I felt that Jurriaan was very thorough in explaining the approach. It's very clear he has a great grasp on the material.
Unfortunately, the talk does require for the listener to know quite a bit of context. I'm new to functional programming and Haskall (and many in the audience, i think) and found that the gap between my frame of reference and the content of the talk made the larger part of the content hard to follow.
It's a pity, because the problem of solving proper error reporting is probably a very relevant one for most of the audience.
Interesting idea and concepts. I like the idea of adjusting a language to your domain.
As Peter mentioned, I think that this talk was less suited for the wider audience. As the audience got thrown into the deep regarding functional programming and EBNF.
But the message of the talk and the idea of domain specific error messages are still very interesting.
This talk's scheduling could have been better, had it come after the intro-to-parsing talk rather than before it.
It was still a nice ad/exhortion to try out Helium (though I suspect only Haskellers will) and introducing the idea of a more refined set of checks within a compiler in order to help it know *what* is going on so it can give a correct reference frame to errors it finds. I think this idea needs to be shared with other static-typed communities, and I don't mean just Java (though I suppose they could really push it).
I found that, despite not knowing the required base material, the talk was given as clearly as possible and I think I grokked a lot of it. Dr Hage seems an excellent teacher, able to put concepts in both simple terms and with more precise examples (BNF examples).
The most advanced and technical talk I heard at the conference. Even though I didn't understand most of it (all of it? :D) I do believe it was good to see and hear it, because it does show what is being done by the academics in computer science.
This talk was certainly interesting! The context and prior knowledge of functionalprogramming is sort of a requirement, but the speaker really tried to get everyone on board with the whole concept!
Its great to see talks that actually are way more technical than the average! Way to go #domcode15 for getting this into the schedule!
Very interesting and dense topic.
Overall the talk was very nice but I would encourage the speaker to split a lot of his slides so that they become much lighter. There was way to much to read on each of them.
Quite in-depth, more than I expected from the description. Would have liked a more general view not specific to haskell or functional programming, more accessible to those who are interested in domain specific languages, but have limited functional programming experience (ie: my students).
Great talk. Though scheduling could've been better. It would have been a great talk to have seen _after_ the parsers talk as they seemed to complement each other a lot.
Really liked the different look on programming languages from the developer perspective, by trying to better error messages.
The talk to me was a bit theoretical, which required me to pay full attention the whole time. One of the reasons for that, was me not being familiar with the language and concepts.
Even though, the speaker did try to make the concepts clear for everybody. Big plus for that!
A point of improval would be to add some more 'breaks', like a quick summary of few extra jokes in between.
Other then that, excellent talk!
Comments
Comments are closed.
I felt that Jurriaan was very thorough in explaining the approach. It's very clear he has a great grasp on the material.
Unfortunately, the talk does require for the listener to know quite a bit of context. I'm new to functional programming and Haskall (and many in the audience, i think) and found that the gap between my frame of reference and the content of the talk made the larger part of the content hard to follow.
It's a pity, because the problem of solving proper error reporting is probably a very relevant one for most of the audience.
Interesting idea and concepts. I like the idea of adjusting a language to your domain.
As Peter mentioned, I think that this talk was less suited for the wider audience. As the audience got thrown into the deep regarding functional programming and EBNF.
But the message of the talk and the idea of domain specific error messages are still very interesting.
This talk's scheduling could have been better, had it come after the intro-to-parsing talk rather than before it.
It was still a nice ad/exhortion to try out Helium (though I suspect only Haskellers will) and introducing the idea of a more refined set of checks within a compiler in order to help it know *what* is going on so it can give a correct reference frame to errors it finds. I think this idea needs to be shared with other static-typed communities, and I don't mean just Java (though I suppose they could really push it).
I found that, despite not knowing the required base material, the talk was given as clearly as possible and I think I grokked a lot of it. Dr Hage seems an excellent teacher, able to put concepts in both simple terms and with more precise examples (BNF examples).
The most advanced and technical talk I heard at the conference. Even though I didn't understand most of it (all of it? :D) I do believe it was good to see and hear it, because it does show what is being done by the academics in computer science.
This talk was certainly interesting! The context and prior knowledge of functionalprogramming is sort of a requirement, but the speaker really tried to get everyone on board with the whole concept!
Its great to see talks that actually are way more technical than the average! Way to go #domcode15 for getting this into the schedule!
Very interesting and dense topic.
Overall the talk was very nice but I would encourage the speaker to split a lot of his slides so that they become much lighter. There was way to much to read on each of them.
Quite in-depth, more than I expected from the description. Would have liked a more general view not specific to haskell or functional programming, more accessible to those who are interested in domain specific languages, but have limited functional programming experience (ie: my students).
Great talk. Though scheduling could've been better. It would have been a great talk to have seen _after_ the parsers talk as they seemed to complement each other a lot.
Really liked the different look on programming languages from the developer perspective, by trying to better error messages.
The talk to me was a bit theoretical, which required me to pay full attention the whole time. One of the reasons for that, was me not being familiar with the language and concepts.
Even though, the speaker did try to make the concepts clear for everybody. Big plus for that!
A point of improval would be to add some more 'breaks', like a quick summary of few extra jokes in between.
Other then that, excellent talk!