Re: Introducing Quorum Programming Language!
Ethin,
First, you aren't the only one with the opinion on traditional for/while. There's a few facts to be aware of with it:
1. Studies on blocks and syntax in text based languages show basically a tripling effect with novice performance. Basically, students can write correct programs faster, along with other properties (e.g., understanding it better). I'm simplifying from the results, which are a little complicated to understand, but that's the idea. Seems to be the same result regardless of language.
2. It's actually pretty normal to believe the other is superior. In fact, it's such a commonplace belief that one of the studies we did years ago studied people's perceptions of it. Notably, turns out that for every year of experience people claim to have (self-reported), they garner a large and statistically significant increase in how much they perceive that syntax to be intuitive. AKA --- there's a standard equation for a bias effect.
So basically, people that learned such syntax tend to think it's fine, but careful measurements of human behavior show that it should not be immune to criticism because it has undesirable properties like being hard to learn and error prone (e.g., Neil Brown's data). We actually see this all the time in studies. We did a recent study on database programming and, similarly, devs often believe they are doing well, or not, but performance doesn't necessarily match their beliefs. That's why we use data on Quorum. We don't want to just "choose" the syntax, we want something external to our beliefs and value systems we can use as checks and balances during the design process.
Second, Adoption:
Short answer is I don't know why people adopt/don't quorum. It's clearly being adopted, but I have no evidence on the reasons. We do follow evidence from Meyerovich and Rabkin on adoption rates, which are related to what we choose to add on to the standard library in Quorum. That probably has an impact, but how much is unclear. Schools probably have their own reasons, but again I don't track these things. I'm just an open source dev and prof that cares about these things and tries to make quorum better. Schools do seem to like it, which is great, but who knows.
Natural Language:
You're totally right on natural language. In fact, the evidence is often tricky. Language that is natural in English often does poorly in studies. The word repeat does well, which may sound natural depending on what you mean, but individual token choices are based on statistical analysis because it's just not intuitive what works and what doesn't without it. For example, Quorum has no left bracket, but does have a right and the paren's are missing. Or, the comparison operator in ifs is the =, which is 8x less likely to have errors compared to =, even in the presence of assignment operators. Pretty weird how the data comes out, or at least it was before the data came out. In any case, what natural even means is not clear, but you're right that just assuming English, like if then, is the answer is definitely not supported by the evidence.
Anyway, hope that helps.
-- Audiogames-reflector mailing list Audiogames-reflector@sabahattin-gucukoglu.com https://sabahattin-gucukoglu.com/cgi-bin/mailman/listinfo/audiogames-reflector
