Well, I think I might've been unclear. I wasn't actually suggesting that upvotes come with authorship labels. All the reasons you list for why this isn't a great idea, I agree with.
I was saying, rather, that the upvote/downvote system is fundamentally missing something; that it can't substitute for expressing explicit verbal agreement. The immediate corollary that should occur to us is: what is voting even for?
Consider a scenario. I write a post about software usability. A hundred people read it, and have a strong enough opinion on its quality that they are moved to click the voting widget. 99 of those people are ordinary LessWrongers, with no particular expertise in the subject. They upvote me. The 100th person is Jakob Nielsen. He downvotes me.
My post now has a score of 99 points. Is this an accurate representation of its value?
No. One “layman” doesn't equal one Jakob Nielsen, when it comes to evaluating claims or opinions about usability engineering. Even 99 laymen doesn't equal one Jakob Nielsen. If Nielsen thinks that my post is crap, and that basically everything I'm saying is wrong and confused, well, basically, that's that. 99 non-expert LessWrongers doesn't “balance that out”, and the sum of “99 LessWrongers think I'm right” and “Jakob Nielsen thinks I'm wrong” does not come out to “a score of +99! what a great post!”. That's just not how that math works.
Furthermore, suppose Nielsen posts a comment under my post, saying “this is crap and you're a nincompoop”. What, now, is the value of that “99” score, to a reader? You now know what a domain expert thinks. Unless other domain experts weigh in, there's nothing more to discuss. That 99 LessWrongers disagree with Jakob Nielsen about usability is... interesting, perhaps, in some academic sense. But from an epistemic standpoint, Nielsen's hypothetical comment tells you all you need to know about my post. The upvote score is obviated as a source of information about my post's value.
And yet, it's the upvote score that would be used, by various automated parts of the system (and by readers who aren't checking the comments carefully), to decide how good my post is. That seems perverse! Now, I'm not suggesting that "sort by experts' opinions, as expressed in comments" is a viable algorithm, of course. But this scenario, in my mind, calls into serious question what upvotes mean, and what sense there is in using them as a way to judge the value of content.
I personally think to grey lines on the side do a pretty good job, but I also think that the boxes on LW 1 are doing something that makes things clearer. I do think that the LW 1 comments boxes do look a little junky though, and I'm very much enjoying the clean look of LW 2.0 overall. Not sure what a good compromise would be. Maybe all top level comments are a little more distinguishable in some way?
I agree there is something nice about being able to see who upvoted or downvoted a comment or post, but I don't think I'd want this to be the default. I expect I'd feel uncomfortable voting on some stuff if I knew that my vote would be public. Maybe after voting, an option could appear that said something like “Make vote public”. Then you could have something pop up on hover (or with a tap on tablets/phones) that showed something like “Malo and 3 other people upvoted this post”. Though that would probably get unweildy if lots of people made their votes public. I think there's a good idea in there though, just not sure about implementation specifics.
Yeah upvotes can mean a lot of different things like endorse, agree, or high quality comment (even though I disagree). This comment thread on another post discussed some potential extensions to upvoting that might help with this.
I don't think this is working for me. I just made a bunch of comments last night, and got a couple replies since then. When I visited the site today I only noticed people had added comments when I saw then in the recent comments section.
How's this supposed to work?
Re: #1: I to am a big fan of Practical Typography :) That's a pretty good point, I actually don't thik we disagree much. I think I may prefer just slightly prefer whiter backgrouds with slightly grey text. But only slightly.
Re: #2: I largely agree with this, though I might lean more on the side of giving the user less configuration options. Like, if you give everyone an option for everything, then the options get real cluttered. But I don't have strong feeling about adding this preference in general.
Re: #3: Totally.
Request 2: tagging people in relevant discussions FB style. IMO this is *uge* but might need some thinking to avoid high status people drowning in the noise (1 straightforward possibility is sorting your inbox by karma)
Request 1: crisper outlining of comments. LW1 uses nicely visible boxes and colors to separate comments; IMO this is good
Having written the parent comment, it occurred to me to wonder whether it was needed or useful; after all, there's already an upvote button, right? (Which I dutifully clicked, of course.) Did I just write the comment out of habit, having spent considerable time commenting in venues with no voting feature?
But what the upvote counter doesn't tell me is who upvoted something!
As a commenter/upvoter, I'd like to (have the chance to) communicate something more than "someone liked this post/comment enough to click upvote"; I'd like to convey "Not just someone, but I, whom you may know, whose views you may be familiar with, whose credentials on relevant topics you may investigate, agree with / endorse / support / etc. this post/comment".
And as a reader, I'd like to know who it is that agrees with, endorses, supports, etc. the post/comment in question. Maybe their opinion carries great weight with me; maybe they mean nothing to me; maybe their endorsement is, for me, an anti-endorsement.
(Then, of course, there's the old problem that it's not actually all that clear what it means, to upvote or downvote a comment. (All of you who disagree, and think it is clear—how sure are you that it's clear to everyone else, or even that everyone who thinks it's clear has the same understanding? Have we checked?) This, for me, was probably the biggest problem with LW 1.0's karma system, because it was fundamental and conceptual, and transcended any issue with moderation or what have you.)
So, in other words: upvoted, yes. But also verbally endorsed, and the endorsement signed.
Fully seconded, on both points.
Thanks!
Re: #1: It is common practice to make body text off-black. Is it good practice? Well, Matthew Butterick’s book, Butterick’s Practical Typography—considered a definitive work on the subject—uses black text.
You may note that Butterick suggests using off-black text—but consider his reasoning: the issue is contrast! As Butterick notes, screens emit light rather than absorbing it, making high contrast potentially painful to look at. Indeed; but darkening the background reduces the amount of light emitted, while lightening the text increases it. The former is superior as a way of reducing contrast. (Just don’t do both! That's wholly unnecessary.)
Edit: Check out readthesequences.com for an example of “black on off-white”.
Re: #2: Something to be A/B tested, I suppose. (Alternatively and even better: have this be user-configurable, via the account settings page, e.g.: "Display vote widget (•) above post only ( ) below post only ( ) both above and below post". "Sane defaults plus comprehensive configuration options" is the gold standard of UX design for such matters.)
Re: #3: This is exactly the point of responsive design. Hover for desktop clients, hamburger for mobile. There is no reason at all to insist on a single, unified solution; web UIs should at all times be appropriate to the platform they're being viewed on.
This is an epic comment with lots of great ideas and observations.
A few comments/opinions:
I don't think the text should be proper black as in #000000. I find that slightly off black makes for a better reading experience, and I think this is pretty standard practice, though I may be mistaken.
I think it's a feature that upvotes and downvotes appear above and below. I may want to see the count at the top before reading, but then again at the bottom so I can vote once I've read the post.
Agree that hamburgers aren't great, but hover based UIs aren't tablet friendly, so I'm not keen on that solution.
Strongly agree with the comment editor not being visually distinct enough. Multible times when writing this comment I've scrolled up to reference your comment, and then found it a little annoying to find where I was typing my reply.
Nice!
Two thoughts:
What about adding a small link icon next to the time that is the link to the comment. Having the time be the link is pretty hard to discover. Facebook does it this way, and it took me a pretty long time to consistently remember, and rediscovering was really annoying.
I think the idea of displaying the linked comment at the top of the page is cool, but I also find it a little confusing (like I instictively think “where’s the rest of the discussion” for a quick sec). I also almost always click the “Show comment in full contetxt” link. Given this it seems to me that brining the user directly to the comment in context might be best. Maybe the comment could be highlighted in some way so that it was easy to see which comment was linked to.
Harvard Law Review also has a pretty classy way of doing footnotes (example post).
Yeah that would be really great. Medium does this kind of well. Chris Olah's blog also has this feature (example post), but it’s implemented in a pretty hacky way using Disqus.
It would be cool if you could highlight some text in a post, and there was an easy way to create a comment that quoted that part of the text. Maybe you could even show some sort of visual highlight on that text in the post if the dicussion is high quality (measure by come combination of Karma and lenght?).
Very nice!
Some comments on the design, pro and con:
Grey text makes it harder to read than plain black. If you want to reduce contrast, consider instead making the background off-white (either #fcfffa or #f9fff5 would be fine — 99% or 98% lightness of the same hue as the theme color green).
A bit hard to detect when one post ends and another begins, on a skim. Some possible solutions:
Add border to posts (on class "comments-item", add border: 1px solid #cbd6c2, padding: 4px 6px); or,
Add background color to posts (on class "comments-item", add background-color: #edefeb, padding: 4px 6px); or,
Greater v-spacing of posts (on class "comments-item", set margin-bottom: 30px)
The hamburger menu only closes with another click on the hamburger menu (best practices is for a click anywhere else on the page—outside the menu—to close it)
Upvote/downvote buttons and vote count appear twice on each post (above and below)
When adding a link in a comment, hitting Return/Enter in the link URL popup does not cause it to close (i.e. does not have the effect of hitting the Submit button) (Mac OS X 10.9, Chrome 60.0.3112.101)
Mini-hamburger menu on each post has just one menu item (Subscribe); consider simply having a Subscribe link on the post itself, in small text
Hamburger menus considered harmful in general; consider substituting an on-hover drop-down (achieves the same effect, but with fewer clicks)
Line spacing ("line-height" CSS property) of body text of posts (not comments) is too large (more than 1.8); consider reducing to 1.5 (with the short line length—which is good!—this should be more than sufficient to ensure readability)
Conversely, line spacing of body text of comments is a bit too low (1.25); consider increasing to 1.35
The bold weight of the body-text font is barely distinguishable from the regular weight. This is, unfortunately, an uncorrectable feature of the "ET Book" font; no heavier weight is available, as far as I can tell. Consider using one of these free alternatives:
Alegreya
EB Garamond
Crimson Text
(or something else; in any case, the kind of font is just right, but this particular font family happens to have this flaw...)
Consider adding something to visually distinguish the comment currently being written, from surrounding comments; perhaps, a lighter, or darker, background (depending on whether you implement one of my suggestions above for visual separation of posts)?
The reason this is helpful is that, when writing a comment, I might scroll around the page—to the OP, or to other comments—to reference other people's words, etc., and then want to continue writing the comment. (Note that the Tab key does not take me to the comment editing field, which would mostly obviate this issue; this, I presume, is just a feature of how this editing UI is implemented...)
Or rather, the lack thereof. To be added in the future, I hope?
Almost everything else!
The body text font is attractive and readable and the default text size is good for readability (on a 1080p desktop monitor, anyway) (but see Cons)
The text column width is good
Layout of comment blocks is attractive and uncluttered
Overall typography is well-done
Uncluttered and aesthetically pleasing visual design
(mostly) excellent front page layout (but see below); puts the interesting stuff front and center
Layout of post pages also solid; easy to navigate
Seems like the layout will transfer quite well to mobile platforms (looking forward to seeing the design of the mobile version of the site!)
Well done on the redesign of the post sort order widget! This is spot-on
Very nice comment composition/editing UI! I haven't been actually impressed with one of these for a long time; is it custom or third-party?
The stripes to the left of the comments (indicating comment hierarchy) are quite useful
Seems good so far; no major issues!
Here's how the front page looks on a 1080p screen…
Ok, nothing too wrong here...
And here's how the front page would look like on a 4k screen…
Hmmm.
Consider allowing the front-page content to occupy two columns, on wide viewports. (Perhaps, Recommended Reading and Featured Posts on the left, and Recent Posts and Recent Comments on the right?)
Currently, when I'm viewing a post or comment page, and I wish to navigate to somewhere else on the site, I have to first go to the front page, then go from there to where I want to go. This is not quite optimal.
As an alternative to removing the hamburger menu (or transforming it into an on-hover drop-down), consider expanding its use; perhaps, for example, put a copy of the Recent Posts/Recent Comments/Feature Posts feeds in there?
Overall, good work on the redesign! Thumbs up :)
This is now done! Interested in your thoughts on the implementation. We haven't yet exposed the sequence editor, but the navigation and style can be seen in the three major collections linked from the frontpage.
I think this http://lesswrong.com/lw/9v/beware_of_otheroptimizing/ is a large part of it, you need the other person to be part of the process to avoid the failure mode.
(But also, I think it's healthy to develop relationships with specific people who can give you "unsolicited" feedback/observations. True they're going to be less nuanced and maybe off-base, but too much value could be lost without being open to these, we're frequently too blind to our own behavior)
Playing poker at higher levels actually requires one to practice this skill a lot.
Pyrrhonian Scepticism as described there sounds like the thing I'm arguing against as not-quite-right: looking for the negation. The idea implies that you're attached to a hypothesis. It sets a low bar, where you come up with one other hypothesis. I won't deny that this is a useful mental tool, but false dichotomies are almost as bad as attachment to single beliefs, and for the same reason, and it sets up a misleading standard of evidence. The idea that you generate experiments by trying to falsify a hypothesis is confusing. It's better than trying to confirm, but only because it starts to point toward the real thing. You generate experiments, and evidence, by trying to differentiate.EDIT: Ok, that seems too strong. "trying to disprove"/"looking for the negation" is a convenient whipping-boy for my argument, because it's a pervasive idea which value-of-information beats. Nonetheless, asking "what if I'm wrong about that?" is more like the starting point for generating multiple hypotheses, than it is an alternative. So, the method is inexact because it is incomplete. It's likely, for example, that the way Peter Thiel employs the method amounts to the whole picture I'm gesturing at. But, there's a different way you can employ the method, where the negation of your hypothesis is interpreted to imply absurd things. In this version, you can think you're making the right motions (not falling prey to confirmation bias), and be wrong.
The steel-man of Pyrrhonian Skepticism is something like "look for cruxes" in the double-crux sense. Look for variables which have high value of information for you. Look for things which differentiate between the most plausible hypotheses.
HT Gwern. The important bottom-line is that with a sample of 147k participants, we can now predict 6.9% of phenotypic intelligence. Relevant quotes below, with my emphasis:
This study had four goals: firstly, to facilitate the discovery of new genetic loci associated with intelligence; secondly, to add to our understanding of the biology of intelligence differences; thirdly, to examine whether combining genetically correlated traits in this way produces results consistent with the primary phenotype of intelligence; and, finally, to test how well this new meta-analytic data sample on intelligence predict phenotypic intelligence variance in an independent sample. We apply Multi-Trait Analysis of Genome-wide association studies to three large GWAS: Sniekers et al (2017) on intelligence, Okbay et al. (2016) on Educational attainment, and Hill et al. (2016) on household income. By combining these three samples our functional sample size increased from 78 308 participants to 147 194. We found 107 independent loci associated with intelligence, implicating 233 genes, using both SNP-based and gene-based GWAS.
The effects of age, sex, and population stratification (7 components) were controlled for using residuals extracted from a regression model.
Using our meta-analytic data set on intelligence we carried out polygenic prediction into Generation Scotland: Scottish Family Health Study and found that 6.9% of phenotypic intelligence could be predicted (Table 2), an improvement of 43.75% on previous estimates of 4.8%
Yes, this also seems related to a lot of the stuff I've been talking about with Fermi Modeling, and also a bunch of stuff that Peter Thiel calls Pyrrhonian Scepticism.
Quick googling returns: Mastering the Core Teachings of the Buddha, an Unusually Hardcore Dharma Book
MCTB?
Thanks. Sorry about the lost comments. :(
I want to talk more about the pain model in the book, but I'm deliberately not writing up the principles. I don't think people would get much from a summary that's more than what I have here but less than the whole book. Being walked through it was really integral to my getting the benefits.
Apparently only the thoughtful ones
Testing if lw is eating my comments.
Nice, thanks for writing this up.
The prediction is that improving the quality of processing via the principles explained in the book can reduce pain and increase your physical capabilities.
Is there a summary of the principles somewhere?
Yes! I agree. I also see that as a key feature.I've been working on this, but apparently forgot to add it to the feature list. This is related to improving search in general by allowing you to not only search through posts but also comments and user-profiles which is high-priority for me.
From a user's profile, be able to see their comments in addition to their posts.
Dunno about others, but this is actually one of the LW features I use the most.
(Apologies if this is listed somewhere already and I missed it.)
This is now done! I am not fully happy with the implementation, but it feels like the best compromise I could find between a few different design goals with comment links.
You can access the link for each comment by clicking on the time the link was posted. When someone visits that link, the comment that you linked gets rendered at the very top of the post page, together with the immediate comment it is replying to (if it is a reply). That comment then has a link at the bottom that when clicked scrolls you down to the position of the comment in the full context of the discussion threads. My hope is that this will both allow comments to stand reasonably well on their own, as well as make it easier to find a comment in the context of the whole discussion.
For example, writing public event summaries. These could be short, and would give a highly visible sign to assure others that the meetup is still happening (and give them a bit of the flavor of it).
A thing I'd be worried about here is people being ashamed that their meetup is just a social club or something, or in general optimizing for doing things that sound good rather than are what the participants want to do.
For recurring events, it would be nice to have some assurance that the event is still happening. Recurring meetups which stop should disappear after a while, but confirmation of continued active meetups should (1) require minimal effort (2) not be easy to forget.
(1) and (2) are somewhat conflicting. Favoring (1) leads me to suggest something like LessWrong prompting the event creator once every six months: "Is this meetup still happening?" Favoring (2) leads me to suggest something which not only confirms that the event is still happening, but is valuable in itself, so that doing it can feel rewarding. For example, writing public event summaries. These could be short, and would give a highly visible sign to assure others that the meetup is still happening (and give them a bit of the flavor of it). It would also cause people to think about what they're going to write, so that they have an open loop in their brain about writing the thing. It can be a rewarding habit. And, these summaries could be useful in themselves.
But, some groups would doubtless fall out of the habit.
Hm. What I want is a tree which is really just a linear structure functionally speaking (no automatic queuing of prerequisites or anything), but which allows more visual organization of a long sequence into parts for sanity's sake. I'd get >90% of the functionality I want just by being allowed to intersperse headers on a "sequence" page to organize sub-sequences. The sequences themselves currently have one more level of organization than that, with "books" and then "sections" -- so you can think of them as one long sequence with six sub-sequences and twenty-six sub-sub-sequences. That's the kind of hierarchical organization I'm asking for.
Hm. That seems really clunky to me, because the author has to decide where comments belong; but it does seem likely to result in more organized discussion.
Arbital 2.0
Blogging / social media platform.
Initially: 1) make math blogging much nicer, 2) help people connect over similar interests. Eventually: change the shape of the social media and information flowing through it.
If you know math bloggers, I'd appreciate a referral so I could tell them about the platform and see what their needs are.
Also asking "how do you feel about that?" helps, although might come off a bit psycho-analytical if asked repeatedly and directly.
This is now done! Interested in your feedback on the implementation!
(For example, I can imagine people wanting to be able to select manually what the new comment date cutoff is, which is what SlateStarCodex has. Open to implementing that.)
We should get cell-tower accurate positioning, which is usually good enough to determine the city and neighborhood you're in (i.e. we should be able to distinguish between Oakland and Berkeley, but probably not central Berkeley and North Berkeley).
Rough googling suggests about 95% accuracy on city-level detection.
Does that work on mobile?
Julia Galef has a book deal, and could probably point you to her agent.
A merge accounts feature is on the dock. Until then, I am happy to give anyone access to their old LW accounts who pings me on the Intercom chat.
My username on the closed beta is based on my email address, not my username from LW 1.0. Will these accounts be merged when 2.0 goes live, or will both of these accounts exist independently, or something else?
I agree, that's a more accurate description. The sense in which "true prisoner's dilemma" is impossible is the sense in which your utility function is the cooperative one you commit to. It makes sense to think in terms of your "personal" (original) utility function and an "acting" utility function, or something like that.
I still think this undermines the point of the "true prisoner's dilemma", since thinking of humans gives decent intuitions about this sort of reasoning.
I won't try and argue that this strategy in particular is ideal. (FYI, this is the strategy called "nicerbot".) However, the general pattern I'm using it to point at, where you give just a little benefit of the doubt, is only slightly exploitable as a rule. This will often be worth it due to a number of cases where the strategy helps force good equilibria.
Well, we have location by IP, which is rough but accurate enough for meetups. Only more detailed location requires permission.
Probably not suitable for launch, but given that the epistemic seriousness of the users is the most important "feature" for me and some other people I've spoken to, I wonder if some kind of "user badges" thing might be helpful, especially if it influences the weight that upvotes and downvotes from those users have. E.g. one badge could be "has read >60% of the sequences, as 'verified' by one of the 150 people the LW admins trust to verify such a thing about someone" and "verified superforecaster" and probably some other things I'm not immediately thinking of.
However, if you cooperate with probability p+0.001, then both people are trying to be a little more cooperative than the other. You'll cooperate 100% of the time with others following the same strategy, while sacrificing very little in other situations.
Really haven't thought much about this, but my brain wants to say that this strategy must clearly be exploitable somehow.
By the way, in one sense, the "True Prisoner's Dilemma" is impossible between agents of the sort I'm imagining. They see the game set-up and the payoff table, and immediately figure out the Nash bargaining solution (or something like it), and re-write their own utility function to care about the other player.
This seems strange to me. My intuitions about agent design say that you should practically never rewrite your own utility function. The thing that "re-write their own utility function" here points to seems to something more accurately described as "making an unbreakable commitment", which seems like it could be done via a separate mechanism than literally rewriting your utility function. Humans seem to do something in that space (i.e. we have desires and commitments, both of which feel quite different and separate from the inside).
I very much agree with the broad gist of the post, but also have many specific points that I disagree with. This feels like a post for which inline-commenting or at a special content block that allows a comment-thread to start from that place would be extremely useful.
In the absence of that, I will write multiple replies to separate parts of the post, so that we can keep the discussion threads apart.
A comment feed that starts from the top-rated comments since my last visit would be much more useful for me than the chronological feed, and would probably increase the number of discussions I become involved in.
I don't think that using up prominent real estate on core sequences is a good idea. I think that /top/ on lesswrong is a better idea, and agree with jim that mixing in top/unread with recent stuff would be pretty reasonable (e.g. if 1/3 of stuff I saw was just the highest rated article I hadn't read or skipped too many times).
People browsing my blog seems much more likely than people browsing my user page. More generally, stuff I post on my blog seems to reflect on me in a way that stuff I post on LessWrong does not (instead reflecting on LessWrong).
Medium removed commenting though, not sure why.
Medium comments are great and beat GDocs comments in particular.
I'm writing a book on cancer research -- how it's stagnated in recent decades, and what new research directions might offer more promise.
My hope is to influence funders (primarily investors) towards focusing on more diverse types of cancer treatments.
How people can help:
I'm looking for an agent. (I have a draft and a book proposal.) If anybody has contacts, I'd much appreciate it.
LessWrong is not an appropriate discussion venue for both of my most important current endeavors, so instead I'll start us off with something a bit lighter weight that might help capture the spirit I'd like the thread to have:
A Complete History Of The Word 'Hacker'
I am attempting to chronicle the complete history of the word hacker, from its beginnings in the 1950's as jargon among MIT students (and possibly even earlier as a term used by Ham Radio operators) through the decades to the present day. This includes the split between the MIT hacker community and the phreaking community (which later evolved into the computer 'hacking' community of intrusion artists). For more information see my blog post summarizing most of my research so far.
The basic impact I expect this project to have is to contribute to history, in particular the intellectually and culturally fascinating history of the two major subcultures chronicled by the proposed work.
You can help if you happen to meet one of the following criteria:
You happen to have been alive during and participated in the early MIT/Stanford/etc AI scene, or were otherwise involved in the early ARPANET. In this case the best way you could contribute would be by going on record and telling your story so that the recollection is not lost to future generations. (Not impossible given the major AI focus of this community.)
You know someone who meets the first criteria, and can convince them to go on the historical record.
You're aware of documents or artifacts which predate the ones I've listed as earliest examples of a phenomena. Your pointing these out to me would be massively appreciated!
You're aware of interesting documents or artifacts that are not already in my bibliography on that page. I know it's quite long, so if you don't feel like reading it feel free to just message me with what comes to mind.
You also have an interest in the subject, and would like to partner up on researching it. In that case message me and we'll discuss where I'm at with things and what avenues are promising.
Yes! I often put this lesson down in my top three most useful things learned at CFAR.
In the pair debugging sessions, if I came up with a solution to my partner's problem, straight up presenting the solution rarely worked (it was often wrong in some way, and then we were conversationally a bit stuck). However, when I asked leading questions around where I thought the solution could be (e.g. "Is there a way to remove subproblem X?"), they found it themselves, or sometimes better solutions.
I now make it my goal to understand their problem and its different parts, rather than to solve it, and this allows they themselves to generate the best solutions.
Putting everything into a giant book is not a reasonable form of organizing institutional knowledge. (At least not beyond a certain point.)
In general, I feel like the sequences approach strangled the intellectual growth of LW 1.0 by sending people the message that unless their content is in the top 0.1% of content they should just go home. So that's what people did.
If you want a thriving discussion you're going to need to have a space for discussing things with reasonable but not incredible quality standards. You do not want to give users the message that for their discussion to be worthwhile it must ultimately be worthy of inclusion into a canon. Trying to build a canon too early just strangles everything. You're overemphasizing it because it's what EY did when you really should if anything underemphasize it.
p.s. Overemphasizing the construction of a canon also has negative PR effects, because it gives the impression of a fanatic group.
What do you feel is the biggest benefit of using your blog instead of LessWrong? I definitely have a sense of why many authors in general prefer writing on their own blogs, but I would be interested in a short summary of what causes that preference in your specific case.
I don't foresee posting to LW in place of posting on my blog without some unanticipated value add. I would be happy to use a shared comment system or something like that.
If interesting discussions are occurring I would be reasonably likely to chime in from time to time, as I currently do on LessWrong but with a frequency determined by the frequency of interesting conversations.
As many of you might have noticed, after a discussion with Malo and some great suggestions by him, the typography for the whole page is updated and looks a lot better!
I will still do a larger typography rework at some point during the closed-beta, and will obviously do adjustments as I notice problems with the current setup, but I am definitely happier with this.
So a few things I've been thinking about to solve these problems:
I hope we can deal with Eugene after we have a better Karma system and better moderation tools. Though we will see how hard this problem turns out to be. But my guess is that even if it turns out to be harder, it is a problem that can be solved with sufficient, but not prohibitive, engineering effort.
This current page already has a more aggressive scoring system that keeps highly upvoted posts at the top for longer, and which applies an exponential decay function over time, which results in you getting a mixture of recent posts and top posts as you scroll down the frontpage, instead of mostly a historical list of posts. I hope this increases visibility. I was also thinking of maybe adding a "promoted" section to the top of the posts list that always shows the top 5 promoted articles (by time-decayed score), so that you have something similar to main, without it being a whole click away.
I am currently hesitant to create more top-level distinctions like main/discussion/forum, both because I think it decentralizes discussion, adds mental overhead and encourages styles in different parts of the page, which makes it harder to promote the best content from anywhere on the page (i.e. if the forum is mostly written in a style appropriate to forum, then it's harder to promote that content to the frontpage with all the long-form content)
First click could be the generic upvote or downvote; then using a second click you could pick a more specific "flavor" of the vote. (Different flavors for upvotes, and for downvotes.)
We could have some scarce resource based on karma. Not paying with karma directly, because I guess losing karma would feel bad, but rather that with each 100 karma points you get 1 "credit".
You could then spend those credits e.g. on visually highlighting other people's comments and articles. Something like when Reddit displays that a comment got "Reddit gold". It could even transfer some karma (but much less than it costs) to the rewarded user, but mostly it would be a costly signal of "I really liked this", with the name of person giving the reward displayed as a tooltip. A costly version of "+1 nice", essentially.
Can individual articles in the sequence be written by different people? In such case, who manages the sequence itself?
Sometimes you want to talk about "A1 vs A2" and sometimes you want to talk about "A vs Z".
Maybe the distinction between A1 and A2 is rarely made, and most people just perceive them both as a general A, but today you want to talk exactly about those differences, so you need to make the difference clear by putting unusually large emphasis on the differing details. (Maybe the language doesn't even have three different words for A, A1, A2; sometimes there is just one word for both A and A1, and people would say e.g. "A in the wider/narrower sense of the word".)
But the next day you want to talk about differences between A and Z in general, and anyone who focuses on differences between A1 and A2 is losing the larger picture... well, unless those differences between A1 and A2 are actually relevant for the "A vs Z" debate you are trying to have; but in such case, the person insisting on the difference should explain how specifically it is relevant.
Essentially, greater precision comes with greater costs, so the person who increases the cost should communicate the cost-benefit analysis to their partners in the debate. Sometimes the extra cost benefits everyone by helping them reach better conclusions. Sometimes the extra cost only benefits the speaker by allowing them signal greater sophistication. Sometimes the speaker is not even aware that these two cases are different (i.e. they may be merely signalling, but it's not an intentional defection, only just blindly copying what they saw other people do in similar situations). Sometimes the speaker is too mindkilled about "A1 vs A2", e.g. because it has huge political connotations for them, so they are emotionally opposed to using "A" as an umbrella term for both.
Similar here. Two things that discourage me from posting on LW:
- knowing that if the debate gets interesting and will have even smallest political connotations, Eugine will likely harass the participants (well, before the downvotes were disabled, but disabling them brought new problems); furthermore, the potential participants are already aware of this, which makes them less likely to participate;
- there is a ton of stuff each week in Discussion, some of that low quality or just a link; regardless how much effort I put into my article, in two or three days it will be scrolled down into oblivion anyway; this was traditionally solved by Main vs Discussion, but the articles in Main paradoxically had less visibility, and the threshold for both of them was constantly lowering anyway, Main becoming the old Discussion, and Discussion becoming the old Open Thread comments.
So I'd like to see the discussion kept nice; and I'd like to see the good articles have a longer timespan.
(Not insisting on any specific technical solution, but for example the problem with "actually less visible Main" could be solved by simply posting links to 5 latest Main posts at the top of the Discussion page. There, everyone would see them, even if they only read Discussion.)
(Similarly, the "Discussion becoming new Open Thread" could be solved by having three levels, corresponding to previous Main, Discussion, Open Thread, but perhaps using different names, e.g. "Promoted", "Articles", "Forum". To avoid having to estimate the category of your own article, all new articles would go to "Forum", and then moderators would move them into higher levels using their own judgement, with karma as a guideline.)
An alternative would be to focus on "communities" instead of meetups.
This is the basic idea; a consideration to keep in mind is that there's both passive new users ("hey, I just moved to Austin, are there any meetups nearby?") and exciting events ("hey, we're hosting a HPMOR wrap party!"), and you want to handle both cases well.
Now, maybe the thing to do here is have something like findable communities (basically, what's on LW is a geolocation and a link to Facebook/whatever else you use) and then location-based pings (either based on IP or them letting us have their location), which the community owners can create. But I don't really want LW to be prompting users to allow us access to their location all the time.
There are word pairs where there are distinctions that seem unimportant to me (less vs. fewer, comprise vs. compose, etc.) and then other where they do seem important to me (a marginal case on this side of the divide is that vs. which), and if someone is conflating the two it seems like to right move is to complain "hey, I was using that distinction!"
I suppose the annoyance I'm trying to point at is something like, "Hey, I was reserving those words for momentary distinctions!" To give a programming analogy, it's like a library which clutters up the global namespace with a lot of function names like "insert" which you just know are going to conflict with something else (especially in a language which doesn't make this easy to handle).
For example, if I define "knowledge vs wisdom" in a way that's useful to the argument I'm going to make, and someone objects because my definitions seem too nonstandard, I would hope that they're suggesting better words to fit my distinction as opposed to "knowledge vs wisdom", rather than better definitions for knowledge and wisdom. Unless they already understand the argument I'm going to make and are trying to improve it, the second seems far less likely to move the conversation in a useful direction.
Yeah, I don't think I've successfully drawn a line around the cases which annoy me. Here are some of the things that come to mind:
The words really are near-synonyms.
The definition offered doesn't fit the usage, even of the person offering the definition. I think the sympathy/empathy example illustrates this; lots of people will give a verbal explanation of the difference between the two, but it seems only shallowly based on usage differences, if at all.
The definitions offered are claimed to be general, rather than special case (which, especially in combination with the previous point, makes me tend to think that the person doesn't know how words work -- or at least, makes me think they aren't a person who is aware of how words can fluidly adapt to the needs of the conversation).
The distinction being offered isn't particularly crafted to fit the situation. (I may develop a suspicion that the person is mentioning the distinction just to sound smart -- something which might not annoy me in other contexts, but does annoy me when I identify it as a mis-used distinction-of-the-moment.)
The distinction offered doesn't strike me as particularly generally useful. I'll forgive all the previous points if someone is giving a distinction which I find illuminating.
I actually looked into the technologies for this in quite a bit of detail during my time at CEA. Creating a map for the meetups is quite doable, either with Google Maps or with Mapbox, and doesn't require that much engineering effort.
The biggest obstacle I see here is both that Facebook is the tool of choice for creating events, and that it is really good at this in a way that we won't be able to beat. Some kind of Facebook integration might save us here, or maybe fully focusing on reoccuring events is better.
An alternative would be to focus on "communities" instead of meetups. I.e. you can register a community on LessWrong, together with some metadata, such as Facebook groups, community greeter contacts, meetup locations, etc. Those communities are then mapped, and allow nearby rationalists to find those communities and engage with them. This generally reduces the amount of maintenance effort required by the local communities, and makes the system more flexible to integrate with whatever the local group uses as their event coordination tools.
Yep, that's part of the plan.
Whatever the data-structure for the sequences is, the goal is to present the user a linear sequence of content (which might be dynamically generated from a dependency DAG). It's really hard to keep track of your position in a tree, and practically all projects that I know of that tried to present complicated information a tree-view either failed or moved to a linear view eventually (for example Khan Academy)
It'd be nice to have a better meetup system than current LW's. I think I sketched my plan out earlier, but I might as well stick it here as well:There are two basic sorts of meetups: one-offs and regular. (Austin's "Welcome Scott Aaronson to Austin" party vs. Austin's 1:30 Saturday meetup) Both have a location, a datetime, and an organizer. The regular meetups, in addition, have a repeat frequency and might have a link to somewhere else (maybe you arrange events on Facebook or Meetup.com). (And if we could somehow automatically import events from Facebook groups, all the better.)Because of the similarity, those both seem like they could be the same data type to me. It also seems like the best way to display them is a map with a bunch of dots, probably colored by how far in the future they are (with something like "up to four hours ago" counting as now, so that meetups don't disappear right when people are desperately trying to figure out what the address was again), with a UI that manages to map a bunch of dots on top of each other not horrible. (Maybe combine a dot view with a list view, like on Google Maps?)
One of the big motivations here is that the LW meetup map is pretty sad (sometimes the closest meetup was in Europe!) when I think there are actually a bunch of regular meetups. (It also seems like we should encourage any of the EA/rationalist/humanist/SSC/MIRIx/etc. groups to have their meetups on the map, maybe with different shapes for different types and filtering, to make it even easier to connect people. "Ah, there isn't a LW meetup in my area, but there is an EA meetup!")
Including connecting social login to the accounts, I assume?
StackExchange has a bunch of badges and gamification and so on; it seems to me like a similar thing should happen here. (There should be a progress bar that gets fuller as you read the sequences!) In particular, we want reading through the sequences to more rewarding than commenting in open threads (which is currently how I would expect a new user to get karma).
This seems much easier to do if sequences are a top-level content type, and also means that anything we do to get people to read one sort of long-form text we can apply to others trivially.
I was originally imagining something more like a list ("decision theory explained in six posts!"); having a DAG of post dependencies might also be possible (which would enable stuff like "okay, you want to read this post? We've put all the prereqs in order into your reading queue.") but it's not clear to me how much people would want to use it.
It seems to me like tags are potentially a good solution here, so long as citizen-editors have the power to create tags and keep them orderly.
The words that came to mind here were "label" and "handle," but I think the longer phrase "distinction of the moment" is clearer (though longer).
A pet peeve of mine is that a lot of people seem to carry around distinctions-of-the-moment forever, as if they were true subtle differences in the meanings of words.
This seems like it's using "true subtle differences" as being a property of the territory instead of the map, which worries me somewhat. An old analogy here is something like a fuel injector and an intake manifold--to a layman, they're just 'car parts', whereas to a mechanic, they're clearly distinct. There are word pairs where there are distinctions that seem unimportant to me (less vs. fewer, comprise vs. compose, etc.) and then other where they do seem important to me (a marginal case on this side of the divide is that vs. which), and if someone is conflating the two it seems like to right move is to complain "hey, I was using that distinction!"
Ah, notifications actually exist, but they are currently disabled by default. You can subscribe to any comments for which you want to get notifications for by clicking the subscribe button.
You can also activate automatically subscribing to your posts and comments on your profile.
Hmm, that seems correct, but I am also hesitant to have people post the same comment over and over again because they haven't read the thread.
I will think about the tradeoffs here more, and we will see what I come up with. Maybe there is something in between that works best.
Yes. Google docs does contain a lame version of the thing I'm pointing at. The right version is that the screen is split into N columns. Each column displays the children of the selection from the previous column (the selection could either be an entire post/comment or a span within the post/comment that the children are replies to.)
This is both a solution to inline comments and a tree-browser that lets you see just the ancestry of a comment at a glance with out having to manually collapse everything else.
Also: you replied to my comment and I didn't see any notifications. I found your reply by scrolling around. That's probably a bug.
Are you thinking of something similar to what Google docs has? That's how I've been thinking about implementing the general inline-commenting thing.
If you want to encourage engagement, don't hide the new comment box all the way down at the bottom of the page! Put another one right after the post (or give the post a reply button of the same sort the comments have.)
One UI for this I could imagine (for non-mobile wide-screen use) is to have the post and the comments appear in two columns with the post on the left and the comments on the right (Similar to the Mac OS X Finder's column view.) Then when the user clicks on a comment the appropriate bit of the post would get highlighted.
In fact, I could see doing a similar thing for the individual comments themselves to create a view that would show the *ancestry* of a single comment, stretching left back to the post the conversation was originally about. This could save a fair amount of scrolling.
Agreed. Also, at some point Eigenkarma is going to need to include a recency bias so that the system can react quickly to a commenter going sour.
I agree with the spirit of this. That said, if the goal is to calculate a Karma score which fails to be fooled by a user posting a large amount of low quality content, it might be better to do something roughly: sum((P*x if x < 0 else max(0, x-T)) for x in post_and_comment_scores). Only comments that hit a certain bar should count at all. Here P is the penalty multiplier for creating bad content, and T is the threshold a comment score needs to meet to begin counting as good content. Of course, I also agree that it's probably worth weighting upvotes and downvotes separately and normalizing by reads to calculate these per-(comment or post) scores.
Testing if name change went through.
I think keeping some dependence on quantity is desirable, but that scaling linearly with number of posts weights it too heavily compared to variation in number of upvotes (I proposed scaling with roughly the cube root of number of posts in my explicit formula suggestion elsewhere in the comment thread).
That's a really neat idea.
Being able to sort by some of those would also be helpful (e.g. Sort comments by 'exceptional insight', don't show me comments with >2 'overly aggressive'.)
You mentioned an alternative comment structure in the features post. Some of the value of that could be achieved by a 2nd tier vote saying a comment is a key consideration (e.g. "This is a crux"), and being able to sort by that.
Yeah, it's pretty unreasonable to expect typography to be dialed in for the closed beta :)
Some quick thoughts/opinions I have for the post text:
I'd consider making the body text a serif font. I find it's a better reading experience.
Body text is too grey. It definitely shouldn't be black, but maybe darker at something like: #2F2230.
I'd differentiate heading a little more, maybe a different font, or real small caps. Also if I was being really opinionated I'd only support 3 heading levels and make them smaller. I think people are overdoing it these days with really big headings in post/article text on the web. It certainly clearly differentiates them from the text but there are classier ways of doing that.
The current line-height is pretty big at 1.846, I'd change it to something closer to 1.6. Maybe even as low as 1.4.
Most sites have their font size to small, so I'm really happy to see you didn't do that, but I think to current body font is two big at 20px. I'd do 18px at most, an no smaller then 16px. Doing this might make the lines a little on the long end though.
I'd also implement hanging bullets. This is were bullet text is flush with the body text, and the bullets are in the margin a little. It's very easy to do with CSS. For bonus points you could do hanging punctuation for quotes, but that's much harder.
Yes, I am a fan of Practical Typography and skimmed Professional Web Typography a while ago.
I haven't yet spend super much time optimizing the typography of LW2, and am happy about input. Rereading both of the books above in the process of that might be a good idea.
Ah, yes, it does feel a bit odd. Actually, I like it a little bit: it puts visual focus on the text I'm composing, while allowing a lot of room for other text on the screen. But at the same time, it feels clunky. I would probably get annoyed at the large text in the comment-composing box if I were writing a longer comment.
I'd generally recommend reading Practical Typography, and Professional Web Typography. I expect knowing that stuff well would be valuable since LW is primary a websites where people read lots of text.
Would the sequences have built-in tree structure?
Yeah, I was planning to reduce the size of the text in the compose-comment box to normal levels. Would that improve things?
Besides that, I am not sure whether we can avoid the large difference in font-size between article text and comments. The problem is that comments need to be small in font-size to allow the reader to understand the thread structure (i.e. they need to be able to see a comment and a reply at the same time), and the article text needs a large font-size to not run into any problems with the number of characters per line, and general readability for long-form text gets a lot better with larger font-sizes.
This is actually a feature, not a bug. The karma threshold isn't just there to limit who has access to features; it's also to increase the cost of creating sockpuppets and of recovering from bans.
Nitpick, but possibly a significant one: there's an unusually wide range of font sizes being used, with the article text and compose-comment box being very large, regular comments being very small. Normally I would pick my preferred font size by zooming the whole page, but the variation within the page makes this not achievable.
(EDIT: Typography settings have changed significantly since this comment was written.)