Whiley

I must be getting old. I no longer want a language with 273 overlapping features to master to become a software wizard. I want a language with 15 features that compose and work well together.

Lower standards so a larger fraction of papers are accepted? That decreases the average quality of publications. An alternative that *increases* quality of publications is to encourage journal publications: polished versions of the best papers. 2/2

Some research areas (like PL) look bad on naive metrics because publication LPUs are larger and researchers look less productive. Incentives for hiring and promotion become distorted when metrics like paper counts and citations are applied. Solution? 1/2

TIL there are multiple people whose names are on 20+ (a max of 33!) papers submitted to a single conference. I am curious what that community chooses to do, or not do, next. twitter.com/guidosalva/sta…

Oh for goodness sake, Apple is still parsing untrusted TrueType fonts using an old pile of C code in 2023. TrueType is not that hard to parse in a safe language if you don't need hinting, folks (which is disabled on iOS). In fact I wrote most of that code a few years ago.

Awesome to see Aztec Zac talking about SNARKs in person!!!

Join @whileydave, a researcher in smart contracts at @ConsenSys, and learn how formal verification of smart contracts has the potential to significantly improve their security and reliability! 🔐 🗓️#program" target="_blank" rel="nofollow noopener">EVMSummit.org/#program

sat down and wrote the “how do I read type systems notation” overview I wish I had when I was learning about type systems langdev.stackexchange.com/a/2693/861

The Dafny Programming Language is pretty amazing at times ... whileydave.com/2023/06/27/pro… #programming

Finally found a good use for the `by method` syntax in Dafny. The method implementation does not stack overflow on large sequences (whereas the function does on the Java backend as its not tail recursive). Mind = 🤯

Life goals: create a programming language and use it exclusively for the majority of my professional career

Pattern Matching in Rust's Neverland! #rustlang whileydave.com/2023/04/16/pat…

From the ICSE23 website: "Equity at its heart is about removing barriers, biases, and obstacles that impede equal access and opportunity to succeed" . Also, I just paid >AUD 3K in fees to attend. I won't be able to ever do this again. #icse2023.

Disappointed this doesn't compile in #rustlang :(

Nice talk looking at Dafny and Whiley! youtu.be/fWSGhxyTG-4

@JAldrichPL @stevemblackburn @samth Yup, basically … rankings and metrics are the problem. No question about that!!!!

@stevemblackburn @samth There are probably a lot of factors contributing to this, but I think the growing popularity of CS Rankings and its inclusion of only 2 PL conference by default is a major driver of this trend. ISMM and ICFP literally do count 0.0 in the default CS Rankings setup.

Perverse incentives promote perverse behavior. Our community's ever-growing obsession with publication at a few select venues is a blight on our field. It is harming quality research in "secondary" areas. It is fueling unethical behavior among researchers.

@stevemblackburn @samth Well, I certainly don’t view venues in that binary fashion. Papers from PLDI can be (and often are) junk. Likewise, papers from a strange workshop can be amazing! Still, I understand the issue. Not sure what can be done about it though 🤷‍♂️

@samth Yes. Scarcity is always a problem that will be with us. That wasn't my point. My point was that we have accumulated a set of perverse incentives that in the limit give a value of 1.0 to a PLDI paper and 0.0 for an ISMM paper (to be concrete). Or maybe even 0 for ICFP.

Dafny 4 syntax is coming very soon. Finally we’ll be rid of the aweful “function method” syntax!! dafny.org/dafny/