Another crazy idea on blockchain: please tell me why that is not possible

[latzer]

So here is the point:

create a standardization of the way we program connected objects so we can use model-checking algorithm to make sure they behave as expected.

With this, everyone will be able to create its own connected object using chips or smartphones or simply a raspberry pi.
So you do not need to buy a complete device from a private company but just build it yourself, since you can simply use open-source code.

Next is using a blockchain, allow these connected objects to communicate together to share informations in order to improve their behavior. They make transactions to upload and download operational data from the blockchain

they can either just make transactions like "ok this is the way I work (using a given set of parameters), and my electricity consumption is X".
or they can be a node and then verify both a transaction and run a light model-checking tool verifying the parameters given in the  transaction:
Given a set of parameters in a transaction, the tool outputs either "yes, I can use the same parameters and my behavior is not compromised" or "no, with this new parameters there might be a bug". This already exists.
Moreover it is the job of other connected objects to ensure your own object behaves well.

Such a network could make some money saving for people using it. Currently with for instance connected radiators that optimize their behavior according to the presence or not of someone at home one can make a 30% electricity bill saving.

With such a collaborative tool to exchange secure informations (in the sense that it brings no bug) we could develop so much "green" processes without the need of any Google, Amazon or any big brother company. Using again the blockchain we could sell informations to energy companies that need so much to know about our behaviors.

More than that, and IMHO I believe it is an active field of research, we can automate communication between different type of connected objects so they work for a common purpose, like the well-being of the human owner. With those shared data we can ensure the well behavior of connected objects that are connected in the same place

sooo what do you think ? is it interesting or completely infeasible using a blockchain or useless ?

disclaimer: I am a model-checking PhD student.

[1714095882]

1714095882

[1714095882]

1714095882

[1714095882]

1714095882

[samy_voc]

Your concept is good. Not sure if it is feasible.

Always cost savings are more when the middle layers are removed. AS you Google amazon make money in ad revenue and selling user information.

Amazon on top makes money on subscription too. market There should be a marketplace in cryoto world to address this to break the monopolies of this big companies.

[latzer]

sure they are. So the idea is to sell only the informations you want to.

What part do you think is not feasible ?

Your idea is this kind of network would create a marketplace ?

[Bourbon44]

This is a very interesting idea. I can definitely see someone like Cybermiles doing this. Granted they are an ecommerce/marketplace token and blockchain, but their blockchain is designed to have network services like this. They already plan on offering lending and verification services. You should contact them or look into them: https://medium.com/cybermiles/what-is-cybermiles-a457f08fa6a7

[949miner]

Obviously that this is very crazy idea, it has been already developed by some people, but i am not sure, i think that i have see it on the announcement board of this forum, but just like i said, i am not sure about that.
If you put a lot of effort to make that project, i am sure that you are going to develop a succesfull coin.
But you should know that you are going to require a lot of funds in order to make a good marketing campaign, and of course, to hire the RIGHT developers, and not the lazy ones.
Good look with this, i hope that you can make a good project from this idea.

[noloco]

First you have the problem of providing sensitive private information on a public blockchain, and second the problem of fake or erroneous information being added by mistake or as part of an attack.

[latzer]

This is a very interesting idea. I can definitely see someone like Cybermiles doing this. Granted they are an ecommerce/marketplace token and blockchain, but their blockchain is designed to have network services like this. They already plan on offering lending and verification services. You should contact them or look into them: https://medium.com/cybermiles/what-is-cybermiles-a457f08fa6a7

thank you for you relevant answer.

I read about cybermiles, and indeed there is some similar ideas:
- the capability to do lots of transactions per second
- the identity user management

Nonetheless, I have read none about verification as I meant it; let me explain:
connected devices clearly lack security norms, especially formal specifications.
It is relevant to talk about formal specifications, since they can be considered as embedded systems.

Flaws during the design concept of embedded systems is an old enemy of the success of a project (back to 1996, Ariane 5 rocket exploded 36.7 seconds after its launch, due to a bad 64 to 16 bits conversion).

Nowadays embedded systems such as satellites or self-driving cars more than ever need a model-oriented development method.
Roughly this means formally model a system and its functionalities before coding it.
Moreover the formal model must be part of the coding-phase, in other words it is necessary to certify that the code written respects the initial model.

The idea of verification involves an open-source model-oriented way of coding combined with the ability to nodes to check on-the-fly properties through the blockchain.

[latzer]

Obviously that this is very crazy idea, it has been already developed by some people, but i am not sure, i think that i have see it on the announcement board of this forum, but just like i said, i am not sure about that.
If you put a lot of effort to make that project, i am sure that you are going to develop a succesfull coin.
But you should know that you are going to require a lot of funds in order to make a good marketing campaign, and of course, to hire the RIGHT developers, and not the lazy ones.
Good look with this, i hope that you can make a good project from this idea.

thanks let me know if you find the topic, I am really interested in reading it.

from now... it is just an idea. so a marketing campaign in what purpose ? interest people in buying connected objects or in using the framework ?

A good start could be to develop a prototype and some pieces of working code, what to you think ? I know a few people who work with solidity  and ethereum in their daily jobs

[latzer]

First you have the problem of providing sensitive private information on a public blockchain, and second the problem of fake or erroneous information being added by mistake or as part of an attack.

Here is something fundamental you pointed out.

Transactions do not include sensitive informations : it is only informations on how you use your object (some parameters) that will appear on the blockchain, and it will be needed in order to validate its good behavior according to these parameters. Let me explain:

you do have a device that is self-regulating your electricity consumption. Assume it works from 7am to 4pm and your electricity consumption is X. Your friend has a similar device, and wants to make some electricity saving. You can upload your parameters (7am to 4pm), with the formal model your device is running via a transaction. Next all other connected devices can validate your transaction + ensure your friend's device is not compromised with your parameters. This example is quite simple but it can be done with more critical and interesting devices.

This also answer to the second part of you remark: indeed, since all nodes are running light model-checking tools all values in transactions will be verified mathematically. Hence you cannot download parameters that will compromise the behavior of your object.
Moreover all devices will be identified on the blockchain so we can think of a way to ban the attacker's device.

The idea is not to upload private informations, but only informations your device has access to. This is why you need to have a complete control over your device, so you program it using open-source code