Discussion - When to start the next application round?

Secured
#1
Everyone,

During the guide meeting (subject 2) on Monday 13th of August we discussed postponing the next ANO-application round until the FCT-price has recovered.

Specifically there was a suggestion by myself and Niels to postpone future elections of new Authority Node Operators (ANOs) to the Authority Set until the price has stabilized above $10.

As this is a discussion that should also include input from the broader community, we concluded that the discussion should be moved to this forum instead so everyone that wants to can provide their input.

The guides would like, for the record, to point out that this is a decision that formally lies with the Factom protocol guides, but due to the severity of the discussion we would like as much perspective as possible.

The governance document states that we shall achieve full decentralization by April of 2020 (65 organizations operating 65 nodes). This is currently 19 months away, which would mean selecting 44 new ANO, including the six planned for this application round, and onboarding 2.3 of them every month in that time frame.

For this final date to be delayed there would have to be a change in the governance document, but as this date is still quite far off it would be possible to delay the next application round for a while to wait for the price to stabilize before proceeding with a higher cadence.

The current Authority Set (AS) looks like this:
- 21 Operators hosting a total of 43 servers (where one extra is run by inc at 100% efficiency to make the AS odd)
- 6 new Operators will be selected later this month. They will also run 2 servers each (another 12 added to the 43 = 55)
- The authority nodes which are not operated by ANOs are «virtual» and running at 100% efficiency; providing all their tokens to the grant pool.
- The average efficiency is just shy of 50%.

Financials:
- Currently 42 servers are running at 50%-ish efficiency providing 561,5 FCT each to the grant pool monthly (total 23583 FCT).
- 23 servers are (virtually) running at 100% efficiency providing 1123 FCT each to the grant pool monthly (25829 FCT).
- Today, every month 49412 FCT is added to the grant pool.
- With the 6 upcoming selected operators included in the AS, 12 of the (virtual) nodes currently running at 100% efficiency will be reduced to about 50%; i.e reducing the amount of tokens added to the pool monthly by 6738, for a new total of 42674.

Some aspects to take into consideration:
- After including the 6 new ANOs the AS will have 27 operators; hosting 54(55) out of a total of 65 servers.
- We can add an additional 5 ANOs after this round before ANOs will have to start giving up servers to newcomers in the authority set.
- When governance was written (in April) the FCT price was around 4 times higher than what it is now.
- At current prices most ANOs should be about break even in reference to costs (all things considered) and are severely limited in regards of contributing outside of hosting the authority servers
- Adding an additional 5 ANOs (10 servers) at 50% efficiency would reduce the amount of FCT to the grant pool by an additional 5600 monthly.
- We should have a decision about postponing or hosting a new application round ready prior to the current round concluding. This makes it easier for the 10 applicants that dont' make it to plan ahead.
- Increasing the number of ANOs increases the complexity of coordination for network issues such as reboots. As the protocol becomes more robust and as we fine-tune our processes this will become less of an issue.

Reasoning behind the suggestion:
- The low FCT-price in combination with almost non-existent awareness about Factom severly limits the numbers of applicants
- A reasonable next step would be to add another 5 ANOs in the next application round; and this would decrease the the monthly amount of FCT to the grant pool by around 5600 FCT.
- At the current price ANOs are not able to do much more than operating their nodes. This is OK for our "infrastructure operators"; but limits development, marketing and expansion funded by hosting authority servers.
- We do not know if the market is still trending downwards. There is a possibility that the price is reduced further in the coming months.
- With 20+ different entities operating the network's nodes we have a minimum acceptable decentralization for the current status of the protocol. Every extra server we add will remove ~561 FCt from the grant pool per month.


We welcome everyone to provide input regarding this issue, and look forward to read it.
The Factom Guides.
 
Secured
#3
I would fully support the delaying of another round of onboarding due to the reduced FCT price. Despite this, I am concerned about how the non-ANO community will view this decision; I think less ANOs = less exposure in their eyes.
 
Secured
#5
As as member of the non-ANO community I also support postponing the next application round. In addition to the reasons stated by the guides in the opening message of this thread, I think postponing the application round until higher FCT prices come about will help incentivize higher quality ANOs applicants to join. Perhaps also, the protocol will have garnered additional awareness which hopefully will lead to high quality ANOs applying. Not saying the current ANOs and applicants from this round aren't high quality, but in my eyes higher FCT price and more awareness of the protocol will in general contribute to a more competitive field of potential ANOs. Since the initial ANOs will probably be around for a while, I think getting as high quality teams as possible in at the beginning will greatly help in producing successful results for the protocol.
 
Secured
#6
Even though it's a pity to delay the full decentralization of the protocol, I think given the current market conditions and the lack of awareness it is wiser to do so. I hope that the wider community also supports this decision and understands that in the long run this is best for the protocol.

That said, I think we should define very clear conditions under which the next round of applications can be opened. If such conditions are defined, I support this decision.
 
Secured
#7
I believe that both ANO vs Grant pool will put pressure on the Factoid price until the bear market ends.

I think having ANO investing directly in the protocol/team/business has the added benefit that they committed money (usually), built a business plan and will in most cases use their skills/contacts to some extent to put something together (developing/marketing/etc) and then even with a Factoid at 0 or close to.

The grant pool is great but I think it is going to be mainly used by people already invested in the protocol i.e. mostly ANOs.

If we push that thought a little more and speculate that, maybe" the Factoid price will reach something like 50 cents. Some ANO, if not most, will still put the time and the sweat equity even though their pay is virtually 0. They will inevitably put less effort because everyone has to pay bills but still. Where as if you only have grants, with a Factoid price very suppressed, you are going to empty the pool very fast for a lot of simple tasks.

I'm of the opinion that we have to get Factoids in the hands of the most people as quickly as possible and delaying the onboarding of very good ANO would be a mistake. The Factoids are going to be awarded no matter the price, the more good team involved , the better.
 
Secured
#8
ANO elections schedules should never have FCT price as a variable. In addition, it's time for election rounds to come to an end and instead allow any Authority Node Candidate (ANC) to apply at any time. They won't be competing against other ANC's, instead they'll be presenting what they bring to the table and the community can ask questions and the Guides can score them. If they get a score above X, they become an ANO. If they don't, then nobody from that team can reapply for Y duration.

Anytime a highly qualified ANC is ready to step up to the plate, they should be allowed to. If we keep the doors closed, they may seek out competitors instead. Now, it's no doubt going to take Guides a little while to work out the exact process of moving to ANC's being able to present when they're ready, but it's the step we need to take if we want to ensure we are competitive in the crypto-ecosystem. If I meet a kickass group of people at a conference who come to understand the value of Factom and want to become an ANO, I shouldn't have to say, "Well, you can't right now. But when FCT is over $XX then you can. Once there's an election round". That's how you get people to lose interest very quickly. If we want to be able to recruit top notch ANOs, I need to be able to say to excited individuals, "Here's the process. It's going to be extremely hard so be prepared to bring your A game. But if you get through, you'll be a part of one of the most exciting blockchain ecosystems."

We always want to work with and build momentum, not kill it.
 
Secured
#9
While I understand all the logic above at the end of the day the discussion is about money and it is based around a speculative assumption that the fct price will recover post bear market. If it did not recover and the proposed delay went on indefinitely then we rely on the current group to finish the job and drive the protocol forward safe in the knowledge that they are in control. If a ANC is willing to apply with the current prices and they are talented then is that not the very people we need.

Good luck with your decision and I will support it either way but I fear your line of thinking undemocratic.
 
Secured
#10
Largely agree with David's post. Other things to keep in mind:

1. The more ANOs, the more decentralized we are. How much does another 10 ANOs move the needle? Probably not a whole lot though. Are their are other things we can do to make the the protocol more decentralized that will have a much bigger impact? Without a doubt.

2. After factoring in the decentralization component, potential ANOs need to be judged against whether the FCT they would receive is better used if dispersed to them via running a node or if it is better dispersed to someone else via the grant pool. Like any business, we need to be as efficient as possible when it comes to managing our resources. We need to maximize our ROI per FCT.
 
Secured
#11
I agree a lot with Matt's and David's posts. I don't think it would be good to stunt the momentum by indefinitely postponing anymore ANO elections if there are awesome groups that want to join. Creating a bar over which the ANCs will have to rise to become an ANO seems like a good idea (as David mentioned). I also think (as Matt mentioned) that the pros and cons of adding more ANOs vs lowering the funds sent to the grant pool must be weighed. It is impossible to say where the price of FCT is going in the next few months so it is important to direct the FCT funds to the most critical areas within the protocol.
 
Secured
#12
I’d like to thank Quintilian and the Guides for opening this discussion up to the ANO’s and the broader community. Many excellent points have been made and I can see a good argument for either side of the debate. One concern I have is in regards to efficiency. ANO efficiency reduction has become much more of a fluid topic than it has been in the past and I know that several ANO’s included language in their application that efficiency reduction may become necessary dependent on market conditions. What happens to the grant pool when ANO’s are down to receiving 561 fct/month (with 1 node at 50% efficiency and lets say the price of FCT is $1)? I believe we would see several ANO’s reduce their efficiency and thus depleting the grant pool.
Maybe the solution can be found in a combination of the answers above. Where we keep the ANO application period open and only accept qualified teams as they apply, but we only accept a maximum of 5 more teams over the next 4 months.
I would further suggest that we as a community do whatever we can in our power to get the grant pool functioning at a high level. Lets say that the ANO application period is closed. One of the many benefits of having a highly functioning grant pool process is that we can always point people excited about getting involved with Factom to the grant pool.
Overall, the protocol needs time to develop and the ANO infrastructure needs some time to develop. Under current market conditions, creating more time and space for these developments could greatly improve the overall Factom ecosystem. With these improvements, we collectively have a better product to market to the greater audience and new ANC’s.
 
Secured
#14
(reposting due to the forum issue mentioned above)

We (Cube3) understand the reasoning behind looking to postpone future ANO applications and if the guides choose to do this we would support their decision. However, we feel it would be useful to give our view of the opposing perspective:

In the short term, any postponement to ANO applications will delay the decentralisation of Factom. As this is a primary objective, we are concerned that it might generate negative press, as Factom could be seen to have “stalled”. One of the big challenges is how to develop the protocol so that it is attractive to as big an audience as possible. Demonstrating that the Factom ecosystem is alive, well and expanding is probably one of the most effective ways of positively affecting the Factoid price at the current time.

There is naturally a lag between application and onboarding and this significantly exacerbates the difficulties of trying to time the application process with market price movements, as it is the price at onboarding that actually matters. As crypto markets are naturally volatile, what do we do if FCT rises over $10 and leads us to start an Application round only for the price to fall again by the time the onboarding is due to start?

I feel that Dave Chapman made a great suggestion that may present us a better solution to move forward, by moving away from the “batch” process on onboarding, and instead focus on individual candidates. This approach could effectively address the concerns that led to the suggestion of postponing application rounds whilst avoiding the negative perception of “stalling”. At the beginning of the decentralisation process, it was necessary to request multiple candidates at once to allow for comparison between them. However, the existing ANOs now represent a benchmark against which to compare new candidates. If an individual applicant is deemed capable and the guides are satisfied their approach won't disrupt the FCT market dynamics under a realistic range of FCT pricing scenarios, there is no reason why they should not be onboarded individually.

Surely the FCT price itself is going to influence the number of authority node candidates at any one time so that the number reduce naturally as the FCT price goes down and increase as the FCT price increases, in other words why interfere with a natural mechanism?
 
Secured
#15
ANO elections schedules should never have FCT price as a variable. In addition, it's time for election rounds to come to an end and instead allow any Authority Node Candidate (ANC) to apply at any time. They won't be competing against other ANC's, instead they'll be presenting what they bring to the table and the community can ask questions and the Guides can score them. If they get a score above X, they become an ANO. If they don't, then nobody from that team can reapply for Y duration.

Anytime a highly qualified ANC is ready to step up to the plate, they should be allowed to. If we keep the doors closed, they may seek out competitors instead. Now, it's no doubt going to take Guides a little while to work out the exact process of moving to ANC's being able to present when they're ready, but it's the step we need to take if we want to ensure we are competitive in the crypto-ecosystem. If I meet a kickass group of people at a conference who come to understand the value of Factom and want to become an ANO, I shouldn't have to say, "Well, you can't right now. But when FCT is over $XX then you can. Once there's an election round". That's how you get people to lose interest very quickly. If we want to be able to recruit top notch ANOs, I need to be able to say to excited individuals, "Here's the process. It's going to be extremely hard so be prepared to bring your A game. But if you get through, you'll be a part of one of the most exciting blockchain ecosystems."

We always want to work with and build momentum, not kill it.
Do you envision this being only for the next five ANOs or would you have us start taking ANOs down to one server as soon as there's a qualified applicant to give the second server to? I suspect that if we went with this process and started it tomorrow we would immediately have 10 applicants, most of whom would pass the minimum qualifications test.
 
Secured
#16
Do you envision this being only for the next five ANOs or would you have us start taking ANOs down to one server as soon as there's a qualified applicant to give the second server to? I suspect that if we went with this process and started it tomorrow we would immediately have 10 applicants, most of whom would pass the minimum qualifications test.
Good questions.

I wouldn't limit it to the 5. If there's HIGHLY QUALIFIED teams out there, we want them within our ranks. If there's 10 HIGHLY QUALIFIED teams out there, fantastic. If there's 10 pretty good teams that would pass, then I'd suggest the bar needs to be raised. I'd also suggest creating a time period where those who have already applied cannot apply again.

The eventual issue we're going to face is which ANO loses their 2nd server first. I know Factomize won't raise its hand to volunteer ;) I personally would support a system where all ANOs adjust their efficiency (the ANO who lost their server would lower theirs) so that we each lose an equal amount of FCT with each new onboarding until we're all down to one server.
 
Last edited:
Secured
#17
The eventual issue we're going to face is which ANO loses their 2nd server first. I know Factomize won't raise its hand to volunteer ;) I personally would support a system where all ANOs adjust their efficiency (the ANO who lost their server would raise theirs) so that we each lose an equal amount of FCT with each new onboarding until we're all down to one server.
Great Suggestion. I would support handling such a situation in this democratic fashion.
 
Secured
#18
I have no clue what the solution looks like at this time, however I would rather see top producers not have resources pulled from them. In fact, we should be doing the opposite: encouraging them to lower efficiencies in order to give them more resources. If you think of it in traditional business terms, do you give each department the exact same budget? Nope. You maximize for output. Different units have different budgets. if you want a sports analogy, you want your best players taking shots, you don't say, "each of you gets 5 shots a game."

Output needs to be maximized if we are going to be successful, we just have to figure out how to do it. :) And before anyone says, "that's what the grant pool is for," please keep in mind their is zero chance that we don't have more people applying for FCT grants than their is FCT. Go Immutable could single-handedly spend the equivalent of the grant pool each month if given the chance to do so (not that we are even considering applying for that many grant, haha).

Like any business, we need to be as efficient as possible when it comes to managing our resources. We need to maximize our ROI per FCT.
 
Secured
#20
The eventual issue we're going to face is which ANO loses their 2nd server first. I know Factomize won't raise its hand to volunteer ;) I personally would support a system where all ANOs adjust their efficiency (the ANO who lost their server would lower theirs) so that we each lose an equal amount of FCT with each new onboarding until we're all down to one server.
There are a couple of problems with this.
1. changing efficiency can be a bit of a pain. How long does an ANO have to do it? how is it enforced? what happens if someone doesn't?
2. The ANO who loses their server also gets to cut their back-end costs by no longer paying for (or by selling off, in some cases) that server. The other ANOs continue paying for their second servers, and after enough new ANOs are onboarded, those second servers cost more to operate than what they're bringing in each month, especially at very low FCT prices.

I don't think there's a simple solution.
 
Secured
#21
There are a couple of problems with this.
1. changing efficiency can be a bit of a pain. How long does an ANO have to do it? how is it enforced? what happens if someone doesn't?
2. The ANO who loses their server also gets to cut their back-end costs by no longer paying for (or by selling off, in some cases) that server. The other ANOs continue paying for their second servers, and after enough new ANOs are onboarded, those second servers cost more to operate than what they're bringing in each month, especially at very low FCT prices.

I don't think there's a simple solution.
1. It wouldn't be that often. Just when a new ANO was onboarded. Social enforcement. It's pretty easy to see efficiency here: https://luciap.ca/#/authority-set

2. There's nothing a little math can't solve here.
 
Secured
#22
I think we should start thinking about new round without dependance on the FCT price – we shouldn't start tomorrow it if FCT price becomes $10+.

The only thing we should trace – high-qualified teams committing to the Protocol for a while.
Not teams rapidly appearing when application round is announced and so rapidly disappearing after round ends (as we seen in 1st / 2nd rounds).
When we see 5-7 good teams making projects with Protocol, promoting it or making whatever else contribution – it's time to start new round and choose new ANOs in the honest competition.

The best if the next round will be the same competitive as the 1st round.
 
Secured
#23
We would distinguish 2 different phases.
The first one (which is primarily discussed in this thread) is about the pace of the new ANOs election until we reach 32 ANOs i.e. before decreasing the number of servers per ANOs.
The second one is about how to rank/select ANOs to decide who will lost its servers first. This is also related to the pace of the new ANOs election once passed 32 ANOs.
We will only discuss the first point. The second point is being discussed elsewhere (see discussions around tiered FCT reward and metrics)

Our opinion for the 1st phase is that we should adjust the number of spots according to the FCT price to get a balance between number of (good) candidates and spots available.
With a low FCT price, the number of spots should be limited. 0 can be an option if this price is really too low. 1 per month or 2 every 2 month could be fine for an average price (15$?). This would add 6 months after the 27 ANOs already elected gets onboarded.
If FCT reach 20/30$ then we can accelerate. Yes, meanwhile the price could decrease after the round has started. That is why this price limit should be sufficiently high (10$ seems a bit too low).

As Brian has said, Factom is a multi-decade project ("I dislike it as much as you, but my perspective is that factom is a multi-decade project with many phases. While there is need for urgency in many things, I sense more urgency in the abillity to serve customers." - from a discussion about the timing for moving away from the Swarm/Portainer setup). We should not precipitate the onboarding because we fear critics about how the protocol is decentralised. 27 ANOs is already much more decentralised than many (successful - Market Cap point of view) projects.
Moreover, with only 20 tps, there is no immediate need to have plenty of ANOS trying to build applications on top of Factom which could not work simultaneously.
 
Secured
#24
Just a quick feedback from us. We are not supporters of delaying application rounds due to some arbitrary market sentiment. As stated earlier in this thread, a "low" Factoid price could even be an argument in itself for having as many quality teams on board as possible.

Also, we like the ANC-idea proposed, but fear it would be a First Come First Serve-concept, ie filling up all ANO-slots after a very short time. If we are to move away from Application Rounds in favor of a Continous Application system, our proposal is to allow applicants to apply at any time to have a decent sized pool of pending applications at any time, and then periodically score and onboard from the top as many new ANOs as needed to meet the rollout plan.
 
Secured
#25
Thanks for participating in the discussion so far everyone.

I'm bringing this up again per the decision in a previous guide-meeting to continue the discussion here, as to be able to conclude this matter and provide some clarity to the teams who are currently waiting on the sidelines for the next application round (sorry, that was a long sentence!).

First of all I just want to comment on the idea about continuous applications. I have thought (hard) about this and decided that it is not an idea I would support. There is one main reason for this:
- It would be a "first come first serve-principle". In the previous round Every single team was deemed "good enough" to provide the minimum required by the guides (minimum 60% support). So if we opened up for a continuous application process then we would basically be "first come".
Alternatively we could make the requirements harder to be accepted; but that would mean that we'd need a new process for measuring how qualified the teams are; and also we would have different standards from the teams already selected. I'm rater ambivalent about that....

I'm also concerned that this would take up too much time for the guides at this point; and the "bunching of applications" means that we get to pick the ones who are deemed to be the most promising amongst many good candidates... Also a lot of filtering happens in "open market competition", and by announcing the upcoming rounds through more channels in the future I believe we will absolutely pull in new, competitive parties to the future application rounds... So for now I would like us to continue with hosting election rounds as we do now.

Regarding the original discussion:
My initial stance was that we should suspend elections until the FCT-price stabilized above $10 again (would need some kind of metrics TBD), but after reading the input here I have changed my mind to some extent.

I now believe that we in fact should continue to host applications round; albeit with longer intervals between rounds.

This ensures:
- Interested teams are not discouraged to participate in the community as they know when the next round will be held (and they can start preparing for it).

- We are following the intent of the governance doc; working towards 65 parties and full decentralization.

- Governance does not provide us with the possibility to reduce the amount of standing parties based on FCT-price (if so it should be dynamic; and changed at intervals in the future as well; maybe dropping from 65 -> 40 etc.)

- I believe that decentralization of the protocol is one of the more important aspects we need to take into account going forward. Currently we have more than the required amounts of "geographic server decentralization" to operate properly; but for governance purposes I believe we still have too much power distributed on our current Authority Set. This is mainly a function of the current ANOs being a continuation of the broader Factom community that existed prior to the ANO application rounds (we are basically a community that has split up into ANOs and are now compromising the Authority Set). Getting new blood into this mix is very healthy, and it also ensures that we reach a broader audience as the new ANOs will bring in even more people by word of mouth etc.


With the above in mind, I still very much see the issues with bringing on new ANOs if the price is so low that it cannot sustain:
1) Healthy ANOs, covering their operational costs, and;
2) A minimum amount of FCT/USD-equivalent transferred to the grant pool every month to cover a minimum of core development.

I do think this question however is broader than the scope of this discussion right here; and I also think we would be able to take on another 5 ANOs without breaking the bank (which is also the amount of ANOs we can elect without having to decide how to handle scale-back from 2-1-servers for current ANOs)...


My suggestion is this:

- At this time we decide to host one more round for now.
- We select another 5 ANO to reach the halfway-point.
- The next application round starts 1.1.2019.


Postponing the round slightly (last time we went 3 months from round-to-round conclusion, while this would be closer to 5) has this effect:

- Buys us time to have the broader discussion about having the price point influence number of ANOs (if its a fact that 65 ANOs can't be sustained at $3-5 FCT this is a discussion we need to have as a community).
- Provides us with time to conclude the next grant round.
- Postponing the reduction in FCT added to the grant pool (adding the 5 ANOs at 50% efficiency will reduce the monthly FCT added to the pool by approx. 2800 FCT per month).
- Even if its a bit far off, we provide the ANO applicants with a date they can use for planning purposes.
- Gives us enough time to get some of the important processes in place before we grow too large; and provides Brian and Inc. with some relief from continuously adding new operators every month (as they have done the past 6 months).

@briandeery @Niels @Julianft @SamuelVanderwaal - and everyone else;
Whats your thoughts?
 
Last edited:
Secured
#26
I completely agree with you. I would see "continuous applications" as something to work to in the future. Once the authset is filled with 65 ANOs a first come first serve risk matters a lot less, since you would need to be better than the at least one of the 65 (with current governance). I am also worried it will bring too much strain in short term without benefits that would severely outweigh the rounds we have now.
 
Secured
#28
I still standby my assertion that FCT price should not be a variable. Yes, with X more ANOs, maybe the price of FCT will drop more in the short term but I can say I'd be backing up the truck in that case as will others. The market WILL determine a fair price and we're going to miss out on some excellent candidates who can make meaningful differences in the medium to long term due to the current batch system. Obviously I'll get behind whatever you guys eventually decide, but that doesn't stop me from speaking up in the mean time :)
 
Secured
#30
@DChapman The problem as I see it is the grant pool.

If the price stays low and we onboard too many people we won't be able to pay out any meaningful grants without spending a long time waiting for the grant pool to replenish. There are three reasons for this:

1. The more that is paid out to ANOs the less that goes into the grant pool.
2. The lower the price of FCT, the more will need to be paid out for any given grant.
3. After we hit the magical halfway point, we will have to begin reducing the number of servers people are managing from 2 to 1, which will inevitably coincide with a drop in efficiency.

Admittedly, both 1 and 3 are true even without a low price. Even so, the low price amplifies the impact of both of these, and when you combine point 2 then the grant pool could become very feeble indeed.

Supposing you accept that argument, then the question is whether we value ANOs over grants. I would argue that they are both equally important for adoption, so I think striking a balance between conserving the grant pool and onboarding new ANOs is preferable to pushing for more ANOs at the expense of the grant pool.