The QUIC Transport Protocol: Design and Internet-Scale Deployment
Proceedings of the Conference of the ACM Special Interest Group on Data Communication, 2017 · paper · Computer science
Presents QUIC, an encrypted, multiplexed, low-latency transport protocol for HTTPS traffic, deployed at Google and estimated at 7% of Internet traffic. The authors report lower Google Search latency and fewer YouTube rebuffers, and share lessons from deployment.
Checked: PDF from storage.googleapis.com, fetched 4 Oct 2026. Page numbers are those of that PDF.
1 finding
Reference or label Reference [11] reads "Y. Cheng, J. Chu, S. Radhakrishnan, and A. Jain. 2016. RFC 7413: TCP Fast Open ... (2016)".
- p. 14 Reference [11]
2016. RFC 7413: TCP Fast Open.
Only the authors can say what is right here.
Judge's note: Rests on outside knowledge: RFC 7413 was published in December 2014. The image confirms the entry prints 2016.
Second check: another model agreed
Report a problem with this finding
7 minor (typos and the like; most checked by the judge only)
Reference or label Reference [5]: "A. Barth. 2015. RFC 6454: The Web Origin Concept. Internet Engineering Task Force (IETF) (Dec. 2015)." The year given may be wrong.
- p. 14 Reference [5]
A. Barth. 2015. RFC 6454: The Web Origin Concept. Internet Engineering Task Force (IETF) (Dec. 2015).
- p. 14 Reference [5]
A. Barth. 2015. RFC 6454: The Web Origin Concept.
Only the authors can say what is right here.
Judge's note: Rests on outside knowledge: RFC 6454 was published in December 2011. The page 14 image confirms the entry gives 2015 twice. The paper itself gives no other date, so the authors should check the year.
Second check: another model agreed
Report a problem with this finding
Wording "a 68% 0-RTT handshake rate on average—a 20% reduction in successful 0-RTT handshake rate as compared to desktop": 88% to 68% is a 20 percentage-point drop but a 22.7% relative reduction.
- p. 9 Sec 6.3, continuing paragraph
only achieve a 68% 0-RTT handshake rate on average—a 20% reduction in successful 0-RTT handshake rate
- p. 8 Sec 6.3
about 88% of QUIC connections from desk- top achieve a 0-RTT handshake
Judge's note: Section 6 defines "% reduction" in the relative sense for the tables. The wording here matches the percentage-point difference, not the relative one, so it is ambiguous.
Second check: another model agreed
Report a problem with this finding
Typo REJ message contents are enumerated "(i) a server config ... (ii) a certificate chain ... (iii) a signature ... and (v) a source-address token"; there is no item (iv).
- p. 3 Sec 3.1, Initial handshake paragraph
chain, and (v) a source-address token
- p. 3 3.1 Initial handshake, REJ message contents
and (v) a source-address token
- p. 3
and (v) a source-address token: an authenticated-encryption
Judge's note: Page 3 image shows the list running (i), (ii), (iii), then '(v) a source-address token'. A text search finds no '(iv)' anywhere. The sentence reads as a complete list, so this is a numbering slip: the last item should be (iv).
Second check: none (judge only)
Report a problem with this finding
Typo Sec 6.1: "has seen limited deployment at clients (seen Section 8)." The cross-reference should read "see Section 8".
- p. 7 Sec 6.1, second paragraph
has seen limited deployment at clients (seen Section 8).
- p. 7 Section 6.1
has seen limited deployment at clients (seen Section 8)
- p. 7 6.1 Experiment Setup, last sentence of the TCP Fast Open paragraph
(seen Section 8)
- p. 7 Section 6.1, Experiment Setup
limited deployment at clients (seen Section 8)
Judge's note: The page 7 image shows 'seen Section 8' as printed. This is a typo for 'see'.
Second check: none (judge only)
Report a problem with this finding
Typo The metric is introduced as "Video Playback Latency" but is called "Video Latency" in Table 1 and Section 6.4.
- p. 7 Section 6 intro
Search Latency, Video Playback Latency, and Video Rebuffer Rate
- p. 9 6.4 heading
Video Latency
Judge's note: The two names refer to the same metric and a reader is unlikely to be confused. This is a minor naming inconsistency.
Second check: none (judge only)
Report a problem with this finding
Typo Sec 6.6: "Table 3 show how QUIC’s performance impact varies by country." The verb should be "shows".
- p. 10 Sec 6.6, second paragraph
Table 3 show how QUIC’s performance impact varies by country.
Judge's note: The page 10 image confirms 'Table 3 show'. This is a subject-verb agreement slip.
Second check: none (judge only)
Report a problem with this finding
Typo Reference [32] lists co-author "J. Greassley", likely J. Graessley.
- p. 14 Reference [32]
J. Greassley
Only the authors can say what is right here.
Judge's note: Rests on outside knowledge: the Minion draft co-author's name is spelled Graessley. The image confirms the entry prints "Greassley".
Second check: none (judge only)