Hacker News new | comments | show | ask | jobs | submit login
Lbry.io – decentralized digital library (lbry.io)
174 points by rmm on Mar 22, 2017 | hide | past | web | favorite | 92 comments

Would love a quick explanation of what this is, especially after clicking through to "Learn More"[0]. I don't want to watch a video, I don't want to read an essay, even the FAQ[1] (which wasn't exactly prominently linked to) doesn't seem to have a "What is LBRY?" anywhere.

After a few minutes on this site, I still have no idea what this is. "Watch, read and play in a decentralized digital library controlled by the community." sounds intriguing, but... what does that mean? What exactly is a decentralised digital library? How can I watch, read, and play in it? How does the community control it?

[0] https://lbry.io/learn

[1] https://lbry.io/faq

Long and short of it:

Its an IPFS clone by a private entity, with a blockchain for a DNS like namespace. And its monetized, and seemingly badly.

Large chunks of the core code just aren't completes. No GUI. Low amount of users although spamming to tech sites is trying to change that.

Tl;Dr. Stay away. Stick with IPFS. Its on its way of being an IETF standard in my opinion.

Edit: I guess the 2 people from lbry.io found this comment. 2 immediate downvotes without comment. Would much prefer a dialogue if the above is factually incorrect... There seems to be a lot of people floundering in here trying to make heads or tails of this stuff.

No downvotes from me, but there is a GUI: https://github.com/lbryio/lbry-app

Ah, that must be something new. Because when I read all the available documentation, I only found command-line manual JSON queries available.

I stand corrected on the GUI side of things.

AFAICT it's a place where content is hosted by its creators and micropayments are used to subsidize the content creation (instead of advertising). There are many comparisons to youtube on this site, although content is not limited to video.

> How can I watch, read, and play in it?

I haven't figured this bit out yet. I suppose something like a web browser plugin should be on their roadmap if it's not already available. I'd imagine the plugin would yield the LRBY credits to creators in order to browse the content.

The FAQ describes how LBRY is distinct from Storj and IPFS et al, but not mentioned is steemit -- it has a similar concept IIRC.

It also says that non content creators can be given credit for hosting and seeding content.

This is a brilliant project if it works, an attempt to consolidate content from private trackers and ad based pay systems into something that resembles a traditional bazaar for paid digital content.

The only problem I have with it is the name as it's not really a library.

> I don't want to watch a video

And that's doubly-so when the video is confusingly-produced.

It's fairly slow, and isn't very information-dense, and, though it sounds like it's scripted, its grammar doesn't actually make any sense. I transcribed a pierce below:


"The bottom line is that LBRY, like HTTP or Bitcoin, it's a platform or protocol that, once unleashed into the world, no one will fully own or control how artists, authors, video and virtual reality producers how they'll use LBRY to discover and transmit, buy and sell content, that'll be up to them and the demanders of content, so watching this spontaneous world evolve, it's going to be very interesting, and I'm delighted to be a part of it."

New video coming soon with public launch! That was made when net money spent on LBRY was < $10,000.

I had a similar reaction. Playing a video is not practical or convenient for me right now. I was hopeful when I saw they had a FAQ link, but it unfortunately lacks what I believe should be the first entry of any "X FAQ," namely "What is X?"

I took me a while to realise this was a blockchain development.

It is tough to strike a balance between a good introduction for technical people and non-technical people.

https://lbry.io/quickstart is good practical introduction.

https://lbry.io/what is a more theoretical introduction.

We've got a WIP new homepage (more of a wireframe) at https://lbry.io/home2 if you want to share any feedback!

Yeah, still no ideas what it is you do. I'm learning more from HN comments then the site.

A plain english explanation (one to two paragraphs?) would be a huge help.

I like this new page better. It gets straight to the point. I imagine readers coming in from HN would require a more technical intro, along the lines of this[https://matrix.org/].

Try and explain why we should be excited about this from a technical perspective... because it seems like technical people will be your first set of users.

EDIT: The TLDR from the essay at https://lbry.io/what looks pretty nice. Why don't you use that?

Fair enough, I recognise that it's hard.

This article does a good job of introducing LBRY, btw: https://bryanalexander.org/2017/03/19/what-is-lbry-and-what-... (h/t @doctornemo).

I also saw it described as "A cross between Bitcoin and Napster", which was fairly illuminating. I'm sure you wouldn't consider it accurate, but as a starting point perhaps something like this would be useful to have?

At the very least, #1 on the FAQ should really be the answer to "What is LBRY?"

I re-ordered the FAQ, this was a great suggestion. Thank you!

from what I can gather, it is a store, not a library and it is based on blockchain with yet another cryptocurrency.

They pitch it as upside for the creators but except for maybe a good feeling no reason why a consumer would want to use it.

The essay you don't want to read answers exactly those questions and isn't even that long of a read.

Sorry, my point was really that "Watch, read and play in a decentralized digital library controlled by the community" doesn't help newcomers understand anything, because there are too many unanswered questions.

I recall lbry.io being posted here before and there were concerns around the naming system, where anyone could take control of a particular name just by giving more money to it at any point.

There's an FAQ up (https://lbry.io/faq/naming) which addresses this concern, and I think the system that they have designed is fantastic. It seems fair, although I'd give a slightly longer time period for a "counter-bid" personally. I'd taking a look at the FAQ before you move away or comment because they answer a lot of different questions

My problem with all of this is I don't care about having the "best name". I want to make a new name, and then work increasing it's value.

This system seems like anyone who wants to make a name but not monitise it will lose. How much of the value of (picking the first things that come into my head) the names FSF, GPL, GCC and clang come from the organisations and products? I bet those names would be worth much more to someone willing to use adverts and spam.

And even if we raised enough money to keep the names out of the hands of spammers, all that's happening is that lbry gets to charge a tax to anyone who has a name that has got even slightly popular.

Isn't that addressed by this FAQ entry?

> Rest assured, we’re implementing permanent URLs that are always yours.


Yes, but they will just be randomly generated strings, that doesn't really sound appealing. Also they don't exist yet.

That FAQ article is old and I'm going to update it to point at https://lbry.io/faq/naming. You can get names as short as:


Where #a can actually be that short.

If you want a name with no identifiers or filters at all, the voting system will continue to be used.

What are valid characters for an identifier? Can I have lbry://foo## for example?

It matches by claim id, which is a much longer hash, but the system will do partial matches. In the case of conflicting partial matches, the first claim (temporally) is resolved.

Edit: Ignore below...you already said earliest claim is preferred. I need to stop skimming. (or I missed a stealth edit?)

Ah, so everything after the # is random then. And if you choose to abbreviate, you are risking whatever the collision likelihood is, in 4 bit increments, assuming the hash is hex digits.

There doesn't appear to be a throttle on claims, so choosing to abbreviate with just one character could be risky? Or does your substring match prefer earliest claim?

Yes, substring matches prefer earlier claims.

So i can still quickly camp on all the single letter extensions to popular names?

Well, a-f and 0-9, and they are randomly assigned.

I had a bit of a tough time finding this previous discussion since it was flagged.


mrkgnao: These two goals -- being subject to the DMCA (and similar laws), and promising a predetermined, immutable set of rules -- are incompatible.

developer2: It's AOL keywords with a bidding system - for any site or service.

The also generated some comments here, though most were focused on the Berkely issue rather than LBRY:

20,000 UC Berkeley Lectures Made Illegal, So We Irrevocably Mirrored Them


Not sure how it ended up [dead] as well, but we've unkilled it.

Am I the only one who looked at the page for 2 seconds, clicked learned more and then went back here again?

It's sad how "good looking" demo pages have turned me into a robot, ignoring everything that doesn't satisfy my preferences and therefore result in me missing out of a huge amount of valuable information.

Most of these "good looking" pages suck at getting their point across. I really have no idea what they're trying to optimize for. Curiosity maybe? Retention? These pages are trying to be advertisements when they should provide actual information.

∙ What is it (in one sentence)?

∙ What can I do with it?

∙ Where can I learn more about what you just described using meaningless buzzwords?

Edit: This page has more information, but no tl;dr either: https://lbry.io/what

The absence of a succinct summary makes me suspect that hiding what it actually does is the intent of the author(s). That's not too strange an assumption in this age of vaporware, or is it?

The essay and FAQ do offer some explanation, but seem to gloss over other things. Does this require a DRM layer? Can I keep (and backup) any content I purchase?

How can I, as a layman, discern the nature of their cryptocurrency setup? In the FAQ LBRY.Inc is shown to allocate themselves 10% of the total amount of their digital currency, and another 20% is distributed by them to partners. In other places they are boasting that this is an open protocol, similar in impact to IMAP — and free from the clutches of evil content platforms like YouTube.

How do I know this isn't another pyramid scheme (the LBC cryptocurrency) or a mule with a carrot on its forehead pretending to be a unicorn looking for a buy-out?

Their essay has a tl;dr at the bottom, I didn't bother reading the actual essay. It was way too long, when all I wanted was the answers to the three questions you posed.

>TL;DR Digital art is one of the first goods to evolve beyond scarcity. This evolution is changing the way content is discovered, publicized, paid for and delivered. Heretofore, the lack of transparency and monetization mechanisms in peer-to-peer sharing networks has largely enabled piracy. By equipping a peer-to-peer protocol with a digital currency and transparent decentralized ledger, the LBRY protocol opens the door to a new era of digital content distribution making peer-to-peer content distribution suitable for major publishing houses, self-publishers and everyone in between.

If LBRY succeeds, we will enter a world that is even more creative, connected, and conservatory. We will waste less and we make more. We will create a world where a teenager in Kenya and a reality star in Los Angeles use the same tool to search the same network and have access to the same results -- a world where information, knowledge, and imagination know no borders.

Good description.

I think the problem with paying for content is not that it doesn't have a fancy protocol based off blockchain and its own coin.

It is that access to content is not straightforward and it is easier to pirate. What guarantee does Lbry offer that someone won't post a pirate copy of copyrighted content for free?

LBRY offers the same or better protections against infringement that HTTP does.

Is it possible that they try to optimize for discussions on forums just like this one? :P

I poked around a bit and shared some notes here: https://bryanalexander.org/2017/03/19/what-is-lbry-and-what-... LBRY's Twitter team was also very responsive.

Small tip: you can add a dark background to the header through CSS for those who don't have the image in cache. Right now, while loading it's white text over white background.

Great suggestion! Done.


If you're in our Slack, DM me a wallet address at @kauffj for a tip!

I'm still going through the docs and all so this probably has been answered before :

How do you deal with takedown and dcma requests? I know ipfs for example has a blacklist for dcma requests.

I'm trying to understand what keeps someone from copying a video and sharing it for less or uploading a movie or TV series.

Similarly. We will maintain a similar list and official releases will respect this list.

This does not remove the blockchain entry, since this is impossible, but it would be irresponsible and illegal in most countries to continue to host content on this list.

Here's a legal memo the wonderful people at Cardozo drafted for us: https://www.dropbox.com/s/70uezh44ct0244c/LBRY-SecondaryLiab...

We've also spent an unfortunate amount of money on legal fees.

AFAIK, ipfs's blacklist is just on their public http gateway node, and the network itself has no such constraints.

Some good discussion between kauffj (LBRY) and someone on steemit [1].

[1] https://steemit.com/lbry/@hipster/stop-buying-lbry-credits

My favorite bit so far from the LBRY FAQ: "we’re basically, fundamentally the same as a toaster"

I believe most here recognize that something like this is needed (disrupting payment for content from "the little guys"), and I personally hope something like this will happen soon. Best of luck to the LBRY team making this happen; their "growth hacker" seems to be getting a lot of stuff on the HN front page this week! (That and/or LBRY is genuinely about to hit the big time...)

I tried to access the developer program (from this (https://lbry.io/quickstart) link on this (https://lbry.io/get) page) and got a blank page.

It's back, sorry about that!

same here. It's not just you. The quickstart site seams to be down...

It claims no censorship but what if someone starts using it to distribute content that I'd say almost everyone would deem offensive. Can content be removed?

Is there privacy built into the protocol? Can everyone see what I've downloaded & published?

What if someone uploads pirated works?

I tried to add myself to the mailing list at https://lbry.io/signup and got a 500 error.

It's back, sorry about that!

Same here :(

Was the video flashing for anyone else while playing? It would flash in bars of black every second or so. Watching it directly on YouTube "fixed" it.

Looks cool! I'll give it a try.

Looks like you guys created your own blockchain for the service. Any reason why Ethereum wasn't used? Any limitations?

Talked to some of the Lbry team members at Coindesk Construct in SF a few months back. You guys are awesome, keep up the good work!

I'm eager to see how IPFS integrates the IPNS & IPLD layers and how it affects the open linked data context...

Blockchain based?

Missing the most important question:

what proof-of-work algorithm does LBRY use?

Searching around for a while, I came across


which shows it to be Hashcash with some mishmash of existing hash functions (SHA256,SHA512,RIPEMD) cobbled together.

Would be nice if the site featured a diagram showing how the functions and inputs are composed into the output hash. If this is an original function, what is it about LBRY that necessitates this new function?

Is there anonymity in LBRY?

Good question. I'd like to know whether pseudonymous interactions are possible, and what protections are in place to prevent snooping on buyers and sellers and de-anonymization or de-pseudonymization of user accounts.

Great job LBRY Team!

> Subsequent attempts to engage sincerely were also met with derision, so I'm not sure anything could have ever been done that wouldn't have been met similarly.

Some people are like that, where if you respond to their snark, your response will be used against you to tarnish your reputation in the eyes of bystanders.

It is important to never let them get the moral high ground.

If someone says something like...

* "Your team is all white."

* "Your team is all male."

* "Your team is all American."

* "Your team is all able-bodied."

A better way to respond would be something like, "That's true. We're always looking for new talent and embrace diversity, just haven't had much luck yet. <hiring page URL here> if anyone's interested."

It addresses the problem, and contains an open invitation for people of diverse backgrounds to apply.

(All of this is assuming you're willing to work with people who aren't white, aren't male, aren't American, and/or aren't able-bodied, of course.)

We detached this subthread from https://news.ycombinator.com/item?id=13930194 and marked it off-topic.

This type of social control is so bizarre to me. Like, here's a literal script for you to read from if you'd like us to stop attacking you in the future.

> Like, here's a literal script for you to read from if you'd like us to stop attacking you in the future.


I'm talking about them, in a third person sense, not a first person sense.

Them, in this case, is meant to refer to trollish individuals looking to cause trouble under the facade of caring about the plight of women, people of color, etc. but actually are just in it for their own emotions. The people who want to stir the pot, not make things better.

Rather than a script furnished in a list of demands from said subset of the populace, this is more of a suggestion to defuse conflicts in a way that doesn't let them win.


Please don't do ideological flamewars on HN.

> Being forced to adopt others beliefs about diversity as their own is the definition of hegemony.


I really don't have any interest in where this conversation seems to be going.

If you're dealing with a troll, doing what I said will disarm 99% of their tactics. If you don't like that, take it up with the trolls.

If you, rather, have a problem with diversity itself, then let's end this discussion here.


> You're taking part in making sure they have the "right" beliefs about business structure and hiring practices, and it's all the more insidious because you claim you're not.

Okay, this is clearly not a discussion worth having.

> Okay, this is clearly not a discussion worth having.

100% agreed.

Or, you could just not feed the trolls.

Company is a bit tone def when it was pointed out they have a all male team https://twitter.com/LBRYio/status/843865332771164160

Why is it even worth pointing out there team has no women? The original "call out" tweet was very snarky.

There are 9 staff, not including a few advisors. So its probably the case that they all knew each other before the comapny existed. If there were no women in their lab, theres not much they can do about that.

What would the correct response be? "Sorry we didn't know any women when we started, we shouldnt have done anything until the right quota of women showed up."

Should they fire and replace 4 staff with women or just hire 9 women even though they might not have enough revenue to do so?

I agree with your general sentiment, but their response does sound a bit weird. It would have been fine with me if they had said "oh it's just by chance, plus most of us knew eachother from before we started". Saying "we have no female roles" and "oh we get paid less so we're all women" (wtf?) isn't handling the situation well.

It was a crass and tone-deaf attempt on our part to signal awareness.

Lesson learned. Subsequent attempts to engage sincerely were also met with derision, so I'm not sure anything could have ever been done that wouldn't have been met similarly.

Your attempt to respond sincerely was to say you don't have any women on staff because you don't have any executive assistants or office managers, which even a generous reader would call a sexist reason:


yeah... clearly they haven't learned :/

To be fair, I think that was even before my reply.

The initial tweet was not even snarky. It was just trolling. As with most trolling, it should have simply been ignored.

The responses are the damaging aspect here. "Our salaries are low" and "we don't have any EAs" are just painfully bad ways to respond.

Regardless of what you think of the original tweet, the response is ridiculous.

A team consisting of four people. Is this really a problem? It's probably just a bunch of friends that decided to try an idea.

Seriously, what are they supposed to do to resolve it? Fire two partners and hire two women just to make an equality quota? Hire four new women just on the merit of being women?

It's a vapid complaint used by people who want to feel superior without actually contributing anything to the project.

> A team consisting of four people.

I count thirteen: https://lbry.io/team

Thanks. Apologies for not looking at it into detail. The Twitter link only showed the first four.

And the converse is...

"Im sorry. We cannot hire yo for employment at this time because you are male."

Yeah... That would be a lawyers fieldday.

They have almost no staff, what is more, even if they have 10000 male staff only, that is up to them. If you don't like it, start your own company and concentrate on your own business filling your own sense of political correctness.

How does someone create an account a few years ago and this is their first comment?

Some people are just really good at lurking.

Although this is actually this account's second comment.

I was imagining someone selling off-the-shelf users with established history, like pre-shelved c-corps that you can buy.... "in business since 1955".

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