Author

Topic: [ANN] FACTOM - Introducing Honesty to Record-Keeping - page 275. (Read 2115876 times)

legendary
Activity: 1960
Merit: 1128
How can i do something similiar to cold storage with factom ?

The site which someone on reddit referred to : http://factom.org/howto is not found ?


They use the "com-domain" now:

https://www.factom.com/howto/

Cold-storage: Papermill
sr. member
Activity: 270
Merit: 250
How can i do something similiar to cold storage with factom ?

The site which someone on reddit referred to : http://factom.org/howto is not found ?
full member
Activity: 196
Merit: 100
Factom Series A closing tomorrow https://bnktothefuture.com/pitches/factom-series-a their seed round was covered widely so I expect Factom to be on top of this some time next week.
sr. member
Activity: 385
Merit: 250
Thanks for the analysis, Haggis. This is really interesting. I think Tempus is right in pointing to the influence of psychological and "soft" variables on what the price becomes, but having this kind of "hard" theoretical background is really valuable. I think most investors already cobble together some kind of hard analysis (even if very informally or causally) when they're trying to determine whether or not a long-term investment is "worth it." But to spell some of that out, and to elaborate and analyze that hard data—even if it's only speculative—is really helpful I think.
hero member
Activity: 984
Merit: 1000
Snow: “The token supply can grow (if speculators drive up the price) until the price stabilizes. And the token supply can fall (if speculators drive the price down) until the price stabilizes. But in both cases, the stable price is when the value of the token matches the money real applications are spending to buy Entry Credits in order to put data into the protocol. That is because the 73K factoids generated each year naturally trends to the value of the factoids drained from the supply to write into the protocol. If people are spending 1 million dollars to put data into factom per year, then 73K factoids should be worth 1 million dollars.”

Can you link to where this quote is please?

It's from an AMA:


https://forum.bitcoin.com/ama-ask-me-anything/i-am-paul-snow-the-architect-of-factom-and-chair-of-the-texas-bitcoin-conference-ask-me-anything-t4026.html

Actually this is an interesting quote to zero in on. We've been discussing off-and-on what a "natural" or fair price for FCT would be in the future, and this quote seems to provide a strong clue. Given the price-stabilization that Paul Snow mentions, it should, I believe, be possible to deduce a FCT price based on various level of system-use. Fsr I'm having a tough time wrapping my mind around how exactly to do this ... but if someone could provide test examples of moderate/reasonable Factom use in the future—based on current usage trends, etc.—it should be possible to reverse-engineer what the prices of FCT would be in order to maintain the price stability Snow mentions. So with light / medium / heavy use over the next, say 5-10 years, FCT should be worth X / Y / or Z. Thanks to anyone able to give this calculation a crack.
I give it a shot Smiley

Exercise: how must the value of 1 FCT be to hold the current marketcap of 21m USD depending on Entry Credit demand per period (73k new FCT per month or year)?

We know:
1 EC: 0,001 USD
1 BTC: 600 USD
FCT supply: 8,753,219
Marketcap (general): supply * value * BTC_usd
Marketcap (current): 8,753,219 * 0,004 * 600 = 21,000,000

As 1 EC costs 0,001$, 1000 EC cost 1$. When someone buys 1000 EC then factoids worth 1$ get burned.
Formula: (demand_in_k_ec * 1$)/(value_fct * 600$)

Insert this into the marketcap formula:
21,000,000 = (supply - ((demand_in_k_ec * 1$)/(value_fct * 600$))) * value_fct * 600

With 73k extra factoids created during this period:
21,000,000 = (supply - ((demand_in_k_ec * 1$)/(value_fct * 600$)) + 73000) * value_fct * 600
value_fct(x) = (21,000,000 + demand_in_k_ec) / ((supply + 73000)*600)

After x cycles of 73k fct creations and fct burning:
21,000,000 = (supply_x-1 - x*((demand_in_k_ec * 1$)/(value_fct * 600$)) + x*73000) * value_fct * 600
value_fct(x) = (21,000,000 + x*demand_in_k_ec) / ((supply_x-1 + x*73000)*600)


As it is unclear whether 73k FCT are newly created each month or each year, here are both variants:




What you see are the needed prices per FCT to hold the current total marketcap at a given EC demand per month.


According to https://docs.google.com/spreadsheets/d/1RUZLxxoAfT3C5jIJ144DLYdztYe_sqrKqbTlZuD05-s/edit?pref=2&pli=1#gid=445871101 the last 10 days had an average of 31000 new entries per day. Which is approx 100k per month. And the system is not yet even in production  Wink



Tip me if you want: FA37G4zPpbaxEZXXqV8HASZFFcJV4jguXkyKR5c1iP7F3AutVhXk
legendary
Activity: 924
Merit: 1000
Wow...wotta thud after a day when the bulls & bears fought each other to a near-standstill. It wasn't short-sellers: from what I could tell, more than a few used the plummet to cover. I wonder if it's a whale (or group of impatient dolphins) who watched FCT go nowhere today and lose patience. If you bought only four days ago, you've got a >50% profit even at the post-dump price.

I know some folks here prefer more involved interpretations, but I'm still a fan of Occam's Razor. Grin
legendary
Activity: 1960
Merit: 1128
I've been pondering the Factom system.

Entry Credits can't be transferred.  As such, do I understand correctly that in order to run an Entry Credit Store, you will need a Federated Server?

What happens if you have an Entry Credit Store and someone buys 1000 EC's off you and your Federated Server gets demoted?  That person loses their money, right?

Are there any plans to certify Entry Credit Stores somehow so people can avoid being scammed by fake stores?

You don't need a federated server.

Example:

You are allowed to run such a site (no clue about the law) and you open a EC-store. You buy Factoids on Poloniex and I'm your first customer. I pay $0.005 per EC (let's say for 1000 EC's) with Credit Card and you send me Factoids on my EC-address. You make $0.004 profit.

But: You also have the risk of a volatile Factoid-market.

Sending FCT to an EC address automatically converts them?

Yes. 
hero member
Activity: 854
Merit: 1000
I've been pondering the Factom system.

Entry Credits can't be transferred.  As such, do I understand correctly that in order to run an Entry Credit Store, you will need a Federated Server?

What happens if you have an Entry Credit Store and someone buys 1000 EC's off you and your Federated Server gets demoted?  That person loses their money, right?

Are there any plans to certify Entry Credit Stores somehow so people can avoid being scammed by fake stores?

You don't need a federated server.

Example:

You are allowed to run such a site (no clue about the law) and you open a EC-store. You buy Factoids on Poloniex and I'm your first customer. I pay $0.005 per EC (let's say for 1000 EC's) with Credit Card and you send me Factoids on my EC-address. You make $0.004 profit.

But: You also have the risk of a volatile Factoid-market.

Sending FCT to an EC address automatically converts them?
legendary
Activity: 1960
Merit: 1128
I've been pondering the Factom system.

Entry Credits can't be transferred.  As such, do I understand correctly that in order to run an Entry Credit Store, you will need a Federated Server?

What happens if you have an Entry Credit Store and someone buys 1000 EC's off you and your Federated Server gets demoted?  That person loses their money, right?

Are there any plans to certify Entry Credit Stores somehow so people can avoid being scammed by fake stores?

You don't need a federated server.

Example:

You are allowed to run such a site (no clue about the law) and you open a EC-store. You buy Factoids on Poloniex and I'm your first customer. I pay $0.005 per EC (let's say for 1000 EC's) with Credit Card and you send me Factoids on my EC-address. You make $0.004 profit.

But: You also have the risk of a volatile Factoid-market.
legendary
Activity: 1960
Merit: 1128
Snow: “The token supply can grow (if speculators drive up the price) until the price stabilizes. And the token supply can fall (if speculators drive the price down) until the price stabilizes. But in both cases, the stable price is when the value of the token matches the money real applications are spending to buy Entry Credits in order to put data into the protocol. That is because the 73K factoids generated each year naturally trends to the value of the factoids drained from the supply to write into the protocol. If people are spending 1 million dollars to put data into factom per year, then 73K factoids should be worth 1 million dollars.”

Can you link to where this quote is please?

It's from an AMA:


https://forum.bitcoin.com/ama-ask-me-anything/i-am-paul-snow-the-architect-of-factom-and-chair-of-the-texas-bitcoin-conference-ask-me-anything-t4026.html

Actually this is an interesting quote to zero in on. We've been discussing off-and-on what a "natural" or fair price for FCT would be in the future, and this quote seems to provide a strong clue. Given the price-stabilization that Paul Snow mentions, it should, I believe, be possible to deduce a FCT price based on various level of system-use. Fsr I'm having a tough time wrapping my mind around how exactly to do this ... but if someone could provide test examples of moderate/reasonable Factom use in the future—based on current usage trends, etc.—it should be possible to reverse-engineer what the prices of FCT would be in order to maintain the price stability Snow mentions. So with light / medium / heavy use over the next, say 5-10 years, FCT should be worth X / Y / or Z. Thanks to anyone able to give this calculation a crack.

I just see that P. Snow made a mistake there: "73K factoids generated each year" - it's per month. Or I'm wrong and that would be very good news. ;-)


What you say is interesting, but it's not really possible. It would be not much more than guessing, because we have too many unknown quantities:

We can't know how many and "who" (which companies) will use it - and if: when and how.


Plus: I'm not sure if that is still the plan, but it's possible that the federated server will set the price for EC's. I hope that will happen, because they will be paid with Factoids, so they have incentive to do it for a high value of FCT's the best they can. A high price would burn more Factoids per Entry, but the demand would decrease. A too high price would make some use-cases too expensive. And the other way around. So, I believe they will react on how Factom will be used.

Besides all what we can't know, that's an additional unknown quantity.


My personal calculation is very simple:

We already have proof that companies and institutions are already interested, even before the system is totally finished. We know that the team is very professional, skilled and also experienced, diligent and honest. We also know that they are well funded. We know that institutional Investors already tripled the value of Factom-Inc. to about $30 Mio.

What I expect is not just a continuing tendency of what we've seen since launch or at least January.

I believe the more objective parts will:

- continue
- increase
- accelerate


.....if there won't be a "black-swan-event".


In general I don't think much about objective prices, but about impact on the market (what I've often called psychological market). Or more precise: Even objective informations, like "partnership with x and demand of y", will only move the price because of impact on the market.

That's why we have to think about:

- attention (quantity)
- knowledge (quality: how informations are interpreted)

- identification
- image
- etc.

...but also: irrationality.


And there is one fact: Buying and selling and demand will always be under the control of humans. Buying and selling will always be based on decisions based on interpretations. That's true for EC-demand, and that's true for the price of Factoids. And that would even be true if we would exclude speculation on exchanges because if we think about Factoids it will be always about a Factoid-market. Objectivity is in fact an illusion and will always be.

That's why I play this game more with a focus on "humans". Teams on one side, their ideas, how they act, how they communicate etc. On the other side reactions of "the market" (we all). And one thing is safe: The team won't stop. Factom won't fall asleep. Attention will increase and knowledge will increase. And that will happen in a general growing market.

My simple conclusion is: Only a catastrophic event could prevent higher prices. The rest is about time.
hero member
Activity: 854
Merit: 1000
I've been pondering the Factom system.

Entry Credits can't be transferred.  As such, do I understand correctly that in order to run an Entry Credit Store, you will need a Federated Server?

What happens if you have an Entry Credit Store and someone buys 1000 EC's off you and your Federated Server gets demoted?  That person loses their money, right?

Are there any plans to certify Entry Credit Stores somehow so people can avoid being scammed by fake stores?
full member
Activity: 196
Merit: 100
FCT sellers/shorts only look at the chart and determine their actions on price alone without considering marketcap while trading.


Not true, some just open a short they never intend to go through with to make the sell side look heavier so that they can accumulate easier.

Example: Open a short at 0.005, put up a few buy walls at 0.0035.
legendary
Activity: 924
Merit: 1000
hero member
Activity: 984
Merit: 1000
Thx for posting the AMA link!

In this context I want to point at this answer:

Quote from: PaulSnow
Factoids are converted to Entry Credits. Then Entry Credits are used to access the protocol. So the accounting is ... odd. Cannot say how many are consumed per day, but we can give you the starting supply and the current supply. You can go to explorer (http://explorer.factom.org) and click the directory blocks to see the Factoid supply at that time.

We started at 8759968.58633800

Currently we are at 8754356.73644919

So about 5000 factoids have been used to buy Entry Credits since September when the protocol went live. At about 10 cents a Factoid, that's about $500.

Right now about 23 to 26 entries are written into Factom every 10 minutes. Most of that is a data feed which documents the Bitcoin and Factoid prices from Poloniex. We are adding more data streams to Factom. These can be used by applications to access market data, and potentially be used as oracles in smart contracts.
I searched within the factom explorer for the current supply but wasn't able to find it. Anyone else?
That quote is from Dec. 16th, 2015. So probably the supply is now much lower because the amount of entry credits used since then grew a lot.

According to cmc it is now at 8,753,219 FCT which is only 1137 less. Maybe this info is outdated?
sr. member
Activity: 385
Merit: 250
Snow: “The token supply can grow (if speculators drive up the price) until the price stabilizes. And the token supply can fall (if speculators drive the price down) until the price stabilizes. But in both cases, the stable price is when the value of the token matches the money real applications are spending to buy Entry Credits in order to put data into the protocol. That is because the 73K factoids generated each year naturally trends to the value of the factoids drained from the supply to write into the protocol. If people are spending 1 million dollars to put data into factom per year, then 73K factoids should be worth 1 million dollars.”

Can you link to where this quote is please?

It's from an AMA:


https://forum.bitcoin.com/ama-ask-me-anything/i-am-paul-snow-the-architect-of-factom-and-chair-of-the-texas-bitcoin-conference-ask-me-anything-t4026.html

Actually this is an interesting quote to zero in on. We've been discussing off-and-on what a "natural" or fair price for FCT would be in the future, and this quote seems to provide a strong clue. Given the price-stabilization that Paul Snow mentions, it should, I believe, be possible to deduce a FCT price based on various level of system-use. Fsr I'm having a tough time wrapping my mind around how exactly to do this ... but if someone could provide test examples of moderate/reasonable Factom use in the future—based on current usage trends, etc.—it should be possible to reverse-engineer what the prices of FCT would be in order to maintain the price stability Snow mentions. So with light / medium / heavy use over the next, say 5-10 years, FCT should be worth X / Y / or Z. Thanks to anyone able to give this calculation a crack.
legendary
Activity: 1960
Merit: 1128
Snow: “The token supply can grow (if speculators drive up the price) until the price stabilizes. And the token supply can fall (if speculators drive the price down) until the price stabilizes. But in both cases, the stable price is when the value of the token matches the money real applications are spending to buy Entry Credits in order to put data into the protocol. That is because the 73K factoids generated each year naturally trends to the value of the factoids drained from the supply to write into the protocol. If people are spending 1 million dollars to put data into factom per year, then 73K factoids should be worth 1 million dollars.”

Can you link to where this quote is please?

It's from an AMA:


https://forum.bitcoin.com/ama-ask-me-anything/i-am-paul-snow-the-architect-of-factom-and-chair-of-the-texas-bitcoin-conference-ask-me-anything-t4026.html
hero member
Activity: 854
Merit: 1000
Snow: “The token supply can grow (if speculators drive up the price) until the price stabilizes. And the token supply can fall (if speculators drive the price down) until the price stabilizes. But in both cases, the stable price is when the value of the token matches the money real applications are spending to buy Entry Credits in order to put data into the protocol. That is because the 73K factoids generated each year naturally trends to the value of the factoids drained from the supply to write into the protocol. If people are spending 1 million dollars to put data into factom per year, then 73K factoids should be worth 1 million dollars.”

Can you link to where this quote is please?
hero member
Activity: 1008
Merit: 505
https://steemit.com/factom/@monty-kirkman/several-reasons-for-factoms-rise

SEVERAL REASONS FOR FACTOMS RISE

Bank to the future had been over subscribed https://bnktothefuture.com/pitches/factom-series-a

The China effect https://twitter.com/factom/status/763388834096812033

Valuation 17% of the company is being sold for $5,000,000, so that's a 30 million dollar company valuation. It's over subscribed and still has three days to go, so it's valuation is higher than 30 million as of right now. There are currently 8,753,219 factoids in existence. They are currently worth $2.45 each (up from $1.50 a few days ago). That put the value at 21 Million. So on this basis, it's still undervalued. You need factoids to buy factoid entry credits which are required to put immutable hash entries into the blockchain. So, without even paying a premium for this company each factoid is worth $3.42 (30 mil devided by 8,753,219 factoids). Once the revenue streams start flowing in from the continuous sale of credits, the value of this company will rise significantly. In 12 months we'll look back at this price and wish we bought more when they were under $5 a factoid. Yes, I'm an investor, and I'll keep buying as much as I can every pay day until it hits at least $3.42 per factoid.

Business Model There is just no doubt that factom is carving out a serious business for itself with this immutable record keeping. The overlooked gem to this project is the 'revenue stream' being created. Millions and millions of records will be written to the block chain by every major company in the world that wants to prove that it is 100% honest with it's record keeping. This will include all governments. At one tenth of one cent per record entry this makes makes it super cheap for any business to participate using factom. In addition, at this price it will make it near impossible for a competitor to come along and undercut them and then try to make money. This is a very smart move on factoms part. They are clearly way out in front with this technology. By the time another competitor starts up, this company will have already (provably) recorded multi millions of transactions to the blockchain. This technology is here to stay and is going to literally clean up the crooks who currently fiddle with internal company/government records. Well done factom, I'm most certainly one of the biggest cheerleaders for this type of technology finally hitting the planet.
hero member
Activity: 1008
Merit: 505
http://www.londonstockexchange.com/exchange/news/market-news/market-news-detail/other/12927809.html

COINSILIUM GROUP LIMITED

(“Coinsilium” or the “Group")

FACTOM ANNOUNCES COLLABORATION WITH BIG DATA INFORMATION PLATFORM DATAYES

Coinsilium Group Limited (ISDX:COIN), the ISDX quoted blockchain technology investment and development company, is pleased to announce that its investee Factom Inc. (Factom) a company in which Coinsilium has a 1.9% equity interest in,  has announced that in collaboration with DataYes they have published the pricing data for the 3,000 most valuable Chinese stocks onto the Factom blockchain.

Factom builds blockchain technology tools that maintain a permanent, time-stamped record of data.  This is used in enterprise software to simplify records management, record business processes, and address security, compliance and governance issues.

Following the announcement, David Johnston, Chairman of the Board at Factom stated, “Some of the largest stock markets and the most valuable companies in the world are in China and so we believe it’s important to make their pricing data available to developers on a blockchain.”

Coinsilium CEO Eddy Travia commented, “I am very pleased to see an additional proof of blockchain’s utility and Factom’s competencies in the world of capital markets data records. It is yet another example of Factom’s ability to find the right partners and strike deals in markets abroad.”

The Directors of Coinsilium take responsibility for this announcement.

From Factom

Factom & DataYes are pleased to announce that together they have published onto the Factom blockchain pricing data for the 3,000 most valuable Chinese stocks. Program developers now have a powerful new tool for connecting their financial applications to the new world of blockchain technology.

This massive dataset of more than 10,000,000 records and growing, offers an amazing way for applications to verify and audit the world’s financial systems with more transparency and confidence than ever before.

“Some of the largest stock markets and the most valuable companies in the world are in China and so we believe it’s important to make their pricing data available to developers on a blockchain.” Said David Johnston, Chairman of the Board at Factom

“Today the exchanges send out the ‘mark to market’ price and we record them in blockchain as a two-step process. As Blockchain technology matures in the future, all transactions will be recorded on the blockchain, and the “mark to market” stock price as we know today will be evolved to ‘mark to blockchain’ naturally in one step.” Continued George Hu, General Manager at DataYes.
full member
Activity: 129
Merit: 100
FCT sellers/shorts only look at the chart and determine their actions on price alone without considering marketcap while trading. That strategy is fine for scalping and short term flips but if you do this too much and ignore the big picture you might miss out on the real move. I recall many traders on polo flipping ETH between .002-.008 but neglected to hold for the move past .01 which would have netted a far better return than catching the tops/bottoms. Seems a bit early to call the top on this; marketcap is less than DOGE, float is 20x smaller than ETH/ETC, and the protocol is starting to exit its proof of concept stage. If you are a daytrader then sure take the trades that can make you some quick btc, but if you are taking a position based on the fundamentals of the protocol its likely a bit early to exit.
Jump to: