Lambda the Ultimate

Here is a list of the POPL *and workshop* papers that may help you spot the paper of your dreams. It has been put together by volunteers, with in particular Ben Sinclair doing an impressive share of the work.

Amen, but what to do about it? Until SIGPLAN gets a high quality open access venue like the JMLR in the machine learning community, I don't see changes happening.

If the conference is of high quality, let it publish its own papers. In the document markup field, the Balisage conference peer-reviews and publishes the papers presented each year as Gold Open Access (2001-2007 as Extreme Markup Languages, 2008- as Balisage). You can publish them elsewhere if you are so minded, but most people in the field have a lot of for respect Balisage publication. Disclaimer: I presented a late-breaking non-peer-reviewed paper there last year, and have been a peer reviewer for many years.

Why don't we start our own open access PL journal?

Handling a journal is a lot of work. People may believe in the value of open access, yet choose not to dedicate so much time to this noble cause.

For a journal to have chances of achieving the goal of publishing good research, I think the people in charge should be well-recognized within the target community. I wouldn't consider myself legitimate to set up such an initiative (...yet?).

A good proxy for both ready-to-invest-effort and legitimacy is being a committee member of existing conferences or journals in the field. Those are the people that are most competent to run these kind of things. To my knowledge, today's active LtU members are not necessarily members of that group.

There is an excellent open access journal in a related area of computer science already, namely Logical Methods in Computer Science. Look at the managing editors. Those are the kind of people you want to run a journal. It's interesting to note that they also obviously have large competence in PL, so maybe people looking for "an open access PL journal" should just go there.

I've been on a few committees and I know how it works. But...

A open access journal done in an innovative way could be really disruptive. Sure, you'll need some of the successful senior researchers involved, if for credibility and their experience. But by going open access, one could also open up peer review and the feedback cycle between content providers and consumers (as well as reviewers). For example, one could allow for early community review of title/intro/abstracts before whole papers are reviewed and even written (with the idea that a paper could never be better than its intro, it could just fail to deliver). Applying the idea of reddit-style karma to the review process could also work (so...partially crowd sourced filtering).

If the feedback loop between content providers and consumers could be improved, content quality could increase, the quality of the community could increase, and critical mass is more achievable.

Also, PL is not just one field: its more like many different fields pushed under to one umbrella. The readership for something like POPL is extremely related, and sometimes I wonder if its more about math than applied PL. Likewise for PLDI, ICFP, OOPSLA, ... Its not completely weird to think of building a new community to address a field that is currently under-represented (e.g. PL design).

A great opportunity for the LtU crowd to post news items about these papers, proving how great open access is.

I posted about Backpack and LVars already, and both are POPL'14 papers. I thought about what you just suggested, but ultimately decided otherwise: if I started posting a notable proportion of the papers I appreciate at a given conference, people could try to deduce which papers I was less interested in, which would be embarrassing.

I think the initiative is excellent, though, and I hope the publicity on LtU contributed, in some small way, to making those papers available to people that previously had to laboriously search the web for them. I regret that the proceedings will not remain available in the future, though (in my experience this is when they'll be most useful). This means there is still value in collecting links to the preprints all over the web.

I started posting a notable proportion of the papers I appreciate at a given conference, people could try to deduce which papers I was less interested in, which would be embarrassing.

Amusing. But if each LtU member attending posted about one paper, I think you'd have an alibi.

people could try to deduce which papers I was less interested in, which would be embarrassing

Why? If a conference accepts 50+ papers, personal recommendations are very useful in finding the gems.

As explained by Jeremy Gibbons above, the proceedings were only available freely for a month around the conference, and they have since be removed.