Welcome to Gaia! :: View User's Journal | Gaia Journals

For my History of Math class I have to watch these final presentations of my peers and some of them are AWFUL. The presentation expectation for length was 15 minutes (conference standard) and one guy did under 7 minutes on Eratosthenes. It was so lacking in information that I decided to peruse wikipedia for 15 minutes and found out a ton of cool s**t. The presentation basically said "he died in (year) in (place), he was born in (year) in (place) and he was a physicist, mathematician, inventor, and geographer".

Like, my guy was done DIRTY by this. Eratosthenes was also a poet for example which is pretty cool. But he also is considered the FOUNDER OF GEOGRAPHY. That's kinda a big deal to omit... He also went blind later in life which made him depressed because he couldn't do anything he loved like looking at the scenary so he KILLED HIMSELF BY STARVATION. How do you leave that out?

And the presenter talks about how none of his works survived, and just leaves it there. But like do you know WHY? Cuz in addition to all of the above titles, he was a librarian at the library of Alexandria, and so all of his written works were in the library when it burned down. AGAIN, kinda a big deal.

He then goes on to talk about the Sieve of Eratosthenes and claims its not that significant because it's "not really used that much". Like, dude, what? It's studied in many computing science classes, especially when time complexity is involved. It is a great example of small improvements to algorithms that can drastically improve the time complexity, and -- I didn't know this before -- in my 15 minute dive I learned that EULER USED THIS TO PROVE THE RIEMANN-ZETA FUNCTION. Not used my a**. ******** do better dude.

AND THEN he describes the Sieve as "counting the number of primes less than or equal to a given number". FALSE. It tells you which numbers less than or equal to a given number are prime. From this you can then easily count them. There isn't much of a need for the ancient greeks, as far as I know, to count the number of primes below a number, though this is related to the Riemann Zeta function and the Riemann Hypothesis thousands of years later, knowing IF a number is prime, and building a table of primes would be very useful. And since the algorithm is infinitely expandable given enough time and a table of primes below a certain number we can always build on top of this. For example if I knew all of the primes up to a number, say m, I could easily use that and the sieve to carry this idea on from m to n where n is some number greater than m. Meaning that researchers and educators could have a table of primes, and then expand this table easily using the sieve if maybe they needed to know if a certain number was prime.

And finally, the reason I'm so heated about this, Eratosthenes was known to criticize Aristotle for the way he "split the world into Greeks and Barbarians" and further criticized him for arguing that the Greeks should strive to maintain racial purity. My guy was an anti-racist too. I mean, I spent 15 minutes on wikipedia, and could've done a more interesting presentation.

BUT THEN, this other presenter... There's one rule: If you talk about Alan Turing in my presence you are likely to get corrected. But his presentation was on Turing's Bombes of WW2, which I'm not that versed in, so I thought "Hey this could be okay". It was not. It was not okay.

Problem #1: Turing Machines != Bombes.
When I did my presentation on Turing Machines, the actual Turing Machines I made a point to distinguish between those and the Bombes. THIS GUY describes the bombes as the way Turing Machines functions: A magnetic tape, with a read/write head and a state machine. This is not at all how Bombes worked, though the implementation could be similar as the nature of Turing Machines implies that bombes can be simulated on a turing machine, but nevertheless, that is not what is special about bombes. Being a Turing Machine is not the brilliance of a bombe.

The Bombes were special because they exploited a fatal flaw of Enigma: the plug board design meant that no letter could map to itself on a press. This meant if you assumed that a mapped to t, then you could verify a rotor configuration by virtue of contradiction: Assume the configuration follow the logic until you find a mapping that requires a character going to itself (eg: a maps to a). Turing then realized we could assume that if we found this contradiction then all of the entrailed statements in addition to the original assumption must be held false. In other words, if the assumption a maps to t implies that p maps to q and that implies x maps to x then p mapping to q must be false in addition to a mapping to t. This drastically speeds up the time it takes to check a configuration and as a result, checking all of the possible configurations was surprisingly fast.

Problem #2: Banburismus was not an unrelated invention of Turings.
The presenter mentioned the invention of Banburismus by Turing, which leads to a scoring unit of Bans. These essentially, as I understand them, score the probability that a certain configuration is the real one given some known word in the sentence. The presenter ignores the fact that these drastically sped up the checking time of the Bombes, effectively breaking it further. It also ignores that these are essentially just bayes theorem, which is a point that Turing's Assistant made to Turing, Cook, made to Turing to which Turing agreed.

Problem #3: Turing Complete != Halting
I dunno where this person got this idea, but they claimed that Turing Complete means that the Turing Machine has completed(?). That is not even close to the definition of Turing Completeness. It's not even an attempt. The more accurate term for that would be "halting", as in, the halting problem, as in the machine comes to a halt. Turing Completeness is how one would describe an abstract machine that can simulate any turing machine. In other words, it can do anything that a Turing Machine can do. This is important because programming languages are kinds of machines, and we need this programming languages to do everything a turing machine can do. Turing Machines are fundamentally the best you can get in terms of solving problems, and Turing Complete machines are just as powerful as Turing Machines. For example, Python, C, C++, Java, Lambda Calculus, Powerpoint, and Minesweeper are all Turing Complete. Not because they halt. But because they are powerful enough to simulate turing machines.

Problem #4: Turing, but not his history?
I find it moderately offensive, as a queer person in the field of computing, that one would talk about Turing, especially the work he did in world war 2, and NOT his conviction for being a homosexual. His work effectively shortening the war had little effect on the severity of his sentence, a sentence that would lead to his suicide, and to gloss over these facts with not even a nod is a little wrong, almost revisionary. Turing's Bombes were so essential, hundreds of them were distributed all over the UK (fearing that a single bomb could take out Bletchley park and they'd not be able to break enigma for a long time). There are massive estimates of Turings effect that cannot be understated. Why didn't they give him any slack? Because the government feared blackmail over having gay people in their military... ffs.

Unrelated to this, I must note, be wary should you ever watch the Imitation Game. I have many gripes with its construction of the character of Turing. Turing was an ecclectic for sure, but the way they portray him as a stereotypical autistic person is problematic. For one thing, this trope in society as painting all genius as autistic and anti-social creates the narative that all autistic people are mathematical and scientific geniuses with no social skills. This is not necessarily true, and is a frustratingly problematic trope. But furthermore, Turing's anti-social protrayal is not even accurate. In actuality, Turing was perhaps awkward, but Olive Bailey (Whom he was engaged to) described him as having a "lovely sense of humor" which never comes out in the movie. She also critiques to movie for not getting his character quite right.

One friend of Turings admitted that he was quite naievely open about his homosexual relationships (which got him found out in the first place, due to a burglary), and he often joked about the conviction leading up to his guilty plead. For example, he had a binder of notes on relevant information relating to the trial he kept which he labelled Burglary and Buggery (bugger here being a slang for sex), and also was quoted as once joking something to the extent of "I'm facing 10 years for sleeping with a man... But if I had buggered a sheep I'd be getting 20".

Or my favorite of his jokes, in a letter to a friend of his admitting his intent to plead guilty:
"I'm afraid the following syllogism may be used by some in the future:
- Turing believes machines can think
- Turing lies with men
- Therefore machines do not think"

Anyways, rant over. Turing is an important figure for me as a queer theoretician in computing science -- Turing is Daddy -- and I don't take kindly to bastardizations, misinformation, or weakly argued omissions of his history.

Did you know Turing used to ride a broken bike to Bletchley Park, where every so often the chain would fall off? Instead of getting a new bike, or fixing it, he just counted the number of turns of the pedal before the chain would fall off, and would fix it right before it fell off.

He also had severe alergies triggered during this ride so he wore a generally administered military grade gas mask to thwart the pollen.