Side N
533 Bytes an Entry, 100 Terabytes a Fleet
Which sources back which section of this article, and what its evidence base looks like.
Which sources back which section Two sources were used. One is the operator's own account of the system it built; the other is the protocol standard it is checked against. The rows say which backs which section.
| Section | Sources behind it |
|---|---|
| What Happened | Sebastiaan Neuteboom, The Cloudflare Blog, for every figure and the rollout dates; RFC 1035 §4.1.4, for the compression-pointer claim confirmed in the same table |
| Timeline | Sebastiaan Neuteboom, The Cloudflare Blog — every date on this timeline is the post's own |
| The Argument | Sebastiaan Neuteboom, The Cloudflare Blog — the design rationale for all five changes is the post's own account of its own work |
| What Others Add | RFC 1035 §4.1.4, read directly for the compression claim; Sebastiaan Neuteboom, The Cloudflare Blog, read again for the arithmetic that re-derives its own published figures |
| Conclusion | Sebastiaan Neuteboom, The Cloudflare Blog — the synthesis draws on its own figures and its own stated plans throughout |
What the evidence base looks like Two sources, deliberately not stretched into more
- 2Sources behind this article
- 5Sections, every one of them sourced
- 2Read in full, at their own page
- 1Written standard consulted directly, independent of the operator
| What the source is | How many | How it was read |
|---|---|---|
| The operator's own engineering account of its own system | 1 | Read in full, fetched directly and its layout claims checked against a working compiler |
| The protocol standard the design is checked against | 1 | Read in full, at the RFC Editor's own copy, for the section it verifies |
This article uses fewer than five sources, so the scatter is not shown. Four points against a four-item source list would leave no point anonymous.