I've been both a vocal critic and vocal supporter of LBRY over the last few months, and that comes with a certain intellectual burden to respond both analytically and personally when they come out with a new, significant update to the user facing plan. Not only that, when things looked to be running positively, it's in my best interest to be vocal in my support in order to maximize the positive result available to me.
Or, in other words, when people do the right thing you should be loud and proud, even as much or more so than when they do the wrong thing and you complain.
About a week ago, LBRY Inc. put out The Future of Monetization and Creator View Rewards, a document which clarifies and expands on what and how will they intend to shift the reward system and monetization of content on the LBRY platform. There is a lot of material in there worth taking a look at. I'll be quoting that original document pretty aggressively in my response and analysis.
(I know that at least one person on earth must be waiting for my response to their previous discussion of implementing some sort of web-of-trust user-level mechanism for content discovery and surfacing, but that's going to have to wait a little bit longer until I'm ready to tackle it. It's more of a mixed bag.)
Fiscreality
Last month, we spent almost 3.5 million LBC on this program, which is not sustainable.
But we never anticipated this program to be the end state of LBRY, just a stepping stone.
I'm going to refer to LBRY Inc. as LBRY for the rest of this article, but that should not be interpreted to mean that I conflate the financial/technical organization with the platform itself. The platform and technology are rightfully referred to as LBRY and the incorporated form is explicit in being separated from the platform thanks to the open source nature.
It's good to see that LBRY can be forthcoming and self-aware when it comes to the non-sustainability of the current/last iteration of the reward program. One of the things that has been all too often ignored or underplayed is that all of the LBC rewards which came through the program were not being generated through "inflationary systems" attached to the mechanics of the platform. They were being distributed from a pool of pre-mined stake owned by LBRY Inc. That means that an entity owned them and was giving them away of their own free will in accordance with a set of rules that entity set and created.
Rewarding Life
There has been an aggressive contingent of inflowing users of the platform who never seemed to understand that. They treated the LBC view rewards as mechanical side effects of simply using the platform and not a singular entity distributing its own money to users which satisfied its own qualifications. I believe that many of the problems with new users could have been avoided if that had been made clearer from the beginning. This document takes some significant steps forward in making that much more explicit – though I doubt that a lot of the people who would be problems will read it, at least it can be referred to.
Looking at that reward curve in particular, it's terrifying. I probably would have left out the June point altogether and let it terminate in May because it doesn't really give a proper view of the situation, but it's bad enough as it is. This is not a trend that can be sustained by any platform in a real sense.
There are no free lunches in this world. While some blockchain companies try to create perpetual motion machines, LBRY will remain firmly grounded in reality.
We believe that by replacing YouTube and Amazon with an open-standard that is far less costly to operate, and with no cut to the middleman, we can design a system that is win/win for all.
But a system that works perpetually must mean that ultimately no money or tokens have to come from us.
This is a great section and it lays out some very specific points.
To be clear, this is an implicit reference to and criticism of the way that the
STEEM and HIVE (and arguably MINDS) have mechanized their reward system for creators. The "reward pool" of those platforms comes from the inflationary issuance of more tokens into the pool, reducing the value of tokens held by all on the blockchain by a little bit every cycle. They are sold as "perpetual motion machines," but in a proper economic sense, they might be thought of more as "Zeno engines."
That which is in locomotion must arrive at the half-way stage before it arrives at the goal.
— as recounted by Aristotle, Physics VI:9, 239b10
If you're not familiar with Zeno's Paradox, the short version is "if you have to cross half the distance every time you want to go somewhere, how do you ever reach that point since you're only crossing half the distance at any given time?" Likewise, with an economic token which is issued as inflationary injections, how do you ever achieve parity value if every injection causes a decrease in value?
(The obvious response is "you don't, you just get increasingly close enough," which probably suffices for Zeno but doesn't actually help much with fiscal valuation. There are a number of ways to talk about/think about the problem, but I've never seen anyone take it up.)
Which brings us to the specifics of where LBRY envisions tokens transferred to creators coming from, and it's an interesting selection:
- Viewers, when buying content
- Fans, when donating
- Fans, when purchasing memberships, badges, and other things
- Advertisers, when viewers don't want to pay
This is an interesting collection of options, because when taken along with the suggestion below which is written from the perspective of creators just as this is written from the perspective of consumers, and which allows a creator to simply put content up without any particular cost, it covers a lot more options and opportunities than any other network platform for content distribution that I've seen, centralized or not.
Two of these mechanisms already exist and are functional within the system (at least at a limited level). Consumers can buy content with a price set by the creator and they can give a creator money as a donation/tip.
One of these things is going to take a significant overhaul of the social media platform mechanisms which are available on LBRY and which actually give me some cause for concern. The "purchase of memberships, badges, and other things" really requires a lot more social media architecture and focus than I've seen be the real interest of LBRY either in documentation, white papers, or public. What they have in mind is probably something like what Twitch and YouTube have available to set apart a particular user in the social context of a creator or brand, but in order for that to have meaning there will have to be a lot more development of real-time or at least more aggressively transactional communication between users in order for those things to be meaningful to show off.
It's not impossible but I haven't seen the interest or design for creating that sort of thing as yet.
As for advertising – that probably deserves its own subsection.
Ad Hoc
As we add richer advertising, viewers will be given choices about what to share with advertisers. Viewers that share more are likely to be more valuable to advertisers. For example, your local pizza place can only advertise to you if they know where you live.
LBC will always be a way to use the network more privately. Users will be able to choose to hand over a lot of info and see fewer ads, less info and more ads, or spend LBC and see none. The key is transparency and choice, and we guarantee that the final system will have these properties.
Remember when I said I wanted to be more positive and I was glad to have the opportunity to discuss things which I see is moving in the right direction?
This is the sort of thing I had in mind.
It's rare to see a platform which goes back to basics on understanding what advertising is actually for and rethink how it can be used to achieve that underlying goal. While simultaneously respecting the one things that the consumers on the platform actually have which they exchange openly: their time and personal information.
Other people, possibly wiser people, have said "if you're not paying for the product, you're the product" – and that's absolutely true. That is the exchange that we make in order to use the platform. We trade one of the most valuable commodities in human society – data.
Google doesn't make their money from hosting websites. Google makes their money from targeting advertisements. Facebook doesn't make their money by charging for access and never will – that would make no money at all. Facebook makes their money by allowing advertisers to put advertisements in front of you in exchange for cash. The more specifically they can choose who to put their advertising in front of, the more the platform can charge for the service being provided to those advertisers.
You may be personally offended by being the product. You may think that using information that you reveal to the platform in order to serve you advertising you never wanted to see in the first place is racist or whatever-ist is cool this week.
You would be wrong.
There are no free lunches.
It's interesting to see that relationship made explicit in this document. It's a real relationship with real requirements that almost never gets spoken of in plain language. Kudos to LBRY for doing so here.
In a very direct way they say "you can have complete privacy but because that makes things more expensive for us (as creators), it's going to cost you."
That's a bold decision. I'm not sure it's one that's going to be received particularly well by a user base that has become used to the idea that everything is "free" (in the sense that they are never told about the costs), but I have always believed that explicit is better than implicit and more honest.
But will it really change things significantly? Let's look at the specific monetization options there currently talking about.
- Fixed monetization, by price. A creator wants to earn a certain amount per video. A viewer must either pay that amount, or watch advertisements until the creator has earned enough money.
- Fixed monetization, by time. A creator wants a viewer to earn an amount of money equal to 30 seconds of the user's time. That viewer must watch 30 seconds of advertisements or pay the amount for the video that those 30 seconds of advertisement are worth.
- Voluntary or suggested monetization. A creator requests a user pay what they want, or suggests a user pay a specific amount or watch a certain amount of advertising, but does not require it.
- Disabled monetization. A publisher specifically says they do not want monetization and attempts to pay or tip would be rejected by applications. This is useful for sensitive or archived content, where payments can be morally or legally undesirable.
Keep in mind, one of these options is "no monetization at all," which is the way that most media has been delivered on the web since the day it was born. If you just want a free distribution network (at least as it stands now, more about that later), nothing will compel you to put monetization on your work. Which is an interesting concept given the lip service paid to "the donation economy" by a lot of the people involved in social media. I look forward to seeing how many people go from fully disabling monetization but repeatedly pitching various ways of donating to them (which is effectively what Patreon and other like sites do) to more creator-directed advertising architectures which effectively erect a paywall between users and content.
Hopefully this is intended to be on a per-post level, because at that point it becomes possible to create, promote, and paywall content in very much the same way that Patreon allows a creator to make much of his work public but gate some of it behind an effective paywall.
Of the four options here, I think the most likely to be used is no monetization at all and, following that, fixed monetization by time. Monetization by time specifically sets out the maximum amount of time that the creator imagines the user will be willing to trade for the experience of enjoying their content. That is a direct assessment of value which is a lot more interesting and easier to do than saying "I want to earn five cents every time this piece of content is consumed by a new person." That's a much harder judgment as to what is appropriate and it doesn't scale with increases and decreases in LBC on the user experience side.
Left unsaid here is that there is an absolute requirement for there to be a biddable open market of advertising slots in order for the value of advertising to be determined at the first place, and that is an entirely different mess of user interface architecture, underlying system architecture, and a reasonable means for users and potential advertisers to get into being engaged with it. I would love to see what people have been thinking about in terms of building that system, because in terms of complexity it's probably at least as hard as building the social network consumption side.
(It does occur to me that advertisements are just content with a negative cost to view. If the system just treated any content with a negative cost to view as "an advertisement," all they'd need to do is provide filter hooks for keywords for creator-desired advertising for their channels/content pieces and the hooks to provide the opportunity to view content with a negative cost before you view content with a positive cost, and most of the architecture falls out automatically. There is still the massive issue of how to "police" (for lack of a better word) advertisements for being properly keyword tagged so that creators don't get screwed with inappropriate advertisements, but it is theoretically possible.
It would invert the current digital advertising paradigm by putting it the selective power in the hands of creators whom the advertisements would be running over. That's a long time coming. As it stands, in most digital advertising architectures, the power resides in the hands of advertisers who select by data abstraction content creators they wish to advertise on. Switching that up would go a long way to providing creators the power they've been denied for decades.)
Self Awareness
Briefly, I want to touch on praise for LBRY for putting a section which explicitly calls out flaws in the current system. All of the flaws they call out are real problems, either for the platform itself or for them as an organization. It's great to see that they recognize and acknowledge problems for themselves as an organization as a problem for the community, because as I have said in other contexts about cryptocommodities in general, it's very difficult to differentiate a single body organization which is literally the biggest token holder and the platform as a separate entity, because in a real sense they are not differentiated. LBRY Inc. is LBRY for all intents and purposes, no matter that the platform is open source and someone else could step in and take over development at any moment. They won't. It's not going to happen. So a threat to the organization is a threat to the platform and a problem with the organization is a problem with the platform.
The current system requires manual intervention from "an authority" in order to have valuation at all. This is justly observed as a real problem in a system that intends to be "decentralized." The theoretical lack of bound in the amount of LBC which could be required to be given away literally runs into the problem of manual intervention by demanding it, so that makes sense. The lack of sensitivity to the current price is directly connected to the fact that it's not really a market; you have a lot of sellers but you don't have a lot of real buyers with funds to spend.
The issue of disproportionate rewards to homepage placement creators is a little bit of a different issue. If I, as a platform user, was the one picking what I saw as my homepage content, this would be a nonissue. The problem is that I am not; the system (a.k.a. "someone who is not me") is picking what should be on the homepage unrelated to what I'm interested in beyond the section on "Channels you follow and Tags you follow." The easy way to fix this problem would be to simply remove any homepage content that isn't directly derived from my expressed interest. Quick, simple, and it means the homepage opens faster.
If the problem is that in order to facilitate discovery, I'm seeing Trending or Top content under those filters by default – I'm not sure that's a problem. It's only a problem if I'm seeing content I'm not really interested in, and for content blocks which are created as a direct result of me choosing Channels to follow or Tags to follow, that's the nature of the beast. It would be nice if I could say that I want to see the newest in those contexts by default all the time.
Really quickly, let's take a look at what they think of as the properties of a good system.
There are only a couple of points here that I think are not self-evident at the level of "I hate war and love puppies."
- Not allow one creator to become too powerful
Not to put too fine a point on it, but what does that even mean? In what sense could "one creator" become "too powerful?" What kind of power are we talking about?
Is this implicitly some sort of statement about limiting how much LBC an individual can have on the platform? The obvious consequence of that would be large creators selling their LBC stake to exchanges in order to take their profits rather than keeping it invested in the LBC architecture past a given breakpoint.
Is it an awareness that as LBC becomes concentrated as a result of people willingly exchanging their funds for content products, successful accumulators will have increasing amounts of influence because they will have enough LBC to float really large Supports in order to surface content that they are interested in rather than an audience? That's a side effect of thinking of Support as a useful or meaningful way to determine what "good content" is from an aggregate point of view rather than doing the real hard work of figuring out what "good content" is from the individual user perspective. That doesn't mean that Support should be limited or log scaled or whatnot (mainly because that just means that large holders will come up with automated means to spread their Support actions across multiple automated accounts which are effectively impossible to track), if that's what's under consideration.
- Reward quality, original, and engaging content
The problem here is that it shouldn't be the job of the reward system to determine quality or originality. Only the last is really measurable in terms of metric. Quality and originality are things that only individual users can judge for themselves and a system that attempts to reward those qualities ultimately just ends up falling back on the arbitrary judgments of individuals whose taste and interest differ from the majority of users.
That doesn't end up serving a good purpose. It just ends up making people doubt the system as a whole.
"Rewarding engagement" is one of those things that people complain about in more traditional social media distribution platforms, particularly YouTube, with the complaint that more controversial, argumentative, and contentious content gets promoted more regularly because it drives "engagement," specified by comments, showing that content to other people, etc. I'm not necessarily one of the people who disagrees with that but I'm also not necessarily one of the people who thinks it's a bad thing. It does lead to a very specific type of content which is rewarded and as we all know you always get more of what you reward.
In With the New
All of which brings us around to a sketch for a new reward system. It's really a sketch; specific mechanisms are very lightly traced but vision ideas are fairly strong. And because it is generally coherent and decent, I'm going to take it on one point at a time.
- A fixed pool of LBC to be dispersed each week
- The pool varies in size based off the number of reward-verified viewers that watched content the week before (when LBRY grows, we all benefit)
- The multipliers used to determine the pool size would vary based off the price of LBC (no 40x YouTube CPMs)
A fixed pool of indeterminate size is not exactly starting off on the right foot. Mainly because it doesn't answer the really important question "how big is it?"
Again, competing social media platforms determine their reward pool size based on an inflationary curve of how many tokens are in play. I'm going to make the assumption here that this "reward pool" would be fully funded out of the pocket of LBRY Inc., just as the current unbounded pool is. This provides a certain problem because while, in theory, it will vary based on "the number of reward-verified viewers that watched content the week before," we aren't given a baseline start. For an inflationary curve, we know at any given time how many tokens there are that exist as a side effect of the blockchain itself. The LBRY proposed pool is suggested to vary in size based on reward-verified viewers without giving us any kind of even soft numbers.
That's a problem. It means that we're back at the original "unbounded pool" problem.
It might be more useful for LBRY Inc. to simply start by setting aside a portion of the LBC they control up front as "full reward pool," and allocate each week from that pool, adjusting for verified viewers. This would have the great advantage of being largely transparent and setting a visible, clear, countdown timer on how long they can afford to run rewards of significant value.
That sounds as though it might be a drawback but it's really an advantage. Nothing keeps them from adding more LBC to the pool in response to community developments they're happy with or deliberately choosing to put any rewards that they get from tips on their own content into the reward pool for everyone – and if it's literally kept in its own wallet or Channel, then nothing keeps community members from donating directly to the reward pool as a show of community support or as the result of a special event.
Adjusting the pool size based on the price of LBC is interesting, but I'm not sure it's necessary. At that point, you might as well just set the reward pool size in US dollars and pay out accordingly. Instead, losing the price of LBC as a modifier to the reward pool means that as the value of LBC goes up, the value of the reward pool size goes up. As the value of LBC goes down, the value of a fairly allocated reward from the reward pool goes down. This matches the expressed desire for LBRY growth driving the growth of the reward pool.
In short, if you base the pool on tokens rather than value, it automatically scales with growth and you don't have to worry about making rules to do so. Less rules means less chance of mechanically broken rules.
- The entire pool is distributed each week, in proportion to views from verified users. Alternatively, the pool could be distributed by the time spent viewing, which would reward longer viewed content more.
Really two points, one of which is good and one of which could be quite the opposite.
Distributing the whole pool each week, split by views – quite functional. Easy to calculate, easy to manage, and easy to do.
Distributing the pool by time spent viewing is a little bit of a different problem. It's not really hard to calculate, assuming that we are only talking about viewers who have data collection turned on, but as noted it definitely skews distribution to longer form video content. I would remind both LBRY Inc. and other creators that video content is not the only kind of content that LBRY distributes. No one tracks how long people "watch" a text article like this one. How would you? How long the pages open for? That doesn't make any sense. Does that mean that text content shouldn't be rewarded because it doesn't keep people on the platform as long as video streaming content? How about images from photographers? Does it make sense to reward them based on how long people keep the image up on a given page without going onto experience more content?
This is a real problem, not a supposed problem. If LBRY wants to shift to being an entirely video-focused platform in terms of creator rewards, that's certainly their prerogative but I definitely think that it would be a bad idea in terms of the passive reward pool.
The counterargument is that longform video content creators might say "why am I getting the same amount of reward from the reward pool as someone who just made a single picture?" And that is a legitimate question, to which the legitimate answer is "if you really need more reward than the passive view reward pool, you need to start canvassing for donations or selling advertising time in your longform show which takes a lot of your time and effort to create." It needs to be kept in mind that the reward pool is not the only mechanism by which creators can be rewarded and it shouldn't be treated as such.
(Likewise, if I thought my passive income from the reward pool wasn't sufficient to keep me creating longform text content for the platform, I should also either start canvassing for donations or selling advertisements in order to be able to afford to continue to create content like this. The idea that one might be able to get advertising revenue for longform content, driven by the content creator, is actually pretty exciting.)
- Add bonuses for content creators that are publishing original content to LBRY before anywhere else.
This makes perfect sense and is well within the capabilities of LBRY as an organization. I recognize that this is specifically directed to video creators putting content up on LBRY rather than YouTube, but I hold out hope that one day it might be almost as important to them to maintain bloggers in the same way.
- No one creator can earn too much of the pool. For example, no one creator could ever earn more than some small percentage (0.1-1%) of the pool each week*.
I find this a little bit concerning. Not because it's unreasonable; it's LBRY Inc.'s money and they can decide to limit how much any individual gets out of the pool at their discretion. What bothers me is that it sets a bottom limit for the number of participants in the reward pool, without concern for the value of LBC.
If this became standard operating procedure, I would request that the "maximum earnings available from the reward pool last week" be of value that was always visible in the interface for your wallet. It needs to be transparent and non-ephemeral so that people were aware of that limit and where they stood in regards to it.
Along the lines of "you get what you reward," this has some interesting implications. Big creators would be less incentivized to depend on reward pool payouts to pay for their content. If it caps and they always hit the cap, then their strong motivation becomes to lean harder on soliciting donations and/or advertising, which creates a demand for advertising and greater presence of advertising. I can't decide if that is a landscape in the aftermath that I feel really great about.
In a sense, I'm already living in it given the really strong preponderance of Patreon pitches and embedded advertising (as opposed to system-tied advertising) on YouTube. It's sort of like living through a recapitulation of early 50s radio.
- Creators must be positively approved by a LBRY Inc. staff member, to prevent fraud (we already made this change, hope you like it).
Absolutely required in order to make the system work at all, so that's good to see.
What isn't so great is a lack of a link to a document which discusses in detail what one needs to do and be in order to be positively approved by LBRY Inc. staff and the list of infractions which can result in the retraction of that approval. Yes, yes, I know those documents exist (somewhat), but this would be a perfect place to link to them as a reminder and as a reference.
- An additional rule would prevent the pool from ever growing too large too fast (or too small), to solve the problem of rewards being potentially completely unbounded.
That is definitely a sketch. Unfortunately, it's too much of a sketch to be more than a vague impulse. This really should have been at the top of the list along with the sketch injuries talking about how that pool will be allocated, grow, and shrink. In fact, as points go, it probably could have been removed altogether and assumed to be part of determining the pool size in point 2.
Moreover, what is "too fast?" What is "too slow?" We don't have anything that even suggests orders of magnitude on any of the definitional parts here, and while I do appreciate that it's a very loose sketch, at least some sort of grounding needs to touch somewhere along the framework to start making sense of it.
- Should we ever generate any income from advertising above the operating costs of lbry.tv, it will be used to purchase LBC on the markets.
The only way this point could be better is if it explicitly said that the LBC so purchased would go into the reward pool, and thus improve rewards for all validated participants of the platform.
It's interesting that it's explicitly tied to advertising and not to other functions of LBRY Inc. as an organization, but that's just nitpicking. Overall, this is an excellent call.
Epilogue
So while we think this plan makes sense, we're not going to anything until we hear what you think of all of this. Does our intent make sense? Does the design? What do you think?
I think I've answered that question both in the specific and in the broad, but overall I think the idea goes in the right direction. It certainly has the potential to pay off for individual creators in a way that other media distribution platforms don't or haven't because of a focus on top down assumptions rather than bottom-up building.
In particular, the inversion of the typical advertising architecture has a lot of promise, although I know that a lot of people are going to be unhappy that advertising can exist at all within the infrastructure. Hopefully, putting the control in the hands of creators in terms of what advertising they accept and what advertisers they except would be sufficient to change the dynamic in a way that improves the experience for everyone. There is a lot of space to explore there and a lot of potential.
Deemphasizing the reward pool is something that will have positive knock on effects through a lot of the platform, though a lot of people who have turned up under the assumption that "LBRY, it's like free money" are going to throw massive hissy fits. And that's terrible, but it is necessary for the growth of the platform overall. Hopefully automatic rewards from LBRY Inc. become an ever smaller portion of rewards distributed by the platform, making them less important going forward. That will be good for not just the organization before the platform, depending less on a single supply of token funding for interacting users.
I'm still concerned, as I often am, with the fact that there are so few ways for individuals to earn LBC outside the scope of "being a content creator." That's going to make it hard to build a distributed reward mechanism that doesn't depend on people pumping fiat currency into it because that's the only tradable commodity mechanism to get into the game. The integration of ability to buy LBC is helpful to get on board, but only if you already have a fiat currency you can afford to trade for that privilege. As it stands, there is no marketplace of interaction to give people the ability to earn LBC for more specific actions. While it's a little bit outside the scope of reward and advertising mechanisms, both of them will only be enhanced if we can get some sort of integration with a digital marketplace like OpenBazaar to let people earn LBC which they then can spend with content creators on the platform.
Otherwise, a largely positive reaction from here.
Addenda
If you’ve enjoyed what you’ve seen here today and want to support my continued efforts to bring engineering, art, and the occasional philosophical divergence to the masses, please feel free to send me a tip. Or thousands of dollars, I’m really not that picky. It’s through your efforts that content like this gets created.
Check out my portfolio if you'd like to get in touch or see more of my work.
- PAYPAL: http://paypal.me/SquidLord
- CASH APP: https://cash.app/$SquidInks
- BITCOIN: bc1qy9wkj3g6ueugu0tjuwmh6dfxxn590r3p9tj9qt
- ETHEREUM: 0xcca4C3A3024d7496bDe06f6Fd865BfD5992Bf10B
- BAT: 0xcca4C3A3024d7496bDe06f6Fd865BfD5992Bf10B
- STELLAR: GBI6AMTGCBUPRBKWY3V246FDBZJILGOZHNOYWJXZNRBLACB6ZWAQTIYW
- ZCASH: t1dhcviaYgQuX4z91jhkjHrmkWNf5agJCNc
- MONERO: 48JHK6G4FyvSktU45t5ra7SALwFJAMEebEShPmk1K7xd95p4mYNJMnRZ27JBmcc97VhrJqEDmCyHtVP8iDTeNAJE1VnaCN8
- PO.ET: 0xcca4C3A3024d7496bDe06f6Fd865BfD5992Bf10B
- TRON: THKkzVBLz4V14dmvpDxkag8piKuJ74aATa
- XRP: rHvFZbRtsW2975zQXYTHDz7VAXPSdozoEs
- LBRY: https://open.lbry.com/@LexTenebris:9
- STEEM: @lextenebris
- HIVE: @lextenebris
Congratulations @lextenebris! You have completed the following achievement on the Hive blockchain and have been rewarded with new badge(s) :
You can view your badges on your board And compare to others on the Ranking
If you no longer want to receive notifications, reply to this comment with the word
STOPDo not miss the last post from @hivebuzz:
Support the HiveBuzz project. Vote for our proposal!