Hacker News Reader

I’ve met their head of data science. They’re doing all that stuff and have been for many years.

Porn has always been leaders in technology. They were the first to embrace home video and can arguably be credited for VHS winning over BetaMax despite being inferior, because it was cheaper and all the porn was VHS only.

>can arguably be credited for VHS winning over BetaMax

This is something that is always mentioned, but sometimes I wonder if it is not a myth, or a simple exaggeration or simplification.

It's definitely a myth, or at least a minor contribution.

BetaMax housings were very small, so they didn't have enough capacity for feature films unless they were run at 1/2 or 1/3 speed, which reduced quality. The thing is, VHS could also be run slower to improve run times, but instead of maxing out at 90m, they could run for almost 250m for tapes about at the same price.

Tapes with longer run times eventually came to market, but well after VHS had established itself.

Something I've heard brought up regarding the length is that early Betamax couldn't tape the average sports game, while early VHS could. I'd imagine that was a big selling point.

The early Betamax recorders didn't have a timer on them. Plus the tapes were super short. So if you had a show at 3pm you wanted to tape, you had to be there at 3pm to push the record button, then be stuck with only 90 minutes of recorded content.

The first Betamax recorder sold in the US was the LV-1901 which included a 24 hour timer.

The first standalone Betamax VCR was the SL-7200 on which the timer was a $45 optional add-on.

Also at the time people were already buying wall socket timers for their reel-to-reel tape decks to record the radio, these would also work with a Betamax deck. https://www.youtube.com/watch?v=XbNgbtZJAcU

Is it true that Sony deliberately did not make feature length tapes so people would not record feature films from TV?

Highly doubtful. They were (mainly) “just” a telecommunications hardware/electronics company up until their purchase of Columbia Pictures in 1989; they wouldn’t have had an incentive to defend the entertainment sector in 1975 when Betamax was introduced.

This is pretty interesting on the history of the Betamax.

https://flashbak.com/product-genius-eulogy-sony-betamax-1975...

Don't forget they continued the same scenarios in DVD's + vs - R!

> Porn has always been leaders in technology.

This is typically repeated but I question if there is actual proof of it rather than one of those '8 glasses of water a day' which just keeps getting passed along or some kind of plausible and believable anecdote.

Given the bandwidth porn has consumed since literally the beginning of the internet, I think it's fair to say that they can be credited with significant technology gains too.

As a recent (but maybe not too important) example, PH created hover-to-autoplay-gif for video thumbnails and then YouTube copied it.

Is that why YouTube thumbnails seem to focus on segments with a woman wearing little clothing, even though it’s a small part of the video?

Of course, along with the very-fucking-annoying Youtube Face [0]. It's so ordinary to mess with human psychology, anything to increase click through rates!

[0] https://boingboing.net/2018/04/11/youtube-face-the-same-thum...

It seems like most of the examples of that are "reaction" videos, and if you're trying to sell a reaction video, you show the reaction, right?

I see it everywhere, not just on reaction videos.

I can count the number of reaction videos I've seen on one hand but the YouTube face is all over my homepage.

The next iteration to this is out, it's using extreme Snapchat-esque filters to comically exaggerate the size of the eyes and mouth of a reaction face.

It's mesmerisingly disgusting, I wonder what the next steps are. More digital augmentation awaits?

> I see it everywhere, not just on reaction videos.

I have seen them in in-store advertising in supermarkets.

> I wonder what the next steps are. More digital augmentation awaits?

Virtual/retouched streaming personas. It's already a thing in japan and among furries.

What software is being used to create virtual personas? Am I right in understanding that this work isn't being done by Western companies?

I haven't dug deeply, but there seems to be a whole ecosystem. An open source avatar data format[0], full body motion capture software[1], this stuff integrates with standard 3d software environments such as unity/unreal/blender/maya/max. So does commercial software[2]. For the hardware it can be a simple web/phone camera on the low end (giving you a talking head) over camera + VR trackers + an iphone for the face sensor to high end rigs with multiple cameras and mocap suits.

[0] https://vrm.dev/en/ [1] https://github.com/sh-akira/VirtualMotionCapture [2] https://www.youtube.com/watch?v=hJ_5d_1NHCQ

I don't think it's 100% the same thing (VRM appears to be full body, not just face), but FaceRig[1] is a similar concept made by a Romanian studio.

[1] https://facerig.com/

Funny related story! At Reddit we built a tool to create thumbnails for webpages. The algo was basically find the biggest “squareish” image and then remove a row or column of pixels. The row/col was chosen based on which one left the image with the most entropy. We kept doing this until the image was a square.

That algo basically cuts off women’s heads and leaves just their cleavage.

Obviously we had to adjust it to try and avoid that, but even today you’ll see thumbnails on Reddit that are just cleavage.

Seam carving gone wrong?

I thought that a face would contain more entropy than two flesh mounds! Color me surprised. Could the algorithm be tweaked to favor removable from the bottom part of the image, since humans generally perceive from top to bottom? Most images contain the 'meat' at the top or middle, I'd assume?

We thought so to! But it turns out that a nose, mouth, neck, and chest have more entropy than a face.

I don’t remember exactly how we fixed it but I think biasing towards the top was involved.

Link from yesterday shows some good examples of the "algo" (seam carving):

https://avikdas.com/2019/07/29/improved-seam-carving-with-fo...

I don't know about formal studies, but I've read many an article (insert appropriate "reading for the articles" joke here) talking about how demand for porn influenced early adoption of most new forms of media:

printing, broadsheets, photos, different forms of film, and definitely computer video. (Here in Seattle I recall reading about some form of early adult film that shows up notably in local laws...some form of -scope, though I don't recall the name)

DVD-HD vs blu-ray? Adult industry influence was not small. When HD rolled around, the adult industries had notable influence. Where did the makeup techniques for dealing with such high def come from? Porn.

I think the comparison to the "8 glasses of water" standard might be off because unlike that, here no one is claiming a precise measurement, merely an observation of trends.

VHS and Betamax was maybe influenced by the adult industry (although sports seems to have mattered as well)

Parroting the same about HD/Blu-ray is kind of hilarious though. Porn was deep into the streaming game long before blu-ray hit the scene.

> long before blu-ray hit the scene

When the dust settled, perhaps, but I recall when the issue first came up in 2000 and just about every discussion argued technical merits...and where the porn industry was looking. Anecdotal data, but in 2000 streaming wasn't viable for the majority of users (and still remains not viable for a lot) so that's no reason to discount the impact in that arena.

Only if massive pixels on an offbrand realplayer that only two sites used and had to be installed to your desktop counts as "deep into streaming."

It also froze every 30 seconds: https://www.youtube.com/watch?v=0Awlgy3qUSE

If people had the star trek holodeck around today what would one of its primary uses be

I mean, there would be holodeck sex, but unless you're a teenage boy, you're not going to do that more than maybe twice a day max, and you're usually going to do it alone. After you get off, you do the same thing even teenagers do today with the internet: you start playing shooting/fighting games with all your buddies.

Another example is VR/remote control toys adoption - Mindgeek was pretty early adopter of VR porn and remote control toys with synchronization of video frames.

> I’ve met their head of data science. They’re doing all that stuff and have been for many years.

This sounds very much in violation of GDPR. Are they just betting on people being too embarrassed to take them to court?

Maybe they had to change their practices after GDPR. I met him before that was passed.

It's not a violation of GDPR. However, is somebody makes a GDPR request to retrieve all of their personal data, or to delete their personal data, then they must comply within 30 days. They have to organize their datasets to make these operations easy.

There are six legal bases for processing personal data under the GDPR. Which do you think applies in this case?

Additionally, sexual orientation is subject to special rules under the GDPR and cannot be processed at all without consent (and some edge cases). That could possibly apply as well.

I think the site would require that the visitor give consent when they visit.

You're not allowed to require consent in exchange for access to content.

Really? Where is this rule stated in the GDPR?

Article 4 section 11 defines "consent" as something that must be freely given. [1]

Article 7 section 4 further clarifies that consent is not freely given if provision of a service is conditional on consent. [2]

Recital 43 is super clear: "Consent is presumed not to be freely given if... the performance of a contract, including the provision of a service, is dependent on the consent despite such consent not being necessary for such performance." [3]

The gdpr.eu website unpacks the legalese: "“Freely given” consent essentially means you have not cornered the data subject into agreeing to you using their data. For one thing, that means you cannot require consent to data processing as a condition of using the service." [4]

[1] https://gdpr-info.eu/art-4-gdpr/

[2] https://gdpr-info.eu/art-7-gdpr/

[3] https://gdpr-info.eu/recitals/no-43/

[4] https://gdpr.eu/gdpr-consent-requirements/

How long do you think GDPR has been around for? Why do you think it was passed?

> [porn] is very easily classifiable with a list of the physical features of actors (with maybe a few behavioral distinctions), the actors in the video, how they are matched ("how" doing a lot of heavy lifting here) and possibly director, producer, and age of content, it would be easy for them to have a very specific dossier on all users. Moreover, it would be financially beneficial, because it'd be easy to maximize engagement with that stuff and a past record of engagement time

See, this was my thinking when I created an account and tried to train the recommendation algorithm.

I was sorely disappointed when the secret sauce turned out to be nothing more than:

    for vid1 in watch_history[:-10]:
        for vid2 in all_videos: // Notice how it does not exclude watch_history
            best[Levenshtein_distance(vid1.title, vid2.title) + Math.random()] = vid2
    return ksort(best)

I would conclude their engagement maximizer is meant to give me the worst possible recommendations just so I spend more time on the site and load more pages (ad banners) before I find something, but then it would have taken the highest Levenshtein distance instead of the lowest. I'm not sure what the sibling comment from jedberg ("They’re doing all that stuff and have been for many years.") is on about, because they are definitely not matching the video content. I'm guessing it's still too computationally expensive. Reverse psychology (letting you think that you are getting personalized hits but giving you unrelated stuff instead) seems a bit too much like a conspiracy theory. And the theory of giving you something somewhat-good does not explain why it always matches the title and never anything else.

I've never really gone to tube sites (I just know how much business they do), but I can't believe it's that bad. I'd also think it would be cheap enough computationally, and iafd.com has pretty much compiled and indexed everything already.

Maybe there's an opening for someone who can do it cheaply, or they've discovered that it's not worth it?

From what I've heard from people peripherally involved in the space (and from what I've observed in practice, though that might be less useful) the big platforms definitely don't do only this.

> Moreover, it would be financially beneficial, because it'd be easy to maximize engagement with that stuff and a past record of engagement time

Maybe they don't recommend the best video for you (i.e. the one you'll like the most) but one that is relatively good fit for your taste, without being too good, so that you spend more time on the site. If you know what I mean.

I was thinking the same thing when I typed that:)

No we aren't doing that. Ads are contextual on TrafficJunky/Pornhub. Too few people actually making accounts and purchases compared to the billions browsing anonymously.

what's the % of users you see using ad-blockers?

separately, has there been any consideration given to googleanalytics (+google fonts) usage on your properties, and Google's ability to track visitors across your sites and the web?

Small since ~80% of our traffic is mobile and we also serve static ads to most adblock users on desktop.

Google Analytics is the only analytics product that can support the data we need processed at a reasonable price. We do use their IP anonymization by default and it's pretty easy to opt-out of GA tracking.

wow, I'm surprised that mobile makes up that much traffic!

it would be interesting to see what sort of UA visit your site... I'm assuming Chrome is the dominant mobile browser (which unfortunately doesn't support ad-blocking)

It’s split Safari/Chrome pretty evenly. We publish a lot of our data on pornhub.com/insights if you’re interested!

Mindgeek uses PHP. I’m not sure how that meshes with “very skilled technologists.”

BTW, I’m a hardcore Python snob who will eventually watch the world dissolve into an oblivion of Golang, Elixir, and Scala.

You can write poor code in any language. Language itself has very little other meaning than how well it does it's job in the given context. PHP is great for fast development, has a lot of developers available, and comes with great tooling for the web.

Traditional downsides like poor threading and overall performance were/are overshadowed in that most webcalls are always I/O bound. PHP7 has made great strides on this though, especially on memory usage.

YouTube has been criticized for it's recommender system tending to push viewers to more and more extreme content. I think the porn tube industry no different.

so somebody here thinks pornhub's recommendation system doesn't also encourage the viewing of more and more extreme content? if so, say so, don't be lazy.