How to avoid development hell

189
SHARES
1.5k
VIEWS

Related articles


Cointelegraph is following the event of a completely new blockchain from inception to mainnet and past by its sequence, Contained in the Blockchain Developer’s Thoughts, written by Andrew Levine of Koinos Group.

We lately launched the third and closing model of the Koinos testnet, which is why I need to speak about one thing few tasks like to speak about: Constructing blockchains is improvement hell. On this article, I’ll clarify why and the way different builders can keep away from getting caught in it.

At first blush, constructing a blockchain doesn’t sound so laborious. A blockchain is only a mixture of well-established cryptographic primitives, which, when correctly applied, permit for the development of a ledger containing a verifiable historical past of transactions by a community. The extra decentralized the community, the extra reliable the historical past.

Blockchain “frameworks”

In an effort to make constructing new blockchains simpler, different groups have launched blockchain “frameworks” that, in principle, ought to remove the necessity for builders to fret about constructing the blockchain itself in order that they will deal with no matter distinctive options they need to construct into the blockchain. Cosmos, EOSIO and Polkadot’s Substrate are examples of such blockchain frameworks.

When our crew stopped engaged on Steem (the world’s first fee-less blockchain), our authentic intention was to leverage an current blockchain framework to construct a blockchain designed to be as accessible as attainable. We had spent 4 years refining Steem’s fee-less design and figured that, by porting that resolution into an current blockchain framework, we may ship a blockchain that was much more accessible than some other blockchain in comparatively little time.

Associated: Inside the blockchain developer’s mind: Proof-of-burn blockchain consensus

Actually fee-less and general-purpose

However we had been stunned to search out that not one of the current frameworks allowed us to create the type of actually feeless consumer expertise we had been trying to deliver to the market. We didn’t simply need to take away charges on a technical stage, we wished to empower builders to construct functions that had been free to make use of. In addition they lacked plenty of different options we believed had been required to ship a suitable developer expertise.

The facility of a general-purpose blockchain stems not from the options the blockchain engineers construct into the blockchain however from the options that builders add to that blockchain as sensible contracts. That is doubly true for a blockchain framework that ought to actually be probably the most general-purpose blockchain possible for the reason that entire thought is to permit folks to construct any type of blockchain they will think about. And but, the present frameworks did not empower us, one of the vital skilled blockchain improvement groups, in our makes an attempt to construct the blockchain we wished to construct in a number of methods.

The prevailing frameworks not solely made it not possible for builders to create free-to-use functions, however additionally they pressured builders to study new and sometimes troublesome programming languages and dramatically restricted the speed at which each functions and the blockchain itself may enhance.

Associated: Inside the blockchain developers’ mind: Building a free-to-use social DApp

Releasing builders

We wished to construct a blockchain that may free builders to construct insanely nice functions that strange folks would love to make use of. That allowed the builders to work within the programming languages they already knew and cherished (what we name “common language assist”); that allowed their functions (and the blockchain itself) to quickly evolve; and, most significantly, it allowed them to construct functions that had been free to make use of.

However to be able to construct that blockchain we first wanted a really general-purpose blockchain framework that may not solely permit us to construct the blockchain of our goals however as a pure consequence of being probably the most general-purpose framework possible, ought to permit anybody to construct the blockchain of their goals.

Koinos is that final general-purpose blockchain framework that can function the muse for the final word feeless layer 1: Koinos mainnet. The Koinos Blockchain Framework (KBF) is designed to be the best blockchain possible, containing solely these cryptographic primitives essential to assemble a blockchain and the best “system calls” to permit for the widest vary of behaviors to be added in-band (and not using a laborious fork) by the importing of a sensible contract.

Infinite upgradeability

The advantage of this design is infinite upgradeability, however the fee is that it makes getting the system calls proper all of the extra vital. Lacking a system name or designing it incorrectly would imply having to cope with an in any other case avoidable laborious fork. As a result of laborious forks are so time-consuming, political and disruptive, they’re the only greatest issue limiting a blockchain’s potential to enhance itself, which is in the end felt by each single developer whose utility is constructed on such a platform.

Now we will see how each utility and blockchain builders can get trapped in improvement hell: working in programming languages they aren’t comfy with, on platforms that drive them to include charges and that enhance at a snail’s tempo. Below these situations, even minor adjustments have extremely excessive stakes, and we haven’t even factored in these eventualities the place there are hundreds of thousands, generally billions, of {dollars} in danger.

That’s improvement hell. Because the core improvement crew behind the Steem blockchain that needed to oversee 23 laborious forks, we all know this terrain higher than nearly anybody else, which is why we had been so dedicated to banishing it, properly … to hell.

Constructing the Koinos Blockchain Framework from scratch with a completely novel microservice structure and getting it to the purpose the place we may launch variations 1 and a couple of of the testnet had been extremely difficult. However integrating the suggestions we acquired from these testnets, fixing the bugs they uncovered, and finalizing the all-important system calls had been an entire different stage. However we went by that improvement hell with the hopes of creating it in order that others gained’t ever should.

Harbinger

Testnet v3 is, subsequently, greater than “simply one other” model of the testnet. It’s the closing model of the Koinos Blockchain Framework, which is why we’ll now check with the testnet, not by a model quantity, however by the identify: Harbinger. It’s the end result of almost two years of working, figuring out and implementing all the mandatory system calls wanted to present blockchain builders the final word stage of freedom and permit their utility builders to learn from a platform that’s enhancing at a extra speedy price than some other blockchain on the market.

In fact, everytime you’re making an attempt to unravel a extremely vital downside, some stage of improvement hell is inevitable. You simply need to just be sure you don’t undertake applied sciences that increase the stakes unnecessarily and that the objective you’re striving for is price the fee.

For us, that objective shouldn’t be solely sparing blockchain builders (together with ourselves) from numerous hours spent in developer hell but additionally leveraging this new expertise to construct Koinos mainnet: the final word feeless layer 1 for empowering builders to construct insanely nice blockchain-based functions.

No extra improvement hell

The KBF goals to remove improvement hell by permitting builders to construct any blockchain they will think about just by writing sensible contracts, which is infinitely simpler than working within the blockchain code itself. And that’s precisely the place we discover ourselves in now. Whereas creating the blockchain framework was extremely difficult, now that it’s over, all that’s wanted to finish Koinos mainnet is simply two sensible contracts: one for our proof-of-burn consensus algorithm and the opposite for governance.

Not solely is writing sensible contracts far easier than blockchain improvement, however this additionally signifies that blockchain builders now get to benefit from the ever-expanding instruments which might be obtainable to sensible contract builders just like the AssemblyScript SDK one among our neighborhood members constructed. AssemblyScript doesn’t simply produce much more environment friendly sensible contracts than these written in C++ (that was a shock), it makes sensible contract improvement accessible to JavaScript builders.

For the reason that KBF turns any sensible contract developer into a possible blockchain developer, which means creating customized blockchains will now be accessible to JavaScript builders with much more languages to come back, reminiscent of Rust.

Should you’re serious about operating Harbinger, we’ve made it insanely simple to run a node so that you could be up and operating in underneath 5 minutes.

This text doesn’t comprise funding recommendation or suggestions. Each funding and buying and selling transfer includes threat, and readers ought to conduct their very own analysis when making a call.

The views, ideas and opinions expressed listed here are the creator’s alone and don’t essentially replicate or symbolize the views and opinions of Cointelegraph.

Andrew Levine is the CEO of Koinos Group, a crew of trade veterans accelerating decentralization by accessible blockchain expertise. Their foundational product is Koinos, a feeless and infinitely upgradeable blockchain with common language assist.