What if we let the registrars set the LTM price?

[abit]

So my current standing is that LTM/AM (and/or shorter membership schemes, e.g. monthly, if they get introduced) are the only subject of this proposal.
As for other fees - it could be done but I don't see a reason why we would need it.

Why don't we make it fully consistence at this point?

Let's have the committe decide the basic_fee for every operation.

Then let's allow the registrar to enter a % value that will be added on top of all basic fee.

E.g.
network basic fee:
   -transfer 10 bts
   -account_upgrade 10k bts

Registrar jakub choose 75% for his "costumers" jakub-cost

jakub-cost will pay:
   -transfer 17.5 bts
   -account_upgrade 17.5k bts

Jakub will take the %cut he charged:
   -transfer 7.5 bts
   -account_upgrade 7.5k bts

And where would this registrar's pricing policy be stored?
Any account can be a registrar. So you would need to allow every account to define the pricing policy in case this account becomes a registrar.
It becomes very complex IMO.

Technically, the "policy" can be stored in the faucet which is running by the registrar; then the parameters can be saved to every account registered, just what we're doing right now.

It makes no change to the result. Either competitors competing on a same base price and different on-the-top margins, or on a same end-user price and different discounts.

[Empirical1.2]

My point is that a user who paid the full price will find out the day after he could have gotten the same thing for free. That is the user experience you are advocating.

In the same way my UX is affected when I buy a pair of nice sneakers from a local online shop and three days later I find out that I could've bought them on Alibaba for half the price.
This is the way things are nowadays and people are getting used to it.
Smart people who have the time to hunt for discounts get lower prices than people who don't have time for that.
Take airline tickets as an example.

Yeah, this is just a fact of life.  One could come up with a zillion examples.

When Nike (Sneaker company) wholesales their product to retailers, their marketing costs are included in that wholesale price. (Advertising and Sales commission.) Retailers then compete on price/location/other for the end user.

In the BTS model, the referral programme is a form of sales commission currently included in the wholesale price. This creates a level playing field for your global sales force.

By removing this you lower the wholesale price which is beneficial to some end user businesses but removes the incentive for a large part of your sales force.

Even some end user businesses that engage in BTS marketing like CCEDK will also be less incentivized & have less profit margin to do so because they will have to compete on price with other providers

As mentioned there is a loophole that makes this possible anyway but it does limit BTS's ability to market itself, other than by relying on low costs and word of mouth.

[puppies]

Couldn't registrars or referrers just give a refund out to purchasers of LTM or AM?  Is there a valid reason to go beyond this?  I don't currently see one.

[abit]

Could we have an estimation about time/cost for these changes?
IMO, they are not so "radical" at the end of the day, and would not cost very much, but I would like to hear abit's opinion that would be much more correct than mine. @abit

We need to hear it from abit.

I feel it would be a lot of work. Sorry right now I'm unable to make a detailed estimation.

It gives the registrar/referrer some conveniences and lower barriers, but the result is no change if let committee but not registrars set the lower limits (and I am against the idea that will give out zero cost AM to new users).

[jakub]

So my current standing is that LTM/AM (and/or shorter membership schemes, e.g. monthly, if they get introduced) are the only subject of this proposal.
As for other fees - it could be done but I don't see a reason why we would need it.

Why don't we make it fully consistence at this point?

Let's have the committe decide the basic_fee for every operation.

Then let's allow the registrar to enter a % value that will be added on top of all basic fee.

E.g.
network basic fee:
   -transfer 10 bts
   -account_upgrade 10k bts

Registrar jakub choose 75% for his "costumers" jakub-cost

jakub-cost will pay:
   -transfer 17.5 bts
   -account_upgrade 17.5k bts

Jakub will take the %cut he charged:
   -transfer 7.5 bts
   -account_upgrade 7.5k bts

And where would this registrar's pricing policy be stored?
Any account can be a registrar. So you would need to allow every account to define the pricing policy in case this account becomes a registrar.
It becomes very complex IMO.

[Bhuz]

So my current standing is that LTM/AM (and/or shorter membership schemes, e.g. monthly, if they get introduced) are the only subject of this proposal.
As for other fees - it could be done but I don't see a reason why we would need it.

Why don't we make it fully consistence at this point?

Let's have the committe decide the basic_fee for every operation.

Then let's allow the registrar to enter a % value that will be added on top of all basic fee.

E.g.
network basic fee:
   -transfer 10 bts
   -account_upgrade 10k bts

Registrar jakub choose 75% for his "costumers" jakub-cost

jakub-cost will pay:
   -transfer 17.5 bts
   -account_upgrade 17.5k bts

Jakub will take the %cut he charged:
   -transfer 7.5 bts
   -account_upgrade 7.5k bts

[jakub]

You should realize that IF "the committee manages to push through flat transfer fees at around $0.02" does it really MEANS the shareholders push it through with their votes.
So, IF the shareholders do not want to save the baby, it is not on the Committee.

I agree with that, I could have worded it more precisely.
The only reason I expressed it in this way, is because I'm afraid it might happen that "the baby's killing" gets bundled with other valuable fee adjustments and thus obfuscate the choice for the shareholders.

Is it correct that the shareholders/network/committee would end up setting the basic/lower_limit fee, and each registrars would *add* another fee *on top of the basic one*?

That's correct.
But please note that the main purpose of this idea is this: in case of AM (and/or shorter subscriptions) to allow setting this lower limit at a level close to zero.
This would allow some registrars to give it away for free *if* they really needed this for their customers (no income lost for the network, as those customers would not have bought AM anyway).

Is this limited to LTM-fee, or it would work for *every fee* on the platform?

Initially I meant it for LTM only.
Then I realized that AM (Annual Membership) and also shorter membership schemes would do the same job even better.
So my current standing is that LTM/AM (and/or shorter membership schemes, e.g. monthly, if they get introduced) are the only subject of this proposal.
As for other fees - it could be done but I don't see a reason why we would need it.

Could we have an estimation about time/cost for these changes?
IMO, they are not so "radical" at the end of the day, and would not cost very much, but I would like to hear abit's opinion that would be much more correct than mine. @abit

We need to hear it from abit.

[Bhuz]

So is this the conclusion?
* committee can set a bottom for LTM/AM and whatever short period memberships, which is paid to network
* we'll still have a built-in referral program, any registrars can set their own % of cut on top of basic network fee, they can set their own % of cut to referrers.
* no vesting on cash-backs.

I'll support it.

Yes, that's a very good summary.
I'd be happy to make a BSIP out of it but the current situation with the committee does not encourage me to do so.

If the committee manages to push through flat transfer fees at around $0.02, I cannot see much need for this concept.
My goal was to offer a sensible alternative which allows us to save the baby (i.e. keep flat transfer fees at the current level).
When the baby is killed, there is little need to have the alternative.

You should realize that IF "the committee manages to push through flat transfer fees at around $0.02" does it really MEANS the shareholders push it through with their votes.
So, IF the shareholders do not want to save the baby, it is not on the Committee.


If I correctly understood the proposal, I would support it.

Is it correct that the shareholders/network/committee would end up setting the basic/lower_limit fee, and each registrars would *add* another fee *on top of the basic one*?

Is this limited to LTM-fee, or it would work for *every fee* on the platform?


Could we have an astimation about time/cost for these changes?
IMO, they are not so "radical" at the end of the day, and would not cost very much, but I would like to hear abit's opinion that would be much more correct than mine. @abit

[jakub]

It's actually not exactly what we have now.  What we have now is a mandatory $100 fee for LTM.  If the business wants to opt out (fully or partially), they have to wait for the 80% share to vest and then split those funds up and send a portion to the referrer and a portion back to the user (or they can reimburse the user immediately and pay themselves back after vesting). 

It's quite inconvenient in terms of technical execution and vesting but nevertheless even now we have this: a registrar can sell LTM for any price between $20 and $100 and offer the referrer as a reward for her efforts anything between $0 to $80.

Also, as it stands, the lifetime membership and referrer program are mashed together and some people can't wrap their heads around giving 80% of the LTM fee to a referrer and only 20% to the network.  That does seem a little odd and I never liked it myself (mostly because it obfuscates the actual costs of each component).  Instead, the proposal I described, the LTM fee is the LTM fee and the network collects it no matter what.  Then, separately, there's the referral program that is optional and it's up to the business to charge their user the additional fee or not.  Doing it this way, no one should have any problem with 100% of that fee being used by the referral program.   So I think what we currently have is totally different than what I was proposing (which was actually just a combination of what you originally proposed and what Xeldal proposed). 

So what you propose is this: arrange things is such a way that there is a clear picture for the user who gets what.
You do not propose to change the financial situation of any party involved (i.e. the network, the user, the referrer and the registrar) but just want to make it clear, simple & easy.
Is it correct?

[abit]

So is this the conclusion?
* committee can set a bottom for LTM/AM and whatever short period memberships, which is paid to network
* we'll still have a built-in referral program, any registrars can set their own % of cut on top of basic network fee, they can set their own % of cut to referrers.
* no vesting on cash-backs.

I'll support it.

Yes, that's a very good summary.
I'd be happy to make a BSIP out of it but the current situation with the committee does not encourage me to do so.

If the committee manages to push through flat transfer fees at around $0.02, I cannot see much need for this concept.
My goal was to offer a sensible alternative which allows us to save the baby (i.e. keep flat transfer fees at the current level).
When the baby is killed, there is little need to have the alternative.

I'd rather wait for the committee to publish their proposal first, which is said to be done in next week.
An good pricing strategy is important for the platform.

[jakub]

So is this the conclusion?
* committee can set a bottom for LTM/AM and whatever short period memberships, which is paid to network
* we'll still have a built-in referral program, any registrars can set their own % of cut on top of basic network fee, they can set their own % of cut to referrers.
* no vesting on cash-backs.

I'll support it.

Yes, that's a very good summary.
I'd be happy to make a BSIP out of it but the current situation with the committee does not encourage me to do so.

If the committee manages to push through flat transfer fees at around $0.02, I cannot see much need for this concept.
My goal was to offer a sensible alternative which allows us to save the baby (i.e. keep flat transfer fees at the current level).
When the baby is killed, there is little need to have the alternative.

[abit]

So is this the conclusion?
* committee can set a bottom for LTM/AM and whatever short period memberships, which is paid to network
* we'll still have a built-in referral program, any registrars can set their own % of cut on top of basic network fee, they can set their own % of cut to referrers.
* no vesting on cash-backs.

I'll support it.

//Update:
After more thinking, now I tend to leave the design of vesting period on large amount of cash-backs unchanged.
So I will still support the 1st and 2nd proposals above, but against the 3rd.

[tbone]

@jakub: I think you're starting to over-complicate it now with AM vs. LTM.  What's this LTM loophole you're concerned about?

Why do you say it's more complicated now?
It's the same idea but I now propose to use AM instead of LTM.
Nothing else changes. Just replace LTM with AM.

The loophole is this: if a user buys LTM (for whatever price) she can then generate other accounts by making herself the registrar and setting the price for upgrading to 0.
This way she could end up with 100 (or more) LTM accounts and only 20% (of the price she paid for the first one) has gone to the network. The network gets nothing from the other 99 accounts.

With AM this situation is not possible as AM does not give you the possibility to create further accounts (as far as I know).

But I thought you agreed with Xeldal's proposal to enforce a $20 fee to the network for every LTM account created.  Wouldn't that solve the problem you're talking about?  So now, by essentially separating the network fee from any referral fee, we can have the optional referral program you were proposing. 

To elaborate, the $20 upfront fee is what the network gets in exchange for offering lower fees and losing its cut of the otherwise higher fees.  So that is mandatory (until we're ready and able to lower fees to $0 per today's mumble).  Beyond the network's mandatory $20, businesses can charge ANY additional amount IF they wish to utilize the built-in referral program.  Businesses that utilize the referral program will attract referrers who would get paid for distributing a URL that points to the wallet of that business.  Businesses that do not use the program will not attract such referrers.  So everyone gets what they need.  Am I missing  something?

What you described above is exactly what we have now.
Currently OpenLedger has the option to offer a refund:
- up to $12 for every AM bought, thus effectively selling it at $4 (as that's the minimum price required by the network for AM).
- up to $80 for every LTM bought, thus effectively selling it at $20 (as that's the minimum price required by the network for LTM).
Obviously OL is not doing this for a very simple reason - there is no competition on the hosted wallet market.

The core of my proposal is to push it a bit further in case of AM: to allow OL (and similar business) to effectively sell AM for $0 (so *no* minimum price would be required by the network when AM is sold).

I appreciate Xeldal's proposal *despite* him enforcing the $20 minimum price for LTM (or $4 minimum price in case of AM) because in Xeldal's world a business can access the lowest prices without using AM/LTM altogether. So in this respect, our proposals are different but the end result is very similar, both for businesses and the network.

My only concern with his proposal is that his model relies on the GUI to calculate the right fee. Anybody who can hack the GUI, can game the hosted wallet business and pay only the minimum fees required by the network. That's a big downside of his approach.

It's actually not exactly what we have now.  What we have now is a mandatory $100 fee for LTM.  If the business wants to opt out (fully or partially), they have to wait for the 80% share to vest and then split those funds up and send a portion to the referrer and a portion back to the user (or they can reimburse the user immediately and pay themselves back after vesting).  The proposal I described above would allow businesses to determine their level of participation and then let the blockchain handle the rest, so the user is charged the right amount (if anything) from the start and the referrer also gets their proper share, if applicable. 

Also, as it stands, the lifetime membership and referrer program are mashed together and some people can't wrap their heads around giving 80% of the LTM fee to a referrer and only 20% to the network.  That does seem a little odd and I never liked it myself (mostly because it obfuscates the actual costs of each component).  Instead, the proposal I described, the LTM fee is the LTM fee and the network collects it no matter what.  Then, separately, there's the referral program that is optional and it's up to the business to charge their user the additional fee or not.  Doing it this way, no one should have any problem with 100% of that fee being used by the referral program.   So I think what we currently have is totally different than what I was proposing (which was actually just a combination of what you originally proposed and what Xeldal proposed). 

See what I mean?

[jakub]

@jakub: I think you're starting to over-complicate it now with AM vs. LTM.  What's this LTM loophole you're concerned about?

Why do you say it's more complicated now?
It's the same idea but I now propose to use AM instead of LTM.
Nothing else changes. Just replace LTM with AM.

The loophole is this: if a user buys LTM (for whatever price) she can then generate other accounts by making herself the registrar and setting the price for upgrading to 0.
This way she could end up with 100 (or more) LTM accounts and only 20% (of the price she paid for the first one) has gone to the network. The network gets nothing from the other 99 accounts.

With AM this situation is not possible as AM does not give you the possibility to create further accounts (as far as I know).

But I thought you agreed with Xeldal's proposal to enforce a $20 fee to the network for every LTM account created.  Wouldn't that solve the problem you're talking about?  So now, by essentially separating the network fee from any referral fee, we can have the optional referral program you were proposing. 

To elaborate, the $20 upfront fee is what the network gets in exchange for offering lower fees and losing its cut of the otherwise higher fees.  So that is mandatory (until we're ready and able to lower fees to $0 per today's mumble).  Beyond the network's mandatory $20, businesses can charge ANY additional amount IF they wish to utilize the built-in referral program.  Businesses that utilize the referral program will attract referrers who would get paid for distributing a URL that points to the wallet of that business.  Businesses that do not use the program will not attract such referrers.  So everyone gets what they need.  Am I missing  something?

What you described above is exactly what we have now.
Currently OpenLedger has the option to offer a refund:
- up to $12 for every AM bought, thus effectively selling it at $4 (as that's the minimum price required by the network for AM).
- up to $80 for every LTM bought, thus effectively selling it at $20 (as that's the minimum price required by the network for LTM).
Obviously OL is not doing this for a very simple reason - there is no competition on the hosted wallet market.

The core of my proposal is to push it a bit further in case of AM: to allow OL (and similar business) to effectively sell AM for $0 (so *no* minimum price would be required by the network when AM is sold).

I appreciate Xeldal's proposal *despite* him enforcing the $20 minimum price for LTM (or $4 minimum price in case of AM) because in Xeldal's world a business can access the lowest prices without using AM/LTM altogether. So in this respect, our proposals are different but the end result is very similar, both for businesses and the network.

My only concern with his proposal is that his model relies on the GUI to calculate the right fee. Anybody who can hack the GUI, can game the hosted wallet business and pay only the minimum fees required by the network. That's a big downside of his approach.

[tbone]

@jakub: I think you're starting to over-complicate it now with AM vs. LTM.  What's this LTM loophole you're concerned about?

Why do you say it's more complicated now?
It's the same idea but I now propose to use AM instead of LTM.
Nothing else changes. Just replace LTM with AM.

The loophole is this: if a user buys LTM (for whatever price) she can then generate other accounts by making herself the registrar and setting the price for upgrading to 0.
This way she could end up with 100 (or more) LTM accounts and only 20% (of the price she paid for the first one) has gone to the network. The network gets nothing from the other 99 accounts.

With AM this situation is not possible as AM does not give you the possibility to create further accounts (as far as I know).

But I thought you agreed with Xeldal's proposal to enforce a $20 fee to the network for every LTM account created.  Wouldn't that solve the problem you're talking about?  So now, by essentially separating the network fee from any referral fee, we can have the optional referral program you were proposing. 

To elaborate, the $20 upfront fee is what the network gets in exchange for offering lower fees and losing its cut of the otherwise higher fees.  So that is mandatory (until we're ready and able to lower fees to $0 per today's mumble).  Beyond the network's mandatory $20, businesses can charge ANY additional amount IF they wish to utilize the built-in referral program.  Businesses that utilize the referral program will attract referrers who would get paid for distributing a URL that points to the wallet of that business.  Businesses that do not use the program will not attract such referrers.  So everyone gets what they need.  Am I missing  something?

[jakub]

@jakub: I think you're starting to over-complicate it now with AM vs. LTM.  What's this LTM loophole you're concerned about?

Why do you say it's more complicated now?
It's the same idea but I now propose to use AM instead of LTM.
Nothing else changes. Just replace LTM with AM.

The loophole is this: if a user buys LTM (for whatever price) she can then generate other accounts by making herself the registrar and setting the price for upgrading to 0.
This way she could end up with 100 (or more) LTM accounts and only 20% (of the price she paid for the first one) has gone to the network. The network gets nothing from the other 99 accounts.

With AM this situation is not possible as AM does not give you the possibility to create further accounts (as far as I know).

[tbone]

@jakub: I think you're starting to over-complicate it now with AM vs. LTM.  What's this LTM loophole you're concerned about?

[jakub]

After discussing it with @Xeldal , I now think his solution is the ultimate solution for BitShares in the long run.

Xeldal's proposal turns upside down the whole referral program:
- the current approach is this: we charge high fees by default and then offer discounts.
- his approach is this: we charge low fees by default and facilitate adding profit margins for businesses on top of that.

Xeldal's proposal looks very elegant and efficient.
However, it has one fundamental difficulty: IMO it will require a major code-base upgrade (possibly similar to the migration from BitShares 0.9.x to BitShares 2.0).
(I'm not an expert in this area, so this is just my speculation, that needs to be confirmed by someone more competent.)

Whereas my solution achieves a similar result for both businesses and users, but requires only subtle changes in the code-base, so it should be doable within a simple worker proposal.
The downside is this: it's not as elegant and generic as Xeldal's proposal.

Therefore I suggest using my solution as a quick patch to the referral program. In the long run, however, we should aim at Xeldal's proposal IMO.

One more thing: this proposal has been revised and no longer LTM is involved in it.
I now think that LTM should stay as it is, as the same goal can be achieved with AM (Annual Membership) and/or shorter memberships.

AM (and/or any other shorter membership scheme) is better for our purpose than LTM due to these reasons:
- AM does not allow users to create subsequent accounts (this would be a loophole if LTM was used).
- AM only offers what we actually need for our purpose: a price discount. This way we do not end up giving away access to advanced features when it's not needed.
- AM costs significantly less, so it is easier to sell it at a relatively good price.
- *if* there are any cases of lost revenue for the network, they will be on a much smaller scale.

I do maintain my position that a revenue can be considered as lost *only* when a user ends up paying less than s/he was willing to pay.
I don't think there will be any significant number of such cases, at least until competition on the hosted wallet market grows substantially, as only strong competition can push the AM to a level that is lower than the level users are willing to pay.
So until that happens, we are sure to preserve the network's income or increase it.

[abit]

EDIT: as @abit and @EstefanTT have noticed, the same goal could be reached by using the Annual Membership (AM) instead of LTM.
So if this idea is too radical for you because it involves LTM, feel free to replace LTM with AM.
The main concept remains the same - let the registrars decide how much AM costs on their platform, in a similar way that they decide how much profit they share with referrers.

I think I don't want to change LTM nor AM as they are too sensitive, instead, I'd like to introduce monthly, weekly and even daily memberships. Maybe I'll set a bottom price for selling these memberships, and let registrar or whoever want to sell them decide how to sell and/or what price should they sell, to bring more competitions and not drive all the prices to zero.

I'm afraid if we go below AM it defeats the purpose of my proposal.
If I understand it correctly, your proposal means that bitcrab would have to renew the memberships of his customers on a monthly, weekly and even daily basis.
That's a lot to work to manage all this stuff for all your customers. I guess he would not want to be bothered with that.

Unless I'm missing something?

I'm not saying only bitcrab can do it. Let all LTMs or maybe AMs to be able to do it in an easier way. Give people some free money. Decentralize it. When a lot of people earn small money, the network will earn big money.

So you suggest expanding the concept to all possible memberships?
Why not do AM and anything shorter than AM, and leave LTM untouched.
We don't really need the advanced features LTM offers, so AM (plus anything shorter) will do the job.

You can check our transaction history to see how many people upgraded to LTM, and how many upgraded to AM. IMO AM is too long for an average Joe to consider, it's hard to sell.

[jakub]

EDIT: as @abit and @EstefanTT have noticed, the same goal could be reached by using the Annual Membership (AM) instead of LTM.
So if this idea is too radical for you because it involves LTM, feel free to replace LTM with AM.
The main concept remains the same - let the registrars decide how much AM costs on their platform, in a similar way that they decide how much profit they share with referrers.

I think I don't want to change LTM nor AM as they are too sensitive, instead, I'd like to introduce monthly, weekly and even daily memberships. Maybe I'll set a bottom price for selling these memberships, and let registrar or whoever want to sell them decide how to sell and/or what price should they sell, to bring more competitions and not drive all the prices to zero.

I'm afraid if we go below AM it defeats the purpose of my proposal.
If I understand it correctly, your proposal means that bitcrab would have to renew the memberships of his customers on a monthly, weekly and even daily basis.
That's a lot to work to manage all this stuff for all your customers. I guess he would not want to be bothered with that.

Unless I'm missing something?

I'm not saying only bitcrab can do it. Let all LTMs or maybe AMs to be able to do it in an easier way. Give people some free money. Decentralize it. When a lot of people earn small money, the network will earn big money.

So you suggest expanding the concept to all possible memberships?
Why not do AM and anything shorter than AM, and leave LTM untouched.
We don't really need the advanced features LTM offers, so AM (plus anything shorter) will do the job.

[abit]

EDIT: as @abit and @EstefanTT have noticed, the same goal could be reached by using the Annual Membership (AM) instead of LTM.
So if this idea is too radical for you because it involves LTM, feel free to replace LTM with AM.
The main concept remains the same - let the registrars decide how much AM costs on their platform, in a similar way that they decide how much profit they share with referrers.

I think I don't want to change LTM nor AM as they are too sensitive, instead, I'd like to introduce monthly, weekly and even daily memberships. Maybe I'll set a bottom price for selling these memberships, and let registrar or whoever want to sell them decide how to sell and/or what price should they sell, to bring more competitions and not drive all the prices to zero.

I'm afraid if we go below AM it defeats the purpose of my proposal.
If I understand it correctly, your proposal means that bitcrab would have to renew the memberships of his customers on a monthly, weekly and even daily basis.
That's a lot to work to manage all this stuff for all your customers. I guess he would not want to be bothered with that.

Unless I'm missing something?

I'm not saying only bitcrab can do it. Let all LTMs or maybe AMs to be able to do it in an easier way. Give people some free money. Decentralize it. When a lot of people earn small money, the network will earn big money.

[jakub]

EDIT: as @abit and @EstefanTT have noticed, the same goal could be reached by using the Annual Membership (AM) instead of LTM.
So if this idea is too radical for you because it involves LTM, feel free to replace LTM with AM.
The main concept remains the same - let the registrars decide how much AM costs on their platform, in a similar way that they decide how much profit they share with referrers.

I think I don't want to change LTM nor AM as they are too sensitive, instead, I'd like to introduce monthly, weekly and even daily memberships. Maybe I'll set a bottom price for selling these memberships, and let registrar or whoever want to sell them decide how to sell and/or what price should they sell, to bring more competitions and not drive all the prices to zero.

I'm afraid if we go below AM it defeats the purpose of my proposal.
If I understand it correctly, your proposal means that bitcrab would have to renew the memberships of his customers on a monthly, weekly and even daily basis.
That's a lot to work to manage all this stuff for all your customers. I guess he would not want to be bothered with that.

Unless I'm missing something?

[abit]

EDIT: as @abit and @EstefanTT have noticed, the same goal could be reached by using the Annual Membership (AM) instead of LTM.
So if this idea is too radical for you because it involves LTM, feel free to replace LTM with AM.
The main concept remains the same - let the registrars decide how much AM costs on their platform, in a similar way that they decide how much profit they share with referrers.

I think I don't want to change LTM nor AM as they are too sensitive, instead, I'd like to introduce monthly, weekly and even daily memberships. Maybe I'll set a bottom price for selling these memberships, and let registrar or whoever want to sell them decide how to sell and/or what price should they sell, to bring more competitions and not drive all the prices to zero.

[clayop]

Why he's not a businessman in this case? Do you think the CEO of Alipay is not a businessman too?

My point is this: if he's a businessman (as I'm sure he is), he'll sell LTM for zero only to those customers who would not have bought LTM in the first place.
So no network's income is actually lost. What is lost is *theoretical* income that would not have materialized anyway.

And your approach is based on the same assumption. LTM fee decrease -> more demand for LTM / Transfer fee decrease -> more demand for transfer

I don't understand why you associate more demand for LTM with transfer fee decrease.

I meant that you argues LTM fee decrease -> more demand, while you opposites transfer fee decrease -> more demand for transfer. It's ironical.

And I get your point - price differentiation. But it cannot be applied when the product is exactly same. If you can get LTM for free, why would you like to pay for it? There can be some reasons, such as informational barrier, but I believe people will move to zero cost registrar quickly.

[jakub]

We can leave LTM fee unchanged, let it be the key to advanced features.

In the meanwhile, we give annual members more discount on basic features(it's currently 0% due to a bug), and give registrar a choice of how much they charge for it.

//Update: maybe we can provide more choices: not only annual membership, but also monthly, weekly even daily membership (for example for trials). When the users found that membership is good, they'll buy when needed.

Hi, anyone noticed my comments above?
I think short-term membership is more attractive and better for competitions .
Better if users can buy membership from any LTM's, but the seller are always need to split some to registrar/original referrer.
If the price not be driven to zero, all parties will happy.

//Update: I think it's OK to leave LTM and annual membership fees unchanged. Play with shorter period of memberships.

@abit , I've given it some thought and I think I agree with you.
We could leave LTM untouched and reach our goal with AM as all we need is the 80% discount for transfer fees, not the advanced features associated with LTM.

And if it's doable, we could also introduce shorter period of membership as a marketing tool.

But the key thing for me would be to fully transfer the control over the AM price to registrars.
This way the transfer fee battle could be finally resolved without pushing the flat transfer fee extremely low.

[jakub]

Why he's not a businessman in this case? Do you think the CEO of Alipay is not a businessman too?

My point is this: if he's a businessman (as I'm sure he is), he'll sell LTM for zero only to those customers who would not have bought LTM in the first place.
So no network's income is actually lost. What is lost is *theoretical* income that would not have materialized anyway.

And your approach is based on the same assumption. LTM fee decrease -> more demand for LTM / Transfer fee decrease -> more demand for transfer

I don't understand why you associate more demand for LTM with transfer fee decrease.

[jakub]

They are given full control and therefore full ownership and therefore full liability.

Full ownership of what?
I don't think you've understood my proposal.

The registrars already decide how much of their profits they share with the referrers.
All I propose is to give them one more option: decide how much LTM costs.

They do not get any ownership of anything.

The current situation gives registrars no control over the network..

I don't give them any control over the network.
I just let them choose a pricing strategy that suits their customers.

They own price control. Control = ownership.

Their profit margin is set by how much they make on the services they are 'selling' at a margin.

'Their customers' not Bitshares users.

Right now I can sell you LTM for $25 or $50 or even $120.
Do I control the price?

[clayop]

If you say he'll sell LTM for zero to customers who could have bought it for more than the current minimum price of $20 - well, you are right, the network would have lost some revenue.
But that would also mean that bitcrab is not a businessman.

Why he's not a businessman in this case? Do you think the CEO of Alipay is not a businessman too?

And your approach is based on the same assumption. LTM fee decrease -> more demand for LTM / Transfer fee decrease -> more demand for transfer

[jakub]

He is most likely to sell ltm at $0. In this case how can you compensate the network income?

If you say he'll sell LTM for zero to customers who could have bought it for more than the current minimum price of $20 - well, you are right, the network would have lost some revenue.
But that would also mean that bitcrab is not a businessman.

[clayop]

The biggest problem that I have with this proposal, is that you are proposing to give a way one of the biggest sources of revenue of the network it self. It's not just the referral program that makes money out of it, but also the network itself. How would you compensate for that?

I think this is a misconception.
I'm not giving away "the biggest sources of revenue of the network".

Actually the opposite is achieved here: I'm trying to maximize for the network the revenue stream from sold LTMs .

What scenario would you choose?
- bitcrab treats the referral program as a "trouble maker" and sells 20 LTMs a year. Network's income: 20 * $100 * 20% = $400
- bitcrab embraces the referral program and sells 2000 LTMs a year (some of them for free but with an average price of $10). Network's income: 2000 * $10 * 20% = $4000

bitcrab will do his best not to give LTM for free, if he feels some of his customers could pay something for it.
So he'll most probably give away LTM for free only to those who would not buy LTM anyway.

He is most likely to sell ltm at $0. In this case how can you compensate the network income?

[BunkerChainLabs-DataSecurityNode]

They are given full control and therefore full ownership and therefore full liability.

Full ownership of what?
I don't think you've understood my proposal.

The registrars already decide how much of their profits they share with the referrers.
All I propose is to give them one more option: decide how much LTM costs.

They do not get any ownership of anything.

The current situation gives registrars no control over the network..

I don't give them any control over the network.
I just let them choose a pricing strategy that suits their customers.

They own price control. Control = ownership.

Their profit margin is set by how much they make on the services they are 'selling' at a margin.

'Their customers' not Bitshares users.

[jakub]

They are given full control and therefore full ownership and therefore full liability.

Full ownership of what?
I don't think you've understood my proposal.

The registrars already decide how much of their profits they share with the referrers.
All I propose is to give them one more option: decide how much LTM costs.

They do not get any ownership of anything.

The current situation gives registrars no control over the network..

I don't give them any control over the network.
I just let them choose a pricing strategy that suits their customers.

[BunkerChainLabs-DataSecurityNode]

I think this structure change would put a HUGE amount of liability all on the registrar.

@BunkerChain Labs
Would you elaborate on this?
How can freeing up the LTM price "put HUGE amount of liability on the registrar"?

They are given full control and therefore full ownership and therefore full liability.

The current situation gives registrars no control over the network.. they are not the ones making any decisions other than what cut of the refer they are giving for providing a service to those that use their wallet. Otherwise they have zero involvement with the account holders and what is paid for because it is being handed by a decentralized body called a committee.

In your scenario.. they are front and center and are setting the price on everything, and therefore are sell all services. Therefore they are in the money transmitter and/or exchange business. It effectively gives every regulator a bullseye to take aim at. Our current structure does not.

[jakub]

I expect the fees taken by registrars to get very low longterm. Whithout a minimum fee for the network that would make the whole LTM thing pretty much obsolete.

You could be right *IF* the LTM reached the minimum price (i.e. $20) already.
But the opposite has happened: after more than 3 months, not a single business has tried to undermine OpenLeger's price of $100.

So there are no reasons to think we are even close to the possibility you describe.

[jakub]

I think this structure change would put a HUGE amount of liability all on the registrar.

@BunkerChain Labs
Would you elaborate on this?
How can freeing up the LTM price "put HUGE amount of liability on the registrar"?

[jakub]

The biggest problem that I have with this proposal, is that you are proposing to give a way one of the biggest sources of revenue of the network it self. It's not just the referral program that makes money out of it, but also the network itself. How would you compensate for that?

I think this is a misconception.
I'm not giving away "the biggest sources of revenue of the network".

Actually the opposite is achieved here: I'm trying to maximize for the network the revenue stream from sold LTMs .

What scenario would you choose?
- bitcrab treats the referral program as a "trouble maker" and sells 20 LTMs a year. Network's income: 20 * $100 * 20% = $400
- bitcrab embraces the referral program and sells 2000 LTMs a year (some of them for free but with an average price of $10). Network's income: 2000 * $10 * 20% = $4000

bitcrab will do his best not to give LTM for free, if he feels some of his customers could pay something for it.
So he'll most probably give away LTM for free only to those who would not buy LTM anyway.

[abit]

We can leave LTM fee unchanged, let it be the key to advanced features.

In the meanwhile, we give annual members more discount on basic features(it's currently 0% due to a bug), and give registrar a choice of how much they charge for it.

//Update: maybe we can provide more choices: not only annual membership, but also monthly, weekly even daily membership (for example for trials). When the users found that membership is good, they'll buy when needed.

Hi, anyone noticed my comments above?
I think short-term membership is more attractive and better for competitions .
Better if users can buy membership from any LTM's, but the seller are always need to split some to registrar/original referrer.
If the price not be driven to zero, all parties will happy.

//Update: I think it's OK to leave LTM and annual membership fees unchanged. Play with shorter period of memberships.

[Frodo]

As I stated before, we already have this problem.
I could "ruin" the OpenLedger business by offering 75% discount on LTM for all OpenLedger customers.

So let's face this problem and embrace the good side of it.

That is actually not a "problem" but the reason to start a competition.

The biggest problem that I have with this proposal, is that you are proposing to give a way one of the biggest sources of revenue of the network it self. It's not just the referral program that makes money out of it, but also the network itself. How would you compensate for that?

 

Just because we don't have a lot of competition right now doesn't mean it will be like that forever.
I expect the fees taken by registrars to get very low longterm. Whithout a minimum fee for the network that would make the whole LTM thing pretty much obsolete.

[BunkerChainLabs-DataSecurityNode]

I think this structure change would put a HUGE amount of liability all on the registrar.

They have gone from being just a hosted wallet to a decentralized network to being the owner/gateway of a network with a decentralized backend.

Because you have placed full control in the hands of the registrar they will effectively be viewed as money transmitters and will be subject to all applicable laws and regulations to this effect within the jurisdiction they are operating.

For this reason alone, I think the structure would be a detriment to the Bitshares network and would reduce availability of hosted wallet providers.

[xeroc]

As I stated before, we already have this problem.
I could "ruin" the OpenLedger business by offering 75% discount on LTM for all OpenLedger customers.

So let's face this problem and embrace the good side of it.

That is actually not a "problem" but the reason to start a competition.

The biggest problem that I have with this proposal, is that you are proposing to give a way one of the biggest sources of revenue of the network it self. It's not just the referral program that makes money out of it, but also the network itself. How would you compensate for that?

[jakub]

Why would I as a registrar not offer a LTM to my customers for free? Who prevents me from doing that?

As I stated before, we already have this problem.
I could "ruin" the OpenLedger business by offering 75% discount on LTM for all OpenLedger customers.

So let's face this problem and embrace the good side of it.

EDIT: It would be much smarter for me not to ruin OpenLedger but instead open a similar business and offer LTM with a 10% discount.
This way I would earn much more and the whole BitShares network would grow.

[EstefanTT]

Won't we upset our most loyal bts members who paid 100$ for them LTM ??

But the point is we have this problem already. If you give me a public key of your choice I can register a LTM account for you for just $25 (this is a serious offer).
Are you upset that you've paid $100 to OpenLedger? 

Would it be possible to do something similar but with "annual members" ? So LTM still have a valued product that anybody is giving for free and they also keep having part of the fee produce by the annual members and the basics accounts below ?

Absolutely, annual membership should be included in this solution.

It's just that it seems that we don't stop loosing loyal old members of this community with these kind of changes.
That said, I find the idea rather interesting and original ?

I wouldn't be upset as long as I perceive the long term benefit for BitShares but I not everybody is watching the far horizon of BitShares as I do. I'm not proposing any solution, just underlining a possible consequence.

I understand what you said about your proposition (25$). You are right !

I still feel that LTM and AM (annual) could be articulate in an interesting way with your proposal. I just don't have it very clear in my mind at the moment. I'll keep thinking about it. [emoji6]

[xeroc]

I might have got this wrong, this was my conclusion from this statement by bitcrab:

as discussed in committee, $0.018 is an acceptable flat fee.

Not sure what let him to post this, but the committee (at least those that are actively participating) have not agreed on $0.018 yet ..

[jakub]

And I've just realized that my offer to @bitcrab is actually much better than $0.018 he's managed to get from the committee.

He didn't actually get anything from the committee and the $0.18 is also not the current proposed transfer fee

I might have got this wrong, this was my conclusion from this statement by bitcrab:

as discussed in committee, $0.018 is an acceptable flat fee.

Anyway, if that's not the case, this only makes my offer to him more attractive.

[abit]

Won't we upset our most loyal bts members who paid 100$ for them LTM ??

Would it be possible to do something similar but with "annual members" ? So LTM still have a valued product that anybody is giving for free and they also keep having part of the fee produce by the annual members and the basics accounts below ?

This makes more sense for me.
We can leave LTM fee unchanged, let it be the key to advanced features.
In the meanwhile, we give annual members more discount on basic features(it's currently 0% due to a bug), and give registrar a choice of how much they charge for it.

//Update: maybe we can provide more choices: not only annual membership, but also monthly, weekly even daily membership (for example for trials). When the users found that membership is good, they'll buy when needed.

[xeroc]

Why would I as a registrar not offer a LTM to my customers for free? Who prevents me from doing that?

[xeroc]

And I've just realized that my offer to @bitcrab is actually much better than $0.018 he's managed to get from the committee.

He didn't actually get anything from the committee and the $0.18 is also not the current proposed transfer fee

[jakub]

Won't we upset our most loyal bts members who paid 100$ for them LTM ??

But the point is we have this problem already. If you give me a public key of your choice I can register a LTM account for you for just $25 (this is a serious offer).
Are you upset that you've paid $100 to OpenLedger? 

Would it be possible to do something similar but with "annual members" ? So LTM still have a valued product that anybody is giving for free and they also keep having part of the fee produce by the annual members and the basics accounts below ?

Absolutely, annual membership should be included in this solution.

[jakub]

And I've just realized that my offer to @bitcrab is actually much better than $0.018 he's managed to get from the committee.
If he chooses to give LTM to his customers for free, than they end up paying $.008 ($0.04 * 20%) per transfer (i.e. 50% less than what the committee offers).

Well, if this is not a generous offer, I don't know what is.

That sounds like a very reasonable compromise for all of us.  Although the final solution really should include %-based fees so that micro-transactions and non-payment applications (e.g. messaging, etc.) will be more feasible.  Do you have thoughts about %-based parameters that would yield similar revenue to the network compared with the $.018/.04 flat fee?

If this proposal gets some support and we settle for the flat transfer fee around $0.04, the minimum %-based limit seems quite straight forward to me: $0.008 (which equals the LTM discount: $0.04 * 20%).

As for the maximum limit and percentage rate - these are quite political decisions.
But I believe they should be chosen in such a way that for most issuers it will be a tough decision to make (this will be the case when the trade-off between flat and %-based scheme is balanced).

[abit]

I would be interested in seeing the reaction of some of our larger referral-dependent businesses, such as @ccedk .

I admit that CCEDK contributed much to the ecosystem, and I appreciate that.
However, I think ccedk currently has unfair advantages when competing with other registrars, since it's the only one registrar listed on http://bitshares.org. Also BM recently suggested that we need more faucets integrated in wallet.
Competitions will make the ecosystem better, right?

Jakub's proposal is very interesting. Right now I don't know if it's good, I need to think more.

[EstefanTT]

Won't we upset our most loyal bts members who paid 100$ for them LTM ??

Would it be possible to do something similar but with "annual members" ? So LTM still have a valued product that anybody is giving for free and they also keep having part of the fee produce by the annual members and the basics accounts below ?

[tbone]

And I've just realized that my offer to @bitcrab is actually much better than $0.018 he's managed to get from the committee.
If he chooses to give LTM to his customers for free, than they end up paying $.008 ($0.04 * 20%) per transfer (i.e. 50% less than what the committee offers).

Well, if this is not a generous offer, I don't know what is.

That sounds like a very reasonable compromise for all of us.  Although the final solution really should include %-based fees so that micro-transactions and non-payment applications (e.g. messaging, etc.) will be more feasible.  Do you have thoughts about %-based parameters that would yield similar revenue to the network compared with the $.018/.04 flat fee?

[jakub]

(A) The committee approach: reduce the transfer fee to a very low level of $0.018 (~5 BTS) and thus effectively remove the transfer fee incentive from the referral program.
Consequences for businesses:
For a blog/website business (e.g. fav or Max Wright):
- there is no incentive to operate as there is almost no income (unless they manage to target advanced users, which is unlikely)
For a UI business (e.g. ccedk, kenCode or merivercap):
- there is almost no income from the referral program (unless they manage to target advanced users, which is unlikely)
- they have only one option: construct their own mechanism of charging higher fees and, if they do that, they get 100% of the profit
- the lowest *flat* transfer fee they can offer to their customers is $0.018

Consequences for the network:
- there is no revenue from non-advanced users (EDIT: I should have said: almost no revenue. The network gets 20% of each transfer fee but as the fee is extremely small this income becomes negligible)
- the number of LTMs sold drops substantially (only advanced users have a reason to buy it)
- as a result, the network's revenue stream drops substantially, as LTM upgrades currently make around 90% of the network's revenue

Conclusions: (1) blog/website business are lost, (2) each UI business needs to construct its own charging mechanism to survive and (3) the network's income definitely drops.


(B) My approach: reduce the transfer fee to a relatively low level of $0.04 (~12 BTS) and keep at least some of the transfer fee incentive in the referral program. Also, allow registrars to sell LTM at any price they want, provided the network always keeps a 20% cut.
Consequences for businesses:
For a blog/website business (e.g. fav or Max Wright):
- there is still an incentive to operate as they can count on the referral income from non-advanced users
For a UI business (e.g. ccedk, kenCode or merivercap):
- they have a real choice: they can opt out from the referral program by offering LTM for free and instead construct their own mechanism of charging higher fees to keep 100% of the profit OR they can stick to the in-built scheme and sell LTM for any price they find suitable for their customers (the downside is this: they only keep 80% of the profit and they risk that some of their users will try to buy LTM cheaper elsewhere and then migrate to use their apps)
- the lowest *flat* transfer fee they can offer to their customers is $0.008 (i.e. 50% less than the committee approach)

Consequences for the network:
- revenue from non-advanced users is preserved, as the network always gets 20% whenever the in-built referral program is used for non-advanced users
- the number of LTMs sold increases substantially as LTM prices are perfectly matched to what each customer is willing to pay
- it's hard to predict the net effect: on the one hand there might be some revenue lost due to getting on average less than 4k BTS for each LTM sold, but on the other hand there will be more LTM sold and the income from transfer fees will be substantially higher

Conclusions: (1) blog/website business are preserved, (2) UI businesses have a *real* choice and (3) the network's income is most likely preserved or even increased.

[jakub]

And I've just realized that my offer to @bitcrab is actually much better than $0.018 he's managed to get from the committee.
If he chooses to give LTM to his customers for free, than they end up paying $.008 ($0.04 * 20%) per transfer (i.e. 50% less than what the committee offers).

Well, if this is not a generous offer, I don't know what is.

[tbone]

My point is that a user who paid the full price will find out the day after he could have gotten the same thing for free. That is the user experience you are advocating.

In the same way my UX is affected when I buy a pair of nice sneakers from a local online shop and three days later I find out that I could've bought them on Alibaba for half the price.
This is the way things are nowadays and people are getting used to it.
Smart people who have the time to hunt for discounts get lower prices than people who don't have time for that.
Take airline tickets as an example.

Yeah, this is just a fact of life.  One could come up with a zillion examples. 

[jakub]

Very interesting thought. Where is the "to the network" money going? Does the BitShares system need it for maintenance, etc.? Because if so, then this suggestion might set up a system where some users are paying for it and some are not.

Absolutely, the system needs funds for maintenance and development. And my proposal aims to maximize the income stream for the network.

My point is this: let's us not be afraid to get rid of the minimum price for LTM. As registrars have a huge 80% sales margin, there will be a natural incentive for them to set the LTM price at such a level that maximizes their profits (and also the network's profits) instead of driving the price to zero.

The net effect should be very positive for the network. No artificial protection (in the form of minimum LTM price) is needed, *if* the incentive structure is set right. The 80% sales margin takes care of this.

A better wrinkle might be to keep that 20% as a minimum threshold price and then allow your suggested flexibility for the remainder of the referral. Or perhaps this is what you have in mind already.

If we set a minimum price for LTM, we'll end up with what we have now (maybe except for the option of selling LTM for more than the current fixed price).

[tbone]

Very interesting thought. Where is the "to the network" money going? Does the BitShares system need it for maintenance, etc.? Because if so, then this suggestion might set up a system where some users are paying for it and some are not. A better wrinkle might be to keep that 20% as a minimum threshold price and then allow your suggested flexibility for the remainder of the referral. Or perhaps this is what you have in mind already.

I would be interested in seeing the reaction of some of our larger referral-dependent businesses, such as @ccedk .

What @jakub is proposing is that there is always a 20% network cut of any LTM fee that may be charged.  But he's also saying that the LTM fee can be set by the business itself to any level including zero.  That makes the referral program optional, as it should be.  Think about it.  How can we force a business to accept 80% of a hefty referral fee from their users? 

And in the case of those businesses that can't or don't want/need to take a referral fee, if we do as you say i.e. enforce at least 20% of some hefty fee that would go to the network, now all we're doing a is charging an exorbitant account creation fee without adding any value.  That doesn't make sense.  What @jakub has come up with is a good, sensible, creative way to solve the problem we're facing, and to reach a reasonable compromise. Kudos to him.

[donkeypong]

Very interesting thought. Where is the "to the network" money going? Does the BitShares system need it for maintenance, etc.? Because if so, then this suggestion might set up a system where some users are paying for it and some are not. A better wrinkle might be to keep that 20% as a minimum threshold price and then allow your suggested flexibility for the remainder of the referral. Or perhaps this is what you have in mind already.

I would be interested in seeing the reaction of some of our larger referral-dependent businesses, such as @ccedk .

[jakub]

My point is that a user who paid the full price will find out the day after he could have gotten the same thing for free. That is the user experience you are advocating.

In the same way my UX is affected when I buy a pair of nice sneakers from a local online shop and three days later I find out that I could've bought them on Alibaba for half the price.
This is the way things are nowadays and people are getting used to it.
Smart people who have the time to hunt for discounts get lower prices than people who don't have time for that.
Take airline tickets as an example.

BTW, currently we are not free from this "problem". What prevents bitcrab from selling LTM with a 50% discount?
The only difference is that currently bitcrab can offer maximum 80% discount, while my solution pushes it to 100%. Not much of a difference.

[jakub]

@jakub:  You've really stuck with this and now I think you're seriously onto something here.  The optional/variable LTM makes a ton of sense and should appeal to all parties, especially when coupled with a sensible compromise on the transfer fee schedule (including %-based fees). 

Thanks for this. Sometimes I have the feeling that I'm the only one here trying to find a solution that can be fair for *all* parties.

The only thing I don't get is the idea of letting the network "keep the transfer fees at a higher level (e.g. 10 BTS)".   Are you talking about a flat fee?  What about the %-based fee proposal?  Also, as you know, to me setting fees in BTS is like talking gibberish.  It's just not helpful.  So let me translate 10 BTS, which is CURRENTLY about $.04. 

So the question is what are you targeting at $.04?  It sounds like a flat fee since $.04 is too high for a lower limit if we want to facilitate micro-transactions...and too low for an upper limit since it would hurt revenues.  So if this a flat fee, can I assume you intend to start discussing %-based fee parameters that would yield somewhere around the same amount of income i.e. an average of $.04 per transaction?  If so, I think overall this would be a very good solution/compromise and I can't think of any good reason why anyone would be dead set against it.

Yes, I'm talking here about the flat transfer fee.
And you're right, we should use fiat when discussing fees.

IMO, something around $.04 makes sense for two reasons:
(1) This way LTM can still be targeted to non-advanced users (especially when %-based fees are implemented) - and this is extremely important for businesses that cannot earn money in any other way, e.g. Max Wright.
(2) This way there is space for the minimum limit in %-based fees (otherwise the choice for issuers will be fictional, and if the committee ends up to be the only entity using it, private SmartCoins will get unfair advantage)

We already know that $.04 is too high for the Chinese market. And that's where the flexible LTM price kicks in. If it's too high for your customers, sell LTM to them for e.g. 100 BTS or even give it for free and the problem is solved. No revenue is lost for the network, as those customers who cannot accept $.04 per transfer would not have bought LTM anyway.

[mint chocolate chip]

My point is that a user who paid the full price will find out the day after he could have gotten the same thing for free. That is the user experience you are advocating.

[tbone]

@jakub:  You've really stuck with this and now I think you're seriously onto something here.  The optional/variable LTM makes a ton of sense and should appeal to all parties, especially when coupled with a sensible compromise on the transfer fee schedule (including %-based fees). 

The only thing I don't get is the idea of letting the network "keep the transfer fees at a higher level (e.g. 10 BTS)".   Are you talking about a flat fee?  What about the %-based fee proposal?  Also, as you know, to me setting fees in BTS is like talking gibberish.  It's just not helpful.  So let me translate 10 BTS, which is CURRENTLY about $.04. 

So the question is what are you targeting at $.04?  It sounds like a flat fee since $.04 is too high for a lower limit if we want to facilitate micro-transactions...and too low for an upper limit since it would hurt revenues.  So if this a flat fee, can I assume you intend to start discussing %-based fee parameters that would yield somewhere around the same amount of income i.e. an average of $.04 per transaction?  If so, I think overall this would be a very good solution/compromise and I can't think of any good reason why anyone would be dead set against it. 

[jakub]

Why buy a cow when you can get milk for free?

Nope, this is *not* an adequate analogy.
If I'm a registrar, why would I give you LTM for free if I can earn an 80% margin by not giving it for free?
After a while, the market competition will settle the regional LTM price, which is appropriate for a given regional or demographic market.

Your idea undermines the entire referral system.

Nope, it makes the referral system stronger, as it introduces real competition among registrars and referrers that cooperate with different registrars.

The current situation is not so much different.
These are the only differences:
- registrars cannot sell LTM for more than 20k BTS, even if they had customers willing to pay more.
- registrars are forced to use the referral program, as there is no way that they can avoid the minimum 4k BTS payment to the network.

As a result, the price of the LTM has a lower bound of 4k BTS, which seems to be still too much for the Chinese users.
This is what we end up with: the Chinese businesses perceive the referral program as an obstacle while it might make perfect sense elsewhere (or even in China, *if* the LTM price could be adjusted to suit their markets) .
IMO, the problem is not with the referral program itself, but with this fixed minimum price that needs to be paid to the network, regardless of regional circumstances.

One might say that this proposal means that the network will lose a huge part of LTM income on the Chinese market.
Well, this is not true. If the minimum price of 4k BTS is too high for them, they will not buy it anyway, so no substantial income is lost.
On the other hand, if there are some rich users in China who are willing to pay a higher price for LTM, I'm sure there will be some BitShares-based Chinese businesses that will reach them - the 80% sales margin incentive should be big enough to make the effort.

As we have it now, the Chinese businesses try to find a way around the referral program and this is what they currently do: they lead this campaign to lower transfer fees to $0.018.
I'm quite sure that if they could sell LTM for e.g. 100 BTS instead of 4k BTS, they would *not* want to lower the transfer fee to $0.018 so desperately.

So this is my offer to the Chinese business: the network will let you sell LTM at any price you want (including zero) but in return you will let the network keep the transfer fees at a higher level (e.g. 10 BTS) which is more globally suitable.

The net income for the network should be positive: even *if* the network loses some income on the LTM sales on the Chinese market,  it will certainly have more income from transfer fees globally.

[mint chocolate chip]

Why buy a cow when you can get milk for free?

Your idea undermines the entire referral system.

Maybe you can have it so the referrer can reduce his take and thus make LTM membership cheaper, but the 20% to the network has to be out of the full amount charged for LTM that is set by the network.

[Pheonike]

Sounds good. The more modular we can make the fee structure, the better it will be for everyone. Then we can have true competition between business instead one accusing the other of handicapping it's strategy.

[jakub]

Currently we have this:
The price of LTM is set globally by the committee and the network takes a 20% cut whenever a user buys LTM.

And we have this problem:
Some businesses say that fees (especially the transfer fee) are way too high for their customers.
The low fees offered to LTM users would be fine for them but upgrading to LTM is currently too expensive to be considered an option for their customers.

What if we did this:
We could allow the LTM price to be set not globally by the committee but instead individually by the registrars (e.g. OpenLedger).
The network's cut would be 20% of any amount the registrar sets as the price of LTM.

If the registrar believes she can sell LTM for 50k BTS, 10k BTS goes to the network.
If the registrar charges 20k BTS for LTM, 4k BTS goes to the network.
If the registrar decides to give LTM for free, nothing goes to the network - and we have the situation desired by China: low fees and no referral incentive.

And this way the referral program becomes effectively optional, as all depends on the sales strategy of the registrar.
Some of them might want to position themselves as high-end platforms and try to acquire clients who will pay a higher price.
Some of them will find medium pricing most suitable.
And some of them (mostly in China) will be happy to skip the referral program entirely in exchange for low fees for their customers.

EDIT: this proposal has been revised and no longer LTM is involved in it.
I now think that LTM should stay as it is, as the same goal can be achieved with AM (Annual Membership) and/or shorter membership schemes.
So whenever LTM is mentioned in OP or the discussion below, please replace it with AM.

AM (and/or any other shorter membership scheme) is better for our purpose than LTM due to these reasons:
- AM does not allow users to create subsequent accounts (this would be a loophole if LTM was used).
- AM only offers what we actually need for our purpose: a price discount. This way we do not end up giving away access to advanced features when it's not needed.
- AM costs significantly less, so it is easier to sell it at a relatively good price.
- *if* there are any cases of lost revenue for the network, they will be on a much smaller scale.

I do maintain my position that a revenue can be considered as lost *only* when a user ends up paying less than s/he was willing to pay.
I don't think there will be any significant number of such cases, at least until competition on the hosted wallet market grows substantially, as only strong competition can push the AM to a level that is lower than the level users are willing to pay.
So until that happens, we are sure to preserve the network's income or increase it.