Jump to content

JackTheRipples

Member
  • Content Count

    497
  • Joined

  • Last visited

2 Followers

About JackTheRipples

  • Rank
    Regular

Recent Profile Visitors

The recent visitors block is disabled and is not being shown to other users.

  1. Once they cross, the signal turns brown. Like a dookie
  2. My left testicle dropped 1.25 inches in 2017, I predict it will drop another .75 inches by July 11th, 2020
  3. I'm confused, so did bg make an appearance somewhere on the internet?
  4. You da man. I'd say neither, rather you are intuitive.
  5. I agree that charts don't dictate price, for me I think they provide outlook on possibilities, and hopefully are accurate at guessing the likelihood of the possibility presented. Now, according to the "TA Rules", do we categorize this TA as unlikely/very, likely, or very likely? If that makes sense lol.
  6. https://cryptodaily.co.uk/2018/12/ripple-fractal-similarities-point-to-explosive-breakout-in-early-2019 Yippy @Nouk is the accurate TA? I am grasshopper of TA
  7. I've been trying to find the definition of "They" for a long time. Now I know, thank you deeply.
  8. Hey all, been a while since I looked at my ledger nano s and what not. I tried sending some xrp from my ledger nano s to bitstamp and it kept failing to send. Tried with and without destination tag and cant figure it out, probably something hugely simple I'm forgetting. I am prepping for moon to have my cash out plan in order lol.
  9. Yeah. This puts all of WU competitors on an even playing field. WU seems to be now in a scramble to figure out if they can find another advantage before giving in. WU will be irrelevant if they dont use xrapid. There are plenty of other appealing psp's in the sea
  10. See the quote below from David Schwartz aka @JoelKatz from a question on Quora regarding how mass xrapid adoption might affect xrp value. "It could create a significant source of additional demand. If xRapid is broadly adopted, that means that a significant number of international payments use XRP as an intermediary asset in the payment. A lot of companies need to make rapid payments throughout the world. The most dramatic example is Uber. If someone needs a few extra dollars for groceries, Uber wants them to drive for them right then. For that to work, Uber has to be able to get the money to them as soon as they finish driving. Similar situations clearly apply to companies such as AirBNB and Amazon. But even more traditional companies like Seagate currently stash millions of dollars around the world to permit them to make faster domestic payments rather than slower international ones. You can think of xRapid as converting an international payment into two domestic ones plus a movement of XRP. So what might companies do in the world where xRapid is widely used? First, they can buy XRP very cheaply, possibly even below market value. Why? Because they can just wait for somebody to make a payment where they have the currency the payment tries to deliver. They can take the XRP the sender got in exchange for the asset they’re paying with and provide the asset they’re delivering to the recipient. This is precisely what xRapid does, and companies can take advantage of it to buy XRP cheaply. Second, they can use XRP to make cheap payments. Why? Because if those payments are using xRapid, they can just use the “from XRP” half of the payment and roughly halve their costs. Again, this is precisely what xRapid does. This only works if they keep a pile of XRP around to match the times when other people need the assets they hold to the times when they need to make an outbound payment. Essentially, instead of keeping many piles of money all over the world to handle their international payments, they could just hold one pile of XRP to convert international payments into domestic ones in all the corridors where there’s sufficient deployment of xRapid. You might think that due to the system’s efficiency advantage over current schemes, those piles wouldn’t need to be as big. That’s true. But it’s possible that this can cause a Jevons paradox. A Jevons paradox is a situation where needing less of a good to accomplish some particular result means that people actually want more of that good. For example, an improvement in the efficiency of coal burning power plants that meant that you needed 15% less coal to get the same amount of electricity wouldn’t decrease coal consumption by 15% but instead would likely increase it as people built more coal-burning power plants instead of natural gas because coal would now have a competitive advantage. The same thing could happen here. If XRP becomes much more efficient for international payments, then companies that currently don’t keep piles of money around the world (because it’s too tedious and inefficient) and instead rely on correspondents for their payments (and possibly make fewer of them as well) would find it cost effective to hold XRP. In other words, as less XRP is needed to make a wider array of payments, more and more organizations would find XRP worth holding."
  11. I tend to agree with you....the thought just crossed my mind that we're now back to similar previous times in that xrp will make a move passed eth marketcap....idk when, could be a while, but I think it is coming
  12. I think @JoelKatz said it best. Important notes by him below. Also of importance is the possibility of jevons paradox. See David Schwartz comments: "It could create a significant source of additional demand. If xRapid is broadly adopted, that means that a significant number of international payments use XRP as an intermediary asset in the payment. A lot of companies need to make rapid payments throughout the world. The most dramatic example is Uber. If someone needs a few extra dollars for groceries, Uber wants them to drive for them right then. For that to work, Uber has to be able to get the money to them as soon as they finish driving. Similar situations clearly apply to companies such as AirBNB and Amazon. But even more traditional companies like Seagate currently stash millions of dollars around the world to permit them to make faster domestic payments rather than slower international ones. You can think of xRapid as converting an international payment into two domestic ones plus a movement of XRP. So what might companies do in the world where xRapid is widely used? First, they can buy XRP very cheaply, possibly even below market value. Why? Because they can just wait for somebody to make a payment where they have the currency the payment tries to deliver. They can take the XRP the sender got in exchange for the asset they’re paying with and provide the asset they’re delivering to the recipient. This is precisely what xRapid does, and companies can take advantage of it to buy XRP cheaply. Second, they can use XRP to make cheap payments. Why? Because if those payments are using xRapid, they can just use the “from XRP” half of the payment and roughly halve their costs. Again, this is precisely what xRapid does. This only works if they keep a pile of XRP around to match the times when other people need the assets they hold to the times when they need to make an outbound payment. Essentially, instead of keeping many piles of money all over the world to handle their international payments, they could just hold one pile of XRP to convert international payments into domestic ones in all the corridors where there’s sufficient deployment of xRapid. You might think that due to the system’s efficiency advantage over current schemes, those piles wouldn’t need to be as big. That’s true. But it’s possible that this can cause a Jevons paradox. A Jevons paradox is a situation where needing less of a good to accomplish some particular result means that people actually want more of that good. For example, an improvement in the efficiency of coal burning power plants that meant that you needed 15% less coal to get the same amount of electricity wouldn’t decrease coal consumption by 15% but instead would likely increase it as people built more coal-burning power plants instead of natural gas because coal would now have a competitive advantage. The same thing could happen here. If XRP becomes much more efficient for international payments, then companies that currently don’t keep piles of money around the world (because it’s too tedious and inefficient) and instead rely on correspondents for their payments (and possibly make fewer of them as well) would find it cost effective to hold XRP. In other words, as less XRP is needed to make a wider array of payments, more and more organizations would find XRP worth holding."
  13. Anybody else getting this in the /rippled reddit page?
×
×
  • Create New...