Zuidaima (“Most Code”) launched in China in 2012. Version 1.0 was a source-code sharing economy settled in Niubi points:
Upload → earn points → spend points to download → points loop (with withdrawal requests).
It solved a real problem: course projects need a runnable package, and long-tail search matched people with inventory.
1.0 won on circulation. 1.0 also capped at circulation. When search thinned and authors felt “nobody downloads / points don’t feel cashable,” the flywheel stalled on the supply side first.
We are moving to a 2.0 trust economy: curated listings, official checkout, verifiable “it runs,” and clearer cash splits. The classic library and points stay as a long tail; premium is an overlay.
1. Who still supplies? Almost nobody is listing
| Year | New shares | Active authors |
|---|---|---|
| 2014 | 731 | 306 |
| 2015 | 621 | 349 |
| 2017 | 641 | 290 |
| 2018 | 479 | 240 |
| 2020 | 391 | 110 |
| 2022 | 137 | 56 |
| 2024 | 48 | 15 |
| 2025 | 7 | 6 |
Supply peaked ~2014–2017; author peak ~2015 (349). 2018 was still a big viewing year, but new listings had already passed their peak. In 2025, both new shares and authors are single digits.
Across ~1.8k sharer accounts and 5,000+ live shares, roughly the top 10% of authors produced about half of all shares.
2. What’s on the shelf: categories & keywords (lifetime)
| Category | Live shares |
|---|---|
| Web site development | 1,803 |
| Front-end | 1,469 |
| Syntax / algorithms | 279 |
| Desktop clients | 194 |
| Text / file processing | 192 |
| Mobile clients | 126 |
| Client games | 110 |
| Browser games | 95 |
Keyword hits in title/body (overlapping allowed):
| Keyword | Shares hit |
|---|---|
| MyBatis | 592 |
| SpringBoot | 332 |
| SSM | 287 |
| Hibernate | 244 |
| Permissions (权限) | 233 |
| SpringMVC | 226 |
| Blog (博客) | 157 |
| SSH | 152 |
| Android | 122 |
| Shop / mall (商城) | 117 |
| Vue | 97 |
| Mini-program (小程序) | 65 |
The shelf is still Web / front-end / Java-shaped. Low point bands remain the majority; 2.0 should win on complete runnable systems, not more templates.
3. How inventory gets seen and taken
Logged-in views peaked in 2018 (~780k views / ~96k viewers) and fell to about 1.3% of that peak by 2025.
Lifetime downloads (one count per user per project): ~695k deduped downloads · ~287k downloaders. Peak year also around 2018; detail in tomorrow’s downloader article.
Worse shelf outcomes kill the next round of uploads.
4. Withdrawal activity: counts only, no money
| Metric | Value |
|---|---|
| Lifetime withdrawal requests | 117 |
| Users involved | 40 |
| Busiest recent year | 2021 (~23 requests / 12 users) |
| 2025 | 1 request / 1 user |
Against hundreds of thousands of downloaders and ~2k sharers, almost nobody reached a withdrawal request. This is not a revenue figure — it shows how weak “cash-out feel” was inside the points loop.
5. Sharer takeaway on 1.0
- Points once drove a real UGC wave.
- Supply broke first.
- Categories/keywords stay course-project friendly, but skewed to templates.
- Downloads were large in aggregate; withdrawal requests stayed tiny — circulation without a strong monetization feel.
- Late author path: less traffic → fewer downloads → weak cash-out feel → stop uploading.
6. Why we move to a trust economy
| 1.0 points economy | 2.0 trust economy |
|---|---|
| Upload for points | Curated listing, official checkout |
| Hope for traffic | Platform helps find buyers; version-locked delivery |
| Rare withdrawal requests | Clearer cash splits |
| Zip and done | Git versions; keep selling |
One line for sellers: Put solid projects up; we help close and pay.
Zuidaima — curated code that saves time.
Official checkout · Sellers get paid · Runs locally
Top comments (0)