Getting to Granada
Hear Perry Metzger of the Tezos Foundation give a behind-the-scenes glimpse into Tezos upgrades, including the answer to the age-old question: how are the names decided for each proposal? You’ll also get a look into the proposed features in the latest Tezos upgrade proposal, Granada.
This week, we were joined by Perry Metzger of the Tezos Foundation for a deep, thoughtful look behind some of the inner-workings of Tezos development, namely the evolving philosophy and strategy behind Tezos upgrades.
With the latest upgrade proposal, Granada, currently in the “exploration period,” Perry shared his insights into how some of the included features function, as well as how they help lay the groundwork for future improvements down the line.
Stream this Session to get a behind-the-scenes glimpse into Tezos upgrades, including the answer to the age-old question: how are the names decided for each proposal? You’ll also learn more about Perry’s thoughts on the features that could occur in the future and just how important the developers’ tireless work is to help ensure the continued growth of the network.
Listen
Session Notes
0:13
TQ Tezos’ Co-Founder and President Alison Mangiero had a moderated discussion with Perry Metzger of the Tezos Foundation
1:00
AM: Why don’t you tell us a little bit about yourself?
1:12
PM: Early experience with computers, education, encounters with functional programming, cryptography.
3:06
PM: “I’ve always been a long-time fan of Tezos.” Work with the Tezos Foundation.
3:54
AM: How has the development of the core protocol changed in the last year?
4:25
PM: Technical teams in the ecosystem. “Every time there’s an injection opportunity (about every 10 weeks) we put out a new version of the protocol.”
5:11
AM: Let’s start off with talking about prior releases. Delphi marked a change from the past - tell us about that.
5:48
PM: Change in the practice from waiting for all features to be ready for the next proposal to targeting specific features and more regular rollouts of upgrades - beginning with Delphi.
9:23
AM: Delphi happens. You decide to want to shift to these regularly scheduled releases. What exactly happens after that?
9:38
PM: Immediately began working on Edo. Discussion of its features. “The great flexibility of Tezos is that we have this evolvability.”
12:48
AM: There were 2 versions of Florence. One with baking accounts, one without. And baking accounts were dropped, right?
13:23
PM: Hoping to get baking accounts back. Explanation of baking accounts.
15:43
AM: What exactly is in Granada?
16:40
PM: Significant reduction in gas consumption. Exploration of features.
18:30
AM: How exactly are these things named?
19:00
PM: The people who help build it get the privilege of naming it. Importance of engineers and developers.
20:33
AM: What might be in the “H” proposal?
21:08
PM: Features hoped to be included (not guaranteed) like Tenderbake
25:36
AM: How much of the development process is driven by community input?
25:54
PM: The goal is to make it more usable and for people to want to use the platform - we spend a lot of time listening to what people’s pain points are.
31:40
AM: What are you doing surrounding bug fixes?
32:08
PM: Improving testing methodology and making more use of formal methods
38:25
Audience Q&A Session:
Moderator Alison asks Perry questions from community members.
Q: What was the rationale for prioritizing Sapling before faster finality / shorter block times? Is anyone using Sapling?
38:44
PM: Sapling was ready, and the thing that’s ready goes in. Did not have a version of Tenderbake we felt was ready to go on-chain.
40:23
Q: The community does not pay much attention to proposed changes before protocol injection. How can we fix this?
40:35
PM: Would like to see mechanisms that would alert people doing development that a new test network has come up with this week’s version of the master branch of the Tezos development chain - test against automatically spun-up version of the chain.
43:25
Q: When I use some of the great apps built on Tezos like Quipuswap or Hic et Nunc, I find them painfully slow at times and I've heard others say the same. Is this an issue with infrastructure those apps rely on in the ecosystem or an issue with those apps specifically needing to better optimize?
43:55
PM: We’re happy to talk to the people who built those things and help them look for performance improvements. I’ll make sure that we try to find out what performance problems there might be.
Closing remarks
Follow TQuorum on Twitter
Sign up for our newsletter to get the latest updates from TQuorum
