SAFE Network versus (vs) Everything (v.2)

To be honest, I don’t know. Better to wait for somebody else to make a suggestion. All I know is PARSEC was deprecated some time ago.

1 Like

4 Likes

Great work. I found this really helpful.

1 Like

Great stuff. @Sotros25 should get a tweet out on this

3 Likes

I only like these kind of tables when they are honest…So also list things the Safe Network can’t do atm, like smart contracts, searching, …

3 Likes

What’s dishonest about it. It doesn’t say Maid can do smart contracts, searching …

1 Like

PARSEC has been completely removed. PARSEC was perfect … but perfection is the enemy of the good and PARSEC made the network too slow.

1 Like

My understanding is that SN allows for “Turing Complete Dapps”. For example you could use Javascript (Turing Complete) to connect to the PUT/GET SN API making any application you desire which would encompass anything a smart contract can do and far more like Games, Web 2.0 Streaming Service, Web 3.0 IOT apps. At the current state Alpha Flemming v.x only allows for storage centric commands but future updates possibly Maxwell or Beta would allow for fully decentralized computation as well. SN is designed to be a file granular and process granular integrated decentralized secure global mono server. Decentralized Ledger Technologies (DLTs) are limited to confirmation centric programming which include the contract parties and the enforcement parties and some simple code (literally an IF THEN statement) to apply them. Sure you can make all kinds of programs on the outskirt that sit on clearnet but the DLT network itself cannot do much heavy lifting. This is why SN is not a DLT because it transcends the term altogether. If you put your smart contract, data, and surrounding app in a fully encompassing way on SN it would actually make it more secure as you would have less end cap attack surface.

1 Like

Ok, your client side app is Turing complete, but that’s totally different than what a smart contact does. Smart contacts are about distributed consensus where all parties can trustlessly verify that the contract executed correctly.

Seems further down the road than that, but hopefully not.

Not really sure where you are getting that from. Bitcoin scripting is a bit limited (although certainly not limited to ‘if then’ ), but Ethereum contacts are Turing complete (apart from gas limits) as you seemed to allude to earlier in your post. Smart contracts are also stateful, allowing additional data to be associated apart from transaction confirmations. For example an ENS resolver contract holds info on which addresses should be associated with a given domain name.

“This ability to authorize state changes without completely setting all coins in an account free, is what I mean by “rich statefulness””
https://vitalik.ca/general/2019/12/26/mvb.html

4 Likes

Safe Network won’t have smart contracts to begin with IMO, that’s a big step down the road from what is currently on the roadmap for beta.

A “Turing complete” dapp would be confusing to many in this space as they’d think it was distributed consensus as opposed to an app that was just stored in a distributed/decentalized manor on the network that you can run in your browser. And could be dangerous too - if they think there is security through decentralized consensus that isn’t actually there.

For smart contracts Turing complete isn’t the end-all be-all either, non-turing complete has significant advantages in the distributed consensus model, so IMO, I’d just not bring that up at all in the comparison.

Sorry if that sounds all negative. I love the work you @Bargo_d are doing here.

4 Likes

I’m not a tech guy but from the comments it seems like the Turing line is a bit hazy and should be deleted. Please correct me if I’m wrong.
It still remains a strong comparisons sheet

1 Like

“manner” - the other is used in “Lord of the Manor”.

2 Likes

With the CLI you can do turing complete applications now. Any language/script that can issue command line operations can work with the CLI

2 Likes

Thank You! Your response is exactly the kind of feedback I’m looking for, details matter. So Turing complete in SN is not as cool or unique as I’ve made it out to be. However, it doesn’t feel right to remove “Turing complete Dapps” section because it’s an important selling point when comparing “Clearnet Cloud Computing” like AWS. The fact that you can do general decentralized computing and with potentially greatest performance comparatively on SN should not be understated. It does seem that smart contracts are very possible on SN according to this thread:

I was thinking about adding a section called “smart contracts” but this could be redundant so maybe rename the existing section to “Turing Complete Dapps (Smart Contracts)” so it is more encompassing. One thing is for sure, I definitely need to add more checks to that section as you’ve implied many DLTs already have it.

edit: me → more (last sentence)

2 Likes

To say that SN has Smart Contracts is not a trivial thing. I’ve already listed other features which are long term and aside from brief (alpha/beta) network tests it seems that SN will not go live until everything that can be added prior will be because of the secure autonomous nature of the network. I really appreciate your feedback and I will treat this as an area to tread cautiously. Maybe put a link to my dynalist on the info-graphic and be as specific as possible.

1 Like

Just again to point out - Turing complete isn’t a requirement for smart-contracts. And it’s unclear if such a functionality (Turing-complete smart contracts) is even desirable on Safe Network. I.e - Blockstream’s non-Turing complete Simplicity language for smart contracts (or similar) on SN might be a much more efficient and predictable means of scaling smart-contracts.

So , overall I still think it would be best to separate these two ideas - smart-contracts and a Turing-complete language and to specify/clarify the difference between a smart-contract and a general purpose dApp.

While being able to run Turing-complete dApps on Safe will be trivial, smart-contracts are not and IMO 99% likely we won’t have smart-contracts on Safe Network at beta launch.

1 Like

Has anyone seen this? Roy from BSV replied to someone about maidsafe. It will be interested we we could have an educational debate here between the two communities.



2 Likes

Got it. “Turing Complete Dapps” will stay and for new section “Smart Contracts”, and instead of checkbox I’ll put “post Beta”. I may also “Post Beta” for “integrated cloud computing” while I’m at it.

2 Likes

The original tweet https://twitter.com/murphsicles/status/1383197391512293380?s=20

1 Like

It doesn’t sound like he is very open to rationale debate. It doesn’t seem like he understands the technology at all.

6 Likes