Generative programming research project

Ola Fosheim Grøstad ola.fosheim.grostad+dlang at gmail.com
Fri Dec 1 20:10:03 UTC 2017


On Friday, 1 December 2017 at 19:43:53 UTC, H. S. Teoh wrote:
> There's a link to the OptiML spec that, thankfully, makes it 
> clear that this is something primarily focused on machine 
> learning and certain categories of algorithms, rather than a 
> one-size-fits-all miracle cure that's slated to take over the 
> entire programming world.

Well, not sure where they are heading, but this seems to focus on 
DSLs:

http://drops.dagstuhl.de/opus/volltexte/2015/5029/pdf/19.pdf

«As mentioned in Section 1 and throughout the paper, we argue for 
a rethinking of the role of high-level languages in performance 
critical code. In particular, as our work and the work of others 
have shown, generative abstractions and DSLs can help in 
achieving high performance by programmatically removing 
indirection, enabling domain-specific optimizations, and mapping 
code to parallel and heterogeneous platforms.»



More information about the Digitalmars-d mailing list