WTF Bites
-
@Benjamin-Hall said in WTF Bites:
@cvi the beauty of bibtex and latex in general is you rarely have to care about the formatting details. Hand it to them in whatever, and they can apply their house style. Or they give you the style files and you apply them. I can't imagine doing anything serious in, say, Word.
Had to do that last winter. Journal wanted Word with “Vancouver” style (or whatever, don’t remember). That style doesn’t even exist in Word! Downloaded some bullshit from GitHub, but you need admin rights to install that, so no installing it on our windows servers.
Put it locally on my Mac, but then things got royally fucked up when I edited the file again on a different machine.Most idiotic process ever. Also, the style sucks anyway.
-
with a good hyphenation engine enabled
Manual hyphenation > automatic hyphenation
However, it’s also rather a lot more work, and people will not even notice if you go to that trouble. However, they will notice poor hyphenation. So it’s really just like with everything else in typography: do it well and nobody notices it, do it badly and most of your readers will, even if they can’t put their finger on what is irking them about the text.
@Gustav why is everyone deducting points for doing it right?
Doesn’t losing points mean you get a better grade in Germany?
(A German once proudly told me he “got a 1” for something in school. It took me a second to realise this was equivalent to a 10 in a Dutch school, where you don’t want to be getting too many 1s.)
-
Journal wanted Word
That's enough reason for me to go to a different journal. (Fortunately, LaTeX is pretty much the standard around here.)
-
Journal wanted Word
That's enough reason for me to go to a different journal. (Fortunately, LaTeX is pretty much the standard around here.)
Yeah. Any anything scientific with figures, tables, and complex equations? Word would be hell.
-
@Benjamin-Hall said in WTF Bites:
Journal wanted Word
That's enough reason for me to go to a different journal. (Fortunately, LaTeX is pretty much the standard around here.)
Yeah. Any anything
scientificwithfigures,tables (which manglement just love to use), and complex equations? Word would be hell.
-
@Benjamin-Hall said in WTF Bites:
Journal wanted Word
That's enough reason for me to go to a different journal. (Fortunately, LaTeX is pretty much the standard around here.)
Yeah. Any anything scientific with figures, tables, and complex equations? Word would be hell.
Figures are easy, just copy and paste them. Tables work too. Equations are annoying as fuck, but the Word equation editor, while still shit, is much better than it used to be. It even understands some trivial latex.
Yeah, all of this still sucks and doesn’t work correctly half the time. It’s doable, but a real pain for inferior results. And, obviously, the journal ends up typesetting it in something sane anyway, so I have no idea why Word is even involved.
-
Doesn’t losing points mean you get a better grade in Germany?
(A German once proudly told me he “got a 1” for something in school. It took me a second to realise this was equivalent to a 10 in a Dutch school, where you don’t want to be getting too many 1s.)
If it's anything like in Czechia, no. We have grades from 1 being best to 5 (only 4 in university) being the worst (fail), but points are the more the better, with the maximum being specific to the test or competition.
So if you have a test with a bunch of questions, you get some points for each correct answer and then if grade is needed, you will have a required number of points for given grade (so say 15 for 4, 20 for 3, 25 for 2 and 30 for 1; different table for each test depending on how many points can be had and how hard it is).
-
@Bulb The Dutch way seems to be much simpler, but also a lot more varied (at least when I was in school, looong ago). Most teachers did it fairly simply, as I recall: things like a test with n questions, each question being worth (n ÷ 10) points, were pretty typical. Of course, this does not fit with the idea that a 1 is the worst grade you can get on a test, but I’m not sure all of them realised that (ISTR teachers who simply rounded any score below 1, up to 1, though — the few times something like that happened probably didn’t warrant tinkering with the numbers to do it properly).
-
@Gurth Usually, the mapping from number of correct answers to grade shouldn't be linear. In most cases even the worst students are supposed to be able to answer a large number of questions correctly, maybe 50% or 60%, otherwise they shouldn't really be in that class at all (even in basic school, which is mandatory so you can't just drop out, you can repeat a year or be moved to a school for mentally ill). So you usually need a lot more than ⅕ points for better than the worst grade, 5, which is considered failure (of course failure in one test does not mean overall failure in the class, except in university where there is just a final exam, but you'd have to catch up for the next exam).
Or if the grading was simple, it was usually a major error means a grade down. That was typical especially in grammar (so five errors in typically about a page is a fail), but sometimes in maths too.
-
@Bulb In the Dutch system, 6 (or 5.5 before rounding) is generally the lowest passing mark.
I've had tests which needed way more than 60% for a passing mark, often deducting grade points from 10 for each wrong answer. Though I also recall a specific one where in theory you could get a 13.6 or so, which would have been rounded down to 10.
-
deducting grade points from 10 for each wrong answer.
I've mentioned before my Linear Algebra class, in which each test was worth 100 points and each question worth 10 points. However, there weren't necessarily 10 questions. I remember well one particular test that had only 2 questions. I completely missed one of them, not even partial credit. 50%. Definitely a failing score.
Except, I got a mark of 90. A. I could have gotten a mark of 80 (B) for handing in a completely blank paper.
-
@Gurth Usually, the mapping from number of correct answers to grade shouldn't be linear.
You’re thinking like a mathematician/statistician there, not like a geography, history, or language teacher in a secondary school ;)
In practice, it depended entirely on how each individual teacher decided to grade the tests they made. Some (teachers and/or tests) would do it like I described above, others would assign a number of points to each question in the test to reflect their perception of difficulty, importance, etc., some might indicate beforehand how much each question was worth while others wouldn’t — and in the end, it all worked out.
I also recall a specific one where in theory you could get a 13.6 or so, which would have been rounded down to 10.
I remember one case where the teacher decided to add a point to everyone’s score because, IIRC, he had made one question clearly far too difficult, or it was about material we hadn’t covered yet in class — something, anyway. Which meant that one kid technically got an 11, because he did get it right, and every other question in the test too. (Of course, this was the kid who, if he got anything below a 9, we felt he hadn’t done all that well.)
-
one kid technically got an 11, because he did get it right, and every other question in the test too.
I have one of those stories, too, which I think I've also told before. (Fake edit: Maybe? I was thinking there were extra-credit questions of the test, but it may have been just a normal question.)
My very first programming class was FORTRAN IV, many years ago. Tests, at least the one I remember, were multiple-choice and/or true-false, so pretty easy if you've studied at all. One of the true-false questions was,
You cannot do *thing* in FORTRAN.(I've long forgotten the value of thing.) The expected answer wastrue; you cannot do thing. However, I'd read the textbook. He hadn't covered it class, nor in the assigned reading, but thing was indeed possible, and the book explained how to do it.My answer was marked wrong. I argued the point with the instructor after class and explained why I had given the "wrong" answer. He admitted I was right, but still refused to give me the point on my score.
So I either missed out on an extra-credit point in addition to a perfect score on the normal questions, or got 99% instead of %100 if there weren't any extra-credit questions.
-
I had a teacher (Calc II in college) who once gave a test. Not an easy one--only 13 questions over an hour, but...
- 1 of those was a full-scale epsilon-delta proof, and not a trivial one.
- 1 of those was a trick question--it was an integral with a divergence in the middle of the range and you were supposed to show that it wasn't integrable by either of a couple methods.
- The remaining 11 were non-trivial calculations.
Points were assigned per question, varying per question. All out of 100% (whether adding up to 100 or just like 30 points, with each point being 3.33%.
The next day, he came back with the graded tests. His opening words were something like "I think that test was a little long." The high score was a 65%, and he ended up putting the mark for an A (normally 93%) at like a 60%. Of the dozen or so of us...2 got As. The median was under 50%.
As a teacher, I heard of other teachers who had students with higher than 100% on their semester grade. I personally found that extra credit only helped those who didn't need it--the ones who did need the help were also the ones who would rarely even do the regular work, let alone anything extra.
-
@Gurth yeah, that happened to me somewhere around 9th grade. Math exam for 90 minutes had 12 questions. Somewhere around half I realized "for fuck's sake, that's a lot of questions". I panicked and hurried up. In the end, I managed to answer 11 questions or so, but the teacher realized afterwards that it was too much and put the 100% mark at around 9 or 10 questions.
Reminds me of the Simpsons episode were Lisa is cheating: you even answered the questions the copier cut off!
-
@Benjamin-Hall oh yeah, I remember something similar to that too. Exam in some theoretical course, Algorithm Engineering I or whatever it was called. Apparently the exam was meant to be a lot more than can reasonably answered, so that you can just choose to answer whatever topics you knew best. Problem was all of the question were fucking hard. (E.g. "Here's some graph problem you've never heard of, prove that it's NP-complete." I thought those weren't bad while studying, but they only picked questions for the exam that I couldn't answer.)
I got a B with something like 25-30%, passing grade was set at 15%.
-
@topspin That reminds me of the qualifying exam as a physics PhD student. We had to take it before the first day of actual grad school, and you had 4 chances total (start of each of your first 4 semesters) to pass it, otherwise you were let go. Very few, if any, pass on their first try.
The test? 12 questions. Over anything and everything you could have learned as an undergraduate physics student. The time limit? 12 hours, broken up into 4 3-hour chunks over two days. One hour per question. And it was given in 3-question chunks, too (for obvious reasons to avoid cheating). So you couldn't even work ahead. Graded by a committee of professors, the same ones who came up with the questions each year.
The passing grade varied each time, since the questions varied. The highest the passing grade had been in the memory of my advisor (who retired after I finished) had ever been? 55%.
Of course, I pulled my usual "who me, study?" thing and didn't realize the test was there until I'd showed up and gotten the intro packet about 2 days before the test. So I just wrote down stuff, figured "meh, I'll retake it." Somehow passed on my first try, to my extreme astonishment. I made sure to not tell my fellow students how much (or little) I studied, lest I be lynched.
-
-
@Benjamin-Hall said in WTF Bites:
12 hours
Yup. Fortunately you didn't actually have to stay the whole time for each session. I think I made it about 1.5 hours per session. I usually was one of the first ones out. Because at some point, there's nothing more you can do.
-
-
@Benjamin-Hall said in WTF Bites:
I made sure to not tell my fellow students how much (or little) I studied, lest I be lynched.
I went up for thermodynamics without ever touching an exercise before the exam. First problem I solved in that course was right at the exam. Initially figured I'd just have a look and retake it later. Ended up barely passing - which I'm still relatively proud of (it wasn't a hard exam, but also not ridiculously easy - probably 30%-40% that failed?).
On the flip side, a fellow student went to an electrical systems (IIRC) exam under similar circumstances. We discussed afterwards and they felt pretty good about it ... e.g., easy test, should go well, maybe even have a higher grade. They got zero on the test, though.
-
Tims Financial
Discover the rewarding Tims Credit Card from Tims Financial. Earn Tims Rewards Points on everyday purchases like gas, groceries, and transit, with bonus points on dollars spent at Tim Hortons.
A credit card provided by the financial services division of Tim Hortons
-
-
@Benjamin-Hall said in WTF Bites:
Graded by a committee of professors, the same ones who came up with the questions each year.
Our physics exams were similar to that. Our exam covered six topics with 3 tasks each and you were allowed to choose 3 topics out of the six. No mix-and-math, though. If you chose one topic you had to do all three tasks within it.
The trick was to recognize the difficult ones where one professor had decided to impress his colleagues.
When training for the exam (there was a special course for it, under the supervision of one professor, using old exams) we were shown some of those questions.
For example, one task asked to show why the moon created the tides. On both sides of the Earth. In case you do not know: There's a high tide on the side towards the moon (because gravity, duh). But there's also a high tide on the direct opposite side.
And that one is an absolute bitch to explain.
-
And that one is an absolute bitch to explain.
I've never seen anything more detailed than "centrifugal force". Caused by the fact the center of gravity of the Earth-moon system is a thousand or so kilometers away from the center of gravity of the Earth alone.
-
For example, one task asked to show why the moon created the tides. On both sides of the Earth. In case you do not know: There's a high tide on the side towards the moon (because gravity, duh). But there's also a high tide on the direct opposite side.
And that one is an absolute bitch to explain.Is it? My understanding is that all it takes is decomposing the situation to a uniform field matching that in the centre of the Earth, which gets cancelled by the Earth's acceleration it causes, and the remainder is near-symmetrical. Not exactly symmetrical, because the force varies quadratically, not linearly, with position, but you do get a resulting force pulling the antipode upward as you needed.
I've never seen anything more detailed than "centrifugal force". Caused by the fact the center of gravity of the Earth-moon system is a thousand or so kilometers away from the center of gravity of the Earth alone.
That's … another way of saying the same. Because the ‘uniform field cancelled by acceleration’ is the ‘centrifugal force field’, just coming from the other side of the GR equivalence principle.
-
-
That's … another way of saying the same. Because the ‘uniform field cancelled by acceleration’ is the ‘centrifugal force field’, just coming from the other side of the GR equivalence principle.
And of course you're supposedly not allowed to say centrifugal force because it is not a force. It is a constant required when working in a rotating (rather than an inertial) reference frame.
-
And of course you're supposedly not allowed to say centrifugal force because it is not a force.
That depends entirely on your point of view. In a very "lies told to children" sort of way, we've privileged inertial frames...when really there isn't any such innate privilege. Sure, it's more convenient most of the time to work in inertial frames. But for things like the movement of air over the spinning earth? Yeah, rotating frame ftw. Things like the Coriolis force, etc just naturally fall out in that frame, while they're a stone pain in an inertial frame.
-
@Benjamin-Hall Possibly the person who told me that didn't quite understand it himself, since they insisted you should instead be using the centripetal force. Which is neither a force nor a consequence of a reference frame. Rather, it is the amount of force you need to have (from gravity, rope tension, whatever) to ensure objects stay in the same relative position in a rotating reference frame.
-
We have defined named constants instead of :spooky: "magic values" in the code. Except all the fucking named constants are just named after the value of the constant, not the purpose of the constant.
public static final String LANG_CODES_EN_JP_KR_DE = "en|jp|kr|de";No reason why this constant exists - here, these locales get differently formed URL patterns. And, surprise, JP no longer needs special handling, so it needs to be removed from the list. So now I have to change the value AND the name of the constant, and verify that this is the only reason this constant exists (it is).
-
@Benjamin-Hall Possibly the person who told me that didn't quite understand it himself, since they insisted you should instead be using the centripetal force. Which is neither a force nor a consequence of a reference frame. Rather, it is the amount of force you need to have (from gravity, rope tension, whatever) to ensure objects stay in the same relative position in a rotating reference frame.
Which is evidenced by the direction you bugger off to as soon as you fail to exert said centripetal force.
-
@Benjamin-Hall Possibly the person who told me that didn't quite understand it himself, since they insisted you should instead be using the centripetal force. Which is neither a force nor a consequence of a reference frame. Rather, it is the amount of force you need to have (from gravity, rope tension, whatever) to ensure objects stay in the same relative position in a rotating reference frame.
If you're in an inertial frame, the relevant "force" is the centripetal "force". It's actually an acceleration (if using F = ma, it goes on the a side in the center-facing direction). Yeah, the nomenclature screws up students quite a bit.
If you're in the co-rotating frame, the relevant force is the centrifugal, which points out, and is a "real" force in that frame--
F_inward + F_centrifugal = 0if you're at rest in that frame.Basically, 90% of physics is figuring out the right reference frame. The other 90% is abstracting the problem correctly and the remaining 90% is realizing that the math doesn't work so you need to play silly math games to cover it up. That last sentence is only partially tongue in cheek.
-
@error_bot xkcd Centrifugal Force
-
-
So now I have to change the value AND the name of the constant, and verify that this is the only reason this constant exists (it is).
So you can name it something sensible! (yeah, codereview (changes required): Please name this after the contents)
-
So you can name it something sensible! (yeah, codereview (changes required): Please name this after the contents)
Or worse, "Please do the same for the remaining 588 constants."
-
I expect the centrifuge to be destroyed by the vibration caused by the unbalanced wheel (due to the weight of Mr. Bond). Whether that happens before or after Mr. Bond is crushed by the rotating reference frame is an unanswered question.
-
@HardwareGeek said in WTF Bites:
Whether that happens before or after Mr. Bond is crushed by the rotating reference frame is an unanswered question.
This needs a YouTube channel like the hydraulic press or Will It Blend?
-
We have defined named constants instead of :spooky: "magic values" in the code. Except all the fucking named constants are just named after the value of the constant, not the purpose of the constant.
public static final String LANG_CODES_EN_JP_KR_DE = "en|jp|kr|de";No reason why this constant exists - here, these locales get differently formed URL patterns. And, surprise, JP no longer needs special handling, so it needs to be removed from the list. So now I have to change the value AND the name of the constant, and verify that this is the only reason this constant exists (it is).
I am astonished that a long-term WTD
member can be astonished by the existence of such code. Actually, you ought to have seen lots of stuff likepublic static final Int ONE = 2;
-
@HardwareGeek said in WTF Bites:
I expect the centrifuge to be destroyed by the vibration caused by the unbalanced wheel (due to the weight of Mr. Bond). Whether that happens before or after Mr. Bond is crushed by the rotating reference frame is an unanswered question.
Were you involved in the destruction of Iran's uranium centrifuges?
-
That's … another way of saying the same. Because the ‘uniform field cancelled by acceleration’ is the ‘centrifugal force field’, just coming from the other side of the GR equivalence principle.
And of course you're supposedly not allowed to say centrifugal force because it is not a force. It is a constant required when working in a rotating (rather than an inertial) reference frame.
Noticed the sentence mentions General Relativity? That changes everything. General Relativity is General, because it works in all reference frames, inertial forces are considered just as real as anything else, and gravitation is an inertial force (and in standard English terminology, gravity actually means the total inertial force, not force due to gravitation).
-
@BernieTheBernie said in WTF Bites:
Were you involved in the destruction of Iran's uranium centrifuges?
: I can neither confirm nor deny any implication or lack of implication in any specific incident, Mr... Bernie, is that right?
-
it’s not even considered “standard” in a lot of hosting environments even though Laravel claims to insist on ext/intl being installed (but frequently doesn’t bother needing it)
I somehow managed to create a form field in Symfony which converted text to
DateTimeusingDateTime::createFromFormat, andDateTimeto text usingIntlDateFormatter(that's what internal framework component uses, but I didn't know that at the time). This is how I learned about the duplicate data because saving the form in any of the Russian timezones causes the time to shift by one hour. That was shortly (~2y lol) after Russia dropped the daylight saving time.
-
@Vault_Dweller said in WTF Bites:
" that one is an absolute bitch to explain."
I thought I knew those. Apparently not.
-
In 2020 and 2021, dozens of scientists were invited to speak at a series of webinars (because pandemic) on the topic of modeling the spread of COVID. Some were medical doctors; others were experts in modeling. All were happy to share their research for the public good.
The webinars were organized by a Polish company, Villa Europa. Villa Europa is now sending the scientists bills for the privilege of sitting in their own homes in front of their own computers in a Zoom call with other scientists. (No one but the speakers attended the webinars.) Well, technically, they are being billed "for webinar debate fees and open access publication required for the debate proceedings" and "editorial work". €790 for the former, and €2785 for the latter. In at least one case, this was after the speaker had specifically asked if he would be expected to pay any fees, and being told "no". With late fees, interest, legal costs, and such, some of the bills are over €20000. One scientist is being billed for €80000, including the cost of software Villa Europa used to edit an animation in his presentation.
The fees are mentioned in an agreement they were asked to sign after the end of the webinar. The agreement was purportedly to give Villa Europa permission to publish the webinar recordings.
These fees are mentioned in a long clause in the last page of the contract, and are written out in words rather than numbers, without any highlighting.
There is a lot more stench of rotten fish. The speakers were recruited by a man named Matteo Ferensby, whose email signature claimed an affiliation with the University of Warsaw. The university has no employee named Matteo Ferensby, and there is no track record of scientific publications by anyone named Matteo Ferensby.
The agreement imposes binding arbitration by a Polish arbitration court. However, that court, Pan-Europejski-Sąd-Arbitrażowy (PESA), also smells rather bad. The director general of the Court of Arbitration of the Polish Chamber of Commerce, says she has never heard of it. Whois says PESA's domain name was registered by Villa Europa after the conferences in 2021. Or it did until recently. (The info is now private, and website of the registrar who.is says to contact for more info is broken.)
Villa Europa is pursuing judgements against some speakers in courts in their home countries, based on arbitration results from this sketchy Polish arbitration "court". Some courts will honor and enforce judgements by courts in other countries; others won't. However, those are real judgements by real courts. In this case, however, it doesn't appear PESA has submitted it's findings to actual Polish courts for actual judgements, so the foreign courts PESA is bothering aren't too likely to enforce the awards, if they're paying attention. Oh, BTW, legitimate arbitration organizations notify the other party and give them the opportunity to tell their side of the story (even if it's in a very inconvenient venue in a foreign country), which is something PESA appears not to have bothered doing.
-
@HardwareGeek Kinda standard shady business. This kind of fraudulent “arbitration courts” was a common problem in Czechia too.
-
So you can name it something sensible! (yeah, codereview (changes required): Please name this after the contents)
Or worse, "Please do the same for the remaining 588 constants."
Job security! You'll be busy for decades!
-
@Benjamin-Hall said in WTF Bites:
Basically, 90% of physics is figuring out the right reference frame. The other 90% is abstracting the problem correctly and the remaining 90% is realizing that the math doesn't work so you need to play silly math games to cover it up. That last sentence is only partially tongue in cheek.
In general relativity the math does work, no? It's QM where you play silly math games a.k.a. normalization.
Ok, and the infamous Navier-Stokes equations that nobody knows whether they have a solution in all the cases we'd like.
-
Ok, and the infamous Navier-Stokes equations that nobody knows whether they have a solution in all the cases we'd like.
That's more like a mathematical curiosity, though, than a physics problem. You can just solve them, as far as your compute power permits. Should they ever blow up in some situations, that's going to involve scales where we know they're not modelling reality anyway (they model a continuum, but at some point way, way below what you normally calculate with them, you reach molecular levels).
123: Centrifugal Force - explain xkcd
