Just FTR, your critique is answered by Stefan Lucks, Norina Grosch, and Joshua Koenig's paper[0] from last year's langsec workshop at IEEE Security and Privacy. They define a formalism, calc-regular languages, to handle length-prefixed strings.

[0] http://spw17.langsec.org/papers.html#calc-regular

Thanks, this looks promising! I'll take a look.

In other words, "if you're autistic and can't read body language, assume that nobody wants to talk to you."

Real inclusive there, pal.

> if you're autistic and can't read body language, assume that nobody wants to talk to you

This is a good point, which is why we exercise judgment. Pretty much all laws are written for an imaginary "average individual" because they cannot possibly account for all the wide diversity of people's mental and physical abilities, but that's why determining whether there's been a violation of the law is entrusted with a person, not an algorithm, who can consider whether there have been extenuating circumstances, and being autistic qualifies. Obviously, many laws aren't meant for people who are unable to comply with them, and every law recognizes that there may be valid reasons for breaking it.

Hi. That's me you're quoting.

> The status quo: highly funded white people in a ranch house, obsessive dreams of white supremacy floating into the night...

Were you there? If not, then you're projecting. As it happens, the participants were at best plurality white. And no, the PoC weren't all Asians either.

Also, because Twitter requires short statements, there are multiple readings of "ignored" there. I see you're taking the least charitable one, i.e., "everyone let it slide." That interpretation is incorrect. People stopped talking to him. Isn't shunning what you progs want?

> I see you're taking the least charitable one, i.e., "everyone let it slide." That interpretation is incorrect. People stopped talking to him.

Good, because I used your interpretation: silence. You silently enabled a white supremacy propagandist.

Of course, you're certainly NOT silent when it's time to support him. (Despite your own evidence of his history of "casual racism" at a professional networking event.)

So you're oddly quiet with white supremacy: you "stopped talking". But god forbid you "ignore" when a conference right next to Ferguson (StrangeLoop) declined to help promote a white supremacist's influence!

(Hell, if I mocked people's favorite programming language or text editor, I'd almost certainly get a vocal reaction. Mild rebuke at least. Says everything about techies' priorities.)

> As it happens, the participants were at best plurality white. And no, the PoC weren't all Asians either.

That's interestingly vague: were these participants to Moldbug's "casual racism"? I imagined he only did this with obviously sympathetic audiences like you.

(And what does "the PoC weren't all Asians" exactly mean? For someone berating others for interpretations you don't like, you leave open a wide range of interpretations.)

Thanks! Those are all excellent papers. I am actually looking into formally verifying Hammer using Frama-C, but haven't gotten too far yet.

It's probably most accurate to say that gender isn't a significant factor in my calculus of attraction. I was born in Houston, though, and lived there for 23 years.

For what it's worth, I don't recall intending anything in particular with that list of adjectives, but you never really know about authorial intent.

The W3C working group will eventually produce a crypto API standard, though whether that standard will meet the requirements you describe remains to be seen. In particular, it exposes primitives (the proposed API can definitely be called in unsound ways), which a whole lot of people think is a terrible idea but which the standard editor seems bound and determined to ship. It's very frustrating.

That's because W3C's goal in having a cryptography standard isn't security, but rather interoperability; they see encryption as another step towards making the web a first-class application development environment. Without it, they can't get Netflix to run on pure "open" web technology.

It's unfortunate, because we could use a secure browser crypto interface much more than we could use better browser interoperability with random non-web technology. But our industry is, of course, fundamentally unserious about security.

Gosh, thank you!

I'd argue that in some cases there's a degree of control that can be developed, but also that this seems to be more the case for sensory triggers than for psychological ones.

I agree it's not the ideal UI. It's similar to what Darrell Issa's keepthewebopen.com was using with Madison, and to the marked-up texts that the EFF has produced as PDFs, but there's a cognitive jump (for me, at least) between "make a comment" and "propose a change".

Fundamentally, I want a user to be able to propose one or more changes by selecting a section of text and annotating it with a replacement. So there's a semantic difference between "comment" and "change". Furthermore, I also want to be able to easily select a changeset, then produce a draft output that integrates all changes from the changeset.

I think we can do this with co-ment's existing features and maybe a little bit of extra Django hackery, but before I go haring off into the internals, is this more like what you're thinking of?

That's a high-priority forthcoming CSS fix, you just bumped it higher. Thanks.