Topic: I hate Coinbase - What would it take to clone it? #opensource/#crowdsource (Read 2018 times)

Don't underestimate the difficulty in replicating Coinbase, coding a site and back-end is the easy part. With the number of users they're signing up each month accelerating like it is, it's actually surprising that there's any possibility of individualized support at all. These are just growing pains - they're eclipsing every other service out there by orders of magnitude in terms of user base.

Anyone who remembers the Paypal launch has seen this movie before. They'll undoubtedly sort out most of their issues soon, but there's no question they've already captured the first-mover position - just like Paypal, Ebay, Amazon, Quicken etc etc have in their respective niches. Lots of copy-cats have all fallen by the wayside in each of those categories over the years.

If you don't run into problems with verification or have API coding issues (i.e. if you're just a typical user with a linked bank account) Coinbase is actually an amazingly easy, inexpensive and secure way to buy, keep and sell bitcoin.

After all this campaign against coinbase, your idea starting to make sense.

Well that is shitty.  I can't think of a bitcoin service that is good at support though, except for this forum, quite honestly...

Centralization.  It's a motherfucker.

Yup, the idea is not unique, and frankly, I have no desire to do it... But if coinbase wasn't the only game in town, there is no way I would stay with them.   I currently have a support ticket open with them with unauthorized transactions that hasnt' been acknowledged for almost 2 days.  That's just unacceptable.

I'm not knocking your idea, but coinbase does precisely this, and provides a secure wallet.

Yeah, there are proven and secure ways of doing that.  Honestly though, I wouldn't provide a wallet service.  It is one of the highest risks aspects of running a BTC business.  The thing the BTC community needs the most IMHO is a trusted and efficient way to convert fiat-> btc and btc -> fiat. 

Something that's simple that could take btcs, pay the customer, move the btc to a market where you'd clear.  On the flip side, keep limited btcs in a hot wallet, let customer buy btc, and send directly to an address they provide.  There are plenty of wallet solutions out there already.

Their customer service is horrid.  Their verification process is iffy at best.  Their testing is non existent.

I'm just curious as to why the sentiment is that coinbase blows.  Coinbase is the only reliable way to convert fiat->BTC for Americans.  All other companies have failed dramatically in some regard.

Yes coinbase blows.

This is a great idea.  My partner and I are in the process of setting up a Coinebase alt. along with an escrow service.  Both of us are currently in New York working at an investment bank.  I have personally been mining both Litecoin and Bitcoin.

The idea would be to setup the company in Singapore (non-US) or someplace else where we have a good long standing relationship with someone tied with the government.

We could get this funded relatively quickly within the community then tap some international $.

PM me if of interest in discussing further.   Would be great to get a team together.

Both of us have been through the experience of starting, funding, and running a company before.

 

I think the government's tone today is pretty positive, and easy going compared to what I expected from the US. I would say read up on it some but otherwise, just launch your product and learn to meander through the regulatory stuff as it comes.  

Whatever coinbase is doing is working fine.  Banks should be eager to work with a startup that emulates coinbase - banks are making money doing this.

If I were going to head up a project like coinbase, I wouldn't loose sleep over FinCen or bank regulations, I'd worry about securing bitcoins actually.  Coinbase goes through some pretty involved procedures of moving coins offline AFAIK to keep things secure.  Developing a strategy for that would be hard if you ask me.  I'd be pretty nervous holding a lot of other peoples coins in an exchange that is mentored by fiat powers.

Well because coinbase allows me to buy BTC at a market price instantly with 1 click and own them to send elsewhere.  I can remove the coins purchased immediately and expect an ACH transfer days later.  That's a nice thing.  Localbitcoins requires me to go outside, and leave my computer, and talk to humans.  And it requires cash, and a fee.  And uncertainty due to fluctuations.  The inability to buy opportunistically.  Just dont keep coins on coinbase if you don't trust internet wallets -  I don't leave them there I send them out as soon as I buy them.  Coinbase also lets me transfer coins in and sell them with zero confirms.  Pretty good service I think, really.

I just don't trust web wallets anymore after inputs.io . Not that I am one of the victims, it's that it's never safe. If you really want to buy/sell bit coins, why not localbitcoins.com ?

Web wallet is too risky in my opinion...

They've spent as much money as anyone in the bitcoin space registering across the country to legally handle the money.

I would look there before worry about how to code the interface itself.

The development part is easy... its all the legal and banking stuff that I don't know about, nor do i think a lot of people do ;(  I guess there's a reason why they get 25M in VC.

Start with legal part. Read Liberty Reserve case before doing anything US-related.

It's going to be tough, but even if you do get the development of it done you're going to have a really difficult time with the legal part. I like your vision, though.

Well, on top of this online wallets are not good for the ecosystem at all. Copying online wallets is not a good solution, and takes capital, legal integration etc. I would say 10'000$ and 3 months of work? As soon as you have bank integration you have centralization.

I am probably gonna code a Simple web wallet that can deposit and manual withdraw btc with API's/IPN's for merchants. I don't really care to involve fiat though cause then it just makes doing a web based wallet that much harder. Cloning Coinbase is doable though, would probably take me 2+ months but I have no interest in a direct clone.

So i've been using Coinbase as a clearing house for the last few months and I absolutely hate their lack of testing and support.  I'm not talking about silly verification bugs, or can't buy bugs.  I'm talking about serious APIs where the sell api was actually BUYING or 503s being returned and the actual call succeeding.   And don't get me started on their customer support which has no SLA of any type and you can go for weeks without a response.

Enough of a rant - my question is, how difficult would it be to clone its offerings.  The way I look at it we'd need:

* Development: frontend - this is pretty trivial.  There's only about 20 pages.
* Development: backend - little more complicated, but not by much.  Wallet creation and management has been done before, and frankly, I don't think its a huge feature on the site.  I think the ability to buy btcs for end users and the quick conversation of btc to fiat are its biggest assets.
* Development: ACH integration - there are for pay APIs that allow for ACH setups
* Legal: FINCEN registration - Know nothing about this. IANAL.
* Legal: User verification/Bank Validation - Related to above.
* Clearing house: I don't know how they solve this, but we'd need an exchange where we could covert fiat to btc and vice versa in a trustworthy way (I don't consider MTGox trustworth)
* Capital: Obviously there would need to be a float capital to handle transactions, but lets assume this is not a barrier to entry.

There would need to be a CS component of this that could be supported with using the minimal profit that would exist. I guess this would be the hardest part since the system integrity would have to be protected. 

Is anyone interested?  Thoughts?