/biz/ - Business & Finance

[Anonymous No.8901987]

I think we can work out the income you could make from a node based on the number of monetised API requests that are being made nowadays and the amount they charge.
>https://nordicapis.com/the-ultimate-guide-to-pricing-your-api/
Let’s take the “hobbyist” from nordicapis.com as equivalent to a NEET running a chainlink node. That’s 688,991 calls per month. Per year it’s 8,267,892 calls. Say you charge $0.01 per call, which nordicapi reports, and is the minimum Oraclize charges. That’s $82,678.92 a year. IBM Watson charges $0.0025 per call which would be $20,669.73 per year. Docusign says for their API that "You may not exceed 1,000 API requests per account per hour". So 1000 per hour is 24,000 a day, at 1 cent per job that's $87,600 per year per node operator. With 19,000 node operators (Sergey's said that's how many interested node operators he had last year) serving 1000 API requests per hour that's a total revenue of all nodes of $1,664,400,000 per year. That’s only 0.08% of $2.2 trillion (estimated total value of API economy in 2018).

Btw, RapidApi Marketplace does 400 Billion API requests a MONTH. They have 8000 APIs available for 500,000 users right now. Apparently they have a revenue of $1.5 Million per year.

Also interesting is the fact that Mulesoft acquired ProgrammableWeb in 2013. Programmableweb surpassed 19,500 APIs recently, and Sergey said he had 19,000 interested node operators in December.

It's basically a sure thing that monetizing APIs via chainlink nodes will be very profitable IF adopted. All the data I have provided above is only for the non-blockchain world. If you believe blockchain WILL be adopted IRL then once you consider the stats presented here, only one option remains: go all in on ChainLink.

[Anonymous No.8902026]

>>8901987
if CL can acquire 3% of api economy one link will be ~850$
calculations have been done already

[Anonymous No.8902073]

In due time fellow marines

[Anonymous No.8902093]

>>8902026
No it’s $188.
Anyway. That is a different issue. That $188 figure comes from the idea that a crypto’s marketcap is equivalent to the market share of a regular company. This is something that may not be entirely true but it does make sense intuitively. The OP was about income from a node based on the the current revenue one can make from API requests nowadays.

Here’s the pasta about 3% of API economy for your convenience though:

Yes, the API economy is huge and it's growing. How huge? Well, the research consulting company Ovum says that the size of the API economy is going to be 2.2 trillion dollars during 2018. There were ~15,000 public APIs available in 2016, growing by 40 new ones per day during 2015. In 2016, Netflix alone received 5 BILLION API requests PER DAY.
>An API request is equivalent to a "job" that an oracle would do.
>For comparison, Oraclize claims to do "thousands" of jobs per day. That shows just how much room for growth this API/Smart Contract economy has.

If ChainLink can capture just 3% of the $2.2 TRILLION global API economy, that gives it a market cap of $66,000,000,000 - 66 BILLION. Consider that Sergey said there are 19,000 people interested in running a node: that's over 100% of the available public APIs in 2016, so 3% of the market is very conservative. Consider that Starbucks holds 40% of the market share of US coffee shops. We’re not even considering what % of the smart contract economy ChainLink can capture, for which I can't find much on predictions for its value or its current value.

Price per coin multiplied by number of coins determines market cap, so marketcap divided by number of coins gives us coin price. So if LINK captures 3% of the API economy, that gives us a price of $188

[Anonymous No.8902100]

Didn't know there was so much money in APIs

[Anonymous No.8902136]

>>8902093
my bad good sir, still a decent price

[Anonymous No.8902297]

>>8901987
That's a lot of IFs my dear linkie, impossible to know.

>With 19,000 node operators
I see this number being thrown around all the time. Basically this number is a meme, there won't be a need for this many nodes and Sergey got this number from the ICO where you had to check whether you want to run the node or not.
Since everyone wanted to get in, they just checked (Yes) to increase their chances (it didn't matter). On top of that, many people had multiple accounts, for example I had 13 addresses registered for the ICO so you can see how that number is misleading.
But honestly that's a good thing, since that would be way too many nodes and it would hurt the overall node revenue. Then again, when something is profitable, a lot of people will want to run a node and ruin it for everyone, especially when nodes do not require LINK. This is something they'll need to think about.

[Anonymous No.8902357]

>>8902297
Actually, if there's any ledditors that post on LINKtrader or official subreddit, maybe you could ask this question.
It would be interesting to see what are their thoughts on the network having too many nodes thus making it unprofitable because pajeets can setup unlimited amounts of nodes with 0 LINK staked.

The economy will probably figure itself out but nonetheless.

[Anonymous No.8902386]

>>8902297
the more link you hold in your node though the higher your reputation and thus your revenue
nodes with 0 link will therefore not pose too much of a threat if you are serious about it

[Anonymous No.8902410]

>>8901987
good post, OP. glad people are trying to work out the math this early.

[Anonymous No.8902440]

>>8902297
You're completely right on the node numbers. 19,000 is far inflated. I think the way to limit the nolink pajeet nodes is setting link stake requirements for higher paying api calls. I think sergey had already mentioned letting the contract parties specify that. So its in our best interest to keep link low until mainnet comes, then let value explode as to price out others of ever having enough link for the big $ contracts.

[Anonymous No.8902465]

>>8902386
Staked LINK is just one of 5 things that affect reputation. I'm sure you could have a high reputation node with no LINK at all.
It will also depend on whether the users of the network will write the contracts so that nodes will require LINK (AFAIK this will be an option to choose from).

I still don't get why couldn't they just set a minimal required stake of 1000 LINK or something. 1000 LINK is not that expensive for a network that will process critical data, it's not some shitcoin masternode bullshit.

[Anonymous No.8902550]

>>8902440
>I think the way to limit the nolink pajeet nodes is setting link stake requirements for higher paying api calls.
That's what I was thinking about. But then the question is, will the majority of API calls on the network consist of some basic low priority calls or high priority critical data?

I guess this depends on whether Sergey convinces the big players to jump onboard.

[Anonymous No.8902607]

>>8902550
i really do not think that sergey will not have thought about that

[Anonymous No.8902955]

>>8902607
>i really think that sergey has thought about that
fixed it for you

[Anonymous No.8902974]

>>8902093
What happens to that figure/the price if you factor in the 60%+ that will be out of circulation? And the FOMO from everyone company in the world wanting to buy /biz/'s tokens?

[Anonymous No.8903101]

>>8902974
then price of token goes up :^)

[Anonymous No.8903109]

>>8902955
sometimes a double negative changes the meaning of what you're trying to convey. not in this case though.