Initial Witness Pay & Number of Witnesses

[montpelerin]

Then 33 is probably the next best number.  33 degrees  ;)

33

After all, it's the magic number...

Plus, we can get free marketing from Adam Curry and John C. Dvorak.

[Helikopterben]

17 should be plenty to begin with.  Even fewer than that would probably work at the beginning stages as long as they are sufficiently geographically distributed.  As Ander said, pay projections should allow for a drop in market cap, perhaps back to previous levels near 1400 sat just to be safe.

[emailtooaj]

IMO I think we should consider starting with 21 Witnesses @$250-300 pay rate, while also have 10-16 "stand by" Witnesses for redundancy.

I also like the idea having the witness count increase in lieu of increases in market Volume, not market cap.
Market Volume is what puts processing strain on the network, not BTS value (though market cap does have some play in this obviously) hence the reason you'd want to have High Quality and Highly reliable Witnesses running the show!
Therefore paying them well enough makes sense to me  :D

In BTS 1.0 we had a unreliable wallet with a kick ass node system.
With 2.0 rolling out, why the hell would we revert to having a kick ass wallet and a crappy low budget node system????
Defeats the purpose in my mind!

To reiterate my position.... I really, really, really like the idea of scaling nodes,  averaged by BTS Daily Volume on a monthly(?) bases.  21min floor witnesses and scale up to 37 (or more when we pass Pluto and are headed to Alpha Omega 9 Sector Z).

[muse-umum]

Having been running delegates since the beginning of BTS1.0, I clearly remember we suffered from low delegates participation a lot. Every time when a new version got published the network worked like a roller coaster.  Is losing 20 among 101 delegates more dangerous than losing 5 among 17 witnesses?

[Riverhead]

Looking at the business partners Graphene has attracted, and what attracted them, it matters little what "bitcoiners" perceive as secure.

How many witnesses do you think we actually had with 0.9.x? 101? 75? 50? I'd say closer to 30 unique nodes.

[alt]

I don't like the way think of witness for money.
we need witness really want to contribute BTS, they must be honest and can provide stable chain service.
please don't worry about the money too much. if the cost is 40 USD, then 40 USD is enough for now.
we still in a hard time, we need to save the money.
I think we should write a list for the witness account first, then decide the number from the reason like: decentralize, efficient ....

[BTSdac]

I think the current witness is 17 is OK,  but the max witness is 17  ,too low ,
I don`t know if system find a mechanism to optimize witness amount

1. initial witness is  17 , and the minimum is 17
2.max witness is 35
3.how many witness acc. to  market cap
      3.1 market cap can calculate in BTS 2.0 system by price feed
      3.2 witness amount increase 2 if the market cap increase 10,000,000 usd, but the max witness is 35
4.  adjustment witness amout  per 15 days

[puppies]

I would prefer something in the 30s range.  While BM is probably right about the security requirements I agree with those that are speaking from a marketing standpoint.  I also think $200 a month is a reasonable payment.  I think we will be able to find 30+ separate witnesses that will be willing and able to secure the network at that price.  Especially since most of us are rather bullish on bts price.

[twitter]

i like the idea to decrease the witness number to 17 

[Brekyrself]

17 might be the most optimal off the bat however we should take in consideration some may go offline or have trouble thus build in some redundancy.  21-31?

[Pheonike]

Then 33 is probably the next best number.  33 degrees  ;)

[donkeypong]

I agree with some others that 17 is too low; outside perception may consider this overly centralized, and I don't think we're going to get a "do-over" this time. Let's get it right the first time, even if it costs a bit more to fuel that perception. I would go with 25 or 31.

[Empirical1.2]

This is the maximum rate, which would be around 6.3% inflation, which we should try to stay well below, imo, because of the perception impact that it has on the price.  If you inflate at 6% and it causes people to dump back to 1400 sats, you will get less money that if we inflate at current 2% level and the price is a lot higher.  The price is our #1 marketing tool right now, and having it collapse again would be really bad for the project.

+5%  2% inflation from a large CAP is worth much more than 6% of small one.

I think 17 is too low and somewhat risky IMO.

+5%

I think BM numbers are good but for people outside of BitShares, 17 may seem too little, even scary. People from Bitcoin use to think they are fully decentralised, even if it's not the case with pooling.

+5%  I believe it's far too low from a marketing & PR perspective.

Just like with inflation, what's OK for the established market leader, Bitcoin isn't OK for up and coming challengers.
They need to be seen as far superior to challenge the status quo.

I think the number needs to be at least double. 37 is probably the best odd number in that range...

Other Suggestions:

1. Pay Standby witnesses
Being able to replace witnesses is only effective if we have people ready and waiting to replace them.

2. $1000 monthly voting lottery
As BTS gains wider adoption, DPOS will only be effective if small shareholders vote. $1000 is a small amount but may incentivise very small shareholders who don't think their votes have impact to vote frequently. 


Costs

Witnesses                 x 37 x $300 = $11100
Standby Witnesses x 14 x $100 = $1400
Voting lottery           x 1 x $1000 = $1000

Total                                               = $12500

I would expect BTS to have a CAP of $30 million + post 2.0 which yields $50 000 per month at 2% annual inflation meaning 75%+/$37 500 would be able to be spent on workers.

However I expect the CAP to rise rapidly, providing more money for workers, if the market sees you are keeping costs below 2% and hopefully moving towards eventual profitability.

37 Witnesses...

some numbers are inherently more appealing than others.
Ask a friend to think of a two-digit number less than 50. Tell them both the digits must be odd and different. Say that they can't have 11 but could have 13. When psychologists tried this in experiments, more than a third chose 37.

http://www.dailymail.co.uk/news/article-2601281/Why-lucky-7-really-magic-number.html#ixzz3mQ6dSRcg

The number 3 (三, Pinyin: sān, jyutping: saam1) sounds similar to the character for "birth" (生, Pinyin: shēng, jyutping: saang1), and is considered a lucky number.
The number 7 (七, Pinyin: qī (Mandarin) "chut" (Cantonese) symbolizes "togetherness".It is a lucky number for relationships.

https://en.wikipedia.org/wiki/Numbers_in_Chinese_culture#Combinations

[clayop]

I prefer 31 with $150 payment

[Pheonike]

I think 21 would be a good number. It sounds about right, that's my reason. We will have one server per million BTC.  :)  Also today is the 21st( in US).