The problem with self-moderation is that it's not a stable solution. Political stories draw in people interested primarily in politics, and since it's so easy to be interested in, it draws in lots of people who are primarily focused on that.
Anyone can have an opinion on politics, whereas finding people to discuss actors vs CSP vs threads as methods of concurrent programming is much tougher, and sacrificing that to reenact liberals vs libertarians for the 37898289261st time on the internet is a tragedy.
Self moderation is augmented by the cycling of stories off the front page.
HN is in a news cycle where there is a nexus between the business of "high-tech" and politics playing out in a new way - or the same old way, depending on one's cynicism.
As always the solution to a low ebb in HN quality is to post interesting content, or even better to write some.
One solution is to write something really fucking great. Doing so is left as an exercise for the writer.
Those economics are not restricted to stories about politics. Remember the days when the front page would be flodded every time the MacBook Pro got longer screws, a couple of megahertz processor speed, and a bigger base hard drive?
There have always been cheap and easy stories. There always will be. If you run a zeitgeist, it comes in the box.
> One solution is to write something really fucking great.
No, it isn't. I thought the logic was pretty clear:
In the time it takes one person to write a great article - assuming they're capable of it, which not all of us are - another person can submit 100's of low value links, some of which will get a lot of votes because they're about high-passion, high-profile topics.
The logic was clear and its ongoing attractiveness as a first approximation of HN empirically demonstrated over the years. Sometimes, I find it an attractive first approximation.
However, I recognize that my dog, Scarlett, is an animal and therefore to a first approximation an insect. Yet, I do not squash her when she's in the bed with me, nor spread chemicals to prevent her entry to the house, nor scrupulously remove all potential food sources.
We agree that there is an endless supply of HN submissions. Though we may disagree about which is which, we also agree that some are better and some are worse. We also agree that there is virtually no barrier to entry in regards to becoming a supplier of articles to HN.
The points at which I find dissonance in the market model are:
() The supply of karma points is also unlimited. I can create a karma point for another HN'er. Of course, I can destroy a karma point for another HN'er, but only to a more limited degree; at least I understand that there's throttling in the software.
() Karma points are not fungible. I cannot exchange them for anything - except in a limited sense snark that I am willing to have downvoted, a practice in which I have occasionally engaged.
() Thus, if there is a market of the sort amenable to economic market theories, it is the sort of market which is for all practical purposes entirely illiquid. I don't know enough economic theory to suggest that such markets are not the subject of voluminous academic research, but supposing they are, that perhaps still suggests dubiousness in the first order approximation.
Teasing out the interesting facts about HN requires, in my opinion, a better abstraction.
I prefer a political model. HN has some democratic institutions - like one account one vote. It even has some aspects of Athenian democracy where the mob gets to vote on daily decisions regarding articles, and as in Socrates day, rhetoric oft sways the passions and ballots of men (and unlike his day, women).
More specifically, I prefer the republic - in the platonic sense - as an abstraction for modeling online communities. Ultimately there is a sovereign with whom we form a Hobbesian compact. Participation in a compact with HN requires us to give up certain rights in exchange for a particular form of security. The rights we give up to HN's sovereign and the security we gain are distinct from what we give up and obtain from the sovereign of StackOverflow or MSDNtechnet, or a professional cycling website.
The security we gain is largely that of reasonable norms. But among the rights we give up is that of not being subject to unreasonable policies at the extreme and policies with which we may niggle in the common.
Perhaps amending to that a filter to that which gives less upvotes for submissions that are repetitive based on url, page title and article keywords during a certain period of time. Though the only problem with that would be if early submissions are of less quality and get upvoted while ones that come slightly later are of higher quality.
Some of it is guff that shouldn't have been subbed. Some of it is interesting and it got ignored. Some of it is not particularly great, but it got a few upvotes then got many more as it hit the front page.
Items 3 and 4 in the list were posted very close together in time, and were unlikely to have been subbed by anyone else. One got no votes, the other made front page for a while and got comments.
* How to Almost Build an Engine With Only Parts From the Hardware Store [video] (howtolou.com)
87 points by DanBC 21 hours ago | 44 comments
*Programming the Vintage Intel MCS-48 Microcontrollers (theresistornetwork.com)
1 point by DanBC 22 hours ago | discuss
And then there are the articles that get upvotes, and also get flags. The responsive URLs article (11 in the list) got 6 upvotes, but was flagged to death. The Wikipedia loo roll article got 9 upvotes, and two comments (with a 3rd comment asking why the hell it had been posted) but was also flagged.
I don't claim my subs are any good, or that they shouldn't be flagged, but it is confusing.
I've gone to bed posting something I thought was mildly relevant and woken up to a big bump in karma.
I've posted something I thought was incredible and seen it wind up with one upvote.
Then again, why do I submit articles? Mainly because I think it will contribute to HN. If it doesn't, it doesn't. I tend to achieve more consistency with my writing anyway, and that's a more valuable skill to boot.
Anyone can have an opinion on politics, whereas finding people to discuss actors vs CSP vs threads as methods of concurrent programming is much tougher, and sacrificing that to reenact liberals vs libertarians for the 37898289261st time on the internet is a tragedy.