Thanks to this mechanism, most of the time, the top comments were really interesting ones. The noise usually ended towards the bottom of the page.
I'm pretty sure that the expectation that this would happen made people - often busy people - spend time to write very thoughtful comments, which could then raise to the top. A positive, self reinforcing loop. Which is getting slowly, but surely, less effective with the years.
I wish there was a simple formula to reproduce the original mechanism. But I'm afraid there is not.
/rant
One day there was a person who like model railways, but wasn't really that into it. However, he noticed the model railway club and realised right away that these people are amazing. He would go to the meetings just to bask in the glory of the incredible things that these people built. Eventually, he couldn't keep it to himself. He told a whole bunch of his friends and they all showed up too.
The model train club started to get bigger and the original members couldn't be happier. When they first started most people looked down their noses at them. Now they were downright popular. One day some of the members said, "This really amazing stuff is really amazing, but I wonder if there is a way for me to approach this without being quite so geeky. I have a family and I don't have so much time. I want my train setup to be super awesome, but I don't really care that much about the details".
Lots of the other members were in the same boat and so they worked hard to have a kind of "scaled down" setup that would appeal to people who didn't have the time or energy to devote themselves completely to model railways. The club became even more popular due to this and lots of people who really liked trains, but weren't that interested in model railways started to show up.
Eventually several of the members started to have outings to visit train museums and to have photo competitions on train themes. To raise money they even made a calendar with pictures of trains. The model side of the club was a little bit less accessible to most of the membership because the core group of enthusiasts were interested in highly technical details that most other members were not interested in.
Eventually, it was decided that the model railway club should change it's name and purpose in order to be more accessible and welcoming to new members. It would now be the train enthusiasts club. Of course model train enthusiasts would have a place, but it would be appreciated if they refrained from having heated discussions about highly technical and inaccessible topics like programming train simulations and the like.
Eventually, the original model railway enthusiast stopped going to the club. He liked the people. They were great and friendly, but he was, after all a model railway enthusiast.
If you take something that appeals to 1% of the population and scale it to include 20% of the population, it morphs into a thing that the 20% group likes, not the 1% group. Can't be helped. It's just a matter of numbers. It happens everywhere.
I guess one way to handle this is to subdivide the field so there is a hierarchical structure and you can zoom into subfields to capture the the heterogeneous interests of a large body people?
A related example is where a town is a great place to live because of lots of walkable attractions (grocery stores, YMCA, parks, a variety of little shops, cheap rent for other small businesses, theaters, music halls, etc.). Then lots of people want to live there, driving up prices. The stores, non-profits, small businesses, etc. all get bought out and replaced by expensive housing. And then one day people who have lived there for decades realize that the town is not that great place to live anymore because there is nothing interesting to walk to there, no jobs, rents and taxes are high, and grocery shopping involves a long drive...
It can take some active planning and foresight to avoid that. Related:
http://shirky.com/writings/group_enemy.html " I want to talk this morning about social software ...there's a surprise. I want to talk about a pattern I've seen over and over again in social software that supports large and long-lived groups. And that pattern is the pattern described in the title of this talk: "A Group Is Its Own Worst Enemy." ... So these are human patterns that have shown up on the Internet, not because of the software, but because it's being used by humans. Bion has identified this possibility of groups sandbagging their sophisticated goals with these basic urges. And what he finally came to, in analyzing this tension, is that group structure is necessary. Robert's Rules of Order are necessary. Constitutions are necessary. Norms, rituals, laws, the whole list of ways that we say, out of the universe of possible behaviors, we're going to draw a relatively small circle around the acceptable ones. He said the group structure is necessary to defend the group from itself. Group structure exists to keep a group on target, on track, on message, on charter, whatever. To keep a group focused on its own sophisticated goals and to keep a group from sliding into these basic patterns. Group structure defends the group from the action of its own members."
http://www.t0.or.at/delanda/meshwork.htm "But even if we managed to promote not only heterogeneity, but diversity articulated into a meshwork, that still would not be a perfect solution. After all, meshworks grow by drift and they may drift to places where we do not want to go. The goal-directedness of hierarchies is the kind of property that we may desire to keep at least for certain institutions. Hence, demonizing centralization and glorifying decentralization as the solution to all our problems would be wrong. An open and experimental attitude towards the question of different hybrids and mixtures is what the complexity of reality itself seems to call for."
Part of the challenge is that human mentality evolved in the context of hunter-gatherer tribal relationships -- growing up in a tribe of mostly related kinfolk you knew from birth. It's a huge stretch to expect the human mind (and its ability to manage social relationships) to functional well in other contexts... Frankly, it's amazing the human mind works in other settings as well as it does.
The reason the internet has more noise than signal is because there are many people, and they don' all want to talk about the same thing as you.
As someone who has only ever succeeded in bringing down the level of every discussion i have ever been in and never in contributing anything useful, it would be very helpful if i could just be automatically directed to the appropriate content sewer by posting my thoughts directly into a search engine that could dump them in the most suitable location.
I've been toying with this idea. My go to would be to do a pagerank on the vote graph. The intrinsic quality of a webpage was transitive enough to make Google win the search engine war; surely people can be viewed as websites that have links indicating quality to other commenters: HN accounts and votes to someone else's comments.
If only HN could release the vote/down dataset...
That could be a privacy violation. Maybe they could do it themselves, or try to disclose it to select trustworthy parties.
Maybe they're doing it internally and they're keeping it to themselves. After all, YC is in the business of detecting precisely individuals that are capable of articulating high quality ideas.
Whether it would improve the quality of the comment sections is an interesting question. I'd be interested to see that tested somewhere ...
To have a focused discussion on Thing X, ask for a focused discussion. The scenario he stated (discussions on the README getting picked up and whitespace preferences being spun into a mini-holy war) is just... it's going to happen unfortunately. But that discussion wasn't one the author initiated for feedback, it was a general comment thread on the repository. If you, as an author, want to spark an intelligent discussion on your product than do just that, present the points you want to debate while allowing some freedom for that noise since it isn't always useless.
Instead of: "Hey HN, I just wrote a badass new streaming protocol" and linking to the github source, specifically mention your concerns about continuing to rely on TCP for video streaming and the buffering lag it induces, link to a blog-ish thing which links to your source but frames a number of questions or doubts you have around your project if you, indeed, are at a loss and want to turn to the internet as a whole to try and solve them.
Basically, as I do in my workplace, refuse to let your discussion/meeting/whatever become a vaguely directed waste of time - set a clear agenda (i.e. a set of specific points in a blog post) to help funnel the discussion toward the useful parts. In the resulting chatter you'll probably get some whitespace criticism (spaces are simply poor manners, #tabs4lyfe) you'll get some chatter about your question (and on HN that'll be floated to the top) and you might get someone who dug behind your question and found that - yea TCP is bad, but if you're going to use UDP you'll need to have some solid check-summing in place which your project currently lacks.
So, you need to direct the discussion, but allow noise, without noise you'll eventually settle the discussion into a local maximum of functionality where everyone is nodding and saying "Yup we did it, it's not a solvable problem but we've clearly come up with the best partial solution."
> Instead of: "Hey HN, I just wrote a badass new streaming protocol" and linking to the github source, specifically mention your concerns about continuing to rely on TCP for video streaming and the buffering lag it induces, link to a blog-ish thing which links to your source but frames a number of questions or doubts you have around your project if you, indeed, are at a loss and want to turn to the internet as a whole to try and solve them.
That sounds exactly like an example of Dan's suggestion to "tell a story."
What I stated above wasn't what I read in the article, it was my proposed solution to what I discerned the issue to be.
People chatting about random topics prompted by the link is the whole point of a link aggregation discussion site. The discussion is not intended to precisely meet the needs of the person who created or submitted the link, and it’s silly to dismiss it because it predictably doesn’t.
If someone wants e.g. concrete technical feedback about a new specialized tool that takes hundreds of hours of preparation to appreciate, they should find a forum consisting exclusively of people with enough background to give meaningful feedback, and then describe exactly what kind of feedback they are looking for.
= = = =
The top comment thread picks on the coding style in a README example. It turns into an argument about indentation with over a hundred replies and a brief history of how different programming languages approached formatting. There are obligatory mentions of gofmt and Python. Have you tried Prettier?
Somebody mentions that open source projects shouldn’t have beautiful landing pages because it’s misleading marketing. What if a junior developer falls for it without fully understanding the fundamentals?
In a response, somebody argues the landing page design is boring. Additionally, it’s broken in Firefox. Clearly, this means the project author doesn’t care about the open web. Is the web as we know it dying? It’s time for some game theory…
The next comment is a generic observation about the nature of abstractions, and how they can lead to too much “boilerplate” (or, alternatively, “magic”). The top reply explains that one shouldn’t confuse “simple” with “easy”. Actually, Rich Hickey gave a very good talk about this. Have you watched it?
Finally, why do we need libraries at all? Some languages do well with a built-in standard library. Is npm a mistake? The leftpad accident could happen again. Should we build npm right into the browser? What about the standards?
Confused, you close the tab.