Re: A controversial topic

new topic     » goto parent     » topic index » view thread      » older message » newer message
axtens_bruce said...
ChrisB said...

so what's the 'grab me and me look at this feature' ... ?

Before we answer that about Euphoria, answer it for Python, Clojure, Go, Kotlin, Rust, C++ etc. None of these languages are "single issue" languages. I'm part way through the Exercism learning tracks for Python, Clojure, Go, Kotlin and Rust (and 21 others (language junkie much)). There's no "one feature" that grabs me about any of them.

This is the wrong way to think of a language, unless you are making an extreme leap away from the norm. Considering how OE has traditionally fought against being outstanding (in many ways), arguing for uniqueness is a dead end. I recall introducing "string tokens", the fight against pretty much every word ("string") and meaning ("word" is like an atom) and origin (mIRC) and concept (Ai processing, small databases), and then it being picked up as a library in other languages, and then OE being formed out of RDS Euphoria and changing what i had done before incorporating it into OE.

OE's best features is the built-in sequence type, as linked data store, and user typecasting, to lock in a fixed format in a program. I would not use OE without these two, but sadly everything outstanding that could be built on them is disallowed because it wasn't already allowed or no one needs it right this minute.

axtens_bruce said...

They all have their strengths and weaknesses. Each requires a particular way of thinking about how to solve problems with programming languages. This "grab me and look at" sounds like marketing-speak. Before Perl was marketable, it worked. Before Python was marketable, it worked. Euphoria mostly works but needs more done before it works enough to be marketable.

Sadly, when OE was formed, a lot of work was put into "eye candy", such as maps, or making case() better than any one else's case(). I looked down the list of "improvements" and was repeatedly saying "we don't need that", "too narrow of focus", etc etc.. I am not saying stuff didn't work (<coff>case, compiled tasks</coff>), but to the extent it worked it could have been an include file. Instead of fixing 100% of the bugs, we got immensely complex preprocessors to twist source code so it fit .. something.

Icy_Viking said...

Wasn't there a attempt or a dicussion about changing Euphoria's name in the past?

I remember that too. When searching for it online, "programming" was not in the results.

ghaberek said...

We need to flank it from all sides: start doing regular releases, update the website, bring in modern features and libraries, and market the hell out of it. All of that needs to be done at the same time, but it's a delicate balance to avoid drawing in too many people who are then turned off by something dated (like the website) and turn away to never return. It's a delicate balance. Personally, I'm trying to "lie low" on marketing anything until we've got releases working and a decent facelift on the website. Then we hit hard with marketing on all fronts.

I agree, no marketting until it's [Ph,f]ixed. Brownie points if you know where that Euphoria syntax is from. Also "killer app" has been mentioned before. But everything people have asked for to make one is in opposition to the fundamental base of Euphoria. To me, the original beauty of RDS's Eu was it was interpreted, NOT compiled, so it could have been less restrictive about what it could do, and how fast it could evolve, as in keeping ahead of the crowd.


new topic     » goto parent     » topic index » view thread      » older message » newer message


Quick Links

User menu

Not signed in.

Misc Menu