November 2018 Po.et Engineering Progress Report
Po.et is building the verifiable web: the decentralized protocol suite for content attribution, discovery, monetization and reputation.
In the weeks ahead, we look forward to delivering to you the mainnet implementation of the Po.et protocol and its related software systems and tools. We’ve been heads down working on technical content to help explain everything happening under the hood of Po.et, including clarity around the Po.et Node architecture, where decentralization comes from and how this milestone sets us up for the future. Most importantly, the foundation of Po.et is being shipped and will allow us to iterate much more quickly from here on out.
If you’re in software development, then you’re probably familiar with the “ninety-ninety rule”:
“The first 90 percent of the code accounts for the first 90 percent of development time. The remaining 10 percent of code accounts for the other 90 percent of development time.”
— Tom Cargill, Bell Labs
We’re just as excited as you are that the last 10 percent didn’t actually end up taking up as much time as the prior 90 percent. In fact, our month-over-month progress has remained steady even as we work through the final tickets of the milestone.
As we wrote in our September engineering update:
“Remember, one of the reasons we look at the month over month progress report is to detect any issues regarding the rate of completion. If we start to backslide, or progress slows too quickly, that could indicate scope creep. …
Sometimes things get exponentially slower as you get closer to completion, which is often a result of underestimating the scope of all the little finishing details.”
And in August:
“Managing scope is an essential part of delivering milestones in a timely manner. It helps us prevent scope creep and keep operating at high velocity.”
Scope has been well-managed, and we have prioritized a separate work stream of post-launch enhancements.
Also, the team has done a great job at maintaining steady progress by chunking down larger pieces of work and managing it in measurable goals. Bravo! And well-done.
The average ticket count has also continued to trend downward since last month, as evidenced in this burndown chart:
As we wrote in the October update:
“Now the burndown chart is starting to indicate that work is beginning to wrap up. It’s much clearer now that the average ticket count is trending down. We’ll still see it tick back up now and then as scope gets uncovered, but those tickets are getting closed faster than new tickets are being created.”
The last spike up at the beginning of the month marked the creation of a number of pre-launch items, so we do not foresee many new tickets being created in the remaining time leading up to the launch. We expect to see the average ticket count decelerate more rapidly in the weeks ahead.
Speaking of ticket creation, the team created a record 100-plus tickets in a month, with the total ticket count increasing from 353 in October to 458 as of yesterday.
This increase in created tickets coupled with a steady progress through the final 10 percent of the project indicates that the team is well-synced and working at a high velocity.
Finally, as promised in August, we are aiming to make the transition process as easy as possible for existing API users. We’re avoiding as many breaking changes as we can at the API layer.
We’ve got just a few tickets left to check off, and then we will release all of the new information on how to use Po.et. But mainnet isn’t a destination; it’s just one step in building a better web.
