I believe that I had read at one point that Mathematica was used to prototype Google(I think wikipedia said this). I'm fairly Sergey Brin did used and recommended Thema as name. Is the Google prototype anyway true/correct?

Shouldn't TemplateApply be TemplateCompose in terms of consistent naming?

I'm asking because we

Compose[
    TemplateObject[TemplateSlot["planet"]],
    <|"planet" -> "Mars"|>
 ]

and not Apply.

@Szabolcs Doesn't this work?: xml /. container : pattern :> (container /. rule). I guess it's just a ReplaceAll version of Michael Hale's Cases idea.

@MichaelE2 Ah yes, I had used the nested Cases for scraping nested XML patterns before, but your ReplaceAll variant looks nice for transformations.

@Kuba The documentation says that Compose was deprecated in 1992

Hey guys, is there a simple way to host an image from a notebook, for sharing with others, without right clicking, saving, uploading somewhere, etc?

ideally I'm thinking of something that's right click->get share link

I'm just going to drop this here for discussion:

It's my thoughts of how Wolfram doesn't support and encourage third-party package development adequately. community.wolfram.com/groups/-/m/t/849709

My biggest gripe is how they gave up on proper namespace use and dropped everything in the System context. That's a step backwards. MATLAB didn't support namespaces at the language level, which was a pain, so now they introduced them. Mathematica did from the beginning, but they chose to stop using them as an end-user feature.

@Szabolcs there is not suppose to be a "third"-party

@Pickett Yes, I know mathematica.stackexchange.com/questions/28778/… the point was that Apply does something different than Compose or @ so I'm just picky about naming.

@Szabolcs On third party packages. I agree. My understanding is that there was a major push for better support for them around like version 6, but it stalled. Not enough interest in the end. Most packages people write are for their own research and it ends up being like packages in R :(

@Kuba Alright, I just thought naming it after something deprecated wouldn't make sense either. If you look at Names["*Apply*"] you get the sense that it is Apply that is the outlier, not TemplateApply. (Apply is of course a common operator in programming, and should be named Apply for that reason, but it must not necessarily preclude any other usage of the word, in my opinion.)

@Szabolcs While I personally wished contexts were actually used, I think they're too difficult for the average user just writing simple scripts.

It's terrifying and awful to see. People for some reason really value being able to use a function without having to "load a package". I've seen people compare Mathematica favorably over Python for scripting for this reason. It's hard for me to understand that mindset.

it's the same mindset of a high school student in their first programming course, wondering why you wouldn't just make everything a global variable.

@Searke "Most packages people write are for their own research" <- That's true, but I'd really like people to share more, and polish their packages more. That's why I was so excited about packagedata.net There was already the Wolfram Library Archive, but I was hoping for something easier and more dynamic, some place where people can share even a version 0.1, get users, get encouragement and hopefully even get collaborators.