I am really hoping the 12.0 FE is less crash prone than the 11.3 one.

I just had a spontaneous crash that wiped out two hours of work and I have no idea why.

I plotted ~10 points...

@C.E. I strongly suspect many of these assignments make rather marginal sense. For instance, the only element assigned to Finland was found at a time Finland didn't exist as a nation, and even the concept of Finns hadn't really formed in any clear sense...

@kirma It all hinges on whether all discoveries made in those time in Finland should be assigned to Sweden or not. I think it would be a huge disservice to the region to do this? :P

@Kuba ah, good :)

@kirma to be clear, I think assigning to Finland that which happened on today's finished territory is the right way to go, and that seems to be the usual convention also in other cases. e.g. Wikipedia calls the chemist a "Finnish chemist". Your argument could also be applied to that.

@C.E. As a Finn I'm mostly disinterested of such competitions anyway... maybe because the ranking wouldn't be particularly high. ;)

@kirma I don't even see it as a competition and if it were a competition I would not identify myself with Sweden or any other country. I have contributed nothing to it... I think the most interesting thing about these visualizations is it makes you think about history.

@C.E. Sure. :)

@C.E. @Kirma Irish people have this issue all the time

Even today Irish people (actors, footballers etc) are claimed by British media as Birith footballers, actors etc

Constant thing, causes a lot of ire here (if you forgive the pun)

ah, for fin, here's wikidata on the subject

query.wikidata.org/…

*for fun

I like it because it has various versions of countries (French first republic etc)

still no finland though :(

@CarlLange Wikidata identifies the same chemist as the discoverer but they assign him to Sweden... oh well.

ah, but now I see that this query is slightly different: I looked up the country of discovery, whereas this query is for the citizenship of the discoverer. Then it makes sense to assign him to Sweden, as well as assigning, as they have, Marie Curie to both Poland and France.

@CarlLange I think there has been a kingdom and a socialist version of Finland in addition to the republic, in kind of historically esoteric sense. Neither existed long enough people to really notice, though, thus practically non-states.

@b3m2a1 I could put a link to your paclet server on PackageData next to packages that exist on your server. If you think it's a good idea. You would have to provide a JSON that maps your packages to my entries or provide information so that I can do the mapping. I think the Github repository address should be enough.

@xzczd you said "This solution is incorrect, just compare it to the numeric solution". Do you think that the problem itself is well posed? I remember the teacher telling us in class that if both ends are Neumann boundary conditions, then there is no unique solution. I get same solution as Maple. But there is arbitrary constant there, the zero eigenfunction. I need to learn how you did your numerical solution. So you really think Maple solution is wrong?

school opens next week and will ask the teacher about this also.

@C.E. sounds good, but I'd suggest this URL: paclets.github.io/PacletServer I've been thinking about ways to add a submission form directly to the server website that'll walk people through getting stuff up there and so hopefully pretty soon the average user doesn't actually need to go to the repo.

@b3m2a1 Generic link doesn't work for me, it would have to be a mapping between the packages on PackageData and the paclet server.

I don't want to send users to the paclet server if the package isn't there and waste their time.

@C.E. Ahhh I see. You want it there. Every paclet gets a page that's <baseURL>/lowercasepacletname.html. I can provide those names in a JSON file sometime soonish. Or alternately they're on the RSS feed if that's easy enough to work with: paclets.github.io/PacletServer/index/rss_feed.xml.

The JSON isn't bad though

What other info would you like?

@b3m2a1 The package may have different names on PackageData and the paclet server, even if just slightly it means it would be hard to automatically find the corresponding packages. The only thing I can think of that would allow automatic mapping is the github repository URL, which I'm guessing you also have stored somewhere?

@C.E. I use it in the build process for packages stored on GitHub so I can make that JSON file with the map

It's not laying around as JSON anywhere since I use .wl files to store the paclet info

@b3m2a1 it would be great if you could generate a JSON like that as part of your build process. I think that will make it possible to identify a fair number of packages, since a lot of the URLs on PackageData are Github URLs.

I think it will be great to be able to provide this to PackageData users :)

@C.E. Happy to do it. I'll get this up sometime in the next small round of edits I'm making.

:)

github/WolframLanguageForJupyter

I am confused by "Properties":mm = MengerMesh[2]

mm["Properties"] is a long list of properties, but

mm["AbsoluteDimension"] doesn't return anything useful even though it is in the "Properties" list.

Shouldn't anything that appears in the mm["Properties"] list be something that I can extract?

btw, everyone please take a moment to appreciate what a nice website the team behind the CoDEx package made for it.

@CraigCarter That's the idea, but this particular property does not exist for me either.

@C.E. Thanks:

Yikes: mm[#] & /@ mm["Properties"] doesn't return anything for all of its "Properties".

Is it worth asking this on the regular StackExchange, or is this something that is too trivial to ask?

@CraigCarter - mathematica.stackexchange.com/q/85344/9490

it looks like the property list was populated, but they weren't all implemented - or something like that

@JasonB., Thanks, I see. How frustrating that they would do that. 1/2 hour checking my sanity...

@Kuba I wonder who'd be interested in using this...

(by mma I mean WL)