@moonman not necessarily. There are plenty of #FreeCode analytics packages (#Matomo etc) that could be built in or integrated using plug-ins.
opensource.com/article/18/1/to

@moonman also, I believe #PeerTube is building in the ability to do accurate view counts and some replacement for 'likes' across a federated platform, which could be re-used for a #fediverse podcasting platform

Follow

@strypey @moonman Not sure how a federated platform really could accurately count views, as any of the nodes could be lying. As far as I can see, it works require some Bitcoin like leger of each view to be verifiable. That would be prohibitive though, as views are much more voluminous than transactions in BTC.

@Blort @moonman the same way
> Not sure how a federated platform really could accurately count views

the same way any other network of #BitTorrent trackers do?

@strypey @moonman

Maybe I'm wrong, but it seems like a different problem to me. I mean, if you want to confirm another node has a bit of a file, all them to send it over and confirm the hash, sure.

If you ask how many views they've served for a video, how to confirm when they tell it's 10,000? This is more like asking how many times a file has ever been downloaded by everyone, than how many seeders there are or similar. Does BitTorrent show that?

@Blort @moonman my understanding is that a tracker can identify leachers who never seed, and give them lower priority on downloads than users who do seed. I'm not sure how the tech works though, and it could be I'm totally wrong. Some discussion here:
superuser.com/questions/211264

@strypey @moonman

So what's interesting about looking at that link is that it confirms

@strypey @moonman

What's interesting about that link is that it confirms nodes can lie, but that they suggest nodes can compare how much has been uploaded to what the other side says has been uploaded to try and catch faking. I still think this could be spoofed by telling each node that you uploaded a ton to other nodes and then incrementing by the actual amount uploaded by the real amount as each node can only check it's own figures. I'm no expert though.

@strypey @moonman

What is nice, is that by using , it benefits from the no-doubt large amount of work already done there to prevent spoofing. It just seems though that there would be a bit less of a financial incentive for people to bother spoofing seed/leech ratios than view counts. If PeerTube really took off, I suspect that it would encourage the spoofers to get more determined and force new anti spoofing measures to be introduced.

@Blort @moonman given that I don't think #PeerTube intends to build in monetization, leaving that to third parties like #Patreon, #Liberapay etc, I think view counts will only matter on PT about as much as they do on vanilla BitTorrent. Who really pays attention to view counts anyway (I don't even look at them unless I have a specific reason). Totally agree that working with mature protocol is good, as there's a large base of existing protocol implementers who can help solve spoofing issues.

@strypey @moonman

Regardless of what code, view count, demographics and engagement will always be important to full time content creators. They're the most important factors to sponsors and advertisers who, like it or not, pay for the vast majority of content creation today.

I'd love to see socially minded alternatives (eg Liberapay) baked in, but sadly there's ample evidence that these alone aren't competitive to fund large scale content creation.

@strypey @moonman

Yes, we can say "screw advertisers and sponsors. Let's keep the PeerTube's platform pure and away from that stuff", however without them, PeerTube loses most of the funding for content creators, and thus most of the content, viewers and social impact. It's essentially saying we'd rather give most potential cultural influence to rather than accept . Satisfying, but ineffective. It's the niche Puritan dilemma.

@Blort @moonman online advertising is a zombie business model. Ads, like phone books and TV guides, existed to solve problems the web solves better. Which is why platforms that try it eventually end up turning to fulltime datafarming (see: blogs.harvard.edu/doc/the-adbl).

@strypey @moonman @blort Online advertising exists because (1) advertisers want to sell product, (2) advertising networks want to sell ads, (3) sites that host ads want to generate some revenue without selling any products or services of their own.

The problem is that almost no one clicks on an ad and buys a product or service. So advertisers and ad networks tried getting more aggressive (pop-ups, pop-unders, slide-overs, interstitials, flashing, moving, speaking), but sites found that those tactics drive people away from their sites, reducing ad revenue.

The switch to personalized data collection and exploitation is because advertising failed to meet expectations.

@lnxw48a1 @moonman @Blort
> The problem is that almost no one clicks on an ad and buys a product or service.

Yes, but underlying that is the problem that websites accepted pay-per-click models that newspapers, radio, and TV never had to. If they had just kept their powder dry, and waited until the web was obviously getting more eyeballs that those three combined, a) they could have insisted on pay-per-display and b) better business models for funding websites may have got out ahead of ads

@lnxw48a1 @moonman @Blort also 1) advertisers could have been sold their own websites, instead of patches of space on other people's pages, 2) ad networks could have pivoted to selling website services (and many now have), 3) websites could have insisted on pay-per-display, or sold commercial website services as a side hussle, and probably made more money that way without creating #SurveillanceCapitalism

@strypey @moonman

That said, these days advertisers are more interested in minutes of attention and engagement than pure viewcount, but verifying minutes watched is essentially the same challenge technically as viewcount.

@Blort @moonman citation please? I haven't see any solid numbers, but an informal survey of the YT creators I have watched, who make anything resembling a living, suggests they do it via Patreon and other micro-patronage platforms, not via YT ads, which I've heard pay appallingly little to creators.

Sign in to participate in the conversation
Mastodon

The social network of the future: No ads, no corporate surveillance, ethical design, and decentralization! Own your data with Mastodon!