I have submitted a grant application, ZIPs For Nerds, a written series that takes one Zcash Improvement Proposal at a time and explains what the spec says in plain language, how the mechanism works, who has to do something differently, and the strongest argument against it. The request is $7,500 for six full depth explainers on the NU7 candidate ZIPs: 218, 231, 233, 234, 235 and 2002.
The series is already running and self funded. The ZIP 318 article is published, about 7,000 words with nine original diagrams, and it is the sample I would rather be judged on than anything I claim in the application: ZIP 318: Orchard to Ironwood Migration · ZIPs For Nerds
Writing it turned up one question I could not answer from the specs, so I filed it as zcash/zips#1366.
Everything is CC BY 4.0 and mirrored open access, including to ZecHub. I would welcome the committee’s and the community’s feedback on scope, topics, and budget.
Fair challenge. I think you’re reading this as content production, which AI genuinely has commoditised. What the grant pays for is reading each NU7 ZIP closely enough to explain it, and the articles are what falls out of that.
Here is what one week on one ZIP looked like. ZIP 318 defines EXPIRY_MODULUS as both “the number of blocks corresponding to 30 days” and “34560 blocks at 75-second block spacing”, in the same sentence. Today those are the same number, so nothing looks wrong. They stop being the same if 218 ships, because 34,560 blocks at 25 seconds is ten days rather than thirty, and ZIP 318 leans on that constant to promise “between one and two months of validity”. librustzcash hardcodes 34_560 and Vizor carries the same value, which is the right thing to do if you’re following the reference implementation. So the ambiguity is sitting in two codebases already, I couldn’t find it raised anywhere, and that’s why I filed #1366 rather than guess at it in the article.
That is one ZIP, one week, on a spec everyone implementing it has already read. None of it comes out of a $0.2 generation either, and I say that having run one. A model paraphrases that sentence and moves on, because it has no reason to check a constant against the reference implementation.
That’s the argument for funding it. NU7 is shipping, and as far as I can tell nobody is reading these ZIPs slowly from outside the group building them.
A short “what this means for wallet users” section would help: what changes, whether users need to do anything, and whether it’s still a proposal or already live. That would make the longer explainers easier to use when answering everyday questions.
Thank you for your grant application. After ZCG reviewed it and the community had the opportunity to share thoughts on the forum, the committee reached a decision not to advance the proposal. We value your work and invite you to remain part of the Zcash community. Meeting minutes.