What if it is and it became unrecoverable ages ago?
I don’t really care about the Confederate thing, but there are two Fort Braggs, one a town in coastal California, and it was kind of obnoxious to have the same name on them.
https://en.wikipedia.org/wiki/Fort_Bragg%2C_California
It’s notable for Glass Beach:
https://en.wikipedia.org/wiki/Glass_Beach_(Fort_Bragg%2C_California)
https://en.wikipedia.org/wiki/Fort_Bragg
So I kind of liked having it not be “Bragg”.
Also, if I remember correctly from the American Civil War military history I’ve done, Bragg (the general, not the private) didn’t actually perform very well across a number of battles. Like, he was high-ranking, but I’m not sure that he’d be someone to name forts after.
kagis
Yeah, actually, sounds like he was outright awful, in fact.
https://en.wikipedia.org/wiki/Braxton_Bragg
At the start of the Civil War, Bragg trained soldiers in the Gulf Coast region. He was a corps commander at the Battle of Shiloh, where he launched several costly and unsuccessful frontal assaults but nonetheless was commended for his conduct and bravery.
In June 1862, Bragg was elevated to command the Army of Mississippi (later known as the Army of Tennessee). He and Brigadier General Edmund Kirby Smith attempted an invasion of Kentucky in 1862, but Bragg retreated following a minor tactical victory at the Battle of Perryville in October. In December, he fought another battle at Murfreesboro, Tennessee, the Battle of Stones River, against the Army of the Cumberland under Major General William Rosecrans. After a bloody and inconclusive battle, it ended with his retreat. After months without significant fighting, Bragg was outmaneuvered by Rosecrans in the Tullahoma Campaign in June 1863, causing him to surrender Middle Tennessee to the Union. Bragg retreated to Chattanooga but evacuated it in September as Rosecrans’ troops entered Georgia. Later that month, with the assistance of Confederate forces from the Eastern Theater under James Longstreet, Bragg was able to defeat Rosecrans at the Battle of Chickamauga, the bloodiest battle in the Western Theater, and the only significant Confederate victory therein. Bragg forced Rosecrans back into Tennessee, but was criticized for the heavy casualties his army suffered and for not mounting an effective pursuit. In November, Bragg’s army was routed by Major General Ulysses S. Grant in the Battles for Chattanooga and pushed back to Georgia. Confederate President Jefferson Davis subsequently relieved Bragg of command, recalling him to Richmond as his chief military advisor. Bragg briefly returned to the field as a corps commander near the war’s end during the Campaign of the Carolinas.
Bragg is generally considered among the worst generals of the Civil War.[1] Most of the battles he engaged in ended in defeat. Bragg was extremely unpopular with both the officers and ordinary men under his command, who criticized him for numerous perceived faults, including poor battlefield strategy, a quick temper, and overzealous discipline.[1] Bragg has a generally poor reputation with historians,[1] though some point towards the failures of Bragg’s subordinates, especially Major General and former Bishop Leonidas Polk—a close ally of Davis and known enemy of Bragg—as more significant factors in the many Confederate defeats under Bragg’s command. The losses suffered by Bragg’s forces are cited as highly consequential to the ultimate defeat of the Confederate States of America.[1]
When I’ve interviewed candidates for software engineering positions, I have never asked any questions about them outside of work. I don’t personally feel that it has a lot of impact on how they perform at work.
A lot wealthier, but that’s in significant part because most of the world is very poor relative to most people reading this. Some guy in Sudan isn’t gonna be reading your post.
https://en.wikipedia.org/wiki/List_of_countries_by_wealth_per_adult
Median global wealth per adult (2023):
North America: $108,918
Europe: $28,611
China: $27,273
World: $8,654
Latin America: $6,341
Asia-Pacific (excluding China and India): $5,176
India: $3,755
Africa: $1,242
I think that you’re going to have a hard time getting fine-grained data globally, though, because lots of countries just don’t gather data and those that do often don’t measure it in the same way. I really wish that the UK had stayed in Eurostat when leaving the EU, because now getting comparable data for the UK and EU isn’t doable.
If it’s the US (in 2021 dollars), on a per-household basis:
https://www.census.gov/content/dam/Census/library/publications/2023/demo/p70br-183.pdf
Median Wealth by Household Characteristics: 2021
Typically someone tends to build up assets during their working life, and spend them in retirement, so age is gonna be a factor.
Age of Householder
Age | Wealth ($) |
---|---|
Under 35 | 30,500 |
35 to 44 | 126,900 |
45 to 54 | 186,000 |
55 to 64 | 276,000 |
65 to 69 | 341,400 |
70 to 74 | 373,900 |
75 and older | 315,900 |
Highest Level of Educational Attainment in Household
Education | Wealth ($) |
---|---|
No high school diploma | 8,460 |
High school graduate | 55,030 |
Some college, no degree | 90,810 |
Associate degree | 139,000 |
Bachelor’s degree | 266,600 |
Graduate or professional degree | 555,900 |
Annual Household Income
Quintile | Wealth ($) |
---|---|
First | 12,000 |
Second | 61,260 |
Third | 145,200 |
Fourth | 269,100 |
Fifth | 805,400 |
Also, just for good measure, since I suppose that “gold” would fall into “other asset holdings”, you’d be pretty high relative to the norm in that category:
Composition of Wealth by Asset Type
Category | Percent |
---|---|
Retirement accounts | 34.1% |
Equity in own home | 28.5% |
Stocks and mutual funds | 11.9% |
Assets at financial institutions | 8.1% |
Other asset holdings | 4.6% |
Rental properties | 4.3% |
Business assets | 3.9% |
Other real estate | 3.6% |
Vehicles | 3.2% |
Bonds | 0.7% |
Student loan and education-related expenses | -1.5% |
Credit card and store bills | -0.5% |
Medical debts | -0.5% |
Other unsecured debts | -0.4% |
I was away from things for a day and when I came back everything as wizards and orbs.
Let this be a lesson to you.
From top-to-bottom:
Skynet from The Terminator
Joshua from WarGames
They are, however, able to inaccurately summarize it in GLaDOS’s voice, which is a strong point in their favor.
to ensure that the technology is “safe, secure and trustworthy.”
None of the really iconic AIs are safe, secure or trustworthy.
If Justin Trudeau gets active on this, you can probably get it to Gulf of Canada (Gulf of Mexico) (Gulf of America).
What you’re describing is called a linter, and they’ve existed for ages.
Yup, and I’ve used them, but they’ve had hardcoded rules, not had models just trained on code.
The only place I’ve seen prices listed that high in the US is in California.
California apparently has some sort of minimum cage size mandate that a lot of the rest of the US doesn’t, so can’t pull in eggs from the rest of the US, which apparently contributes to California’s problems, since it fragments the US market. Which is probably pretty great if you’re an egg producer in California who hasn’t been hit by bird flu – you’ve got a protected market, and a lot of your competition has been wiped out – but sucks if you’re an egg consumer.
Bird flu continues to play a part in higher egg prices in California.
The U.S. Department of Agriculture, in a Jan. 10 report, said a dozen large shell eggs in the state rose to $8.97.
Some states, like California, are being hit especially hard by the egg crunch, and part of that is likely a result of state-level legislation.
California’s Proposition 12, also called the Farm Animal Confinement Initiative, places restrictions on how hens, sows and veal calves can be kept.
The bill, which took effect in recent years, in part banned confinement of egg-laying hens (chickens, turkeys, ducks, geese and guinea fowl) in certain areas with less than 1 square foot of usable floor space per hen.
Other states, including Arizona, Colorado, Massachusetts, Michigan, Nevada, Oregon and Washington, have similar laws that specifically provide animal welfare protections to egg-laying hens.
That limits how eggs can be produced and what can be sold in each state. Those that allow only cage-free products already face fewer suppliers and farms (a little more than a third of U.S. egg layers are cage-free, according to the USDA). Manufacturers and sellers also are facing a slowdown as they change operations to comply with such laws.
Find out what people are upset about. Go put on a large performance and get in the news in such a way that you give the impression that you’re doing that. Most people will not go looking for data, rely on their gut impression from the tone of news coverage.
His campaign for his first term was primarily run on protectionist US trade relations and stopping illegal immigration via building a wall.
He put on a major act out of “killing” TTIP and TPP, which had negotiations that had already failed. He slightly tweaked and renamed NAFTA. He used very aggressive phrasing, broke norms for communication, acted extreme, kept himself constantly in the news, and after the fact, stated that he did a great job. Few people go look at actual data. People who wanted these things had the vague idea that he’d dramatically changed things. What he’d done was to provide them with political theater aimed at giving voters the impression that he’d done that.
He spent a lot of time in the news talking about The Wall, fighting for The Wall. The typical voter got the impression that he was going to build a wall along the whole border, and talking about how he wouldn’t spend anything on it, how Mexico would pay for it. He broke social norms for communication, used aggressive phrasing, acted extreme, kept himself constantly in the news. The federal government paid to build a little wall, principally maintenance of existing fencing. Mexico did not pay for it. What he’d done was to provide them with political theater aimed at giving voters the impression of what he’d done.
He spent a lot of time in the news talking about illegal immigrants, fighting them, staying in the news on it. He broke social norms for communication, used aggressive phrasing, acted extreme, kept himself constantly in the news. And what actually happened:
In 2020, the removal of illegal immigrants from the interior of the United States was the lowest as an absolute number and as a share of the illegal immigration population since ICE was created in 2003
What he’d done was to provide voters with political theater aimed at giving voters the impression that that’s what he’d done.
Right now, Trump is constantly in the news for breaking communication norms, using aggressive phrasing, constantly staying in the news over eliminating waste. I think that Musk did say that he’d eliminated some dollar amount of waste, albeit without specifics as to that waste. Ditto for illegal immigrant deportations – lots of photographs and strong phrasing, not a lot of data. Probably other stuff, didn’t pay attention to his campaigning for second term, so don’t know what supporters were promised. Oh, right, foreign aid.
https://www.brookings.edu/articles/what-every-american-should-know-about-u-s-foreign-aid/
Opinion polls consistently report that Americans believe foreign aid is in the range of 25 percent of the federal budget. When asked how much it should be, they say about 10 percent. In fact, at $39.2 billion for fiscal year 2019, foreign assistance is less than 1 percent of the federal budget.
This means that a lot of Americans feel that they’re being taken advantage of, are upset about that 25% of the federal budget.
Trump “destroyed” USAID, was constantly in the news for doing so, said a lot of extreme things. Musk said something about “feeding it into the wood chipper”. In practice, he was in the news for suspending operations and a small number of layoffs, and IIRC there was a reorganization that placed it under State, so for some definitions, it ceased to exist. I have not seen anything about long-run actual dollar value change to foreign aid, but if I had to guess based on his past MO, it won’t change a lot. But I bet that at the end of his term, a lot of voters will have the impression that it has.
“Tech workers” is pretty broad.
Tech Support
There are support chatbots that exist today that act as a support feature for people who want to ask English-language questions rather than search for answers. Those were around even before LLMs, could work on even simpler principles. Having tier-1 support workers work off a flowchart is a thing, and you can definitely make a computer do that even without any learning capability at all. So they definitely can fill some amount of role. I don’t know how far that will go, though. I think that there are probably going to be fundamental problems with novel or customer-specific issues, because a model just won’t have been trained on it. I think that it’s going to have a hard time synthesizing an answer from answers to multiple unrelated problems that it might have in its training corpus. So I’d say, yeah, to some degree, and we’ve successfully used expert systems and other forms of machine learning in the past to automate some basic stuff here. I don’t think that this is going to be able to do the field as a whole.
Writing software
Can existing LLM systems write software? No. I don’t think that they are an effective tool to pump out code. I also don’t think that the current, “shallow” understanding that they have is amenable to doing so.
I think that the things that LLMs work well at is in producing stuff that is different, but appears to a human to be similar to other content. There are a variety of uses that that work, to varying degrees, for content consumed by humans.
But humans deal well with errors in what we see. The kinds of errors in AI-generated images aren’t a big issue for us – they just need to cue up our memories of things in our head. Programming languages are not very amenable to that. And I don’t think that there’s a very effective way to lower that rate.
I think that it might be possible to make use of an LLM-driven “warning” system when writing software; I’m not sure if someone has done something like that. Think of something that works the way a grammar checker does for natural language. Having a higher error rate is acceptable there. That might reduce the amount of labor required to write code, though I don’t think that it’ll replace it.
Maybe it’s possible to look for common security errors to flag for a human by training a model to recognize those.
I also think that software development is probably one of the more-heavily-automated fields out there because, well, people who write software make systems to do things over and over. High-level programming languages rather than writing assembly, software libraries, revision control…all that was written to automate away parts of tasks. I think that in general, a lot of the low-hanging fruit has been taken.
Does that mean that I think that software cannot be written by AI? No. I am sure that AI can write software. But I don’t think that the AI systems that we have today, or systems that are slightly tweaked, or systems that just have a larger model, or something along those lines, are going to be what takes over software development. I also think that the kind of hurdles that we’d need to clear to really fully write software from an AI require us to really get near an AI that can do anything that a human can do. I think that we will eventually get there, and when we get there, we’ll see human labor in general be automated. But I don’t think that OpenAI or Microsoft are a year away from that.
System and network administration
Again, I’m skeptical that interacting with computers is where LLMs are going to be the most-effective. Computers just aren’t that tolerant of errors. Most of the things that I can think of that you could use an AI to do, like automated configuration management or something, already have some form of automated tools in that role.
Also, I think that obtaining training data for this corpus is going to be a pain. That is, I don’t think that sysadmins are going to generally be okay with you logging what they’re doing to try to build a training corpus, because in many cases, there’s potential for leaks of sensitive information.
And a lot of data in that training corpus is not going to be very timeless. Like, watching someone troubleshoot a problem with a particular network card…I’m not sure how relevant that’s going to be for later hardware.
Quality Assurance
This involves too many different things for me to make a guess. I think that there are maybe some tasks that some QA people do today that an LLM could do. Instead of using a fuzzer to throw input in for testing, maybe have an AI to predict what a human would do.
Maybe it’s possible to build some kind of model mapping instructions to operations with a mouse pointer on a screen and then do something that could take English-language instructions to try to generate actions on that screen.
But I’ve also had QA people do one-off checks, or things that aren’t done at mass scale, and those probably just aren’t all that sensible to automate, AI or no. I’ve had them do tasks in the real world (“can you go open up the machine seeing failures and check what the label on that chip on the machine that’s getting problems reads, because it’s reporting the same part number in software”). I’ve written test plans for QA to run on things I’ve built, and had them say “this is ambiguous”. My suspicion is that an LLM trained on what information is out there is going to have a hard time, without a deep understanding of a system, to be able to say “this is ambiguous”.
Overall
There are other areas. But I think that any answer is probably “to some degree, depending upon what area of tech work, but mostly not, not with the kind of AI systems that exist today or with minor changes to existing systems”.
I think that a better question than “can this be done with AI” is “how difficult is this job to do with AI”. I mean, I think that eventually, pretty much any job could probably be done by an AI. But I think that some are a lot harder than others. In general, the ones that are more-amenable are, I think, those where one can get a good training corpus – a lot of recorded data showing how to do the task correctly and incorrectly. I think that, at least using current approaches, tasks that are somewhat-tolerant of errors are better. For any form of automation, AI or no, tasks that need to be done repeatedly many times over are more-amenable to automation. Using current approaches, problems that can be solved by combining multiple things from a training corpus in simple ways, without a deep understanding, not needing context about the surrounding world or such, are more amenable to being done by AI.
Too bad there aren’t any non-profit search engines you could promote instead of the one that charges people money in order to make a profit.
If I remember from prior discussions, you prefer Duck Duck Go. If you want to mention that you use Duck Duck Go, I have no problem with that. I think that that’s great.
giving a corporation that would happily fuck you over in a second like every other corporation
I think that Kagi has considerably less-incentive to do so than Duck Duck Go does, because they have a viable revenue model that doesn’t involve datamining me the way Google does or showing ads to me the way Duck Duck Go does. Yes, you can use an ad-blocker on Duck Duck Go, but then you’re offloading the costs onto other users who don’t do that, and in the long run, Duck Duck Go has an incentive to block users using ad blockers.
You may disagree with my assessment. But I’ve made that decision, I’m happy with it, I like the fact that Kagi added a Threadiverse search feature, and I am not going to change search engines to your favorite search engine, nor do I intend to stop telling people that I use Kagi.
You’re free to comment every time if you want. I have no intention to change what I am doing, because I happen to like them, and my use of the term predates my use of that engine – I wrote googles prior to this.
If you want to ban me because you cannot tolerate my writing style, do so, and I’ll go use communities other than those you that you moderate. Trying to harass me into changing what I write is not going to have an effect.
Honestly, I’m a little surprised that a smartphone user wouldn’t have familiarity with the concept of files, setting aside the whole familiarity-with-a-PC thing. Like, I’ve always had a file manager on my Android smartphone. I mean, ok…most software packages don’t require having one browse the file structure on the thing. And many are isolated, don’t have permission to touch shared files. Probably a good thing to sandbox apps, helps reduce the impact of malware.
But…I mean, even sandboxed apps can provide file access to the application-private directory on Android. I guess they just mostly don’t, if the idea is that they should only be looking at files in application-private storage on-device, or if they’re just the front end to a cloud service.
Hmm. I mean, I have GNU/Linux software running in Termux, do stuff like scp
from there. A file manager. Open local video files in mpv
or in PDF viewers and such. I’ve a Markdown editor that permits browsing the filesystem. Ditto for an org-mode editor. I’ve a music player that can browse the filesystem. I’ve got a directory hierarchy that I’ve created, though simpler and I don’t touch it as much as on the PC.
But, I suppose that maybe most apps just don’t expose it in their UI. I could see a typical Android user just never using any of the above software. Not having a local PDF viewer or video player seems odd, but I guess someone could just rely wholly on streaming services for video and always open PDFs off the network. I’m not sure that the official YouTube app lets one actually save video files for offline viewing, come to think of it.
I remember being absolutely shocked when trying to view a locally-stored HTML file once that Android-based web browsers apparently didn’t permit opening local HTML files, that one had to set up a local webserver (though that may have something to do with the fact that I believe that by default, with Web browser security models, a webpage loaded via the file://
URI scheme has general access to your local filesystem but one talking to a webserver on localhost does not…maybe that was the rationale).
I’m fairly sure that I’ve read some articles by this guy on Forbes before, because I remember that he had some article on something, many years back, that I really liked and I distinctly remember thinking that his thumbnail looked kind of frumpy. I believe that he’s British. Looks like he hasn’t been at Forbes in almost a decade, though.
kagis
Yeah, was apparently born in Torquay, England.
https://www.timworstall.com/2008/07/about-tim-worstall/
I was born in Torquay in 1963, grew up mostly in Bath (with a couple of years in Naples, Italy as a result of my father\’s Naval career) and was educated at Downside Abbey.
EDIT: Also, a “ton” and a “tonne” aren’t the same thing – that’s not just dialect. A “ton” here in the US means a short ton, 2000 US lbs. A “tonne” is a metric ton, 1000 kg. I don’t know what Brits normally mean if they write “ton”, whether it’s a short ton or a long ton or metric ton. In the US, we’d normally write “metric ton” instead of “tonne”.
If we want to have artificially-generated demand for zinc – if we really need to ensure domestic production capacity – there’s no requirement for it to be the penny. I’m sure that we can find something else to make out of zinc.
The penny itself wasn’t always zinc. I don’t remember the changeover year.
checks WP
https://en.wikipedia.org/wiki/Penny_(United_States_coin)
The current copper-plated zinc cent issued since 1982 weighs 2.5 grams, while the previous 95% copper cent still found in circulation weighed 3.11 g (see further below).
https://en.wikipedia.org/wiki/Zinc
Zinc is most commonly used as an anti-corrosion agent,[123] and galvanization (coating of iron or steel) is the most familiar form. In 2009 in the United States, 55% or 893,000 tons of the zinc metal was used for galvanization.[122]
Zinc is more reactive than iron or steel and thus will attract almost all local oxidation until it completely corrodes away.
We can just subsidize zinc production, or purchase something that requires those anti-corrosion properties.
I also am not at all sure that this was in fact the rationale. I can’t find a reference online to this being the rationale. I do see reference to zinc being useful because it’s particularly inexpensive. And the numbers given on this article seem to support the idea that pennies don’t really work out to generating a very substantial demand for zinc.
It’s Not Big Zinc Behind The Campaign To Keep The Penny
To run through the numbers, a penny coin weighs 2.5 grammes. Let’s call that all zinc (it’s not, but close enough). There’s 5 billion made a year, meaning that we’ve got 12,500,000,000 grammes, or divide by a million to get 12,500 tonnes. Now, if that were 12,500 tonnes of gold being made into coins every year, with global virgin production being around 3,000 tonnes, then sure, that would be a contract worth, umm, influencing the political process, to secure and keep running. The same would be true of many metals in fact. But it’s just not true of the zinc industry. Using the USGS, the correct source for these sorts of numbers, we find that US production of zinc is around 250,000 tonnes a year, global production 13.5 million. Even if we assume (as we might, sounds like the sort of thing that might be true) that US coins must be made of US produced metal this is still a very marginal part of the total market.
Further, zinc runs about $2,200 a tonne at present, meaning that we’re talking about maybe $25 million a year as the zinc cost of our pennies. And we’re told who and how much is paid to keep lobbying for the penny:
But his written statement did not mention that Weller is actually a lobbyist and head of strategic communications for Dentons, a law firm representing the interests of zinc producer Jarden Zinc Products, a major provider of coin blanks that are made into currency.
…
Jarden Zinc Products spent $1.5 million from 2006 through the first quarter of 2014 lobbying on such things as “issues related to the one-cent coin” and represented by Weller when he worked at B&D Consulting and, more recently, Dentons.
No, the important point here is not the zinc industry, nor “Big Zinc”. The important part is this “a major provider of coin blanks”. If your business is making coin blanks then obviously you’re very interested in the continued existence of coin demoninations that are made from coin blanks. That they’re made from zinc is an irrelevance compared to that.
Believe me, you don’t spend the best part of $200,000 a year in lobbying expenses in order to sell $25 million’s worth of zinc. This metal is a commodity, you can sell that amount in one ten minute phone call to any London Metals Exchange ring member. Heck, give me a couple of days to get organised and I could sell it for you at the market price. I’d also charge rather less than $200,000 to do it.
EDIT: WP seems to also support the author’s argument.
https://en.wikipedia.org/wiki/Jarden_Zinc_Products
The company has resisted past efforts to eliminate the penny in the United [1] through an astroturf lobby organization called Americans for Common Cents.
The company’s largest source of revenue comes from the production of coin blanks, having produced over 300 billion blanks at their Tennessee facility.
The penny, one of the first coins made by the U.S. Mint after its establishment in 1792, now costs more than two cents to produce, Trump said in a post on his Truth Social site shortly after departing the Super Bowl game in New Orleans.
“For far too long the United States has minted pennies which literally cost us more than 2 cents. This is so wasteful!” Trump wrote. “I have instructed my Secretary of the US Treasury to stop producing new pennies.”
While I think that it’s probably something that we should have done a long time ago, I don’t think that the major cost is actually the fact that the production cost is higher than the coin’s face value, but from the cost of needing to handle and process pennies.
Don’t they support video posts?
kagis
Hmm. Apparently so.
https://help.tumblr.com/knowledge-base/posting-video/
Well, it’s not an open platform, but I guess it may be a platform with the resources to serve some serious video, and it’ll be on the Fediverse.
We have been having a lot of discussions about what it would take to get video on the Fediverse at more than PeerTube scale.
I don’t use tumblr, don’t know if they provide a video-centric interface, but I imagine that one could always write a software package to index those videos and link to them. Maybe PeerTube can already do that, haven’t played with it enough to know.