Ask HN: Should I use version control or blockchain? | Hacker News Reader

Ask HN: Should I use version control or blockchain?

10 pointsQueensGambit21 comments

I am building an app marketplace where apps can be published by anyone and can be modified/republished by anyone. Initially, I was thinking of using a version control for this. But, I am wondering if I can use blockchain for this, since it might have monetization builtin.

Whats your advice? What are the pros/cons of using VCS vs blockchain that I should know as a blockchain newbie?

View on HN

Aaarrrggghhhh blockchain.

The current purpose of blockchain is to be fashionable and to obtain funding from the ignorant masses.

Use the appropriate technology. If you want to manage versions, version control sounds about right! If you really want audit-ability, you need a significant number of independent blockchain miners (>>1). If you control the only miner (and so the source of all the blockchain), blockchain is providing no security (you could change whatever you like, and adjust the signatures to lie about it).

Also, blockchain gets very expensive - all that useless "proof of work" has to be done and paid for.

Thanks for the advice! I really don't need distributed copy of the app or miners. But, the app can be edited by anyone. So, everyone who is using the app should be able to approve the latest change. Will blockchain work for that usecase - allow strangers to edit the app and "trust" they did it correctly?

Avoid blockchain at all costs if you don't understand when and how to apply it. Unless you just want to ride the passing hype train

I am willing to learn blockchain if it works for my usecase, but not for hype. Can you please explain how "trust" is solved in blockchain and why it won't work in my case?

Agreed, avoid blockchain unless you know everything about it (and in this case you’d know your idea doesn’t make sense with a blockchain). Here’s a blog post I wrote some time ago about a similar problem which might be of interest: https://rjevski.io/blockchains-for-non-cryptocurrency-applic...

What type of blockchain do you want to understand trust about? The two major classifications would be public/permissionless (a.k.a. the crypto space) and private/permissioned (like Hyperledger)

Trust has not be "solved" by blockchain, do I trust the Bitcoin cartel, or any other deciders / developers of the other chains?

Sorry! I don't know which classification will fit my usecase. But, imagine building an application with strangers you don't trust. No one person is the owner and anyone can modify the application in their local repository and send a pull request to merge it into the master branch. Since no one owns the application, I want atleast 50% of the application collaborators to accept the pull request. This way, integrity of the application can be maintained.

Is this a good fit for one of the blockchain classification?

No. Imagine needing N/2+1 people to sign off on a PR. What happens when the project gets to 10 or 100 contributors? How long would a PR take, how many of these people would have the context to sign off?

Why are you intent on block chain?

GitHub already allows for strangers to create PRs, but lets the owners maintain control of how those PRs are approved and merged.

Blockchain only provides trust in the source of the signature and integrity of the bytes. It does nothing for trusting individuals behind the keys or code.

Centralized control is still better

They're actually identical, data-structure-wise. If you add monetization to your VCS you could defensibly call the result a blockchain.

Blockchains don't really have monetization built in, though. Only specific cryptocurrencies (e.g. Ethereum) do.

That's exactly what I thought when I read about blockchain for the first time. When GIT is not used as a distributed version control (as intended), why did they invent blockchain? Is blockchain a solution in search of a problem?

You're missing the consensus part, git lacks this.

Yeah: permissionless, p2p, mining.

I heard that blockchain is actually more in a sort of brainstorm phase right now rather than being a viable technology and that it may be until the 2030s until it becomes more reified and usable.

The reason standardized blockchain globalization won't be realized until many years from now is because of the fact that any decentralization requires a unifying synthesis of the whole that can establish a connection between any two nodes.

It's because it doesn't actually have utility. Mainstream adoption will happen when govt's and banks use it, but will be very different from Bitcoin.

Look at how Hyperledger is being used, it is the most deployed, in production, blockchain project.

Handshake is actually a good application of public, permissionless blockchain.

Re: ICANN sells .org off to for profit group

Version control may implemented through some sort of block chain. Block chain may be used for version control. You should get back to basics!

If you explain why you think block chain would be a good idea, you would probably arrive at some form of actionable specification. Then implement that in the simplest way, trying to strike a balance between minimum effort and maximum value.

If you have a tight deadline, I would suggest going with what you know i.e. version control. You can launch it first, get traction and then change it later to blockchain.

If you're storing the information yourself, and you trust your database, you don't need a blockchain.

Git IS a blockchain.

For version control? Use version control.

Reply on news.ycombinator.com