I am here to answer any questions people have. In showing BigMap we intended to highlight three things for folks:
1. DEVELOPER EASE - This is a concrete example of what we mean by being able to build "without databases" from the developer's POV. This is an example of how we see the developer experience actually improving from a protocol-based compute with blockchain underneath... this is very counter to people's expectations working with blockchain so we like to say this ;)
2. STORAGE COSTS - We are showing we can host data on-chain (inside the Internet Computer) with costs you would expect from a serverless compute platform. While I cannot speak for ALL projects, it would cost "hundreds of thousands or millions" to store so much data in Ethereum, while only a few dollars in the Internet Computer.
3. UPLOAD TIME - As you can see at timestamp 7:55 (https://youtu.be/ylL4cZG2TiE?t=476) in the demo, Sasa had his canister seed such a large amount of data from within the canister (for the purposes of the demo). In practice, each subnet can receive MBs per second worth of UPDATE calls so it would take minutes to upload hundreds of MBs as he showcased. This is still orders of magnitude of the days (or weeks) it would take in traditional decentralized compute.
Really enjoyed the live online event the other day. It feels like the entire IBM cloud team has left to join the DFINITY foundation in developing the Internet Computer Protocol. The medium article contains the videos on what they're building.
One is called "BIGMAP" and "BigSearch" which could be a potential google search competitor. People's thoughts?
The Sodium event was on another level, four hours flew by! Agreed, BigMap/Search is fascinating, there were so many great presentations and unique solutions. I thought Origyn's elegant sound signature, for certification was very clever.
very excited for the mainnet and the actual in production costs for storing data on the icp. the current price of 3-6 million $ for storing 1 gb of data on ethereum is a little bit to high in my opinion. lets see how many orders of magnitude they can reduce it
I am here to answer any questions people have. In showing BigMap we intended to highlight three things for folks:
1. DEVELOPER EASE - This is a concrete example of what we mean by being able to build "without databases" from the developer's POV. This is an example of how we see the developer experience actually improving from a protocol-based compute with blockchain underneath... this is very counter to people's expectations working with blockchain so we like to say this ;)
2. STORAGE COSTS - We are showing we can host data on-chain (inside the Internet Computer) with costs you would expect from a serverless compute platform. While I cannot speak for ALL projects, it would cost "hundreds of thousands or millions" to store so much data in Ethereum, while only a few dollars in the Internet Computer.
3. UPLOAD TIME - As you can see at timestamp 7:55 (https://youtu.be/ylL4cZG2TiE?t=476) in the demo, Sasa had his canister seed such a large amount of data from within the canister (for the purposes of the demo). In practice, each subnet can receive MBs per second worth of UPDATE calls so it would take minutes to upload hundreds of MBs as he showcased. This is still orders of magnitude of the days (or weeks) it would take in traditional decentralized compute.
Hope that helps!