Comment Permalink

KenChen14y20

I commit to working on modifying the code to organize meetups in a way that makes more sense, if someone else is able to provide direction and authority. I haven't worked on the code before, and I can't promise anything except that I will attempt to work on it.

I have just read the github wiki. I will try to get an instance of the site running. What's the next step after that? Who else should I be coordinating with?

matt14y30

KenChen - kudos to you for publicly making this commitment.
Unfortunately for you you're a member of reference class "low site karma people I don't know". The next step is to follow through with your commitment. Please let me know when you've successfully made the code run on your machine, and made even a trivial change to the site (that worked on your local machine).

See in context

36 Do meetups really have to go on the front page?

by InquilineKea

11th Apr 2011

1 min read

36

I've been browsing LessWrong less often as of late, and I believe that one of the reasons is because I don't want to get all excited all over a new insightful LW post, only to discover that it's a meetup in a city that I'm not even in.

And for newbies to the site in particular, they're not going to see all the awesomeness of this site if they simply look at the frontpage and think it's a community just for meetups.

I'm sure meetup discussions can go somewhere prominent. Maybe a separate section can be devoted to them.

Site Meta

Personal Blog

36

New Comment

54 comments, sorted by

top scoring

Click to highlight new comments since: Today at 9:14 PM

[-]Louie14y280

newbies to the site in particular, they're not going to see all the awesomeness of this site if they simply look at the frontpage and think it's a community just for meetups.

Even without meetups, the promoted feed is still a pretty terrible way to browse the site for a newbie. It would be like if Wikipedia's main page was the recently created pages. LW needs a homepage. I've asked for a volunteer to help create one and promised to implement it if it's designed. I'm not a designer but I'm offering to help with the implementation... but no one has even tried.

There's karma for complaining, but none for fixing problems. I think when Yvain, the LW admins, and I helped create the entire discussion sub-reddit, we got 3 karma apiece in comments related to it. You've already gotten more karma just for complaining about meet-up announcements for the 10th time without offering to help.

[-]Scott Alexander14y310

Have you considered a top-level post, possibly by you but even better by Matt from Tricycle, saying:

"Hello! I am the official contact for Less Wrong administration. If you have something you want to code for the site, message me and I will tell you whether you have official permission and whether it will get added when you are done. The answer is probably yes.

Here is the language you should code in, the standards you should follow, and so on. Here's where you can find the rest of the open-source code to see what you should be integrating it with. And here is the process to get your code running and up on the site when it is done, even if that process is just "email it to me".

In the comments to this post, please discuss your programming projects for Less Wrong, check that they're community-supported, and recruit people to help with them. If you can't program, this would also be a good place to suggest projects and recruit people who can."

Then stick a permalink to it similar to the "About" link going to Eliezer's first explanatory post.

All this information is sort of available right now, but not in one place and not with explicit permission to work on things, so the trivial inconveniences add up.

[-]matt14y10

We'd tried to make that stuff clear from https://github.com/tricycle/lesswrong/wiki

[sad face]

In my humble opinion the population of those who will follow through with site improvements is small enough that that link doesn't deserve a very prominent position. It's linked from the About page (and better linked now than it was a few mins ago).

One of the things that holds new contributors back is that the original Reddit code… is really ugly and isn't covered by unit tests. We've (Trike) poked it forwards but never taken the (huge amount of) time required to make it into a clean codebase. That's going to make it hard for inexperienced programmers to contribute inside the capacity of their goodwill reservoirs.

Those goodwill reservoirs are a challenge in most open source development, too. Posts to open source lists that start "tonight I will…" are much more common than posts that start "last night I…". Collecting intentions to work on features sounds useful, but unless it's followed up by hounding it's a thankless task (and hounding is a thankless task).

This wasn't a very constructive comment. Sorry about that.

Nitpick: We (Trike) fund our own work, but act strictly on Eliezer's authority - I don't consider myself to have the authority to grant official permission for anything. I sometimes presume to guess at what Eliezer will authorise and act before he grants permission, but only when my confidence is high.

[-]KenChen14y20

I have just read the github wiki. I will try to get an instance of the site running. What's the next step after that? Who else should I be coordinating with?

[-]matt14y30

[-]SilasBarta14y10

Thank you. The world needs more of those. (For what class I mean by "those" see here.)

[-]Alexandros14y70

I've recently done the work to extract 92 broken links in the sequences. I also offered to extend the scraper to figure out where the links should be pointing to, if someone with edit rights would offer to implement the fixes I turned up. I got very well upvoted (most upvoted link in discussion section right now), but heard nothing back.

[-]Vladimir_Nesov14y30

I've got your post bookmarked, and planning to go through those links when I get the chance.

[-]Alexandros14y00

Thank you.

[-]jsalvatier14y40

I hereby commit to donating $50 (which must be spent on pure hedonism) and a big hug to whoever designs the front page.

Perhaps a contest would be a good idea? That seemed to work well for the charity essay thing. It might get people more interested.

[-]SilasBarta14y20

I hereby commit to donating $50 (which must be spent on pure hedonism)

Aha! I knew your economic theories were secretly trying to get people to make more short-term purchases, and now I've caught you red-handed! ;-)

[-]jsalvatier14y20

Curse you, CommodityMoney Man!

[-]SilasBarta14y10

Don't be too proud of this economic terror you've constructed. The ability to expand a money supply is insignificant next to the power of the ore!

*ducks*

[-]jwhendy14y00

Dang. I wish I knew css. My hedonism funds are low...

[-]Armok_GoB14y20

This is really a problem that needs fixing. Maybe some kind of pool people can put karma in and then is given to the ones who implement the thing the pool is for? Sort of like a bounty.

Edit: do try to come up with other solutions to everyone, I forgot to hold of on proposing solutions.

[-]jsalvatier14y10

I really like the idea of creating bounties. Any kind of official way of creating a bounty would also serve as a focal point for people to offer more material rewards. I don't have a problem offering a cash reward for people donating their time for important tasks, and I suspect there are others like me too.

[-]KrisC14y00

I believe whoever might design a significant addition to the site would already understand that a post explaining that project when it is implemented would garner significant karma.

[-]Armok_GoB14y00

the parent post of these is strong evidence otherwise.

[-]KrisC14y00

Agreed. However a search for "sub-reddit" and no alternate spellings suggests that karma is not so sparsely spent on this topic.

[-]matt14y10

G'day Louie,
Trike has a LessWrong design review sitting near the top of our list of design projects for the awesome Jimmy Gleeson - we'd hope to get to it (the design) inside a month, and to get it implemented soon after. We were planning on just tweaking, but a new post trying to collect design priorities would garner you a large pot o'karma and guide Jimmy's (and my) work on the review.

[-]KenChen14y10

As a quick alternative, would it make sense to create a promoted top-level post containing the Sequences, and "sticky" that post so that it's the first thing newcomers see?

You could create a preference to not display it for logged in users, if that's an issue.

[-]FiftyTwo14y00

The current about less wrong page would make a decent front page. As it contains a brief description of the goals of the site, it's systems and links to good sample articles.

[-]KenChen14y20

You've been telling everyone at this cocktail party about this fantastic new rationality club, and how it's changed your life. You manage to get a few people interested enough to actually listen to what you're saying, a smaller subset to remember the name of the site, and a smaller subset to take the time to actually check out the site for an entire 90 seconds.

That's all you got.

So, after telling a room full of people about the awesomeness of rationality, maybe you get one person who decides to visit the site. They are looking for a justification for why they should spend more time on your site. What do they see? What do you want them to see?

Hopefully something good. Something that will make them hunger for more. Something that will actually make them think (which is kind of the point of all this).

The about page is currently structured like this: Science is good. This site has science. Meta. Meta and code. Meta. Meta. Meta. Meta. Brought to you by viewers like you. Grid of juicy content. Link to Sequences. History. Code. Code.

The about page is great if you have been on LW already and want to know more about it. However, it is not a suitable front page. Unfortunately, it is not great even if you are already motivated to find out the core tenets of rationality.

[-]FiftyTwo14y00

I agree its not perfect, but it would still be a significant improvement on the current front page, which is almost always very brief comments about meet ups and tells you nothing about the content of the site or the community.

The ideal would be a post that combined some of the content from it, the 'welcome to less wrong' page and had a good list of introductory articles.

[-]mutterc14y00

Seems like a homepage could be made out of the Wiki's front page, right?

If that got to be good enough quality LW could have "lesswrong.com" point to the Wiki and there have a link "Blog" that points to the posts.

[-]prase14y00

As for karma for fixing problems: I promise to upvote 20 comments of anybody engaged in the successful home page creation (great idea). I would like to help myself, but have almost no idea how it can be done. It would be great if there was some basic description for inexperienced LWers of how changes are designed and implemented - maybe I could do something even with my very limited abilities, but I'd need some initiation.

[-]Alicorn14y280

I support the idea of a separate subreddit for posting meetup posts, with its own segment of the sidebar so it can be monitored during browsing of the rest of the site.

[-]Richard_Kennaway14y90

I think this would be the best solution. To compress the information further, it would also be good if the sidebar block only showed meetups yet to happen, and all others relegated to a "past meetups" link. As of today, April 11, it might look like (but with nice fonts and everything):

Meetups  
    Edinburgh LW Meetup Saturday April 16th  
    Ottawa LW Meetup Saturday April 16th
    past meetups

That would need some automatic way of discerning what the date of a meetup is.

[-]KrisC14y20

Perhaps a form to fill out to generate meetups.

[-]Alicorn14y00

This would be good. "Region", "Date", "Time", "Location", "Identifying Features", "Notes".

[-]KrisC14y00

"Identifying Features"

Of the contact person?

[-]Alicorn14y30

Yes, like "I look like this" or "I will be wearing thus and so" or "look for the enormous paperclip sign".

[-]KenChen14y00

IMO, there's a distinction between regular meetups, and one-shot meetups. It wouldn't make sense for a newcomer to see only one-shot meetups on the list, when a community in their city already exists.

Perhaps some sort of calendar would make sense, where anyone can post one-shot meetups, and regular meetups are represented by a recurring event.

[-]Barry_Cotter14y120

Geolocation to show only meetups near you?

Preferences option to hide meetups unless they're within X km of you?

I do not support removing Meetups from the front page until such time as it can be assumed that there is one in every major city where large portions of the poulation speak English, conditional upon meetups not completely taking over the front page.

[-]khafra14y30

Significant coding required for those. A LW frappr or other social map would be great, but

I do not support removing Meetups from the front page until such time as it can be assumed that there is one in every major city where large portions of the poulation speak English, conditional upon meetups not completely taking over the front page.

Counterpoint: I'm certain enough that there are no other LW members in Central Florida that I've refrained from posting anything regarding meetups. If there were a meetup section, however, I would at least throw something out there (although perhaps after I've had a chance to visit the NYC group).

[-]Jolly14y20

Er, I'm in Central Florida...and would be interested in a meetup. (Assuming it does not conflict with my crazy travel schedule, of course! :)

[-]khafra14y00

If you have a crazy travel schedule, perhaps you'll be able to get to NYC before I do and ferry back the smoldering igniter for a local group. Although "local" may be a stretch; I live in St. Petersburg; the space coast is about a 3 hour drive.

[-]prase14y00

Some people travel frequently and could be interested in distant meetups. So this ought to be switchable, and that needs more coding. First we should have a basic meetup management working, and later such refinements can be added, if there are volunteers willing to code them.

[-]falenas10814y80

I think the main reason meetups are on the main page is that some people don't look at the discussion section, and we want as many people to see meetups as possible.

[-]prase14y100

As a data point to the contrary, I now mostly look only at the discussion section, because there is nothing on the main page except the meetups.

[-][anonymous]14y00

Same here.

[-]radical_negative_one14y00

Agreed.

[-]mindspillage14y40

How about if meetups were linked from the front page in a separate section so people see them, but the links led to Discussion area posts?

Forgive my quick and ugly graphic.

[-]mindspillage14y20

(Note also that if I had been thinking more, I would have pasted in older front page content that was not meetups...)

[-]Alexandros14y10

This looks good to me. If people could agree to post meetups in a very specific format, I could make a scraper to extract the times, dates, locations, (and whatever else) directly from the discussion posts, and produce text like what you show directly for the front page, say displaying meetups in the next month. We'd still need someone to edit the front page template to include said text though.

[-]lukeprog14y40

I'm not sure what's best, here. One nice thing about meetup posts on the front page is that it shows this is a very active community with lots of real people meeting in physical locations and having fun together.

[-]KrisC14y30

Putting them in a sidebar with an indicator separating past and future meetups might be even more dynamic without being obtrusive.

[-]CronoDAS14y30

Personally, I'd prefer that the meetup posts not be on the front page...

[-]SilasBarta14y00

Why not? Like lukeprog says, one of the goals here is community-building.

[-]mstevens14y20

I too dislike meetups on the front page. I want tasty nutritious content!

[-]Louie14y20

Might I suggest that you (the person reading this) help the problem by actually writing a promoted post for the front page?

If you don't like there being nothing insightful to read, why not provide something?

[-]InquilineKea14y90

Well, like most of the readers here, I appreciate many of the things that other people write, but I'm still not sufficiently advanced enough to write posts that most people like (I'm only a college undergrad, after all). Most of my "intellectual" posts get very few upvotes, and surprisingly enough, my "neurotic" posts are the ones that get the most upvotes.

[-]prase14y60

Promotion is not done by the author. The only reliable way to write a promoted post is to start a Rationality Quotes thread or to announce a meetup. Nor do insights come on demand.

People are clearly more comfortable with writing posts in the discussion section. It may be caused by loss aversion with respect to karma and perceived higher standards for the top level posts. Telling that we should write more will probably not fix that; it is analogous to saying that the criminals should be less immoral as a reaction to complaints about high level of criminality. While the criminals being less immoral would solve the problem, it is so unlikely to happen soon that this solution can be rejected as counterfactual. We should seek an institutional solution instead.

The problem consists of two parts. First, there are too much meetups to fit on the main page, and second, there are too few promoted posts. The first part may be solved by creating a meetup bar somewhere above the recent comments bar or a special meetup section. The demand was expressed several times lately and I think there is enough consensus that people who can code it can just start. (Better first propose one concrete implementation and vote on it.) For the second part, I suggest:

We have criteria about what is unsuitable for the main page, but everything seems suitable for the discussion section. No wonder the discussion section is a safe bet. Longer coherent posts should be demanded to be placed on the main page.
Write down (for instance on the wiki) that criteria specifically. They should not be vague as "high quality insights", but rather specific: Length at least 30 lines - main page. Shorter than 10 lines - discussion. Only links - discussion. Open conjectures - discussion. Explanations of finished ideas - main page. Meta-discussions of internal LW workings - discussion. Or so.
Automatic promotion based on karma (e.g. after receiving 15 or 20 karma, promote). Promotion of the lower-karma posts still possible manually by the administrators.
Reconsider the ten-fold karma multiplier for the main section posts. Its intended purpose might be to encourage high quality posting on the main page, but actually the frequency of the main section posts has decreased after the change was made. It is quite probable that risk aversion may play a role here.

[-]kpreid14y00

I was just noticing this myself. I read LW via feeds, so the contents of the front page are normally irrelevant to me, but I went there this morning and I noticed that the entire visible content was three body-hidden meetup posts; and scrolling down showed a rationality quotes thread, two more meetups, and then some actual LWian content. It's not only not showing anything interesting to newbies at the top, it's also showing a degenerate form of the intended site structure (no post text).

I suggest that even if it is decided that meetup posts should remain on the front page, they be given a more compact format. Specifically, either given them a more vertically compact presentation (e.g. it makes less sense to show the vote/comment/save/hide/report bar when all of the article text is below-the-cut) or put them in a separate column. (Though columns have their own problems.)

Moderation Log