The Kopime Vector | Hacker News Reader

The Kopime Vector

halfcupofsugar.com

22 pointscanadaduane33 comments

We'd need a universally agreed on coding or tagging system in order to copy files that were actually different, rather than just different rips. Probably want to include quality in there too, so each person would be upgraded if the tag matched but one person had a quality higher than the other.

That's a nice-to-have but certainly no must-have. Even if each work of art had an average of say, ten rips - unlikely, given the cost of making a rip and the loss of utility for making the 10th rip - the cost for yet another petabyte is hardly a problem (consider the analogy with software performance and additional resources).

As for the "best rip problem" ad hoc meta-solutions like NFO aggregation sites will arise where tagging would be of most use. Good old regular expression can also do some good for those who wish to have a smaller and cheaper hard drive.

I'm not talking about scene ripping, where things are pretty much already perfect for that sort of thing - I'm talking about all of the other rips or original content that tend to start as personal use and come from (16mm, VHS or Laserdisk -> Various equipment ->) Numerous DVD editions -> Any number of programs -> Any number of formats.

Maybe a half-automated/half-manual registration of files and statistics into a somehow curated distributed database? Maybe magnet links could double as file fingerprints?

You could narrow down the possible matches to just a handful by comparing movie lengths – say, within 10 seconds of the same length is a possible match. Maybe throw in some basic title comparison to narrow the search further, and then do a simple check to see if the audio tracks seem to line up (an exhaustive comparison isn't necessary) to rapidly detect and exclude duplicates with a more than adequate accuracy rate. It would require a bit more complexity in the Kopime device, but not so much as to make it impractical.

Are you aware of any (free or open) code out there that has implemented a reasonably efficient fingerprinting system? I'd love to take a look at it and/or use it for a project I'm working on.

I don't know how efficient it is, but here's an opensource perceptual hash library: http://www.phash.org/

Very interesting. Thanks!

New to me. Thanks a lot.

Will I be carrying a flash drive around in five years? I use my current one (that lives on my keychain) no more than once a week, and that's only to interface with computers that are offline.

How about a system where I hold my phone up to someone else's and speak a semantic description of which files to send and then wait a few seconds. Then, later, when I go home and am near my computer, my phone talks to my computer and automatically uploads the content to my computer/home server, and catalogs/saves it in the correct location.

All of these individual pieces already essentially exist, modulo some of the semantic NLP--nobody's put together the entire package.

Keeping actual files around, eh? What a quaint, old-timey idea.

Seriously. I don't have a single music or video on any media in my home. I stream music to my Sonos and listen to a more awesome range of incredible music than ever before (thanks Pandora and Rdio). This morning it was ethiopian jazz, romantic era piano, ska and early reggae.

True that streaming movies are a weak selection today. But dealing with media is such an annoyance, I'd rather just wait till the selection gets better than deal with all the file formats and wasted time copying files around. I have a more interesting life to lead than that.

Surely you're not implying people who copy files on occasion have uninteresting lives, worth more no more than a passing sneer? That would be a terribly absurd thing to imply.

Surely in this day and age, you can't actually have trouble with file types?

Surely you don't think everyone who copies files from one volume to another just sits around with baited breath waiting for the transfer to complete?

Of course not. Thanks for pointing it out! I'm just expressing my appreciation for great software and systems that make the process effortless. Like Pandora and Netflix.

And no I don't have trouble with file types. I just like to watch or listen immediately, without a lot of fussing around, and I'm happy to pay for it because good content is expensive to produce.

And of COURSE file sharers don't sit around idly, waiting for a copy to complete. They can read comic books while they wait. Or walk up to the kitchen from the basement bedroom in their parents' house for a FREE bite to eat.

Aesthete file sharers frequently have access to movies and music that simply can't be purchased. In the case of movies, subtitles are made every day in a myriad of languages for movies that have never been subtitled. Things that have been lost to the mainstream due to lack of general marketability thrive in the niches that have been created by the maintenance and curation of digital files on hard drives.

You make a great point. For example, I love Yasujiro Ozu movies, but very few of them are available on DVD. It would make so much more sense for those movies to be available to everyone through a streaming system, rather than have to sneak around exchanging files.

Streaming has always struck me as nothing more than a compromise with limited bandwidth and storage. If you can download a two-hour movie in two seconds, and it takes .001 percent of your storage device, why wouldn't you download?

Unless you're trying to appease the entertainment industry, of course.

Whether the data is displayed live off the wire, buffered in RAM, buffered on disk, cached on disk, or beamed into my skull by space aliens, it makes little difference to me.

I only care that the system is responsive with good quality.

But all this misses the whole point of the article: that easy local sharing and storage will make copyright completely unenforceable. If the files in question are pirated, they're not likely to be available for streaming.

I just signed up for the Netflix trial: looked for three movies, none of them available for streaming. Then I tried Hulu: they want $7/mo and still show commercials. Then there's Amazon, but their prices per movie are comparable to brick-and-mortar rental stores (aside from a few things in Prime), and that's for 3-day availability.

Compare to spending a couple seconds copying the file for free, with blu-ray quality and no commercials, and being able to watch as many times as you want with never a "buffering" message. If the industry doesn't step up their game, there will be a market for that.

Most consumers put a huge value on convenience, and don't feel inclined to sneak around copying files. The studios cause the file sharing by making it inconvenient to buy, as you yourself describe.

Look at mobile apps. Easier to pirate than a movie, but piracy is not a problem because it's so easy to buy the apps. Look at the growth rate of app revenue.

Look at Starbucks and how much people are willing to spend on a coffee.

We'll see how it shakes out, of course. But the problem now is that the studios fail to make their product available, not that people don't want to pay for it.

How about the Kopimi Vector? http://www.kopimi.com https://en.wikipedia.org/wiki/Piratbyrån#Kopimi

Some inspiration was gathered there. :)

Seriously though, the word "vector" is used here in the sense of "What are the vectors that enable viral transmission?"

http://en.wikipedia.org/wiki/Viral_vector

The Kopime Vector is an interesting idea, but I don't think that we will see a widespread use of this technology over the next decade. $1200 is still a large amount of money, and I think that trading files from a single physical location will be inconvenient.

I think the use of tools like VPN and Tor that obfuscate location information are much more likely to be the future of piracy.

At $1200 or $400, no. But think a little further out, and remember it's not a single physical location.

Instead, think mobile phone + NFC.

A bit off-topic, but NFC is really not the tech you'd use for this. It requires devices to be very close to each other, and has very low bandwidth. Even low-power Bluetooth would give you about twice the bandwidth, at much better range.

Very close together is a feature here, not a bug.

I kind of agree. There's a bit of specialization worth noting. For example, Tor will likely be used to find and download the newest and latest works of art. But if you want an archive of "art throughout history" it would make sense to just go sync with your local archivist.

It's far more likely in 2017 YouTube, NetFlix, Hulu, iTunes etc will have instant, global, legal, affordable access to massive catalogs - most of those pieces are already there.

Is a wizard coming in 2016 to magically make executives in the movie and music industries understand the shape of the modern world?

In 2016, any executives in the movie industry who prefer that people copy movies around manually on BitTorrent can easily make that happen by blocking their product from convenient streaming access.

I'd like that (and I'd like to pay reasonable prices). But would the movie studios like that? And cable companies? I think the last 90 years have given copyright behemoths a war chest that they will continue to use for some time, to the detriment of innovation and reasonable compromises.

It's a natural extension and obvious next step for what we already have right now, the only major missing piece is "global".

What size have you assumed for each movie? Surely we expect this to increase with (if not be the driving force behind) the cost of storage?

1 GB per movie.

Reply on news.ycombinator.com