Wikipedia has banned its editors from using AI to create articles, @404mediaco reports. @emanuelmaiberg talked to the Wikipedia editor who proposed the guideline about why.

https://flip.it/fggYt0

#Wikipedia #AI #ArtificialIntelligence #Technology #Tech

Wikipedia Bans AI-Generated Content

β€œIn recent months, more and more administrative reports centered on LLM-related issues, and editors were being overwhelmed.”

404 Media
@TechDesk @404mediaco @emanuelmaiberg
The problem is Wiki already contains a lot of wrong info. I saw an article yesterday which I'm pretty sure was mostly AI slop regurgitaed from a very wrong Wiki page. The problem with it is your next-door-neighbour Joe Blow, who has forgotten the rules of Maths, is totally allowed to admin a page about Maths. Welcome to why I post it here instead. Wikipedia is "like an encyclopedia" in the same way that Madonna is like a virgin
@SmartmanApps @TechDesk @404mediaco @emanuelmaiberg for every troll who edits an article, there’s 5 dedicated editors standing by to correct it. the beauty of wikis is that they’re constantly being fact-checked by tons of experts. @wikipedia even has a dedicated counter-vandalism team!
@SmartmanApps @TechDesk @404mediaco @emanuelmaiberg @wikipedia the problem with traditional encyclopedias is they tend to be biased and can be hard to update. Wikipedia and its many policies bypass all that

@GroupNebula563 @TechDesk @404mediaco @emanuelmaiberg @wikipedia

"for every troll who edits an article" - Professor Rick Norwood isn't a troll https://www.researchgate.net/profile/Rick-Norwood and yet his Maths corrections keep getting backed out by admins. Welcome to why I post my Maths facts on Mastodon, where no-one can back them out https://dotnet.social/@SmartmanApps/110968910722113903

@GroupNebula563 @TechDesk @404mediaco @emanuelmaiberg @wikipedia

"they’re constantly being fact-checked by tons of experts" - and getting backed out again by admins. See previous comment. None of the Maths pages ever cite any Maths textbooks, despite the fact there are many available for free on the Internet Archive

"many policies bypass all that" - including bypassing fact-checking πŸ™„ so either the policies don't work, or aren't followed. Either way Wikipedia has a facts problem

@SmartmanApps @TechDesk @404mediaco @emanuelmaiberg @wikipedia the policies do work, and are followed. they work quite well, in fact. the articles don't cite textbooks, but they *do* cite papers from credible mathematicians/universities, which IMO tend to be far more authoritative. if this rick norwood is a highly credible mathematician, he himself would likely have a wikipedia article, and unless he died in 1675, he does not. https://en.wikipedia.org/wiki/Richard_Norwood
Richard Norwood - Wikipedia

@SmartmanApps @TechDesk @404mediaco @emanuelmaiberg @wikipedia the only "rick norwood" i can find on wikipedia, unless he goes under another username there, is an avid contributor to the project and seems to quite enjoy it: https://en.wikipedia.org/wiki/User:Rick_Norwood
User:Rick Norwood - Wikipedia

@SmartmanApps @TechDesk @404mediaco @emanuelmaiberg @wikipedia ...or he was, until he was diagnosed with alzheimer's and begun making disruptive edits to pages, which are probably the ones getting reverted that you're talking about. not to mention, there's zero indication i can find that the wikipedia editor rick and the researchgate rick are the same person
@SmartmanApps @TechDesk @404mediaco @emanuelmaiberg @wikipedia if you have any further concerns about wikipedia or any other wikimedia projects, i highly suggest taking that up with @LucasWerkmeister, the only person i can think of off the top of my head who is more qualified than i to answer this sort of thing

@GroupNebula563 @TechDesk @404mediaco @emanuelmaiberg @wikipedia @LucasWerkmeister

"the policies do work, and are followed" - clearly not, given pages like https://en.wikipedia.org/wiki/0.999... exist

"IMO tend to be far more authoritative" - I see you haven't read any of their blog posts then, where they can't even get order of operations right (spoiler alert: they don't teach it at university, it's taught in high school, which they've long since left - high school textbooks are the references to use)

0.999... - Wikipedia

Elements of Algebra

Algebra is abstract mathematics - let us make no bones about it - yet it is also applied mathematics in its best and purest form. It is not abstraction for its own sake, but abstraction for the sake of efficiency, power and insight. Algebra emerged from the struggle to solve concrete, physical problems in geometry, and succeeded after 2000 years of failure by other forms of mathematics. It did this by exposing the mathematical structure of geometry, and by providing the tools to analyse it. This is typical of the way algebra is applied; it is the best and purest form of application because it reveals the simplest and most universal mathematical structures. The present book aims to foster a proper appreciation of algebra by showing abstraction at work on concrete problems, the classical problems of construction by straightedge and compass. These problems originated in the time of Euclid, when geometry and number theory were paramount, and were not solved until th the 19 century, with the advent of abstract algebra. As we now know, alge bra brings about a unification of geometry, number theory and indeed most branches of mathematics. This is not really surprising when one has a historical understanding of the subject, which I also hope to impart.

Google Books

@GroupNebula563 @TechDesk @404mediaco @emanuelmaiberg @wikipedia @LucasWerkmeister

""you can see right here where it cites two textbooks" - nope. I can see quite clearly they are NOT Maths textbooks, as I said

"explaining exactly why the article is correct" - now go read about limits and/or decimal representations in Maths textbooks and you'll discover why it's wrong. Here's a free head-start explaining why 1/3 isn't actually equal to 0.333... and is only an approximation (now multiply by 3)

@SmartmanApps @GroupNebula563 @TechDesk @404mediaco @emanuelmaiberg @wikipedia @LucasWerkmeister you have to make the line for "period" over the 3 of 0,3.

@fuchsi @GroupNebula563 @TechDesk @404mediaco @emanuelmaiberg @wikipedia @LucasWerkmeister

"you have to make the line for "period" over the 3 of 0,3" - yes, and is an approximation of 1/3, since it's literally impossible to have an exact decimal representation of 1/3 in base 10, since 3 isn't a factor of 10, as per the textbook

@SmartmanApps @GroupNebula563 @TechDesk @404mediaco @emanuelmaiberg @wikipedia @LucasWerkmeister but it's 3 until infinity. That limes should do that, no?

@fuchsi @GroupNebula563 @TechDesk @404mediaco @emanuelmaiberg @wikipedia @LucasWerkmeister

"but it's 3 until infinity. That limes should do that, no?" - I'm not sure what you mean. Even at 3 to infinity, it's still only an approximation of 1/3, so 0.9 to infinity is only an approximation of 1. ALL non-terminating decimals are only approximations, again as per the textbook. Only terminating decimals are exactly equal to a fraction, such as 0.25 is exactly equal to 1/4, as per the textbook

@fuchsi @GroupNebula563 @TechDesk @404mediaco @emanuelmaiberg @wikipedia @LucasWerkmeister

In other words, no matter how many steps you do your long division to, you're still left with remainder 1 - that remainder 1 literally never disappears, even at preceded by infinite zeroes. It's because 3 isn't a factor of 10, and we're doing it in base ten. In base 3 you can exactly represent 1/3 - it's 0.1 - in base ten you can't. The same thing happens in the other direction converting back.

@SmartmanApps @GroupNebula563 @TechDesk @404mediaco @emanuelmaiberg @wikipedia @LucasWerkmeister That's not how ifinity works. You don't have the remaining 1. That also gets divided by 3.

@fuchsi @GroupNebula563 @TechDesk @404mediaco @emanuelmaiberg @wikipedia @LucasWerkmeister

"You don't have the remaining 1. That also gets divided by 3" - which then also gives a remainder of 1, divide by 3 again, another remainder of 1, ad infinitum. That's EXACTLY how infinity works - a never-disappearing remainder of 1, infinitely repeating 3's. Again, every non-terminating decimal is only an approximation, only terminating decimals are exactly equal to fractions, as per Maths textbooks

@fuchsi @GroupNebula563 @TechDesk @404mediaco @emanuelmaiberg @wikipedia @LucasWerkmeister

"I am not sure that's how it works" - That's exactly how it works. Sit down with pen and paper and start doing long (or short) division. 3 doesn't go into 1, add a zero. 3 goes into 10 3 times with 1 remainder, repeat ad infinitum. Again, every non-terminating decimal is only an approximation due to a never-disappearing remainder, due to not being a factor of 10

@fuchsi @GroupNebula563 @TechDesk @404mediaco @emanuelmaiberg @wikipedia @LucasWerkmeister

And the limit is defined as the number it can never reach, hence the name, limit.

@SmartmanApps @GroupNebula563 @TechDesk @404mediaco @emanuelmaiberg @wikipedia @LucasWerkmeister In your text: read the line with the long term. On the right side you see "=4".
That means if you take the infinite term you can make an integer out of it.

@fuchsi @GroupNebula563 @TechDesk @404mediaco @emanuelmaiberg @wikipedia @LucasWerkmeister
Now read all the underlined parts, especially the double-underlined part. You can lead a horse to water...

"That means if you take the infinite term you can make an integer out of it" - no, had you read it, you would understand that means if you take THE LIMIT you can get an integer out of it. The "infinite term" is still less than it, still less than 4, as per the textbook...

@fuchsi @GroupNebula563 @TechDesk @404mediaco @emanuelmaiberg @wikipedia @LucasWerkmeister
"as far as I understand it it's 4" - the limit is 4, yes, the infinite sum isn't - see double-underlined part (I'm starting to think you didn't read any of it and are just a time-waster)
@SmartmanApps @GroupNebula563 @TechDesk @404mediaco @emanuelmaiberg @wikipedia @LucasWerkmeister And I think you ignored the not underlined parts behind.

@fuchsi @SmartmanApps @GroupNebula563 @TechDesk @404mediaco @emanuelmaiberg @wikipedia @LucasWerkmeister

I don't think he's ever said what he thinks the actual value of any recurring decimal number actually is. 0.333... must be (if not actually β…“) one of:
1. a weird alternative notation for 0.333, and it should be understood that an equation like "β…“ = 0.333..." means "β…“ β‰ˆ 0.333". (In this option, the actual number of "3"s matters!)
2. a representation of the entire series (0.3, 0.33, 0.333, ...)
3. completely undefined. It's just nonsense symbols. This is suggested by some of his other writing which claims you "can't use 0.999... in a proof".

As soon as you say 0.333... is a real number greater than every finite decimal approximation to 1/3, the Archimedean property of the reals forces them to be equal. So I suppose he has a fourth option:

4. it's a real number, but the real numbers are not Archimedean.

when I tried to argue this he ridiculed my assertion that, if a number is less than any of 0.1, 0.01, 0.001, etc, then it is zero, complained I hadn't cited it in a textbook, then blocked me.

@fuchsi @[email protected] @GroupNebula563 @TechDesk @404mediaco @emanuelmaiberg @wikipedia @LucasWerkmeister guessing he blocked you now, too?

He seems to have much less patience now than when I first tried to debunk him. Maybe that's a good thing. Sadly he posted another one of his screeds today to his personal collection of rubbish.

No consensus required for posting to mastodon.

I think his concept of a recurring decimal just be a finite decimal where you haven't chosen the number of digits yet. It's not an actual mathematical object to him. Fascinating.

Ich muss mich entschuldigen, dass er jetzt meinetwegen an dir wieder schreibt.

@SmartmanApps read 'em all. Don't highlight people you're blocking, you intellectually vacant coward.

I'm ready to hear what object you think 0.999... is at any time.

@fuchsi @GroupNebula563 @TechDesk @404mediaco @emanuelmaiberg @wikipedia @LucasWerkmeister
"And I think you ignored the not underlined parts behind" - so rather than leave us in any doubt, you decided to confirm you're a time-waster who couldn't take issue with anything the textbooks said and made up an ad hominem instead. Got it. Bye then.
@SmartmanApps @GroupNebula563 @TechDesk @404mediaco @emanuelmaiberg @wikipedia @LucasWerkmeister Sorry, but I don't have math textbooks in my house. My study time was 20 years ago.
@fuchsi @GroupNebula563 @TechDesk @404mediaco @emanuelmaiberg @wikipedia @LucasWerkmeister
There's plenty of textbooks available for free on the Internet Archive, none of which have been referenced in the Wikipedia page, which is contradicted by what textbooks say https://dotnet.social/@SmartmanApps/115365316775919939 The 2 textbooks I referenced in this thread, which say that infinite decimals are only approximations, and the infinite sum never reaches the limit, are also available for free online, also unreferenced by Wiki
πŸ’‘πš‚π—†π–Ίπ—‹π—π—†π–Ίπ—‡ π™°π—‰π—‰π—ŒπŸ“± (@[email protected])

Attached: 1 image 4/5 We can see in this textbook, it uses a dot above an equals sign to show "x approaches the limit a", and says we can also write it as "lim x=a". I have also seen textbooks say x:=a, where := means "is defined as". Note that in all 3 cases, the textbook has quite conspicuously never said x=a, only variations on the limit of x is defined as a, which we do so that we have a finite number that we can perform further calculations with...

dotnet.social

@SmartmanApps @GroupNebula563 @TechDesk @404mediaco @emanuelmaiberg @wikipedia @LucasWerkmeister Wait, what? Are you saying I should only read books that are not referenced by Wikipedia?

Isn't that a bit extreme from "I would change one sentence about infinity"?

@fuchsi @GroupNebula563 @TechDesk @404mediaco @emanuelmaiberg @wikipedia @LucasWerkmeister
"Wait, what? Are you saying I should only read books that are not referenced by Wikipedia?" - umm, no, and I have no idea why you would say that! I said read Maths textbooks, and the actual issue I'm pointing out is that Wikipedia never manages to reference any. BTW here's some more explicitly stating the the sum is always less than the limit...

@SmartmanApps @GroupNebula563 @TechDesk @404mediaco @emanuelmaiberg @wikipedia @LucasWerkmeister I just looked into the page of Limit analysis and there are 15 sources. 10 of it are math books, 4 are study materials for universities, 1 is a wiki intern link to mathematical proofs.

So I think you are very wrong when you claim Wikipedia doesn't link math books.

Link in my language: https://de.wikipedia.org/wiki/Grenzwert_(Funktion)

Grenzwert (Funktion) – Wikipedia

@fuchsi @GroupNebula563 @TechDesk @404mediaco @emanuelmaiberg @wikipedia @LucasWerkmeister
"So I think you are very wrong when you claim Wikipedia doesn't link math books" - Now go look at the pages that we were actually talking about as being wrong πŸ™„

@SmartmanApps @GroupNebula563 @TechDesk @404mediaco @emanuelmaiberg @wikipedia @LucasWerkmeister you didn't link any pages and I talked about limit analysis. So this is a page I was talking about.

Which pages do mean? I just scrolled up, but there were no links. Am I blind?

πŸ’‘πš‚π—†π–Ίπ—‹π—π—†π–Ίπ—‡ π™°π—‰π—‰π—ŒπŸ“± (@[email protected])

@[email protected] @[email protected] @[email protected] @[email protected] @[email protected] @[email protected] "the policies do work, and are followed" - clearly not, given pages like https://en.wikipedia.org/wiki/0.999... exist "IMO tend to be far more authoritative" - I see you haven't read any of their blog posts then, where they can't even get order of operations right (spoiler alert: they don't teach it at university, it's taught in high school, which they've long since left - high school textbooks are the references to use)

dotnet.social

@SmartmanApps @GroupNebula563 @TechDesk @404mediaco @emanuelmaiberg @wikipedia @LucasWerkmeister ah thanks.

Hm ... There are 69 sources and every one of them is wrong?

@fuchsi sag mich, wenn ich ein paar holen sollte ;)

Wahrscheinlich versteht er sie sowieso nicht...

@fuchsi @GroupNebula563 @TechDesk @404mediaco @emanuelmaiberg @wikipedia @LucasWerkmeister
"There are 69 sources and every one of them is wrong?" - have a look at them and you'll see there's blog posts, and media articles,... and yes, they're all wrong, as proven by Maths textbooks and literal proofs (one of which was bizarrely included as a proof of them being equal, even though in fact it proves they're NOT equal - dude writing it doesn't even seem to understand what he's writing!).
@fuchsi @GroupNebula563 @TechDesk @404mediaco @emanuelmaiberg @wikipedia @LucasWerkmeister
Well, as the Gaslighter pointed out, one of them is another book by Apostol, which he said says the same thing as the one I quoted, and the one I quoted quite clearly says that infinite decimals are only approximations! As I said, dude writing it didn't seem to understand any of it

@SmartmanApps @GroupNebula563 @TechDesk @404mediaco @emanuelmaiberg @wikipedia @LucasWerkmeister Just to make sure:
You do know, that new scientific studies and scholarly articles are published in academic journals, not in textbooks, right?

So it may be very plausible that some niche Wikipedia articles only quote well-known journals or even blog posts of experts in the field?

Edit: I looked and the first two sources are already math books. So exactly what you wished.

@fuchsi
He's saying the only textbooks that are correct are the ones that agree with him, of course ;)

God knows how old the one (or maybe it's more than one) he's using there is; the dotted equal symbol isn't used any more. The symbol S_∞ introduced there has no different meaning than S_n (that is, both are said to "approach" the limit) whereas in a modern textbook (A level maths textbooks for example) say S_∞ = 1 (rather than "approaches 1").

Again he confuses the terms of a sequence with its limit. This is genuinely interesting though, because this textbook uses the ellipsis to mean an unwritten finite number of terms rather than infinitely many, so could really explain why he has this misconception.

@FishFace So I assume math books wording did change?

@fuchsi @[email protected]
"So I assume math books wording did change?" - no, they didn't. Here it is from a 2019 textbook - stop listening to the Gaslighter. πŸ™„ He just proved, yet again, that he lies about what's in textbooks...

"in a modern textbook (A level maths textbooks for example) say S_∞ = 1 (rather than "approaches 1")" - no they don't liar. They say approaches, same as always

@fuchsi Regarding recurring decimals, yes the language used in that book is very much not standard. Actually we should be clear that, in the language of that book (Second Algebra by Hawkes, Luby and Touton, published 1911), SmartmanApps is correct; in that book 0.999... is not 1, because 0.999... does not represent a real number at all. From his screenshot of p176: "The numerical value of 0.666... is a **variable** depending on the number of 6s annexed to the right". (My emphasis)

So this is where he's coming from: for him, 0.999... has *no* fixed value but is more like a variable symbol that is restricted to taking values in the sequence of finite decimals 0.9, 0.99, 0.999, etc. In normal mathematical language we would say it refers to the sequence (or perhaps set) of such values, because the idea that a variable can take on multiple different values within the same piece of mathematical discourse is unnecessary and confusing. (Because if x can take on multiple values, you are forced to accept the statements "x = 0.9" and "x = 0.99" in the same argument, because they might represent different values of x at different times. Can we therefore conclude that x - x = 0.09? No! You must at all times pay attention to whether your statements about a variable were only true contingently. Normal mathematical argument would enclose such sub-scenarios in conditional language, making this explicit.)

I'm not convinced that the wording here "changed" in the sense it being normal in 1911 and rare now. Euler spoke of recurring decimals as definite, not variable quantities in the 1700s, and the textbook Introduction to Algebra by G. Chrystal first printed in 1898 clearly does so too (first screenshot)

Regarding S_∞ actually I think the language hasn't changed, but that book he's quoting is a little sloppily written here. The second screenshot is the modern development, including the "S_∞ = a/(1-r)", from the A-level textbook he's using.

The third screenshot is from the bottom of page 172 of Second Algebra which he clipped, and the top of 173. Notice in fact that the authors abandon the language of "S_∞ approaches..." in favour of giving it a definite value. Their treatment on p173 is in fact quite clear and in contradiction with the Smart Man's conclusion: "when we speak of the sum of the series, we **mean the limit**". From what he says, I don't think he is quite capable of understanding what a mathematician is saying when they say, "by these words we mean this thing", else he wouldn't have underlined all those words which contradict him :P

Further to this, on p175 there is the fourth screenshot, which says that (when x < 1), "the sum of the series 1 + x + xΒ² + xΒ³ + ... is a definite number". "Definite numbers" are just themselves; they don't change and cannot "approach" anything.

I am ashamed that I skipped over his screenshot of p176, because that finally actually clarifies what he thinks recurring decimal notation means in a way that his posts dating back to 2024 failed to. Presumably he also thinks infinite summation is similarly a "variable" waiting for you to decide how many terms to add up.

Sadly he doesn't have the ability to explain this: he doesn't write, "0.999... β‰  1 because the notation on the left does not denote a specific number, but is a placeholder for one of 0.9, 0.99, 0.999, ... and 1 is not any one of those". Instead he ridicules and dodges questions. And sadly, while he'll probably never respond to this way of putting things, even if he did he would not be able to understand or admit that his preferred language, from one textbook written over 100 years ago, is not standard and that in the language used by everyone else, 0.999... denotes a single, definite number, and that number is 1.

I think this way of writing maths invites such imprecise thinking though - do you agree? If we see a symbol used *as a number* the way recurring decimals universally are, then I think it will always lead to people treating them as a particular number.

@fuchsi haha, it's very unfortunate that he denied the existence of the "Key Point 6.12" mere minutes before I linked it.

@FishFace I think you both should kiss and make up.

I think you are both more passionate about mathematical problems than I ever was. (Even though I competed in math olympiads as a child.)

@fuchsi that would indeed be lovely! We used to have such fun together.

I think my passion stems from being confronted with a disagreement on the internet that, unlike almost all others, suggests that it can be fully and definitively resolved solely by rational argument. How wonderful that would be!

But the reality is that all mathematics requires language to communicate, and no language is immune from being interpreted in more than one way, as we have seen.

@fuchsi @[email protected]
"passionate about mathematical" - I'm passionate about #debunking #Gaslighters and their #bullying. He's just passionate about Gaslighting/bullying people. Note in the following how often he edited out the references which contradict him...

"The numerical value of 0.666... is a **variable** depending on the number of 6s annexed to the right". (My emphasis)" - also your EDIT, where you deliberately omitted V is the Variable being described, hence your lack of a screenshot

@fuchsi @[email protected]
"Notice in fact that the authors abandon the language of "S_∞ approaches..." in favour of giving it a definite value" - notice that your screenshot in fact deliberately omits the very next paragraph, which says that the sum of the terms is approaching the limit πŸ™„

"when we speak of the sum of the series, we **mean the limit**"" - says person who can't provide a screenshot of them saying that anywhere, given they make the quite clear distinction between them repeatedly

@fuchsi @[email protected]
""Definite numbers" are just themselves; they don't change and cannot "approach" anything" - and, yet again, your screenshot leaves out the very next paragraph, which clearly states "the sum of whose terms approach a limit". Gaslighters gonna gaslight

@fuchsi

BITTE sag mir, wenn es dich Àrgert, dass ich dich als Postkiste benutze. Ich weiß ja, dass es nicht nâtig ist, alles hier *dir* zu sagen. Wenn du so sagst, hâre ich sofort auf.

He is right that there are many on archive.org. I actually already went through and found six more examples which explicitly write that some rational number is equal to an infinite decimal ending in all nines:
* Elementary real analysis by Kenneth W. Fisher (p158)
* Basic real analysis by Houshang H. Sohrab (sec. Binary Ternary, Decimal, etc Expansions)
* Intermediate real analysis by Emanuel Fisher (problem 3.2)
* Elementary real analysis by Brian S. Thomson (exercise 2.3.6 - screenshot is of the hint)
* The Fundamentals Of Mathematical Analysis Volume I by G. M. Fikhtengolts (p8)
* Elements of Algebra by Stilwell (exercise 3.3.2).

Screenshots of the first 4 are attached, should anyone wish to confirm and add those references to the Wikipedia article.

We can only await what creative interpretations of these he'll come up with. Beware though: he doesn't actually care about references; we have 8 now that treat the subject directly, with an equation written out that contradicts him. He can't produce a single reference that writes "0.999... β‰  1". Every single textbook he will either ignore, dismiss as wrong, or torture its *English* content so badly as to invert its meaning. It's impressive.

Did you also notice that he said that Wikipedia doesn't reference the textbooks he's using, but one of them is Apostol's Calculus, and the article references Apostol's Mathematical Analysis which uses the exact same section on decimal expansions. Of course, it is not available to borrow on the Internet Archive, so he didn't check it, but still complains that the identical source is missing XD

@fuchsi @SmartmanApps @GroupNebula563 @TechDesk @404mediaco @emanuelmaiberg @wikipedia @LucasWerkmeister
Bahahaha he can't understand the textbooks he holds in such high esteem
@FishFace
Let me help you out with the parts you tried to distract people away from with your underlining Mr. Gaslighter πŸ˜‚