todd8

I agree that it is unlikely to be increasing precisely linearly, but do you have a more specific claim?

It is my understanding that the GCMs in use have many “tunable” parameters some of which are strongly stochastic, the component of the models are coupled, there are feedback loops that are poorly understood—some positive some negative, these feedback loops operate over all possible time scale from minutes to centuries. The systems of differential equations describing the models will not be analytic and worse will have chaotic solutions.

The dozens of existing GCMs don’t agree, but I’m not an expert so I can’t know which ones to trust. Furthermore, expert climate researchers very careers are at stake so there is no effort that I can detect to make all the data sets, source code, and design justifications open in ways to allow inspection by outsiders like myself.

I have worked professionally on environmental models only twice (over 40 years ago). One had the worst code I ever had to review (and I’ve taught CS at the university level!) and the other made simplify assumptions so ridiculous that the results were meaningless.

I used to be able to download pictures of hand written recorded weather station temperature data from 100 years ago. Now, I can no longer get to it from NOAA or NASA websites. What happened to it? (Perhaps, my Google-foo is failing me.)

I think many would agree that climate research is important or even existentially important for humankind. We already fund the research with our tax dollars. Why can’t the research be performed as openly as free software foundation projects? If I’m curious about the kind of LRU algorithm used by the ZFS file system cache, I can just clone the repo and read it myself. If I don’t understand why an Emacs feature has been deprecated, I can peruse the emacs-dev mailing list archive.

I want everyone (and especially those that are climate scientists) to have easy access to the data and climate models.

There is a lot of discussion here about renewables. I recommend a recent paper published in Nature Communications titled "Geophysical constraints on the reliability of solar and wind power worldwide"[1]. It considers mixes of Solar and Wind, with or without excess capacity, and storage facilities zero, 3, or 12 hours. The paper's model (it looks like the code is available on GitHub) makes optimist assumptions (e.g. no transmission losses within a single country and feasibility of 12hrs of storage for the whole country). Nevertheless, the results are interesting.

Figure 3 (see[1]) indicates how much power outage must be tolerated across the entire country for different mixes renewable capacity and storage depending on the country:

US -- Generation 1.5 times the capacity needed, 12 hours of storage, wind and solar have a power supply gap of 15% when the goal is to tolerate 10 hours of power outage.

Germany -- Generation 3 times the capacity needed, 12 hours of storage, wind and solar have a power supply gap of 60% when the goal is to tolerate 10 hours of power outage.

The larger the power supply gap, the more additional dispatchable power that must be provided.

[1] https://www.nature.com/articles/s41467-021-26355-z

The first laser device was demonstrated in 1960. By 1961, patents on lasers already indicated applications in communication and meteorology[1]. Subsequent patents included uses for range finding, chemical analysis of gases, drilling, and other uses.

Amateur's were constructing working lasers in their basements within a year or two after the laser was invented.

I grew up reading a monthly column in Scientific American titled The Amateur Scientist which in the 1960's described a number of laser projects that could be constructed by amateurs. 1964[2], 1965[3], 1967[4], 1969[5], 1970[6]. Naturally, projects using lasers were discussed in other books and magazines during this time as well.

To me, Quantum computing seems very different than the history of lasers.

[1] https://patents.google.com/patent/US3353115A/en

[2] Scientific American, THE AMATEUR SCIENTIST, Vol. 211, No. 3 (September 1964), pp. 227-242 (16 pages)

[3] Scientific American, THE AMATEUR SCIENTIST, Vol. 213, No. 6 (December 1965), pp. 106-113 (8 pages)

[4] Scientific American, THE AMATEUR SCIENTIST, Vol. 216, No. 2 (February 1967), pp. 122-134 (13 pages)

[5] Scientific American, THE AMATEUR SCIENTIST, Vol. 220, No. 2 (February 1969), pp. 118-125 (8 pages)

[6] Scientific American, THE AMATEUR SCIENTIST, Vol. 222, No. 2 (February 1970), pp. 116-121 (6 pages)

US paper sizes:

ANSI A == 8.5 x 11 inches (letter)

ANSI B == 11 x 17 inches

ANSI C == 17 x 22 inches

ANSI D == 22 x 34 inches

ANSI E == 34 x 44 inches

There are additional sizes frequently used in the US too: business card, legal, half letter, and specialized sizes for architecture.

The article gets some dates wrong:

> The age of the amber monitor lasted from about 1987 until 1990, a mere moment in the history of computers.

I remember the first time I used an amber monitor. It was in a dimly lit lab at MIT in my last year there. It was 1973. The amber monitor was hooked up to a PDP-7, an 18 bit minicomputer, and the monitor only had 40 columns. It wasn’t the first time I had used a monitor, but before that most of my programming was on cards or teletype style terminals that typed output on fan-fold continuous paper.

This amber colored terminal used a primitive rasterization technique that didn’t have a fixed refresh rate so the frame rate slowed as more characters were needed on the screen. Despite the very long persistence of the amber phosphors the monitor flickered badly, and when too many characters were on the screen it would dim as the earlier lines would get harder to read, hence the dim room lighting. The monitor appeared to be some contraption rigged up by the EE department (which CS was just a part of).

After that, I would continue to occasionally run into commercial monitors that were amber, although most used green phosphors or white like the ADM-3 family and the DEC-VT100 [1].

Notably, I had access to the University of Illinois PLATO time sharing system in 1979. It used monochrome touch screen amber plasma display terminals at that time [2].

[1] https://en.m.wikipedia.org/wiki/ADM-3A

[2] https://en.m.wikipedia.org/wiki/PLATO_(computer_system)

First two paragraphs of History section found on DragonFly web site[1]:

> A technical introduction: The ultimate goal of the DragonFly project at its inception was to provide native clustering support in the kernel. This type of functionality requires a sophisticated cache management framework for filesystem namespaces, file spaces and VM spaces. These and other features eventually culminate in the ability to allow heavily interactive programs to run across multiple machines with cache coherency fully guaranteed in all respects. This also requires being able to divide resources, including the CPU by way of a controlled VM context, for safe assignment to potentially unsecured third-party clusters over the internet. This original design direction, although no longer the primary goal of the DragonFly project, has influenced many of the design decisions made in the intervening years. While full cache coherency is no longer a top level goal, filesystem coherency is, and that direction continues to guide the project in a number of ways.

> DragonFly BSD was forked from FreeBSD 4.8 in June of 2003, by Matthew Dillon. The project was originally billed as "the logical continuation of the FreeBSD 4.x series", as quoted in the announcement, but this description has long since become obsolete. From a performance perspective DragonFly's only real competitor these days is Linux.

[1] https://www.dragonflybsd.org/history/

Might be time to learn org mode, just in case.

Barbra Liskov’s CLU language had multiple returns in 1973, ten years before Common Lisp. I was a student of hers back then.

I’m pretty sure I saw multiple returns used in pseudo-code on a class handout the semester before I started working on CLU. The class was “Structure and Interpretation of Computer Programs” that would lead to the writing of SICP.

Cameras in classrooms might help.

Naturally, the “curse of Turing” means that virtually any programming language can, with enough effort, do anything. I wouldn’t classify Prolog as appropriate language for general purpose vs say Python.

In theory, I could write a replacement for the Linux kernel in machine code or COBOL or even Prolog. This doesn’t mean that it would be wise to do so. (I’ve written code in professional settings in all of these ways.)

This isn’t an option for US citizens. Income made anywhere in the world is taxable and requires reporting and payments to the US. Furthermore, giving up one’s US citizenship requires an exit tax that taxes all unrealized gains as if you sold all assets the day before renouncing citizenship.

If the IRS suspects that you are renouncing citizenship to avoid future taxes, I believe that they can tax your next 10 years of (worldwide) income as well.

US tax laws and regulations are complex and I’m not a lawyer so don’t take my advice. (I plan on keeping my US citizenship.)

[1] https://www.irs.gov/individuals/international-taxpayers/taxa...

[2] https://www.escapeartist.com/blog/us-taxes-worldwide-income/

That makes sense, thanks.

I found Jack London's story so compelling as a kid when I read it. Perhaps it was because I grew up in snowy, cold winter part of the country. Maybe that's why I have around a dozen books on survival, living off the land, tracking, nuclear bomb shelters, etc.

This is an interesting list. I'd add hashing and hash tables to the list. There are lots of interesting variations of these as well.

I would put hash tables ahead of piece tables because of their generality. For text editors I've always thought that gap buffers were so straightforward that ropes and piece tables seemed like extra unnecessary work. I see that VS Code uses piece tables so perhaps someone can explain the performance differences to me.

I've never sat down and benchmarked various implementations of text buffers. However, it seems to me that memmove(3) on modern processors would be so fast that moving a buffer gap on even a 1GB file wouldn't be perceptible, and in return operations over the whole buffer like string searching or parsing would be much more efficient in a gap buffer because the program could count on the buffer being in contiguous memory locations (after closing the gap). In buffers implemented using a part table or ropes, every character access involves extra layers of indirection. Furthermore, gap buffers are going to be more cache friendly for the way that most editing operations take place in a text editor.

Perhaps someone that has actually done some comparisons could enlighten me.

> In contrast, wind seems to already be making 10 percent of the US demand[4], versus solars 2 percent.

The numbers can be misleading; I'm not suggesting that jldugger doesn't understand this, just that I find the numbers and the way they are quoted a bit confusing. If we look at primary energy consumption by source from the US Energy Information Agency we see [1] that for 2021 the US used 98 quads of energy.

The breakdown into primary energy consumption by source for 2021 is

    77   quads from fossil fuels
     8   quads from nuclear 
     5   quads from biomass
     3.3 quads from wind
     2   quads from hydro
     1.5 quads from solar
     ===========================
     98  quads total

These number are rounded, see original for details (quad == Quadrillion BTU). It appears to me that wind plus solar are providing close to 5% of US consumption of energy.

[1] https://www.eia.gov/totalenergy/data/browser/index.php?tbl=T...

This is so terrible. It makes me think about comments I heard from John Gretton "Jocko" Willink, an ex-Navy Seal podcaster on YouTube. During the height of the defund the police movement he suggested that doing so would make things worse. As I recall, he said it would be better to increase the funding for police departments so that there could be more police. Increasing the police force by one third would allow patrol officers to be off the streets for one week out of every month. During that week they could decompress from the stress of being on patrol and could spend the time in training classes that could alternate through various subject including de-escalation, Spanish lessons, stress training, etc. (These are my recollections of what he said, take them with a grain of salt.)

I'm sure that a Navy Seal has very different training requirements than a police officer, so Mr. Willink's thoughts may not be relevant, but at least to me they make sense. I would be willing to pay more for police if it meant that they would be better trained, that the job paid well enough that marginal officers wouldn't be hired, and that there was enough time off from stressful work to keep the police officers lives more balanced. I don't know if this would have saved Raj Moonesinghe's life, but what we are doing now is not working.

When I started running longer distances to train for marathons, I ran into lots of other runners that iced after running say 10 miles. I did this too, but as time passed I got too lazy to pack up some ice to use at the end of long runs because these were always scheduled in the very early mornings at sometimes distant but interesting locations.

What I discovered was that cold did feel good and did alleviate some of the discomfort from running, but for overuse induced discomfort it wasn't overall helpful to me.

Now a bit off-topic to explain what did help me with overlong runs:

What helped me the most for the wear and tear caused by long runs was getting the right shoes: some would make the lateral (furthest from the midline of my body) sides of my knees hurt, some would make the medial (closest to midline) sides of my knees hurt and when I found the right ones my knees didn't hurt. Shoes without enough padding under the ball of my foot also ended up causing me more pain than shoes with more padding.

I always alternated my runs between shoes from two different manufacturers because I found that the slight differences in the ways the shoes flexed prevented overuse injuries from running in the same shoes each day.

Finally, for some reason I would curl my toes down as I got tired and bruise a couple of my toenails. I fixed this with little soft silicon toe sleeves that I could pick up at the pharmacy.

Thanks I was just on that website!

What is AOC? It’s hard to keep up with all the acronyms.

My story: my first real job after grad school was at Texas Instruments. It was a good job, and I enjoyed working there.

A fellow new hire and I were tasked with fixing a machine that ran in a clean room where semiconductor wafers were made. On weekends while the line was down, we would go in, crank up one section of the line and let waste wafers travel through the line where very rarely they might get stuck in a multi-lane machine that would etch the wafers with some sort of acid.

The machine had over two dozen asynchronous motors, actuators, pumps, sensors and so forth. All generating interrupts and I/O events that were sent to a computer that ran the whole line and controlled all the machines.

We couldn’t slow down the machine, it had to run at full speed. The program controlling the machine was thousands of lines of assembly language—everything was assembly language, including the homemade OS that ran the computer running the line. It took like an hour for us to bring up the line and two more hours to see the machine do something strange.

The computer running all this had no user interface other than a some front panel switches and some panel lights that would reveal 16 bits of it’s 128K of memory at a time. This was in the 1970s before Ethernet had been invented.

It felt a bit like those escape room events where you know there is a solution, but you don’t know if you will ever get out. Without my coworker cracking jokes about our plight, I’m not sure we would have ever triumphed over that stupid machine.

What the heck! We were participating in it before? I had heard of the terrible practice and had no idea that the US was doing it too.

From the article: “None of the animals inside the saltwater aquarium, which contained clownfish, teira batfish and palette surgeonfish, survived.”

:(

I’m not an Economist, but doesn’t such a tax lower a firm’s ability to pay its employees (by hiring more or increasing wages) or decrease the firm’s ability to pay dividends to shareholders or force the firm to raise prices.

Generally, it seems like a way to collect more revenue from the population while hiding it as an indirect tax. And it seems a bit like a regressive tax that hits everyone. The fact that they want it to be global makes me feel that they don’t want to compete with other countries that may have “better” economic policies -— whatever that would mean.

I grew up in the 1950s and 1960s in Detroit. It was a place where a high school diploma could get you a secure job that would support a middle class family. It’s all changed and Detroit is now half the size it once was while cities like Houston and Austin have grown tremendously.

Why did that happen? Did corrupt unions kill the goose that laid the golden eggs? Or was it poor management of the automotive giants that allowed overseas companies to take market share. Did wages across the board fall behind inflation or was the automotive fundamentally changed by automation.

I wish I understood what happened since I actually lived through that period.

Only the head of the Prince Rupert drop is tough, due to high levels of stress in the glass (its made by dripping viscous melted glass into water). Interestingly, the whole drop with shatter into glass dust if the filament like tail is broken!

Wikipedia has a good description of this. [1]

[1] https://en.wikipedia.org/wiki/Prince_Rupert's_drop

VSCode is impressive, but to keep it in perspective, consider that an editor with a bug might corrupt a file, an OS with a bug might corrupt an entire file system and at worst a bad OS could actually damage hardware. Windows, MacOS, and Linux or BSDs are incredibly impressive software.

Emacs has around 6000 packages available from the most common repositories. Nine years ago Emacs had over 1.6 Million lines of code. It would be much larger now. How many lines of code in VSCode? I don't know.

However, consider really large projects in lines of code (see [1]):

     Average iPhone app: .... 40,000

     Space Shuttle: ........ 400,000

     Windows 3.1: ........ 2,300,000

     World of Warcraft: .. 5,250,000

     Android OS: ........ 11,800,000

     F35: ............... 24,700,000

     Windows 7: ......... 39,300,000

     Facebook: .......... 61,000,000

     Google: ......... 2,000,000,000

Where does VSCode fit in this list.

[1] https://thenextweb.com/news/google-requires-5000-times-code-...

I hesitate to dive into comparing editors, but here goes. I've used editors for a very long time, I believe the first was TECO sometime in the 1970s at the time I stopped punching cards. My preferred editors now are Emacs, Neovim, Jet Brains IDEs, and VSCode (in that order). All of them are really great professional grade tools.

Emacs supports editing remote files via ssh, ftp, or scp; one simply refers to filenames like: /ssh:host:filename or /ssh:user@host:filename. It's pretty simple and transparent. One can browse remote directories, etc.

Likewise editing a local file under a different user id is also supported using paths that look like /sudo::filename.

See section 18.5 Remote Files in the Emacs user manual[1].

[1] https://www.gnu.org/software/emacs/manual/html_mono/emacs.ht...

I took typing for one semester in high school, 57 years ago. Despite so many years of touch typing, I’ve never been a fast typist. I haven’t slowed down much, but I type around 45 WPM just as I did over 50 years ago. I’ve tried to type faster, but I think my hands are simply not very dexterous.

Because of my slow typing, I probably don’t notice any decline in ability compared to those that were unusually fast in their youth.

Depending on the type of work you do, perhaps a more powerful text editor would be useful. I would suggest Neovim or Emacs. Both are excellent in their own way and both have extensive facilities for handling auto completion of common phrases or markup text and use of abbreviations, etc.

Neovim is a modern version of an editor (vi) that has stood the test of time as one of the fastest editors for navigating over and manipulating text. Shifting between its different modes, each optimized for particular tasks, means it will take a couple of weeks if use to feel natural. The editor itself won’t increase your basic typing speed, but you will be able to fly though your documents correcting and rearranging words and sentences faster than any other tool out there.

Emacs is another ancient editor that is hard to master. Moving and editing text are out of the box a bit slower than Neovim, but the hundreds and hundreds of available packages and extensions make it a high power tool for managing any files or folders on your computer. Like the keys used in Neovim? Emacs can be configured to work the exact same way with a couple of packages. Need to rearrange your directories and rename a bunch of files? There’s really no file manager as good at this as Emacs. Outlines, agendas, check lists, bookkeeping, version control, taking dictation, editing 100 files at the same time, all of this is possible with Emacs. Basic Emacs is probably a bit easier to pick up than Neovim, but it will take time to settle on your own configuration composed of the packages you like best.

You don’t need to be a programmer to use either of these editors, but if you are, they can be easily tweaked to operate however you’d like. Both of them and the packages that expand their capabilities are open source. They run on MacOS, Windows, Linux and even other OS’s and cost nothing.

These editors may not increase your raw WPMs, but they may provide a platform that you will find almost as productive as before due to their many features.

What I want from full self driving is something at least close to hiring a professional chauffeur. I’ve had to do this a few times in Europe. The driver pulls up in an immaculate Mercedes sedan, he drives precisely and carefully with an expert understanding of local traffic and road conditions. This is something I’ve also experienced in Tokyo in every taxi I’ve hailed on the street (except for the brand of vehicle).

Out of politeness I engage in chit-chat with the drivers, but I would be completely comfortable reading hacker news in the backseat or even sleeping. I hope self-driving reaches this level of ability soon; I’m going to be to old to drive in 15 years.

To get a perspective of what this requires, reflect carefully on what your brain is attending to while you drive. Does it ever require higher levels of reasoning than pattern matching? It does for me. I regularly encounter issues that would confuse current technology: bad fog, reckless drivers, icing on overpasses, drivers going the wrong way, a marathon blocking my route, boisterous junior high-school students walking along the curb, detours, temporary lanes inconsistent with actual markings, broken traffic lights, and damaged or vandalized traffic signs. I could sleep soundly on my way being driven by a professional driver under such circumstances.

Not really.