a thoughtful web.
Good ideas and conversation. No ads, no tracking.   Login or Take a Tour!
comment by kleinbl00
kleinbl00  ·  4250 days ago  ·  link  ·    ·  parent  ·  post: Hubski Update: Domain pages

An essential tool for spam moderation on Reddit.

On Hubski?

Click on a domain. See if it's constantly being submitted by one user. Block or follow, your choice.

Does raise the question - So are you going to let us follow/block domains?





mk  ·  4250 days ago  ·  link  ·  

    Does raise the question - So are you going to let us follow/block domains?

I knew that was coming before I even submitted. :) syncretic requested it previously. TBH I'd like to think on it and see if/how this changes things. It would have a pleasing symmetry, wouldn't it?

kleinbl00  ·  4250 days ago  ·  link  ·  

Think away. Let's see what happens.

mk  ·  4249 days ago  ·  link  ·  

I was thinking, and had the notion that I might follow a couple of domains. There are some rare gems that pop up that I know that I will forget, and this would increase the chance that I would stumble upon them again. Also, it might present someone worth following at the same time.

Do you think you would follow any?

syncretic  ·  4248 days ago  ·  link  ·  

I hadn't even considered the ability to follow domains. That sounds like a wonderful idea. I would definitely follow my favorites.

mk  ·  4247 days ago  ·  link  ·  

Just to be clear, this would mean that you would get those posts in your feed if someone posted with a link to that domain. Basically like tags. It wouldn't be full on RSS.

That sounds like something you would use?

syncretic  ·  4247 days ago  ·  link  ·  

Sure. Personally, I'd subscribe to a lot of the major news websites (nytimes, washingtonpost, etc), and then by ignoring specific tags, hopefully I'd be left with only the interesting stuff. On the flip side of that coin, I'd also love the option to ban domains from my feed as well ;)

kleinbl00  ·  4248 days ago  ·  link  ·  

I probably would. Especially with the death of RSS.

ecib  ·  4248 days ago  ·  link  ·  

The best use of RSS is actually this specific use-case, -compiling a feed full of sources that do not publish regularly. Nowadays, between Twitter, aggregators like Hubksi, Reddit, HN, etc, and even "dumb" aggregators like personal magazine apps, it's completely out of the question that I will fail to be exposed to any story of interest to me that is published in any of the medium to large outlets.

An RSS feed comprised only of interesting sites that publish irregularly is a goldmine. I can't really think of another way to make sure to capture those posts. Even bookmarking doesn't work, as you fail to follow up. RSS is great for this.

If Hubksi does this it needs to balance the weight these posts would receive in your main feed. You'd have the way most people use it (incorrectly imho), -following large and medium websites vs. using it well, following sporadic publishers of quality. Might make the most sense to make followed domains separate from the main feed in some fashion, or risk losing those rare posts in the noise with an algorithm that mitigates the firehose that results in somebody following wired.com or whatever.

kleinbl00  ·  4248 days ago  ·  link  ·  

I'm working through Trust Me, I'm Lying ("blogs are the new yellow journalism! If it bleeds it leads!" Who knew?) and in amongst a lot of angry, bitter, "I was a dick to readers because blogs are bullshit" rhetoric, Ryan Holliday spends a few pages talking about the death of RSS.

He makes the point that direct links to any website are usually a small fraction of that sites traffic, the majority of which comes from Google, Facebook and Twitter (in that order). He further argues that RSS never stood a chance because it puts the power to see or not see the content in the hands of the reader/subscriber, which is exactly the opposite of what SEO/traffichounds want. It's far more lucrative for any website to publish something sensational and have it linked across the universe for you to see it than for them to consistently put out articles you want to see.

It's a good point - A thin site such as Hubski as a compiler of rarely-publishing blogs that aren't to be missed *CoughDerekBellCough* becomes a powerful tool for people smart enough to install adblock, install ghostery and bypass the noise.

thundara  ·  4248 days ago  ·  link  ·  

<3 Derek. He's been my first and favorite source of information on the pharma industry that doesn't just yell about big corps, big profits.

ecib  ·  4248 days ago  ·  link  ·  

Yup. The shift to the web as a service and advertisement platform over the years is completely at odds with RSS. The tidal wave is only going to get worse as ISPs increase the throttling of content of their choosing (speeding up yours or your partner's pipeline is throttling everybody else's even if you do not reduce their bandwidth) to support their business models. Applications and platforms that do nothing else well save letting users consume content with a large degree of control over the noise will always be fighting against the tide. They can totally exist, but they will always have trouble scaling as standalone services.

A feature that mimics what almost universally mis-used RSS readers do best seems like a great feature to roll into another platform, like Hubski. But if it is a feature that reflects the value a platform is providing, I think it would be worth it to act like a big brother and control the implementation such that you can't misuse it like everybody does with RSS readers. If you like tech, subscribing to a bunch (or even one!) of RSS feeds from popular sites can be fairly useless. I'm not even talking about advertisements, -the content itself becomes noise. Social aggregators actually are one of the solutions to this, but even they run into problems. The feature seems like kind of a natural fit for a social aggregator though.

mk  ·  4248 days ago  ·  link  ·  

So people filtered RSS, you think that's worthwhile? Because that's basically what it would be if we could follow domains here.

ecib  ·  4248 days ago  ·  link  ·  

Well...I actually think people don't filter with RSS is the problem. They follow feeds that dump articles into their reader like a firehose, creating noise, esp when everyone sees those same articles on aggregators, twitter, Facebook etc. I think the value of a feature that lets you follow domains rests in only letting you follow domains that post sporadically, if that makes sense.

mk  ·  4248 days ago  ·  link  ·  

Hm. So I think we might be talking about two different things. I was imagining that adding domain-following would be akin to following a tag: if someone posts with that domain, it shows up in your feed. I didn't envision porting RSS into Hubski. I think that would be bad, unless as you said, it was from low-frequency sites.

ecib  ·  4248 days ago  ·  link  ·  

An I was thinking that you'd have content from followed domains not posted by a human. If its just following a domain that people also post, that seems like it would be decent, although with enough users, it approximates the RSS "problem." A lot of users probably wouldn't even define it as a problem though...

mk  ·  4248 days ago  ·  link  ·  

Yeah, I was thinking about that too.