Bitcoin Core Developer says Bitcoin Too Fragile and in its Infancy

[Fearless]

I still have a feeling that bitcoins is in its early stages. The philosophy of bitcoin rising to the moon can seem true if bitcoin is infant now. Moreover BTC is spreading fast now.

[HELP.org]

What exactly are the requirements to be labeled as a "Bitcoin Core Developer"? They seem pretty minimal to me. Either that or people are just trying to lend extra credence to Hearn's words.

https://github.com/bitcoin/bitcoin/graphs/contributors

Perhaps "contributor" would be more honest.

Too true. For those who can't find him on the list, it's because he's way down at #81. Do an edit -> find "hearn"

I believe he worked mostly on the BitcoinJ implementation and he is the #1 contributor there

https://github.com/bitcoinj/bitcoinj/graphs/contributors

[Ardenyham]

Technically, the big problems with Bitcoin in its existing form are:

- Mining is too centralized.

All are technically fixable, and fixes have been proposed. All of those, though, are incompatible with the present blockchain.

Most small minners connecting to big pools, instead of p2pool or solomining. But in the GPU age most miners were in the big pools as well... Does something really changed in the last two years? I think not much...

[arxwn]

I guess infancy compared to the transactions being made.
The numbers are still too small. The security is solid but the transactions do not compare to the smallest of countries.

[hhanh00]

With due respect to Mike, the human element is what ultimately makes bitcoin resilient. No matter what current problems exist, there is a huge body of people with incentives & knowhow to keep bitcoin going.

Real world example: March 2013 chain fork recovery. https://github.com/bitcoin/bips/blob/master/bip-0050.mediawiki

Theoretical example: If the core SHA256 hashing algorithm is compromised, software cannot automatically recover from that. It will be humans who decide the best course of action for the network to take. It will be other humans who evaluate that judgement, and agree (by downloading & using new software) or disagree.

But while we figure it out, the entire system is offline and what does the economy do in the meantime?

If the software protocol (rule set) is broken or not working effectively enough then you need to stop all connections and implement a change. Bitcoin is a set of rules that connects clients together to make magic happen because all the clients know what to do and they do it all by the rules that are set up. You may have a little thing that needs to be fixed that is in the wrapper. You have to stop everyone's connection (shut down the network) put in your little fix (give everyone worldwide a new copy of the fix) and then restart everyone. If the change is small and you tested it well for every outcome then we all march along happily ever after. What if the change is big? What if the base set of rules that make your release work needs to be changed because it was broken from the start? Your little test net group most likely won't be big enough to discover every problem and everyone around the world will be running a broken Bitcoin. What the economy will do in meantime is sit static waiting for the fix. There will be issues that can't be corrected because it's to late the damage is done (think Mt Gox - malleability). But what are the draw backs of not fixing the problem, not taking the risk? Then you end up with an even bigger problem because your are shutting down an even larger economy that can have even more damage done and even more lost money. There is no logical choice other than to fix the problems now and that's not really a good choice it's just the only thing you can do. That's the problem with letting this little experiment get so large before you fix big issues.

As a dev when you work at a big company then you benefit from a group of sometimes thousands of computer scientists and software engineers working on the same problems. We have a test-net and a handful of volunteer and paid professional and semiprofessional programmers some working only part time. They have done a great job so far making sure that things stay running and fixed but they have a working disadvantage. Bitcoin has been balancing on the edge of sword since it's release. You need a lot of users to drum up interest in development and you need a lot of developers before you have a lot of users spend money.

I was just trying to point out that any suspension of service would be catastrophic for the economy. If bitcoin is used by everyone and no one can do any purchase while developers fix and deploy a new version, we would be back to the stone age. The user I was replying to didn't seem to realize that software that must run 99.999999999% of the time must be treated very differently.
Bitcoin has a long road ahead but hopefully it can grow slowly just like the internet did.

[QuestionAuthority]

With due respect to Mike, the human element is what ultimately makes bitcoin resilient. No matter what current problems exist, there is a huge body of people with incentives & knowhow to keep bitcoin going.

Real world example: March 2013 chain fork recovery. https://github.com/bitcoin/bips/blob/master/bip-0050.mediawiki

Theoretical example: If the core SHA256 hashing algorithm is compromised, software cannot automatically recover from that. It will be humans who decide the best course of action for the network to take. It will be other humans who evaluate that judgement, and agree (by downloading & using new software) or disagree.

But while we figure it out, the entire system is offline and what does the economy do in the meantime?

If the software protocol (rule set) is broken or not working effectively enough then you need to stop all connections and implement a change. Bitcoin is a set of rules that connects clients together to make magic happen because all the clients know what to do and they do it all by the rules that are set up. You may have a little thing that needs to be fixed that is in the wrapper. You have to stop everyone's connection (shut down the network) put in your little fix (give everyone worldwide a new copy of the fix) and then restart everyone. If the change is small and you tested it well for every outcome then we all march along happily ever after. What if the change is big? What if the base set of rules that make your release work needs to be changed because it was broken from the start? Your little test net group most likely won't be big enough to discover every problem and everyone around the world will be running a broken Bitcoin. What the economy will do in meantime is sit static waiting for the fix. There will be issues that can't be corrected because it's to late the damage is done (think Mt Gox - malleability). But what are the draw backs of not fixing the problem, not taking the risk? Then you end up with an even bigger problem because your are shutting down an even larger economy that can have even more damage done and even more lost money. There is no logical choice other than to fix the problems now and that's not really a good choice it's just the only thing you can do. That's the problem with letting this little experiment get so large before you fix big issues.

As a dev when you work at a big company then you benefit from a group of sometimes thousands of computer scientists and software engineers working on the same problems. We have a test-net and a handful of volunteer and paid professional and semiprofessional programmers some working only part time. They have done a great job so far making sure that things stay running and fixed but they have a working disadvantage. Bitcoin has been balancing on the edge of sword since it's release. You need a lot of users to drum up interest in development and you need a lot of developers before you have a lot of users spend money.

I was just trying to point out that any suspension of service would be catastrophic for the economy. If bitcoin is used by everyone and no one can do any purchase while developers fix and deploy a new version, we would be back to the stone age. The user I was replying to didn't seem to realize that software that must run 99.999999999% of the time must be treated very differently.
Bitcoin has a long road ahead but hopefully it can grow slowly just like the internet did.

If done correctly and soon I think even major changes don't have to be catastrophic for the economy. The way changes are implemented determines the effect. It's better to know about a vulnerability and fix it now than to recover from the exploit later. Some things will always need to be fixed. Isn't it better to tackle the large problems now rather than wait until the economy is even larger?

All Hearn is really saying is that Bitcoin needs more developers. I'm saying it needs more professional paid full time developers and fewer free time hobby developers. Of course he's not going to say that because it's insulting to his colleagues.

[master-P]

With due respect to Mike, the human element is what ultimately makes bitcoin resilient. No matter what current problems exist, there is a huge body of people with incentives & knowhow to keep bitcoin going.

Real world example: March 2013 chain fork recovery. https://github.com/bitcoin/bips/blob/master/bip-0050.mediawiki

Theoretical example: If the core SHA256 hashing algorithm is compromised, software cannot automatically recover from that. It will be humans who decide the best course of action for the network to take. It will be other humans who evaluate that judgement, and agree (by downloading & using new software) or disagree.

But while we figure it out, the entire system is offline and what does the economy do in the meantime?

If the software protocol (rule set) is broken or not working effectively enough then you need to stop all connections and implement a change. Bitcoin is a set of rules that connects clients together to make magic happen because all the clients know what to do and they do it all by the rules that are set up. You may have a little thing that needs to be fixed that is in the wrapper. You have to stop everyone's connection (shut down the network) put in your little fix (give everyone worldwide a new copy of the fix) and then restart everyone. If the change is small and you tested it well for every outcome then we all march along happily ever after. What if the change is big? What if the base set of rules that make your release work needs to be changed because it was broken from the start? Your little test net group most likely won't be big enough to discover every problem and everyone around the world will be running a broken Bitcoin. What the economy will do in meantime is sit static waiting for the fix. There will be issues that can't be corrected because it's to late the damage is done (think Mt Gox - malleability). But what are the draw backs of not fixing the problem, not taking the risk? Then you end up with an even bigger problem because your are shutting down an even larger economy that can have even more damage done and even more lost money. There is no logical choice other than to fix the problems now and that's not really a good choice it's just the only thing you can do. That's the problem with letting this little experiment get so large before you fix big issues.

As a dev when you work at a big company then you benefit from a group of sometimes thousands of computer scientists and software engineers working on the same problems. We have a test-net and a handful of volunteer and paid professional and semiprofessional programmers some working only part time. They have done a great job so far making sure that things stay running and fixed but they have a working disadvantage. Bitcoin has been balancing on the edge of sword since it's release. You need a lot of users to drum up interest in development and you need a lot of developers before you have a lot of users spend money.

I was just trying to point out that any suspension of service would be catastrophic for the economy. If bitcoin is used by everyone and no one can do any purchase while developers fix and deploy a new version, we would be back to the stone age. The user I was replying to didn't seem to realize that software that must run 99.999999999% of the time must be treated very differently.
Bitcoin has a long road ahead but hopefully it can grow slowly just like the internet did.

It is not actually possible to "turn off" or "disable" the bitcoin software. That is not a feature of bitcoin. The closest thing to this that could happen is to recommend that users not spend any of their bitcoin. This did happen somewhat when Gox announced that they were suspending withdrawals due to malleability as other exchanges followed, however mining still very much continued and people that had bitcoin in wallets they controlled personally still were able to and did spent their bitcoin.  

[hhanh00]

It is not actually possible to "turn off" or "disable" the bitcoin software. That is not a feature of bitcoin. The closest thing to this that could happen is to recommend that users not spend any of their bitcoin. This did happen somewhat when Gox announced that they were suspending withdrawals due to malleability as other exchanges followed, however mining still very much continued and people that had bitcoin in wallets they controlled personally still were able to and did spent their bitcoin.   

Theorically, not possible. In practice if the software crashes or worse produces unexpected results it will be down.

[hhanh00]

If done correctly and soon I think even major changes don't have to be catastrophic for the economy. The way changes are implemented determines the effect. It's better to know about a vulnerability and fix it now than to recover from the exploit later. Some things will always need to be fixed. Isn't it better to tackle the large problems now rather than wait until the economy is even larger?

All Hearn is really saying is that Bitcoin needs more developers. I'm saying it needs more professional paid full time developers and fewer free time hobby developers. Of course he's not going to say that because it's insulting to his colleagues.

I was replying to someone who had a fairly casual approach to bug fixing. If it breaks, no worry we'll fix it.

Whether bitcoin needs full time developpers or hobbyists is open to discussion but we need to a way to rollback. I'm sure that everyone in the software business will agree that using production to test out high scale deployments is bad idea. Doing so without rollback is crazy. Now the economy is already pretty large but still minuscule compared to the real world yet we are not generally concerned about the product stability.

The topic is "Bitcoin Core Developer says Bitcoin Too Fragile and in its Infancy". Immediately some people jumped and called him ignorant, biased, etc. Ready to put his head on a pike for blasphemy.
In my opinion, if we want to have it reach out to everyone and be a serious competitor to Fiat or even VISA it is too fragile and in its infancy. This is not a bad thing. Everything has to start small. It's a serious problem if we think it's ready when it's not.

[snappa4ever]

It is not actually possible to "turn off" or "disable" the bitcoin software. That is not a feature of bitcoin. The closest thing to this that could happen is to recommend that users not spend any of their bitcoin. This did happen somewhat when Gox announced that they were suspending withdrawals due to malleability as other exchanges followed, however mining still very much continued and people that had bitcoin in wallets they controlled personally still were able to and did spent their bitcoin.   

Theorically, not possible. In practice if the software crashes or worse produces unexpected results it will be down.

Everyone who is running QT has the software running independently of eachother. if the software crashes on one person's computer, it will not necessarily crash on someone else's computer. Also the remedy to software crashing like this would be to simply restart either the software or the computing running the software.

[marcus_of_augustus]

Hearn the "visionary" is much over-rated methinks, we shall see.

Got to admire him for having a go with his Lighthouse venture, how is that going btw?

Hint: guys who know about networks and build protocols don't do java ... except maybe for a hobby.

[digitalindustry]

Jeebus. Not this again.

the fact is that Bitcoin has no flaws, the only implementations it needs are the Global ID and the Central Authority

'Bitcoin is perfect! All it needs is to abandon its most essential characteristic - its decentralized nature!'

 

Moving on...

Devs develop. It's kinda 'what they do'. If the code was in an ideal state, development would cease. It would no longer have developers. Yeah, that's a tautology. But is seems that some must be told this essential fact.

Of course the primary developers see room for improvement. Otherwise it wouldn't need devs.

ahh and it kind of has in case you haven't noticed ...

which means its perfect as it is , but now the monopoly is breaking down, so if Bitcoin loses the initiative and drops the ball with the monopoly kiss the price valuation goodbye.

so thus a central "price maker" authority is needed and this will require an essential "verification system"

[lovegood]

Every time I hear this story I wonder.. Why aren't big vested partners -- like Coinbase, BitPay, etc. -- in the bitcoin space are not doing more to address development.

I totally agree. All those huge, richly-backed companies should be providing multiple core developers to work on Bitcoin. It's not fair that they're riding on the shoulders of a small number of core developers who are basically doing all the grunt work for free.

[Robin_Good]

Every time I hear this story I wonder.. Why aren't big vested partners -- like Coinbase, BitPay, etc. -- in the bitcoin space are not doing more to address development.

I totally agree. All those huge, richly-backed companies should be providing multiple core developers to work on Bitcoin. It's not fair that they're riding on the shoulders of a small number of core developers who are basically doing all the grunt work for free.

Weee! Let's talk more about what other people should do, while we spend time browsing forum

[unexecuted]

Every time I hear this story I wonder.. Why aren't big vested partners -- like Coinbase, BitPay, etc. -- in the bitcoin space are not doing more to address development.

I totally agree. All those huge, richly-backed companies should be providing multiple core developers to work on Bitcoin. It's not fair that they're riding on the shoulders of a small number of core developers who are basically doing all the grunt work for free.

I have a feeling most of the core developers are early adopters and have a substantial amount of Bitcoin already, so they are incentivized to contribute and thus increase the value of their holdings. You don't just quit a cushy job at Google out of compassion. Also, throwing money at a problem does not equate to solving it.

[CoolBliss]

Mike Hearn is an incredibly talented and creative dev. He truly gets the wider reaching applications of the blockchain and has added tremendous value to Bitcoin. I find his level headedness refreshing and necessary. You don't have to agree with everything he says. Give the guy a break and more importantly some fucking respect. /rant

[jbreher]

ahh and it kind of has in case you haven't noticed ...

Nonsense. Looking at GitHub/bitcoin, I see that pull request 4839 was merged a mere 2 hours ago.

which means its perfect as it is , but now the monopoly is breaking down, so if Bitcoin loses the initiative and drops the ball with the monopoly kiss the price valuation goodbye.

What monopoly are you speaking of? Bitcoin in relation to other cryptos? That natural (not enforced) monopoly is trending greater all the time. Do you know how natural monopolies come about? By the monopolizer being vastly superior as compared to all the wannabe contenders.

so thus a central "price maker" authority is needed and this will require an essential "verification system"

Ha. Haha. Hahahahahaha. Perhaps you have not noticed, but the community is unified in the principle that your 'centralized authority' model is A Bad Idea. You have scarcely more chance of convincing otherwise than you do of cracking satoshi's private key on a 1998 era palm pilot.

As others have pointed out: rather than tilting at this windmill, your time would be more productively spent in creating a spinoff alt with the centralization you desire.

[BittBurger]

Core developers like Hearn, Andresen, Garzik and Maxwell are sane voices whose words of realism drown in the noise spewed by the bulltards. Core developers actually seem to understand the inherent limitations of Bitcoin, and the fact that we're a very long way off (5+ years) from mainstream adoption, if it ever starts.

Nonsense.  

Its hilarious to me that they think mainstream adoption is going to come first, and *THEN* they will add the necessary features to support it.   Smh ....

It is now that we need the features for the products 5 years from now.  Only a developer would use the logic that we can put these enhancements off 5 years.   That is suicide.  Any idiot with an ounce of knowledge of product development, marketing, and business trends knows:  You create it now.   Always.

You wait 5 years for mainstream adoption and there wont BE any mainstream adoption.   Incorporate the changes now.  Give the financial industry and innovators in the Bitcoin space what they need to start creating products for 5 years from now.  Do nothing, and nothing will happen.  Bitcoin will only go as far as it can go, and then it will stop.   We have already established that consumers have zero incentive to use Bitcoin in its current form.   So how exactly does Gavin think mainstream adoption will be coming?  

Developers should not be making product development decisions.  Period.

Andreas has always said Bitcoin is fine the way it is, but it is also the first "programmable money". Capable of evolving and addressing market needs. After 2 years in this space I have zeen zilch as far as that goes.

-B-

[jbreher]

It is now that we need the features for the products 5 years from now.  Only a developer would use the logic that we can put these enhancements off 5 years.   That is suicide.  Any idiot with an ounce of knowledge of product development, marketing, and business trends knows:  You create it now.   Always.

Generally, I agree. The sooner you can add features, the better.

But what features do you think are required? The only thing that I can think of that would be an unqualified improvement is a solution to the 51% quandary. But that would not be a feature - it would be a complete re-architecting of the system. Further, it is dependent upon a breakthrough invention.

[Bigbear8]

After its launch few years before, bitcoin raised its value from beings few dollars to more than 1000 dollars by 2012. According to an investor this price variation is too fragile. I do agree that bitcoin might still be in its infancy due to much variation going on..