Common Programming Languages To Program Pld
Welcome!This subreddit is dedicated to discussion of programming languages, programming language theory, design, their syntax and compilers. Post your ideas and get constructive criticism.Be nice, contribute, and stay away from useless flame wars.This subreddit is about programming language design, not programming per se. If you want to ask 'what programming language should I learn', 'what language would be best for X project', or any question like that, please post to or.Related subreddits.For language-specific discussion, take a look at the (rather old). Related online communities.Links.If your post was removed without notification by the spam filter,. Perhaps you can make the case that there are only about 30 places in the world to do serious PL research(split evenly between corporate labs and some universities.I could dig deeper here to come up with a more precise answer—I’m actually very interested in my own question haha), and for everybody else, there’s this sub.
This place is the equivalent of the water cooler in a PL research lab. A place to get advanced as well as a place to ask dumb questions. Perhaps it’s a sign of the future direction of distributed research, and worth studying what makes this place special for things that the academic community could adopt (and vice versa—what do they do in the labs that we don’t do here?). One benefit of belonging to a lab/project is that you've got other people around you, often at the next desk, working on the same thing. It's a lot easier to ask for help and get un-blocked. Other than that the resources and such are the same, you just have more access to mentors & your cohort (when you're a student). When you're an advisor you get minions;-).
Oh and you're all working on a common goal, that might be limiting if you've got your own ideas but it's also great for feeling motivated and part of a team.I was a PhD student at Melbourne university working on Mercury, access to peers was great but there were only two of us (my supervisor and I) working towards our goals with a majority of our time (he of course had to share his time with teaching).Now I'm working for Mozilla on SpiderMonkey, the JS engine, I'm remote so I don't have the 'access' to peers at the very next desk. But they're only a video-call away. So it's less-good for access but I still have an affiliation.This community (reddit, discord, irc) has an interesting mix of people with different backgrounds and experience. I'm not sure if it's actually all that different from a research lab with a couple of senior people and 6-12 PhD/masters students. Some of the same levels of experience exist, the difference may be in our goals/motivations rather than in experience/knowledge.I think there's an almost separate group too, the industrial PL implements. People paid to work on the LLVMs, clangs, SpiderMonkey, etc.
I feel like academia may get more value in talking with the industry groups. But the hobbiest group would get the most out of talking with academia,(Never really thought about it befiore but I'm a member/alumni of each group, academia, hobbiest & industry.). Not sure if it's feasible, but I have a suggestion.Many academic papers describing language features or algorithms are published without a working implementation in code. That makes them harder to explore, since one cannot simply run the program to check its correctness or make changes.Having a community maintained repository with (maybe multiple) implementations of each of these algorithms would make their access easier to the general public while also making changes and evolution possible for future research.As an example, while learning about I got great insights from reading, complementing what I got from the paper itself. Maybe we could setup a “PL Techniques Implementation” wiki, which could be organized by category, date, authors, etc. And then have links to implementations in various languages.To make it a bit more feasible we could start by prepopulating with a) some papers that do have author provided implementations that are at least decent. We could perhaps mine these from research conferences that also separately evaluate artifacts.
B) older pretty popular things that other people have already made implementations of, such as Pratt Parsers, Parser Combinators, etc.After doing some work on setting it up and prepopulating it, perhaps it will look decent enough that the community will be interested in re-implementing papers and submitting their own implementations.Thoughts?. I think there were some people who started a paper reading group here, which is a cool idea. Did that go anywhere?I've read at least a couple hundred papers on programming languages, with varying degrees of understanding). (And a few hundred on non-PL topics.) Some papers are old, but many are also recent (1-2 years old).And I would like to write blog posts about many of them. That would be a good way to get some discussion going,But I have so much programming work to do on Oil that I've largely abandoned any ambitions of blogging about all these papers!I tend to skim a lot of them, and put them on the back burner.
Then something I'm working might remind me of them, and I pull them out again. Idea: it is already a common thing for people to bring up new ideas and ask for feedback in here. Despite the fact that, as they are from an actual research community, saying that we could provide feedback could be a little pretentious, some decent part of this community has enough knowledge to understand the content of many papers/ideas and ask questions about it. This means giving them a chance to explain to some people who, despite being mostly hobbyist, have at least a lot of interest in the content and are eager to talk about new research. They'd benefit by a) explaining their ideas to a different, but equally interested, crowd and b) having the chance to receive ideas from outside of the academia.We, obviously, would benefit by receiving more content.Perhaps they could monthly bring a post for discussion here?
Something that is trending inside the research world and not getting too much publications i.e not reaching out to us? (I realize this could compromise their research, so they'd have to respond if that is even possible).I really hope, again, that this does not sound pretentious. I'm not saying we can enlighten them with our knowledge, far from that, lol.They could also post one blog post per month from some original content posted in here. Might be a stretch. A mindmap , editable using source-code representation (so it can be improved and branched using a git-like approach) describing:. different PL paradigms.
different Intermediate Representations (IR) with links to relevant papers, and their relationships. for each PL paradigm a list of relevant IR with comments/critiques (i.e.
IR usefulness from the point of view of a particular PLD). important projects using a certain paradigm, IR, etc. promising ideas, etc.Then from this common and rather objective/neutral mind-map there can be personal and/or group branches with subjective comments.
For reducing useless conflicts, every branch must:. clearly specify that it is a subjective/personal point of view (probably suffices an automatically generated title + disclaimer).
comments must be added as explicit distinct nodes near the neutral/objective parent mindmap nodes, so it is clear we are adding personal annotations. for reducing useless wars, every comment node must specify the purpose of the comment, using the Six Hats approach : informative comment; emotive comment not neccesarly supported by objective facts (i.e.
Like/dislike); devil advocate/negative comments not considering other positive aspects but underling possible problems; optimistic comments assuming all things go in the best possible way; a new/possible solution to a specific problem/commentEvery branch has an ID and only PL creators listed on PLD website can vote for them, and they will be displayed sorted by votes. In case of negative votes, they will be not displayed.The votes to branches can be done reading some YAML or JSON file retrieved from the website of each listed PL. I reflected better. The mindmap approach requires some adaption of existing tools, while there is already MediaWiki with semantic web extensions that can be used instead.I will try to create a quick demo for seeing if it can be used for:. linking useful info/papers.
link papers to real applications/code as suggested here by. discuss papers, PL, etc. In the 'discuss' section of MediaWiki the papers. import other info and navigate textually and 'graphically' using semantic web extensions. Academic outreach, huh?
Personally, I'm still sore over the rejection of a paper I submitted to SLE in 2017. I have much more I'd like to say about that, but am not sure it would be wise to comment further.But I will say one more thing, about a more general problem, that of research still being too difficult to access. We have this wonderful thing we call the Internet. Yet so much research is still paywalled.
Will the ACM ever throw open their entire Digital Library? What do people think of Sci-Hub? Freedom fighters or pirates, or what? RIP Aaron Swartz. Take a look at SigPlan's.
This is the for the blog. The audience is largely academics and researchers.We are a different community of people interested in PL design, some even building our own programming languages. Some are part of academia, but many are hobbyists and employed in industry. Some even work on mainstream languages.We have been offered an opportunity to write up a post for the Perspectives blog that introduces our community to their audience, and suggests mutually beneficial synergies between our communities. For example, perhaps professors of courses on PL design might want to list this community as a resource that their students can take advantage of to further explore PL design topics.So yes, ideas that would be communicated in a single post on that topic. I hope that clarifies it for you, and others. This is the third clarification about the request, and every time the target changes a little.Many PL projects overlap, so we are more competitors than collaborators.
But rarely a PL became mainstream (i.e. It is like a lottery), so from a practical point of view there is no competition, because we are on the same boat.It is also unlikely that someone can find contributors here, because every one is focused on his project.So we are a community sharing a common passion, but not a common project. The target has not changed, and it really is an open-ended question. Language and understanding is imperfect, and I am doing my best to be clearer as people have questions and confusion.I don't see the communities as competitors or collaborators. There is some overlap in domain where synergies might yield mutual benefit.
Types Of Pld
Why not explore those without worrying too much about structure at this time? Every journey starts with some first steps. What do we think good ones might be between our communities?Thank you for your suggestions. So we are a community sharing a common passion, but not a common project. We are here not for collaborating, but we are here because it is like a showcase/expo of different projects and we want to see what the others are doing, for comparing, being excited and so on.I wonder if that's true for everyone here. I certaily feel the most passionate about Plasma. But I work on other language implementations too.
Programmable Logic Array
I feel like it's Plasma that is the member of and not Mercury or SpiderMonkey though. I usually come here to chat Plasma or see what others are up to.Maybe what we could learn from academia is better ways to do mentorship.