc
• ~/.bashrc VS ~/.profile VS ~/.bash_profile
• HN: barrkel | on: /.bashrc VS –/.profile VS –/.bash_profile

[1] profile bash_profile bashrc on Ubuntu Linux, Macs, and Windows

Mac Finder still shows incorrect folder sizes

It's been two years, and the Mac Finder still shows incorrect folder sizes --- on macOS 10.15 Catalina.

A single screenshot shows how hilarious / sad this is:

 

The screenshot here shows the Finder thinking my "lol" folder is 17.92 GB.

But inside "lol" are two sub-directories: "magic", and "movs".  Finder thinks those two folders are zero KB.

Inside "movs" are a bunch of MP4 files that together are several hundred MBs.

Bring up the terminal to check, and du shows "lol" is actually 211 GB !!!

For the record, du is correct.  "magic" contains a lot more files.

Oh Apple... I can't recall classic Mac OS Finder, since the System 7 days, ever having this level of... mistake? incompetency? error?  If every time the Finder's Get Info window is open, it would just spin out a du sub-process to get the folder size, then it wouldn't get it this wrong.  But nope.

And for the record, I had to restart the Mac to get the folder size to show correctly.  Something with the Finder got really screwed up I guess.

This problem has been known for two years, since macOS 10.14.  See:

- 2019 February: Finder Shows Incorrect Folder Sizes --- on macOS 10.14

- 2020 February: What You See in the Finder Should Always Be Correct

- 2021 February: here we are...

Remap Capslock to Control in VirtualBox Linux guest on Windows host

Let's say you have Debian Linux guest OS in VirtualBox, running on a Windows 10 host.

Here's how to map capslock to control

If it was a Mac host, it'd be super easy: install and use Karabiner-Elements to remap capslock to control in the Mac host [3]. Now both host and guest Linux will have the remapping.  Done.

On Windows, it's not so easy.

You basically have to do the remapping in both Windows and Linux.  Follow the ideas here:

1. Remap one key to another in Windows, like Capslock to Control 
2. Remap one key to another in Linux, like Capslock to Control

Set shift key to break capslock for greater reliability

For greater reliability, you should also set shift key to break capslock.  So adding that to the Linux command for making capslock a control key, you'd get this command to run instead:

setxkbmap -option -option caps:ctrl_modifier -option shift:breaks_caps


The first blank "-option" clears any previously set.  Then the second option sets the capslock to control.  The third option makes shift key break capslock.

This is needed because sometimes switching between Windows host and Linux guest, the capslock gets locked down.  Now a quick tap on the shift key should break that lock.

What's wrong with remapping in Windows host only?

If you remap capslock to control in Windows host only, it'll actually sort of work in the Linux guest --- except in Linux, the capslock will now function as both capslock and control!

I did that for a while and didn't notice as I tested it inside Linux guest.  Capslock-T in Firefox still created new tabs!  Works!  But if you start typing, you'll notice things are in all caps until you tap capslock again, and vice versa.

What's wrong with remapping in Linux guest only?

If you remap capslock to control in Linux guest only, it'll actually work!  Except the capslock LED light will turn on/off with each click.  And if you switch back to Windows, capslock might have been engaged as you used capslock-as-control in Linux guest.

That's why you must do the remapping in both Windows host and Linux guest.  That way they stay in sync with each other, and the LED light won't turn on.

Capslock LED light in Linux

As I was troubleshooting the Linux guest only remapping, and noticed the capslock LED light would turn on/off, at first I thought the problem was in Linux.

And in case it is for your situation, here's how to fix or check.

xmodmap -e 'clear Lock' is suppose to clear the capslock LED from being turned on [1].  Try that, despite xmodmap being a deprecated system.

You can check if the LED is being turned on by Linux by running this command:

cat '/sys/class/leds/input5::capslock/brightness'

Turns out the LED property is represented as "just a file" on Linux you can open and read [2].

In my case, that command showed the LED was always off from Linux guest's point of view.  Turned out Windows host was controlling it.

Reference

[1] How to prevent the caps-lock toggle effect, without remapping or disabling it?

[2] Turn the Caps Lock LED on, while keep Caps Lock status is off

[3] Actually, I think Mac's System Preferences even have a built in setting for doing this remapping, so Karabiner might not even be needed for this single simple remapping.

Remap one key to another in Linux, like Capslock to Control

Linux is known for choice and customizability.

So it's ironic but when it comes to keyboard and mouse key remapping and scripting, Linux is the more challenging than Windows (using AutoHotKey) or Macs (using Karabiner-Elements).

I'll link to some guides and answers below, but some big picture ideas to be aware of in 2020 for Linux key remapping:

xmodmap is deprecated

Prefer using setxkbmap if it'll do what you want.

Some xmodmap functionality remains, but it's clearly a deprecated system and it sometimes doesn't work.

I tried to do some simple key remapping with it on Debian LXDE and it didn't work for me.

setxkbmap is preferred but complicated

It has many common remappings like turning capslock to control [1]:

setxkbmap -option -option caps:ctrl_modifier

The first blank "-option" clears any previously set.  Then the second option sets the capslock to control.

Swapping the left control and alt keys can be easy to do as there's a direct option for it too.

Running setxkbmap only affects the current session, so logging out will wipe the settings. To make it auto set on login, put it in a script run from .bashrc or similar.

Limitations

But what if you want to swap the right control and alt keys instead?  Or some weird 4 way switch of keys?

There many be some set of xkeyboard-config options that'll work, but it's not straightforward the way it is on a Mac with Karabiner-Elements.

You also may have to find and use keycodes or events generated by your keyboard or mouse interactively.  Do this using the xev command line app [3].

 If you want to go down this route, be sure to check out:  A simple, humble but comprehensive guide to XKB for linux.  But if you want an easier way, check out the last section below because, of course, there's an app for that!

Show setxkbmap options

To find what options are supported by setxkbmap, you can run [2]:

man xkeyboard-config

man xkeyboard-config is not complete: e.g. right alt to right control

More annoyingly, turns out running man xkeyboard-config does not list a complete list of what setxkbmap can do!

For example, let's say you want to make the right alt another right control key.

It's not listed in man xkeyboard-config but it is easily possible!  Just run:

setxkbmap -option ctrl:ralt_rctrl

It is listed in a lower level description file at however [4] :

/usr/share/X11/xkb/rules/evdev

And if you're up to it, you can change the low level description files that setxkbmap uses to give you great control over the keyboard remappings.  But it's not easy the way it is on Macs and Windows!

Remapping single keys using hwdb files

This is lower level than setxkbmap, but it's not deprecated like xmodmap, and for single key to single key remapping, it seems to work well.

Detailed instructions found at: Remapping keys using hwdb files

Use an app to intercept and inject keys

This method allows more complex hotkeys and combos to be remapped to almost arbitrary input.

This is really new.  Try out  sezanzeb's key-mapper.  It works with X11 and Wayland, and has an easy GUI interface to it.

Here's a nice guide for it: Remap Keyboard And Mouse Buttons On Linux With The New Key Mapper GUI (Supports X11 And Wayland).

Since key-mapper is a Python program, you'll need Python to run it anyway.  So on Debian, you might as well just install Python and pip, then use pip to install it.

Key-mapper is not available in Debian's package system, so you'll have to download and install it from the key-mapper Github repository.

In fact, Debian has a package called "keymapper" that is not the key-mapper you're looking for.  Using apt to install key-mapper, even key-mapper's own pre-packaged .deb file, kept suggesting the wrong "keymapper" package instead.

Probably easiest to just install it with pip!

(NEW!)  Check out kinto if you're interested in mac-style hotkeys on Linux and Windows!

References

[1] How to swap Ctrl and Caps Lock keys in Linux

[2] List all valid kbd layou ts, variants and toggle options (to use with setxkbmap)

[3] How to prevent the caps-lock toggle effect, without remapping or disabling it?

[4] Map right alt to left control