Thank you Copenhagen - Polkadot Decoded 2023 was an eye opening gathering! Dr. Gavin Wood’s long term commentary on Polkadot was an important call to reflect on where we are and where we are going. We look forward to continuing to take in others’ reactions to it. In my five-year history as a supporter of developers in the Polkadot ecosystem (and 24 years as a student of new technologies) it added to my conviction that this community of technologists is on the right path to delivering the next version of the internet - Web3.
I walked away thinking about the controversial launch of Amazon Web Services in 2006 and its claim to provide storage for the entire internet, “designed to make web-scale computing easier for developers.” Could this new perspective be similar, only with security as the shared offering to the world? And I’m wondering if the idea of parathreads is a lot more important than I had originally thought after reading the initial Polkadot white paper.
As a reminder, Polkadot is not meant to solve for any one use case, vertical, or technology feature, and it is not meant for a small number of people, but for all people. Resilience is a key tenet. We have often commented that web2, mostly fueled by advertising spend, has let us down because of the conflict of interest baked into its design. People consume content from creators of content through a lens of billions-of-dollars of advertising wedged between the two. This lens not only clouds the information we consume, it steals our privacy in the meantime. Polkadot aims to be resilient and to meet user expectations that the internet facilitates unbiased interaction, a place where we can transact in a highly secure way.
Dr. Wood explained that perspective shifts are crucial in technology development, almost as check-ins to ask if we are on the right path in the context of the world around us. He has experience with this, having built Ethereum as their first CTO. As a check-in on Polkadot, he quoted Marcel Proust => “The only true voyage of discovery would be not to visit strange lands but to possess other eyes” - a perspective shift.
The only true voyage of discovery, the only fountain of Eternal Youth, would be not to visit strange lands but to possess other eyes, to behold the universe through the eyes of another, of a hundred others, to behold the hundred universes that each of them beholds, that each of them is. – Marcel Proust
In the spirit of community governance, Dr. Wood introduced a proposal. Polkadot currently has slots with time leases (6 monthly, for up to two years) that can be leased by what we call parachains. The slots are "cores", and the chains are applications. In the "slots/leases model", Polkadot dedicates one core to one application, for the duration of a lease. One’s phone/computer has a CPU with probably 8-16 cores, with many more than 16 applications, because an operating system can allocate a core to one application for some time, but then allocate it to another application at another time. The number of CPU cores just limits the number of applications that can run at the same time.
Could it be the case that we should be thinking about space allocation in this big pot of underlying compute resources that we have available for applications? Might we accept that these slots are simply cores in a ubiquitous computer and that we should be thinking about “coretime” or blockspace?
Co-Founder of Polkadot, Rob Habermeier, expanded on this point as follows -
Blockspace and Coretime are two sides of the same coin.
Blockspace: the raw material created by blockchains - decentralized security and code execution
Coretime: Polkadot's measurement and allocation mechanism for blockspace on a global machine, akin to a decentralized CPU
In Gavin’s presentation, he asked us to think of these 50 parallel slots as 50 CPU cores that are continuously running in parallel (growing to 500-1000 over time.) Each Polkadot core has characteristics that can be described similar to a CPU core - bandwidth at ~1MB/s, computational power according to Geekbench 5 SC score at ~380, and latency at 6 seconds. Also, bandwidth should track hardware and go up over time.
Could it be the time to reconsider the slot auction model, and improve its agility? Could these “slots” be creating unintended and artificial barriers to developers (let’s invite tinkering!) and could these slots be leaving unused capacity on the table that could be allocated to other developers with other applications? What if we take Polkadot, in its initial design, and realize that core time can be bought and used, and cores are agile so procurement could be agile too.
And if we are thinking about these as cores, could they be used not only to secure parachains, but also to host smart contracts?
On the idea that parathreads could be more important than many in the Polkadot ecosystem originally thought (including us), we went back to Joe Petrowski’s blog post in 2019 explaining the concept. Parathreads are defined as “an idea for parachains to temporarily participate (on a block by block basis) in Polkadot security without needing to lease a dedicated parachain slot. This is done through economically sharing the scarce resource of a parachain slot among several competing resources (parathreads).” Compared to parachains, parathreads provide the same security and interoperability between chains. Throughput should be equivalent, but while parachains have a dedicated core, a parathread only takes coretime when needed. This agility comes in handy for teams without sufficient capital to lock up the required amount of DOT for two years and for teams that do not need frequent state updates. The “pay as you go” structure of parathreads, without the need for full throughput, significantly adds to the accessibility and optimization of resources for Polkadot. Dr. Wood showed us how coretime can take a similar agile approach in the future, with a bulk and instantaneous model for usage. The concept of coretime which makes Polkadot sound more like a series of parathreads, except with more predictability than a block-by-block auction.
Parathreads seemingly could be just the thing that is needed in the context of the New Perspective as described above given their inherent fungibility and their ability to meet demand as it is needed.
While this was a discussion of the long term vision, it was also helpful to read Rob Habermeier’s response with respect to the nearer team as he explained,
“In practice all the proposed ideas will be incremental - first on-demand blockspace will come, removing barriers to entry. Then new models for core-time will be implemented. And as it matures, the secondary markets and trading of blockspace/core-time will emerge.”
History often rhymes - the presentation led me back to a blog post (ironically called Perspectives!) on Amazon’s controversial March 2006 press release announcing Amazon Web Services simply as “storage for the internet, designed to make web-scale computing easier for developers.”
Blog author James Hamilton, an early employee of AWS, referred to it as “the beginning of a new era in computing.” AWS was so controversial because shareholders were wired to cheer the hoarding of high margins. As James explained, this move would have had “most enterprise IT companies calling for immediate management change. This really was different. A different supplier, a different model, a low friction provisioning path, and a fundamentally different price that starts low and falls rather than escalates over time.” Side note: the concept that transaction costs should go down, not up, as a tech platform scales was one of the reasons we were initially attracted to Polkadot instead of Ethereum, with its gas fees that would go up with usage back in 2018.
We look forward to learning more from the reactions and discussion commentaries of Polkadot developers, the heart and soul of the Polkadot ecosystem!