- This system USED to be controlled by the public network. -

[theshoemaker]

The security of the Bitcoin system relies largely on the fact that it's open source.  Is there anything to stop people from taking the critical aspects of the source code and creating a functionally legit application to compete with the open source client? If not, then as with all software, the popular choice could shift overtime, to non opensource clients, causing the control over the whole Bitcoin system to shift away from the public.  
Correct?

[1713574114]

1713574114

[1713574114]

1713574114

[AnonymousBat]

Hopefully people aren't stupid enough to use a closed source client.

[imperi]

If hypothetically Apple integrated a client into their latest phone, people would use it.

[theshoemaker]

Hopefully people aren't stupid enough not to use a closed source client.

I think you mean "...stupid enough to use...".   And I think the populous is far more concerned with convenience to make correct descisions.  And they trust in that "others are doing it".

Imperi has a great point.  People are being conditioned these days to trust in platforms.  Applications are cheap "try as you like" things with far less significance.  

Heck, most people don't even know what open source is.

[imperi]

If people are rioting and throwing bricks into windows and starting fires, random bystanders often join in without even knowing the reason. It's herd mentality, extremely common, engrained in our most basic behaviors. If a closed client became slightly popular, tons of people would join in even if it feels like it's maybe not the best idea.

[rezin777]

If hypothetically Apple integrated a client into their latest phone, people would use it.

And exactly what kind of hash rate would an iPhone get? You guys realize miners are the network, right? If you want to change the network, you need the miners to go along with you.

[theboos]

Closed-source clients have the ability to send your Bitcoins where you don't want, or simply steal your wallet.dat, but can't affect the network as a whole. Even open source miners simply hash what is handed them by a server. The vulnerability lies with a powerful pool (Deepbit for example) verifying invalid blocks and splitting the chain.

[Serge]

If client has great features such as easy swapping of wallets, encoding, sorting, categorizing, accounting, billing,  etc etc I don't see why a company could not sell their closed source client software, although I would feel safer if it would go through some sort of Bitcoin certification program if that's necessary.

[AnonymousBat]

Hopefully people aren't stupid enough not to use a closed source client.

I think you mean "...stupid enough to use...".   And I think the populous is far more concerned with convenience to make correct descisions.  And they trust in that "others are doing it".

Imperi has a great point.  People are being conditioned these days to trust in platforms.  Applications are cheap "try as you like" things with far less significance.  

Heck, most people don't even know what open source is.

Yeah I fixed it before you replied :p

[Quantumplation]

The security of the Bitcoin system relies largely on the fact that it's open source.  Is there anything to stop people from taking the critical aspects of the source code and creating a functionally legit application to compete with the open source client? If not, then as with all software, the popular choice could shift overtime, to non opensource clients, causing the control over the whole Bitcoin system to shift away from the public.  
Correct?

I lost interest when you used the word "legit".

All joking aside, yes, this could be a problem.  It's a problem with any closed source application.  Internet explorer could start blocking webpages it didn't like.  It would make it vastly unpopular very quickly, but it could be done.  It's more subtle with bitcoin, as the effects might be more subtle and take longer to get noticed.  Solution: Don't use closed source clients, and hope that someone develops a really kickass open source client. =P

[ctoon6]

If you ever wanted to test a closed source client you could put it inside an emulated environment, where it would look like real btc was at stake and such, but all emulated.

[Insti]

If you ever wanted to test a closed source client you could put it inside an emulated environment, where it would look like real btc was at stake and such, but all emulated.

Good luck emulating 5000Ghash of mining power.
I guess you'd just need to be patient.

[ctoon6]

If you ever wanted to test a closed source client you could put it inside an emulated environment, where it would look like real btc was at stake and such, but all emulated.

Good luck emulating 5000Ghash of mining power.
I guess you'd just need to be patient.

you would not have to, if would be like a brand new network inside the emulated environment. one issue i just thought of is if the client were to have an address hard coded in it that it sent btc to, but even if no one ever "claimed" it yet, the address would then be "created" and you would know the client is malicious. im not sure how secure that reasoning is, but there you go.

[Quantumplation]

If you ever wanted to test a closed source client you could put it inside an emulated environment, where it would look like real btc was at stake and such, but all emulated.

Good luck emulating 5000Ghash of mining power.
I guess you'd just need to be patient.

you would not have to, if would be like a brand new network inside the emulated environment. one issue i just thought of is if the client were to have an address hard coded in it that it sent btc to, but even if no one ever "claimed" it yet, the address would then be "created" and you would know the client is malicious. im not sure how secure that reasoning is, but there you go.

There's something like this on a larger scale, called the Test Network.  They use it to test various protocol changeover situations and such.  It'd be interesting to see a project for a smaller, emulated environment on a single machine for client-testing.

[FreeMoney]

So you are worried that the public might choose closed clients resulting in the public not controlling things? And what they'll be dissatisfied and not switch clients?

Miners and merchants have the actual power anyway. People can store their coins a bunch of different ways and it doesn't matter at all as long as they are satisfied.

[ctoon6]

So you are worried that the public might choose closed clients resulting in the public not controlling things? And what they'll be dissatisfied and not switch clients?

Miners and merchants have the actual power anyway. People can store their coins a bunch of different ways and it doesn't matter at all as long as they are satisfied.

This can be compared to BT, μT easily has 40% of the market share of BT clients, and the many people also don't know anything about BT, mostly use μT. when they say where they got a file, they typically cite μT and will say they never herd of BT before. however the same can't be easily said about BTC because the name of the thing your trading is "bitcoins" and the name of the program is "bitcoin". in order to break this, you would somehow need to separate these 2 things and lead a crowd of blind people around to never know anything else, much like our good friends in north korea.

[theshoemaker]

... "bitcoins" and the name of the program is "bitcoin". in order to break this, you would somehow need to separate these 2 things...

Uh. I'm not sure where you are going with that.  Seems irrelevant.  No doubt that most closed clients would use the name bitcoin.  


I see companies putting competing clients out, and people using them even just out of curiosity to start.  A few years from now it seems probable that there could be dozens of privately owned clients and one that is greatly gaining in popularity.  Then before you know it we're headed toward centralization again, especially if the government were to get involved in the mix.

It seems that testing the worlds closed clients in emulation would be an aweful lot of work.  And successfully finding ill will would be totally dependent on the plan of the culprit and when they decided to implement the plan.  Besides if destroying confidence in the currency is part of the plan, identifying the problem in the their software may only give us the certainty to say, "yep, they're going to screw us".

On the "miners have control" point.  I'd argue that whoever controls the user experience controls the system as much as anything else.

[rebuilder]

Using a closed-source client is equivalent to running executables attached to spam. If you do it, you're asking to get robbed.

I'm pretty sure there will be so much malware targeting Bitcoin wallets that people will learn to be very wary of what they install on the computers they use for Bitcoin activities.

Well... actually I think we'll see centralized banks taking the burden of security off the shoulders of mainstream users, but the point is, the mental connection between running "nice programs" and having your coins stolen will be pretty strong by then. It's already forming.