BrandonS113

I really wanted to like Typst. No more latex would be fantastic. Decided to use it for a project, and had to give up and return to latex, just too many corner cases. Both things its missing from latex, and lack of Pandoc convertibility. Really hope it gets the last 10%

I have European debit cards. They get rejected in many Albert Heijn when Dutch cards (debit and credit) are taken. Jumbo takes foreign cards. Jumbo gets my (and a lot of my colleagues who pass though) business, AH not. Many restaurants take Dutch credit cards and not non-Dutch European credit cards. I went to a restaurant that takes Bitcoin but not EU debit (and Bitcoin transactions costs are much higher)

So if the fees on non-Dutch European and Dutch cards are the same, what explains this? Xenophobia? That is what my colleagues in Holland say.

I agree. But plenty of Dutch vendor only accept Dutch mastercards and not foreign. The fee s the same. WHY? And foreigners can't get IDEAL or Dutch mastercards. I have gotten used to carry a lot of cash when there. Fortunately, Holland is safe. Someone looked at me very funny when I payed them EUR 500 in cash last week.

But we also know why a European home grown payments system is not happening. The incumbent banks hate it and have enough power to block such solutions.

One reason to like the visa mastercard duopoly. They work everywhere. I go to the Netherlands a lot for work, and it is a real pain as they have their own payment system, IDEAL, one needs to live there, that is closed to foreigners. Many vendors, especially outside of main tourist areas will not take foreign cards. Guess they don't want to pay 0.1% fees. Yesterday, was left embarrassed in Albert Heijn where they only took Dutch cards and I did not have enough cash. (some AH do, most don't)

AMEX unusable in Europe? Hardly. I payed for my dentist Friday with it, both supermarket and lunch today, and all Amazon, etc. Some vendors dont take it, but in my European country most places

I much prefer parallel in R with mclapply() to the Julia implementation of parallel. One of the few areas where I prefer R to julia (other being R data.tables to julia dataframes)

I think that is exactly what is happening. Most of my code is much much faster in Julia, and the code is nicer. But R has its moments. Which is good since this particular app has 3K lines, and I do not want to port it to Julia.

And data.tables in R is faster (and I think nicer to write) than DataFrames in Julia. And since data.tables feed my optimization, R still wins.

Great idea, I do lurk there, just might do that next time I look at the code. My hunch is that simd is the low hanging fruit julia brings. But, and I am not an expert, both might end up doing BLAS anyways, which is why they are so similar.

No, its 4 lines of code. I just benchmarked for myself. All it is for 2 vectors x and y of average length 50, and 5 parameters, with exponential, addition, and multiplication, and ultimately sum to return to the optimizer. I was also surprised as I expected c to be much faster. And with Rccp, its actually slower than the R, overhead I guess. When I looked into it, apparently R has really fast code for such vector calculations. With julia, admittingly I did not use simd which would likely make it faster.

Now, I generally use Julia for heavy computes, and usually its much faster than R. But not always.

And this little bit of code runs for hours on the largest instance on AWS every day. Why I was looking so speed it up.

no, the function it calls is pure R and that is where the the code spends all its time.

I used to think so, but I have a function that gets called about a billion times each and every day as new data comes in, and and takes about 0.01 seconds to evaluate (optimizaiton with nlopt). I tried to code it in c (30% speed improvement) python (twice as slow), Julia (about the same speed). Reason is that call has 5 parameters that operate on a vector of length 50 to return a value to minimize. Turns out R is pretty good at such vector calculations.

Well I exported to ps this morning, since I used the pstricks package in latex.

It is the best way I have seen to make diagrams with code, which can be very convenient.

so python takes instrument data and writes latex with pstricks, then to ps to pdf/svg/png/* to insert into whatever doc my colleagues are making

We are continually evaluating alternatives as latex is a heavy dependency, but nothing works as well, yet.

and even more importantly it has worked for 10 years and likely will work for the next 50 years with exactly the same code.

I disagree, there are at least 2 reasons why eps can be better. a) you can exactly match the fonts in an eps figure to a latex file, which will look much better. b) you can use \psfrag to change text to get proper mathematics in figures, like gamma2 in an eps becomes $\gamma^2$ c) maybe also, the pstricks package is very nice, but does not work with pdflatex if pdflatex could do both, I would switch today but

yes, the book Ultra-Processed People https://www.nytimes.com/2023/07/07/books/review/ultra-proces... does that quite well.

Can believe that, but there is more to such additives than cancer. The book Ultra-Processed People https://www.nytimes.com/2023/07/07/books/review/ultra-proces... makes clear the danger from such chemicals in our food.

The demographic most likely to fall for scams is the 20 year old. So. no.

Its doomed to failure.

Latex is horrible but makes beautiful documents and the relevant world either uses latex or Mircosoft Office. Why use anything else? Latex lives by the community of people who use latex for their work and to collaborate. All know 20 year old latex documents will be compilable in 20 years. No competitor outside of Microsoft can survive that.

I have to disagree to "arguably not a very good fit, such as for presentation slides." I have to make regular work presentations with a lot of math. There are two choices: Powerpoint and Latex/beamer. While the latter is far from painless, powerpoint takes much more time and is much worse to revise.

It is easy to say this doesn’t matter, and personally, I couldn’t care less which is used. However, professionally, I have twice in the past two months had a deal with text that was edited by line editor for my organisation, where they strongly criticised our use of these punctuation markers.

And, after much cursing, and my team spending time changing the text, I reflected, and came to like those punctuation markers. Took me a long time, but I have been converted.

I just hired a R engineer, so not that tiny. For a data science job with a lot of stats and math. R still beats Python for stats/data science programing. In my market, R people have more math/stats than Python applicants,

Last time I tried Octave it was several times slower than Matlab. Why use Octave instead of R, Python, or Julia? There is nothing Octave/Matlab can do that they can't. And anybody needing Matlab surely can afford to pay the licence fee?

No. This week, the treasuries and other assets in capital are effectively cash equivalents.

You are conflating two things. Reserve requirements are not same as capital. And there is a modern capital requirement, as per Basel III, in effect since 2022. Capital historically was calculated very differently.

Yes, unfortunately were not exactly working in Amsterdam, and needed a tram to the local station and then a train to, perhaps Amsterdam.

That said, I would love to know why 's-Hertogenbosch is spelled that way. I asked a few locals, and none knew.

yes. I dont want to make much out of this. One of the team came from a country close by and he brought his car. Problem solved.

But just about every Dutch person we worked with had a car. They would meet us on weekends in villages, and most drove. Public transport is great in Holland, but cars have their place too. And unlike some countries close by, Holland has a well maintain and well designed road system. Its great for car travel.

Yes, private bike rental firms are happy to take my money. But they are not common. But the public OV are ubiquitous and more strict.

One needs iDEAL for many things in Holland, and needs to be a resident for that. My colleague tried to get a personal OV-chipkaart, https://www.ov-chipkaart.nl/en/apply-for-personal-ov-chipcar..., needed for bikes, he went though the process, and got nowhere. It was 2 years ago, so may have changed. But he said it stranded on need for a local bank account.

I know you can get anywhere by public transport, much better than where I live. But going though tram-train-train (+even bus or a 3rd train) to visit some village that is less than an hour away by car, will multiply the transport time.

I had to spend a few months in Holland for work. We we left with the problem of what do do on the weekends. You can't rent bikes if you are not Dutch, and visiting the lovely villages and other places Holland is full of is a time consuming exercise in futility unless one has a car.

A train is great for Rotterdam-Hague-Amsterdam, not so good for visiting the Bosch museum in 's-Hertogenbosch.

Holland has a great road system and parking everywhere. The perfect place for a car owner.

R looks pretty much the same as Julia here. Don't think its unstable. Its all float. Nlopt passes parameters as float, data is float. Can't see how it could return int.

But I'll take a second look at the julia version of it. We optimised R as much as possible. might have missed a julia trick. simd perhaps. Except we also run it on ARM

Yes I follow this. The load time keeps getting better. And am looking forward to 1.9.

I really don't want to come across as negative, Julia is a fantastic language, and my hope is that that it will continue its impressive improvement path.

But to follow form the thread's sentiment, I have the feeling Julia lives in an unstable equilibrium. It is lower level than R/Python but doesn't quite deliver the benefits of rust/c/fortran/c++. I find my colleagues gravitate to one of the 2 equilibria.

Maybe your last paragraph crystallizes it. If one lives in the REPL, Julia is wonderful. Not how I work. I prefer the command line. Have new data, run code on it. Data changes in real time, code not. My code may run millions of times on different operating systems and only infrequently change.