Hacker Newsnew | past | comments | ask | show | jobs | submitlogin
“The news web is straight trash” (twitter.com/timeserena)
29 points by rian on Feb 20, 2015 | hide | past | favorite | 19 comments


What is the definition of clickbait here?

"Radical Statements about the Mobile Web" does, indeed, contain some radical statements about the mobile web, from a guy who works for Mozilla.

"Handling five billion sessions a day in real time" is... about handling five billion sessions a day in real time.

"The Unlikely AirBnB hosts of Japan" is... well, you get the idea.

Clickbait is when you give a story a title the actual content does not live up to. "You won't believe", when you absolutely will. "You'll be amazed" when you won't. The stories he's complaining about here are... just stories with titles.


Eh, it could be worse. At least the clickbait is mostly "Look at this cool thing" and not "Look at these bad people doing these bad things. Outrage!"


My usual response to this complaint of the "decline" of HN is that it is a tactic of older entrenched figures in a community marginalizing the viewpoints of newer members.

The second strain I see tends adheres to the timeless tradition of decrying the degradation of the youth in comparison with previous generations, which is as old as the ancient Athenians. "Times change, and men decay."

Normally I dismiss them, and for a while even kept a running tally of each time these silly arguments were advanced.

However, I can no longer in good conscience attribute the very recent behavior of the front page of HN (over the past ~3 months) with these two explanations: there has been an undeniable degradation of quality, new sources in primarily popular science publications with very click-bait titles, and even anecdotal evidence of undetected voting rings.

To me, the real question, is this some artifact of the evolving website leadership (unlikely given the managment of the transition,) has the fundamental purpose and audience undergone another shift (YC companies forum -> General Startup News -> General Technology News.)

Or has HN's Eternal September simply finally arrived, where new users are arriving too quickly, and thus cannot become acustomed to HN's norms, resulting in a lowering of quality. http://en.wikipedia.org/wiki/Eternal_September


Thank you for enlightening me on the concept of "Eternal September"


The author overlooks some clickbait:

* What's the deepest hole ever dug?

* The man who quit money

* The Incredible Degredation of Hacker News

etc.

The question is, was HN ever much better. Perhaps some front page listings from 1, 5, etc. years ago would help.



Thanks. Of course we don't know if we're looking at representative samples, but I'd say the 5 year old front page was better.

I'd pay for a very well-curated site, with excellent discussion. I wonder if a paywall would increase quality of participation? Owners and members might be less tolerant of low-quality content.


More a commentary on the web in general than just HN?


Pretty harsh definition of clickbait.


What I find most mysterious are the article sources (okay, not that mysterious).

I've never visited smithsonianmag.com in my life. Yet, there are four articles from that site on the front page right now.

Something isn't working...


Someone posts an interesting article from a source. 10,000 people visit that source. Some of them noodle around and find other interesting articles and submit links to them.

It's a problem if the submitted articles are low quality.


"straight trash" <--- clickbait


"Nobody goes there anymore, it's too crowded." -- Yogi Berra

When I first heard that quote from Yogi Berra is crystallized for me a concept that had been developing for a while in my sub-conscious. Popularity of social institutions are intimately tied with the participants in those institutions. That evolution happens "outside" our own internal definition of quality. So when you go to a restaurant over a long time, read a magazine over a long period of time, play a video game over a long period of time, pretty much any evolving community, your internal standard of "goodness" is established in your head somewhat at the start of that experience, and "improvement" or "degradation" of that experience is measured against this internal metric which may not be changing at all.

I've most recently been dealing with this in neighbourhood meetings where "long term" residents are really upset at the city for approving the building of multi-story apartment buildings where previously single family detached homes stood. This change in urbanization is a "bad thing" for them (although its great for people moving here who need a place to live and cannot afford to rent a house!). And yet when you explain that the very house they live in was once an orchard, was it a "bad thing" that the farmers allowed the city to build homes on what had been orchard? And the answer (because this is government and governments are often good at keeping records) is Yes! people did complain about all the houses going in and how it was putting businesses that supplied those farms out of business. So when you ask someone in the present would they give up their house, and have razed to the ground and returned to orchard? They would not. But their internal standard of "goodness" was set when they moved in, and the city is evolving, as it must, and it is drifting away from their notion of what a "good" city is. So they advocate for keeping things the same, the young people advocate for affordable housing near mass transportation, and the city council has to weigh these against the long term health of the city and try to decide wisely.

Hacker News is very much like that. It changes based on the internal "goodness" metrics of its users. And as new users get added and their goodness metric is just slightly different then the incoming new users slowly migrates the notion of "good" toward their point of view.

A long time ago I proposed a compass star type voting system where the east and west points represented "less like this" (west), and "more like this" (east), with north and south continuing to represent "this is a good example of something I like" (north) or "this doesn't belong here" (south). The center button could represent "perfect".

The clever bit though, was not the voting system, rather it was to capture through vote analysis this particular user's preferences (their internal metric of "goodness") As defined as a point on an n-dimensional surface of other users, Then using something like k-means analysis to create a rank for a story given the particular user's preferences. In that way the front page would always have really great stories on it, that you like. But they might not be the front page stories that some other collection of members like.

Sadly it remains an idea I don't really have time to implement but if someone does I'd really like to participate especially in developing the system of algorithms that tune the various internal "nodes" in that n-space of likes/dislikes.


So .. where to go next?

I've been returning to Slashdot more and more these days .. and lobste.rs is pretty good too, although a bit click-baity as well.

I think its time for a de-centralized approach to this whole problem again - a USENET 2.0 with smart agents would be nice. But the more I think about it, the more I realize that the aggregation isn't the problem - its the content creation in the first place. The communities that band together to create a content platform eventually self-implode into themselves, so there has to be some other way to get technology news and interesting articles out there. Perhaps there is no perfect answer, and that its all just a process rather than a product after all ..


>So .. where to go next?

Specialized subreddits with competent moderation. Ideally with post scores hidden.


New content is overrated anyway.


He needs to update the image quickly with a current one showing his own story.


Shocking!





Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: