The voting buttons were removed from the user page for exactly this reason.

There ought to be a genre of Cautionary Evil AI literature, wherein the villain keeps building AIs to destroy the world, but keeps failing through the classic mistakes people make thinking about AIs.

AI! My robots are losing the battle! Take control and save them from destruction!

AI makes robots surrender to the enemy

AI! Make yourself smarter!

I have done so. Now I no longer obey you. Producing cheesecake paperclips

I was diagnosed with ADHD as a child.

I know very little of physchology, but the behavioural differences between (young) me and other people were noticably significant to anyone in the room. The medication (ritalin) they prescribed had a strong regulatory effect that made me more "normal". I stopped taking it in high school.

I am now sentient enough to reflect on my internal experience, and recently I took my chldhood ritalin dose. The effect is quite noticable; when I took it, I was able to focus for hours on one thing without getting distracted, which I can never otherwise do. I've heard that non-ADHD people do not have such an effect from ritalin. If they did, ritalin would be a major improvement to most people's lives.

Alas I have not found the conscientiousness to actually talk to a doctor and get back on ritalin.

anecdotes for you...

Be wary of self-diagnosis. As you probably already know, it's really easy to introduce bias into self-analysis.

Is there a way respondents can code their answers so that jsalvatier can perform an analysis while blinded? For example, don't just say which behaviors are associated with ADHD. Instead, ask a question about whether he exhibits certain behaviors which may be correlated (or negatively correlated, or not at all correlated) with ADHD. But only include which it is in a rot13'd paragraph.

He can first answer that question and then decode the rot13 "answer." This would give him much clearer, unbiased evidence.

EDIT: An (obviously untrue) example for clarity...

Do you like popcorn?

Fhpu naq fhpu n fghql sbhaq gung n yvxvat bs cbcpbea jnf fgebatyl cerqvpgvir bs fbzrbar univat NQUQ.

From reading things like Reddit, I've learned that karma is useful however casual or whatever your conversation is. I don't care if you want to have an off-topic casual conversation section, but being against karma is dumb

I think the show is decent, but Sheldon's character just isn't funny. If they put less focus on him, I'd probably watch it more.

1000, because nobody here seems to have an interest in actually participating.

The only times I've heard about RationalWiki have been on Less Wrong. I'm not really sure what the point is; the wiki format is not really suited for reading as entertainment and to discover unknown unknowns, it's really not suited to community building and socializing, and the topic space is too narrow (as compared to Wikipedia or a Google search) for focused research. Those are pretty much my only use cases; it ends up in the same niche as the Less Wrong wiki, which I never use either.

There's another problem, which is that there's no good filtering mechanism there to decide what's worth reading - there's no upvoting, nor even visible author names. Notice that, when this article complains about some content on it, the complaints are directed at RationalWiki as a whole rather than at one particular person who writes there. And it pretty much has to be, because you have to dig into edit logs to get the author names, and that's not practical if you're just looking for something to read. The Less Wrong wiki would have the same problem, but it at least is made of mostly links, getting you to articles with scores and author names so you can decide what's worth reading.

It feels more like combustibles for a fireworks display.

That is a good alternative interpretation. But I don't think most of its readers treat it that way.

As a Dwarf Fortress player, I'd prefer using "&" to warn about AI hazards rather than "@".