This post will clarify a few of my ideas that people have concerns or misunderstandings about.
Politics
Freedom and Capitalism
I want a small government with limited power. The proper function of government is protect people against force – e.g. military, cops, courts. I want a society with tons of freedom including economic freedom (which is what “capitalism” actually means). See my essay liberalism. I’m not necessarily opposed to all anarchist ideas (though most are awful), but I think we should aim for minimal government and try that for a while before thinking we know in advance what further reforms would be a good idea.
I respect thinkers like Ludwig von Mises and Ayn Rand. I do not respect lots of their followers or the people commonly associated with them (e.g. I disagree with Hayek, Rothbard, most libertarians and most Objectivists). I also disagree with most Republicans and most Democrats.
In general I like individual thinkers, not groups.
I think political philosophy and economics are more important than politics. By politics I mean stuff like current events, news and election issues. Current political issues include abortion, gun control, immigration, racism, feminism, rent control, tax policy, government-run healthcare and environmentalist policies. People should learn how to think effectively about general principles before trying to debate those specific controversies.
People are partially right to complain about corporations and Wall Street. Many of their arguments are incorrect, but there is shady, unfair, exploitive stuff going on. But the problem is mostly government involvement in the economy and lack of economic freedom. For example, the main source of monopolies is government laws that make it harder to compete with existing companies, e.g. by increasing barriers to entry.
Trump
I like Donald Trump better than Hillary Clinton or Joe Biden. I think Trump has moderate political views and likes America. I disagree with his protectionist economic ideas (like tariffs and trade wars), though I do agree with his intuition that there’s some problem related to trade with China and that some sort of action should be taken. I think Trump screwed up by hiring a bunch of establishment Republicans and he handled coronavirus badly.
I appreciate that Trump is somewhat, partially challenging the ruling elite class of journalists, media pundits, unelected political influencers, professors, politicized non-profits, lobbyists, bankers, administrators and bureaucrats. Most politicians in both parties are part of that elite which is oppressing and ripping off the American people (the middle and lower classes, and the wealthy people without friends in high places).
Tribalism
Most people bring a tribalist, follower mindset to politics. They cheer for their team, just like sports. They’re super biased. They don’t understand the other side(s) very well. They don’t rationally study or debate the issues.
This doesn’t mean each tribe is equally right or wrong. Currently, I agree with Republicans more than Democrats. That’s despite being an atheist and growing up in a Democrat family in a heavily Democrat area. I do disagree with lots of Republican ideas.
Immigration
Immigration has been used for decades to try to dilute Western civilization by bringing in people who think in other ways and have other values. Western countries have been doing a bad job of standing up for their values and assimilating immigrants. There are ongoing debates about whether Western values are worth standing up for. While this debate is ongoing, I think immigration should be slowed way down. I don’t think bringing in immigrants who agree with you, and vote for your tribe, is a legitimate way to resolve a debate.
I don’t think white people have a monopoly on Western values. I don’t think genetics are destiny. I know many other people who criticize immigration are racist in some way (that doesn’t merit all the tribalist hatred they’re receiving, which often comes from people who are even more racist). They think there are race-related IQ genes and that there is such thing as a biological “human nature” which is controlled by genes and therefore can vary by race. I have a strong “left wing” position – shared by right winger Ayn Rand – that ideas and culture are what matter, not biology. (On a related note, I think gender roles are socially constructed, but I don’t believe some of the other ideas commonly associated with that claim. While males and females biologically differ in some physical characteristics, I don’t think biology is the cause of observed mental differences like personality traits or math success.)
I disagree with many economic arguments against immigration. In a capitalist society, immigrants don’t drive down wages. That’s because, as the workforce gets larger and wages go down, it’s easier to start a business (because you can hire employees more cheaply), so more businesses get started, which pushes wages back up. However, currently US regulations are hostile to starting new businesses. When it’s a huge burden to start a business and hire people, then immigration can drive wages down. But “they took our jobs” is a bad argument.
I think the USA should screen immigrants better. Instead of letting so many immigrants in by lottery or extended family ties (including birthright citizenship granted to babies born here by tourists), I think immigrants should be admitted more based on having American values and being ready and able to do productive work. Although I think IQ tests (and the concept of IQ itself) are highly flawed and culturally biased, I think they’d be better than nothing for an immigration screening method. English language proficiency tests would also help.
Identity Politics
I’m opposed to identity politics. I think we should stop looking at people’s skin color, rather than doing affirmative action or having race-based groups like “Black Lives Matter”. I want a more color blind approach.
I do not think racism, sexism, homophobia, white privilege, etc., are solved issues. There are significant problems there (by both Democrats and Republicans). The current activism – like riots and cancel culture – is making things worse and is making it harder to reform anything.
PUA
Lots of “pickup artists” are idiots. Sometimes their idiocy crosses the line into assault. People like Good Looking Loser, Russel Hartley and RooshV are awful.
In 1994, the alt.seduction.fast (ASF) discussion group was started on Usenet. Some people there figured out some good ideas about how dating and social dynamics work. Of course some people there were dumb, too. Representatives of that group, which I respect, include:
- Mystery Method (Mystery’s Reality TV show))
- The Game: Penetrating the Secret Society of Pickup Artists
- Girls Chase
- Real Social Dynamics (YouTube, website)
It was a discussion community. Many people participated productively and there are archives of ideas people liked, e.g. Classic PUA Writings. Many of them also went out and met people in person. They weren’t just armchair philosophers.
The ASF people aren’t perfect but I think they have some genuine knowledge. They managed to analyze, describe and understand social dynamics in ways that other groups haven’t. This information is useful for all members of our culture, male or female, in order to better understand the unwritten rules of our society. And although the ASF focus relates to dating, many of the social dynamics principles apply to other social situations too.
Other so-called pickup artists vary. Some learned a lot from the ASF crowd or participated in those discussions. Others didn’t and are usually clueless. Some of the ASF knowledge has spread elsewhere but it’s often mixed up with bad ideas too. The “red pill”, “mgtow (men going their own way)”, “MRA (men’s rights advocates)” and “manosphere” groups often have some ASF knowledge mixed in along with some of their own bad ideas.
Because so many fakers try to sell their pickup artist advice (advice that doesn’t work and is often offensive), the ASF people pretty much stopped using the terms “pickup artist” and “PUA”. I’ve been using the term “PUA” anyway but I’ll consider calling them the ASF community or specifying individuals in order to reduce confusion.
Claims about how our culture works are not claims that it should be that way. I’m not in favor of social climbing, promiscuity or pandering to whatever other people want. I’m also against lying to or tricking women (or anyone). The ASF people, contrary to some of the attacks on them, are more anti-lying than the typical person.
FI Members
There are no senior members of FI who have been around a while (years), learned FI well, and who make good role models. Don’t try to copy anyone or assume they’re good and you should try to be like them.
I don’t endorse anyone’s learning behavior, and I certainly don’t endorse their lecturing behavior. Some newer members have potential (and older members could change) but none have established themselves as doing a great job.
Don’t try to copy me either. That will lead to cargo culting. You have to learn things yourself and follow your own judgment. I’m too different in too many ways. You should expect to misunderstand me a lot, not to be able to do what you think I do and have it work in your own life situation.
Being a Discord moderator is not an endorsement of someone’s ideas. Being in a video with me or having a guest post on curi.us is not an endorsement of a person in general.
On a related note, I think everyone but me should be posting anonymously. (Because of cancel culture. And by posting anonymously I mean use something that isn't your real name or connected to your real name.) I’d prefer to be anonymous myself. I think it’s way too late for me to change (and maybe too hard to stay anonymous when e.g. selling stuff, meeting people IRL, and developing a reputation as a public intellectual) but everyone else should go anonymous. What’s the benefit of using your real name?
I’m going to stop posting on Twitter in general. Most of what I posted was just retweets without me writing anything. I dislike Upvotes and Likes in general (pointless) but I found retweets ok (shows stuff to my friends/fans) and tried them for a while. Retweets were not endorsements. I never treated Twitter like a discussion forum or serious place. I will continue to read Twitter because I like a few people there. I’m going to stop retweeting because Twitter has an awful, tribalist political culture which I don’t want to contribute to. Plus Twitter shadowbanned me and is part of the cancel culture which is trying to suppress right wing speech.
I think FI people like Khaaan and Justin are tribalist tweeters who don’t understand the other side(s) of the debate well. That doesn’t stop them from being right or sharing good info over 50% of the time. But they ought to learn how to think rationally instead of doing so much politics. Even if they were going to do politics stuff, their approach is basically unproductive because they’re so biased for their tribe.
Offensive Comments
I’m not careful about what I say all the time. I don’t believe in political correctness. I think misunderstandings will happen whatever you do and it’s not worth the effort to walk on eggshells around everyone. Better if people mostly have thick skins rather than police their own speech.
If you dislike something I say, you can ask about it or criticize it. (Try to understand what it is before attacking it, please.) We might disagree. If so, I’ll have a thought-out position that you can hopefully respect, even if I didn’t explain it all upfront. I can’t preemptively explain everything I think every time I mention a topic. People can ask questions or read my writing to find out more.
I often disagree with all mainstream positions on a topic in some way. When disagreeing with one view, I don’t always communicate what I think about all the other positions. This leads to misunderstandings because people assume if you criticize one tribe then you must be part of an opposing tribe.
Lots of “jokes” reveal genuine racism or other bigotry. Speech is meaningful. I’m open to rational questions and criticism – I won’t just automatically dismiss issues as minor. But please try not to assume what I think and don’t begin the conversation in an adversarial way.
Messages (45)
Quitting Twitter
Should I disable automated notification tweets (videos, newsletters) or leave them alone?
I wouldn't want to mirror stuff on Twitter to enable people to read my stuff at Twitter. That'd be encouraging people to use Twitter and stay there.
I think it's fine to post a link to my website once on twitter.
But what about automatic posting of one link per new thing? E.g. if each blog post was sent to twitter with a title and link. Would that be giving twitter content and helping them too much, or would that be ok? the newsletter and video notifications are similar to that.
Also what about manually posting links to a few of my favorite articles (of my own) on twitter? Is that giving them too much content or ok? I do think that's pretty pointless if that's all you tweet since your twitter account won't be getting much attention anyway in that case.
I'm leaning towards disabling auto posting to twitter.
This is kind of a tangent and I don't have a clear answer to the question curi asked re: quitting Twitter. These are just some related thoughts springing out of something I wondered about:
From a moral perspective, what's the difference between helping Twitter with content via automated posts versus, for example, helping a business that supports BLM (say, Amazon) by patronizing them (as a small scale individual consumer, not some biz with a big contract)? How do you make that sort of evaluation? That question seems to be coming from an opinionated place - the opinion being that maybe not automatically posting stuff on Twitter is sort of like an ineffective boycott. I'm not sure about that opinion though, but I seem to have some intuition along those lines.
Offhand I think in evaluating this issue you'd have to look at things like: 1) what's the benefit to you of engaging in the activity that you might want to abstain from? 2) what's the cost to you of engaging in the activity? 3) what are your viable alternatives?
For the Twitter case,
1) there's not much benefit since anyone who cares much about following your stuff can find and get it elsewhere. Your post already detailed some problems you had with Twitter in terms of not liking some of the mechanics, tribal culture and censorship.
2) OTOH the cost of automated posting is very low.
3) You have other places to promote your material, so you're okay on that.
For e.g. not wanting to help Amazon with business cuz they support BLM,
1) being able to buy stuff on amazon is a huge, significant benefit for many people. they have lots of items and fast delivery.
2) Getting stuff from Amazon has some direct cost (you have to pay for stuff you buy).
3) there's some viable alternatives to Amazon, but if you apply the principle of not wanting to shop at pro-BLM corporations consistently, you basically can't buy anything...so looked at from a principled POV, there's no alternative to the "set of pro-BLM retailers" which includes Amazon.
Hmm according to the framework I just made up and laid out, the evaluations of the two cases come out pretty differently. That's interesting. So what's the big picture answer to the question I asked? I think the answer is that the cases are different because one (Amazon/BLM) involves giving up major values according to a principle which leaves you with no practical alternatives for action. OTOH, the Twitter case, for curi, is maybe giving up a tiny value in the face of many alternatives.
Ah so maybe then in the Twitter case, the issue of stuff like whether some action is helping Twitter takes on a much greater significance in the analysis, because the value involved for the actor is pretty small.
I wasn't trying to give anything up. I thought I might be better off disabling the auto tweets. It'll help communicate to people that they should actually move to better notification systems and that Twitter sucks.
Try it from a different angle. Suppose I'd never had a Twitter account. Would I be sacrificing by not creating a bot account that's only purpose is to auto tweet my stuff? Does anyone actually value such an account? If they do, well, I never intended to provide that (and I never made e.g. auto tweets for blog posts or messages) and I disagree with them. I think they shouldn't want that or organize their life and attention allocation the way they are. Twitter should be a minor side note if used at all, not a primary way to get information.
> I wasn't trying to give anything up. I thought I might be better off disabling the auto tweets. It'll help communicate to people that they should actually move to better notification systems and that Twitter sucks.
Ah heh, I see. OK. I'm not sure how effectively it would communicate that point to people btw. Although I guess it might communicate the point to the people who care more, though, and I guess you care more about those people anyways. The people who were actually using Twitter to follow your stuff and would quickly notice the lack of posts would presumably figure out some other notification method.
BTW DH follows you still, right? So that's maybe a person who you care about who might actually see your stuff. IDK how likely that is, how many people he follows or any of that stuff, though.
> Try it from a different angle. Suppose I'd never had a Twitter account. Would I be sacrificing by not creating a bot account that's only purpose is to auto tweet my stuff?
No, I don't think so.
> Does anyone actually value such an account? If they do, well, I never intended to provide that (and I never made e.g. auto tweets for blog posts or messages) and I disagree with them. I think they shouldn't want that or organize their life and attention allocation the way they are. Twitter should be a minor side note if used at all, not a primary way to get information.
There are a couple of things that I was looking out for on Twitter (Mises Institute and Reisman stuff, specifically). I've now switched over to RSS thanks to your efforts/suggestions at building a list of suggested feeds, so my own use of Twitter as an information source will be going down.
I actually have been making conscious efforts to use Twitter less, have taken "Twitter breaks", and that sort of thing. I don't think we are fully in agreement on the role of Twitter but I think I am moving more towards the minor side note POV.
> I think everyone but me should be posting anonymously.
Do you think you have a bias based on the poster?
Also do you mean for both the FI list and curi.us, or just the FI list?
The main reason I can see to not posting anon on FI is that you either need an email address set up (which is like pseudonymous then), or need to get credentials for the anon FI account.
> What’s the benefit of using your real name?
I mentioned the pseudonymous thing b/c you said this.
One benefit is being publicly accountable if you're public in other ways. Also like demonstrating consistency, and other philosophically/morally important things.
Most people aren't in that category, though.
> There are no senior members of FI who have been around a while (years), learned FI well, and who make good role models.
> Don’t try to copy me either. That will lead to cargo culting. [...]
Initially I thought "does curi include himself in that category?" ('that' referring to "make [a] good role [model]")
The paragraph which the second quote is from made me think "yes". I'm not so sure thinking about it now, though.
If you want to answer, do you include yourself in the category I mentioned?
#17790 By posting anonymously I meant anything other than using your real name (or something connected to your real name). I'll edit the article with a clarification. That's for both curi and FI.
> There are no senior members of FI who have been around a while (years), learned FI well, and who make good role models.
I was not including myself in this.
#17792 thanks for clarifying.
Does
> There are no senior members of FI who have been around a while (years), learned FI well, and who make good role models.
conflict with
> On a related note, I think everyone but me should be posting anonymously. (By posting anonymously I mean use something that isn't your real name or connected to your real name.)
?
someone could have an alias and maybe be a role model. I'm not sure if there are limits to that which would be absent if they used their real name. making your voice or face public can be sort of like using a real name; at the very least it makes verifying/de-anonymizing/doxing potentially easier.
would it be easier for someone to be a role model if they used their real name?
I guess it's sorta moot because no-one is in the position atm - if they were and posted under an alias they could always make their real name known (tho you can't undo that easily)
> What’s the benefit of using your real name?
all my stuff is tied to my real name and i don't want to make new accounts everywhere. using my real name is less work.
Other benefits to using your real name: You don't have to remember which account or name to use where. People who know both your real and fake names don't have to remember which one to use where.
A benefit to me to posting as Anonymous: It's easier for me to ignore social stuff and just talk about real content when I'm not subtly worried about what affect my post or people's responses to it will have on how readers view me as a person.
>> What’s the benefit of using your real name?
> all my stuff is tied to my real name and i don't want to make new accounts everywhere. using my real name is less work.
What do you mean new accounts "everywhere"? Wouldn't you just need new accounts for FI stuff? gmail and maybe discord.
> Other benefits to using your real name: You don't have to remember which account or name to use where. People who know both your real and fake names don't have to remember which one to use where.
IME if someone *actively* uses two names, it's hard to 100% accurately use the right one at the right time. Occasional accidents can happen. But if they use a single name and I know of other names, it's fine. The rule "always use name X" is easy enough for me to remember and make a habit.
> Do you think you have a bias based on the poster?
I think having an *identity* with a *reputation* at FI is broadly good for people (but it's fine not to use it every time). It allows more customized responses, and it allows me to be more friendly and generous when I think a post is bad.
If I think a post is problematic and it's from a fresh anonymous with no reputation, I don't want to invest much time into it because 1) it could be a troll (some people try to harass me and waste my time, on purpose) 2) the person is likely to leave or change to starting a new topic soon. Even attaching a fairly negative reputation to a bad comment can address those concerns.
I'm more aggressive about ignoring anons or giving only brief responses.
> What’s the benefit of using your real name?
If I post anonymously I can't take credit for an idea afterwards. It's possible with an alias but harder than using my real name.
> > all my stuff is tied to my real name and i don't want to make new accounts everywhere. using my real name is less work.
> What do you mean new accounts "everywhere"? Wouldn't you just need new accounts for FI stuff? gmail and maybe discord.
Yeah I guess so.
> On a related note, I think everyone but me should be posting anonymously. (By posting anonymously I mean use something that isn't your real name or connected to your real name.)
Why do you think we should all be posting anonymously? How is this related to you not endorsing our learning behavior?
> Why do you think we should all be posting anonymously?
e.g. b/c cancel culture
> How is this related to you not endorsing our learning behavior?
It's a thing some people are doing wrong, IMO.
>> Why do you think we should all be posting anonymously?
> e.g. b/c cancel culture
I've got a long enough history with FI and other "cancellation-worthy" ideas that if someone wants to cancel me, they won't need a fresh example to do it - they can just go through a ~20 year archive of stuff. Given that they cancel people for stuff they wrote over 30 years ago, I might be dead before I hit any kind of cancellation statute of limitations that might exist.
If I was a newer person I might go anon/pen name, but my cancellation risk already seems baked into my life. I still post anonymously anyways sometimes - sometimes cuz I want privacy or am touchy regarding some particular subject, sometimes because I forget to sign my name when posting on this blog :) But I don't see how a general anon policy would help me much at this point from a cancellation-risk perspective.
I'm also fairly low status so I'm not a very big target for the woke people. And if I get more status it won't be by climbing the corporate ladder or some other very cancellable method, so I'm not too worried.
Somewhat separate point: I've already got an online identity associated with my actual name. Building up a new one would involve time and transition costs.
#17892
> I've got a long enough history with FI and other "cancellation-worthy" ideas that if someone wants to cancel me, they won't need a fresh example to do it - they can just go through a ~20 year archive of stuff.
This is a bit like saying that if someone really wants to break into your house, they can just break your windows, so why bother locking your doors? It is ignoring the people who don't actually care that much about breaking into *your* house, but would do so if they had the opportunity.
There are 2 main ways people get cancelled:
1. Targeted cancellation: someone wants to cancel you specifically, so they look through your history to find something cancel-worthy.
2. Opportunistic cancellation: someone sees some post you made and doesn't like it, so makes it go viral.
You are saying that you are already at risk of #1 happening, so no point being anonymous now. You seem to be ignoring #2.
Opportunistic cancelling is a much higher risk on new posts than on old posts. Those people aren't usually digging through unknown people's old posts to find things to cancel them for. They just see a new tweet or post they don't like, and then spread it around.
> I'm also fairly low status so I'm not a very big target for the woke people.
Right, so there's not much reason anyone would be digging through the last 20 years of your online history. Your biggest risk would be someone seeing a new tweet or post you make and trying to make that go viral.
> #17892
>> I've got a long enough history with FI and other "cancellation-worthy" ideas that if someone wants to cancel me, they won't need a fresh example to do it - they can just go through a ~20 year archive of stuff.
> This is a bit like saying that if someone really wants to break into your house, they can just break your windows, so why bother locking your doors? It is ignoring the people who don't actually care that much about breaking into *your* house, but would do so if they had the opportunity.
> There are 2 main ways people get cancelled:
> 1. Targeted cancellation: someone wants to cancel you specifically, so they look through your history to find something cancel-worthy.
> 2. Opportunistic cancellation: someone sees some post you made and doesn't like it, so makes it go viral.
> You are saying that you are already at risk of #1 happening, so no point being anonymous now. You seem to be ignoring #2.
> Opportunistic cancelling is a much higher risk on new posts than on old posts. Those people aren't usually digging through unknown people's old posts to find things to cancel them for. They just see a new tweet or post they don't like, and then spread it around.
>> I'm also fairly low status so I'm not a very big target for the woke people.
> Right, so there's not much reason anyone would be digging through the last 20 years of your online history. Your biggest risk would be someone seeing a new tweet or post you make and trying to make that go viral.
Ok you make a good point regarding targeted versus opportunistic cancellation. I still don't want to create an anonymous identity right now though. I'm planning to just focus on less controversial stuff for a while anyways in my public material so hopefully that will at least partially address opportunistic cancellation issue.
BTW I'm the same person as #17892 but not the same as #17888. I actually was gonna say something like that as an immediate follow up to my previous post but I did not for some reason, not sure why. I was pretty tired and went to bed immediately after so maybe that had something to do with it.
possible solution to the mess caused by lots of people using the Anon name
cc @curi
re #17928 (in another thread) curi said:
> idk why a bunch of people want to share the name Anonymous and make a mess. it seems like Anonymous got more popular a couple days ago when i suggested people use *anything other than their real name*. i don't understand why.
first, some labeled groundwork:
1. "a bunch of people want to share the name Anonymous" - I note 'Anonymous' is the default name and the *author* field isn't required to make a post
2. "make a mess" - I suspect you mean: the mess is b/c it's harder to keep track of individual conversations and whether you're talking to the same person or who each party in the thread is. I think we agree on this, and also that community/anon accounts can be important in other ways i.e. it's not the case ppl should avoid sharing names always. *tyranny of the default* as such.
3. "Anonymous got more popular" - I've used it more than usual, and it looks like other ppl have, too.
I think the reason for (3) is your post + (1).
I agree with (2). it can be a mess. I have an idea for a ~small change to solve the problem of it being a mess.
I think curi should consider adding some tools regarding author-names to curi.us. I don't think it's reasonable for anyone to expect/demand he's the one that does it.
I think a potential "solution" - if done wrong - could be pandering and otherwise bad, but there are some good reasons to avoid (2) e.g. it's easier to keep track of convos. less "housekeeping" -> less mental load -> more mental energy or time for good stuff. It'd also make extracting a particular conversations easier if there's a reason to do so (also: someone else could do it, not necessarily curi).
I have two ideas for things that might be good to do.
I am willing to implement the frontend stuff for these if crui thinks they're worth adding.
@curi if you're interested in this or want more detail: the next thing I plan to post is a draft interface with no logic behind for idea 1. just like screencaps of what it might look like.
I imagine the end result (of all of idea 1) is either like 2KB of JS, or some HTML for the template + like 1KB of JS. I'd write it straight in JS so it's small; no 200KB JS file nonsense.
**1. basic and easy-to-use tooling around author names (no changes to backend/API/DB stuff)**
purpose: ensure that *by default* conversations and threads are more organised by making it easier to tell which actors are which and who's talking with whom.
aux goals: low impact and easy to implement. no significant side effects. no new required actions for users to do. broadly: hopefully find out more about discussion forum theory.
- note: any info that needs to be saved can be done in local storage (so it's only stored by the user's browser and only frontend code)
- note: you can hide as much/little of the UI via like a little "(options)" button or something
* auto-populate the author field on page-load (with something)
* (optional) auto-generate a name as a default if none exists
* it can look real like "Andrew Ryan" or fake like "limit-trial-grain" which is 3 random 5-letter words; however, there needs to be lots of possible permutations. (nb: I think the second option is better)
* 2x checkboxes: "save name for thread" (default 'yes') and "save name globally" (default 'no')
* a drop down of past names you've used in 2 groups: names used in *this* thread, and names used globally. also: an adjacent "forget name" button which deletes whatever's selected (permanently)
* users can set a default name per thread or globally (this takes precedence over generic defaults)
in essence it means:
* global default: still anonymous + no login required etc
* new thread default: ppl keep the same name (they're free to change it if they want)
* it's easier for ppl to manage multiple names
* it encourages ppl to have unique names, multiple names, and different names for different things
however
* people can easily impersonate someone else
* even if that hasn't been a problem it could potentially become one b/c now everyone has a unique name by default -> potential benefit of impersonation goes up
**2. added security around author names (small change / extension to backend/API/DB)**
purpose: prevent ppl from falsely claiming to be other ppl
* use some basic methods for users to "sign" their posts so that it's obvious if two ppl are using the same name but not the same person. in essence there'd be a bit of "unforgeable" text alongside the post which is unique to *that* author (curi/server admins would be the only ppl who could forge this). something like "dg9f87y" or whatever. users would be able to get new ones whenever but it would be super hard to choose one w/out knowing the secret.
* (optional) reject posts an empty "Author" field (note: it looks like an empty string is sent currently). ppl can still put in "Anonymous" if they want, but now it's deliberately chosen like other name.
the only backend changes (I think) would be like:
- accept an extra field with 32 bytes of data (output of a hash alg like sha256)
- calculate another hash via an HMAC -- note: this requires the DB to store a constant secret but it's impossible for ppl to brute force without making like 1000s of posts
- store that 2nd hash with the post
- send that 2nd hash with the post
- render part of the 2nd hash alongside the author
The client would generate a random secret and hash it to produce the 32 bytes they send with the post. So as a function it'd be like:
author_hash = sha256(author_secret)
// send (comment, author_hash) to server
author_fingerprint_hash = hmac(sha256, author_hash, server_secret)
// gets saved to DB with comment
pretty_author_fingerprint = trim(encode(author_fingerprint_hash))
when rendering the page you just have like: "Andrew Ryan - d9s0843v"
important features for *frontend*:
- author_secret should be linked to the particular name and have the same scope (e.g. per thread / global); info shouldn't get leaked if someone uses 2 diff names
- user can forget secrets
- user can generate more secrets
- user can *share* secrets (or set it to "") so ppl can share identities via like community accounts or things
All the frontend stuff is as easy as the stuff in the first idea, and similarly doesn't need and backend changes besides what I've mentioned here (I think?)
I'd prefer not to do user accounts or do author names based on user tracking.
> I'd prefer not to do user accounts or do author names based on user tracking.
"tracking" is maybe contentious but I don't think anything I suggested counts as tracking -- it's all voluntary and under user control. There's also some deliberate stuff to avoid defaults that would de-anon people across threads.
There aren't any user accounts in what I suggested. I don't think they're similar to user accounts either (one reason being that people never authenticate to the server).
Also, if it wasn't clear, option 2 is meant to be an optional extension to option 1 (not mutually exclusive), and I wanted to separate them completely because I thought there was qualitative differences around stuff like privacy and burden on site admin
oh ok. maybe. how does this work with mobile or private browsing mode or ppl who comment from multiple devices?
also i probably don't want ppl claiming generic or common usernames like Roark, Galt, Rand, etc. i don't want those taken on a first come first serve basis by whoever wants to. if people use more distinctive names then it's fine if they claim them.
also if we're talking about adding cryptographic stuff to posts, do you have any ideas that'd make it easier to detect or ban longterm persistent harassers?
idk if there's an important problem to solve here and i think a fair amount of people use Anonymous on purpose and prefer to stand out less.
> oh ok. maybe. how does this work with mobile or private browsing mode or ppl who comment from multiple devices?
Nothing is shared between browsers/computers unless you do it manually (sorta like now). Option 1 is just automatic name generation and local (browser-level) caching, so it'll just remember names you've used, and set names by default.
Nothing stops people using the same name between multiple devices.
Private browsing will naturally save nothing after the session.
> also i probably don't want ppl claiming generic or common usernames like Roark, Galt, Rand, etc.
I think I've been unclear - maybe. There aren't usernames in option 1 or option 2. No one has ownership over any name (besides you with 'curi' in green/teal, but that's a whole different thing). There aren't any "passwords" that let you "log in" to anything. Option 2 has that secret idea, but it's different to a username/password situation. It's more like a cryptographic signature, except on the author name rather than the message.
The only purpose of option 2 is to ensure that impersonation is not possible (unless deliberate). Usernames/accounts are overkill for that.
> also if we're talking about adding cryptographic stuff to posts, do you have any ideas that'd make it easier to detect or ban longterm persistent harassers?
Yes, but not in a way that's simple and easy to implement. One exception might be bots, but if harassers are actual people with browsers and keyboards, etc, then it's (potentially) a reasonably hard problem.
I'll think a bit about this and let you know if I come up with something.
> idk if there's an important problem to solve here and i think a fair amount of people use Anonymous on purpose and prefer to stand out less.
Yeah, I'm not sure either. I don't think it would be *bad*, though. If more people posted anonymously I think the problem would get worse, though, and something like this would fix it.
I think people use Anonymous because it's disposable and default. I don't think they care much on the whole about whether the name is apples-count-lucid or "Anonymous".
I mean one simple experiment would be to auto-populate the "Author" field with 3x-random-words like "apples-count-lucid" and see how many people change it. If people keep using "Anonymous" then they have to be putting that in manually. I don't think you'd see many people changing it back to Anonymous.
This might help to explain what I mean:
For the record (and b/c I haven't saved it) this was the HTML I used for the above (just pasted it into chrome dev-tools)
```html
<p>
Author
<input id="comment_comment_writer" name="comment[comment_writer]" size="30" type="text">
<sub><a>(hide options)</a></sub>
</p>
<div id="comment_tools_author_options" class="comment_note_2" style="border: 1px solid black; padding: 4px; margin: 4px;">
<label style="display: block;"><input type="checkbox" checked="true"> Remember name?</label>
<label style="display: block; padding-top: 2px">Set "Andrew Ryan" as:
<button>thread default</button>
<button>global default</button>
</label>
<label style="display: block;">Previous names:
<select>
<option> -- Thread --</option>
<option>lucid-apple-dance</option>
<option>Andrew Ryan</option>
<option> -- Global --</option>
<option>sails-seems-tango</option>
<option>Anonymouse (squeak)</option>
<option>oh my god its turpentine</option>
<option>AnneB</option>
</select>
<button>Forget "Andrew Ryan"</button>
</label>
</div>
```
Does it save the data in cookies?
I'm wary of having extra code to deal with or maintain.
Hmm. Formatting didn't come out well. Paste this into a JS console to recover the original string with spacing, etc. (It's just base64 encoded)
let encodedHtml = "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";
console.log(atob(encodedHtml));
#17950
> Does it save the data in cookies?
Nope. This is what I imagine the data structure would look like.
Notice it's all in local storage. That's private per browser, and only settable by code that runs on the page. It's never set to the server (**not** like cookies)
One thing that is annoying is forgetting to post under a name, then I need to think "is it important if my name is on it or is anon okay?"
I mean it's easy to guess that last post was from me, but there's no way for me to keep a name by default.
One dangerous thing about defaults is accidentally posting under the wrong name. I think a *thread level default* is okay for this, though. If someone wants to post under two different names consistently, they can always use two different browsers (or profiles on the same browser), anyway.
It's like how you use a different screen for streaming. I use multiple chrome profiles for AWS so that my "production" browser profile can use a bright red theme (that way I am not going to accidentally do something bad on a live system)
In general, autofilling a username (or other form field) strikes me as something a browser (core tbh, or extension/plugin) should handle. Browser autofill is definitely a widespread thing. I imagine there is some way for people to handle this already.
> I'm wary of having extra code to deal with or maintain.
I would be too. For the basic idea (option 1) the JS code would *just* be enough to glue together those UI components I mocked up.
The HTML I posted has a footprint in javascript on the order of 500B. I think I could probably do the rest in about 500B-1000B depending on optimisations (smaller => harder to read).
Like I mentioned above, all native JS, no libraries etc.
I don't want to work on it if you don't want to use it, though. It's fine if you're interested to see how it works and don't want to commit to using it; I just don't want it to be like a complete waste of time.
#17950 Alternate proposal: add a "Generate Random Name" button that fills in the Author field with a randomly generated name, kinda like AutoAdmit does. Whenever the user wants to keep the same identity in future comments, they can copy and paste (or retype) the name.
The button could be implemented in pure JavaScript. No server-side code changes would be required. Also, adding that button to the form wouldn't complicate the UI much.
> In general, autofilling a username (or other form field) strikes me as something a browser (core tbh, or extension/plugin) should handle.
Plugin could work for this, but then it's not the default (it is for that one user, but not for comments *in general*). Also plugins can't be used on (or much harder to dev for) safari or mobile.
> Browser autofill is definitely a widespread thing.
Not full blown autofill (like auto-populate on page load) - that's a security risk for stuff like names/addresses/etc b/c JS can react to form-filling.
There have been issues in the past with websites putting up hidden inputs with field names/IDs like "address". When ppl click the drop down that chrome shows with names/email combinations, etc, chrome would *also* fill in the hidden address field (which the user did NOT know they were sending).
So there is some automation, but even if a browser remembers inputs it won't auto-populate the fields.
I don't know how I could tell chrome "On curi.us find the field named 'comment_comment_author' and fill it with 'Andrew Ryan' by default on page load"
> I imagine there is some way for people to handle this already.
Plenty of ways, tampermonkey/greasemonkey can work too.
I was making the above suggestions **particularly** b/c of how it would **change the *default* around names**, tho. Like by default you keep the same name in the same thread, but it's an anonymous and disposable name.
The rest of the stuff is just some quality of life features, really.
#17957 It also probably wouldn't be too hard to make a bookmarklet that generates a random name and fills it in.
Another idea for implementing Andrew Ryan's proposal would be to make a browser extension for posting on curi.us. The extension would be responsible for remembering the user's preferences, such as favorite Author names and global/thread Author name defaults.
#17957
> #17950 Alternate proposal: add a "Generate Random Name" button that fills in the Author field with a randomly generated name, kinda like AutoAdmit does. Whenever the user wants to keep the same identity in future comments, they can copy and paste (or retype) the name.
> The button could be implemented in pure JavaScript. No server-side code changes would be required. Also, adding that button to the form wouldn't complicate the UI much.
Yup, I think that's a good MVP. I think it should also auto-generate the name if you *don't* put one in (i.e. it doesn't default to 'Anonymous'). You can use 'placeholder' as a way to show the grey-looking anon name if you don't put one in. I'll do a mockup now.
*However*, this doesn't solve the problem of knowing who's who in a conversation with lots of anons. You just get lots of random looking names.
*Maybe that should just be solved with community norms and expectations, tho?* Like if you want to have a conversation as anon, you should use the same alias during the convo. If you keep using new names ppl can just not pay as much attention or ask if you're the same person / why you're acting like you were the previous poster but using a diff name, etc.
It could look like this?
#17959
> bookmarklet that generates a random name and fills it in
Yeah, the list of names might be an issue. if you gen random names from a particular list (esp if no one else does) then you're leaking your identity because no-one else will be able to generate those names.
WRT to anonymity it can be easy to make little mistakes that lead to complete de-anonymisation (potentially *much* later)
That's another reason the browser extension isn't good for anon stuff -- you should *never* be able to tell if someone is using it because that info is leaking stuff about them and which other posts they might have made.
> The extension would be responsible for remembering the user's preferences, such as favorite Author names and global/thread Author name defaults
This makes more sense for like long term posters with established identities.
#17959 Here's an extremely simple bookmarklet that fills in the curi.us Author field with a random name: https://pastebin.com/raw/CCNKcN2f
#17963 I copied the name generator source from here: https://gist.github.com/tkon99/4c98af713acc73bed74c (it's not tricky code).
Then I added code to fill in the Author field and wrapped it all in a function to avoid name clashes.
dope
#17946
I thought this was funny to see in the GH comments (from the author)
> tkon99 commented on Mar 28, 2018
> Hi, nope you're free to use them for whatever you like, I took them from two websites listing the most popular nouns/adjectives.
> If you do use them, please attribute to my Github :D. I also have this as an npm package called "naampje".
I think I'll leave it alone for now. I think random names would be worse than people clearly labelled Anonymous.
Cool, I think that makes sense if the issue is just a local artifact of this post (2374).