The Web, the Users, and the MOS: Influence of HTTP/2 on User Experience

Bocchi, Enrico; Cicco, Luca De; Mellia, Marco; Rossi, Dario

doi:10.1007/978-3-319-54328-4_4

Cited by 27 publications

(25 citation statements)

References 13 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Our analysis relies on a dataset with 3,400 Web browsing sessions where users explicitly rated the quality of the session. This dataset extends our previous effort [9] and we make available to the community at [28]. We conclude that expert models for Web QoE can easily accommodate new time-related metrics beyond PLT, and that their accuracy is comparable to that of data-driven models.…”

Section: Introductionmentioning

confidence: 52%

“…To assess the impact of application QoS on QoE, we extend our previous experiment on measuring Web user experience [9]. We gather 8,689 Web browsing sessions, that we make available at [28], where 241 volunteers rate their browsing experience with the Chrome browser using the Absolute Category Rating (ACR), where 5-Excellent, 4-Good, 3-Fair, 2-Poor, and 1-Bad.…”

Section: Datasetmentioning

confidence: 99%

“…We gather 8,689 Web browsing sessions, that we make available at [28], where 241 volunteers rate their browsing experience with the Chrome browser using the Absolute Category Rating (ACR), where 5-Excellent, 4-Good, 3-Fair, 2-Poor, and 1-Bad. For lack of space, readers are referred to our previous work [9] for a detailed presentation of our experimental setup.…”

Section: Datasetmentioning

confidence: 99%

See 2 more Smart Citations

Narrowing the Gap Between QoS Metrics and Web QoE Using Above-the-fold Metrics

Hora

Asrese

Christophides

et al. 2018

Lecture Notes in Computer Science

Self Cite

View full text Add to dashboard Cite

Abstract. Page load time (PLT) is still the most common application Quality of Service (QoS) metric to estimate the Quality of Experience (QoE) of Web users. Yet, recent literature abounds with proposals for alternative metrics (e.g., Above The Fold, SpeedIndex and variants) that aim at better estimating user QoE. The main purpose of this work is thus to thoroughly investigate a mapping between established and recently proposed objective metrics and user QoE. We obtain ground truth QoE via user experiments where we collect and analyze 3,400 Web accesses annotated with QoS metrics and explicit user ratings in a scale of 1 to 5, which we make available to the community. In particular, we contrast domain expert models (such as ITU-T and IQX) fed with a single QoS metric, to models trained using our ground-truth dataset over multiple QoS metrics as features. Results of our experiments show that, albeit very simple, expert models have a comparable accuracy to machine learning approaches. Furthermore, the model accuracy improves considerably when building per-page QoE models, which may raise scalability concerns as we discuss.

show abstract

Section: Introductionmentioning

confidence: 52%

Section: Datasetmentioning

confidence: 99%

Section: Datasetmentioning

confidence: 99%

See 1 more Smart Citation

Narrowing the Gap Between QoS Metrics and Web QoE Using Above-the-fold Metrics

Hora

Asrese

Christophides

et al. 2018

Lecture Notes in Computer Science

Self Cite

View full text Add to dashboard Cite

show abstract

“…Moreover, since we already found a significant impact of the congestion control in [24], we further use a variant of TCP and QUIC that utilized BBRv1. 2 A website visit in our testbed is done using a fresh Chromium browser with an empty cache. Starting from scratch has several implications on the protocols: it means that QUIC will not perform a 0-RTT connection establishment but a 1-RTT handshake.…”

Section: Repeatable Protocol Performance Evaluationsmentioning

confidence: 99%

Perceiving QUIC

Rüth¹,

Wolsing²,

Wehrle³

et al. 2019

Proceedings of the 15th International Conference on Emerging Networking Experiments and Technologies

View full text Add to dashboard Cite

QUIC, as the foundation for HTTP/3, is becoming an Internet reality. A plethora of studies already show that QUIC excels beyond TCP+ TLS+HTTP/2. Yet, these studies compare a highly optimized QUIC Web stack against an unoptimized TCP-based stack. In this paper, we bring TCP up to speed to perform an eye-level comparison. Instead of relying on technical metrics, we perform two extensive user studies to investigate QUIC's impact on the quality of experience. First, we investigate if users can distinguish two protocol versions in a direct comparison, and we find that QUIC is indeed rated faster than TCP and even a tuned TCP. Yet, our second study shows that this perceived performance increase does mostly not matter to the users, and they rate QUIC and TCP indistinguishable.

show abstract

“…In other words, the longer we have to wait for the Web page to load (or transactions to complete), the more dissatisfied we tend to become (as discussed by Egger et al [29]). It should be noted that recent moves to HTTP/2, such as studied by Bocchi et al [136], are pushing to optimize the Web by decreasing PLT using (amongst other features) a server push mechanism. We note that the issue of loading times has also been addressed in the context of certain Web usability studies such as Flavian et al [31] and Palmer [86].…”

Section: System Influence Factorsmentioning

confidence: 99%

Survey of research on Quality of Experience modelling for web browsing

Baraković

Skorin‐Kapov

2017

Qual User Exp

View full text Add to dashboard Cite

A significant amount of research has been published to date studying various measures and influence factors related to the user experience when browsing Web content on different devices. For the most part these studies come from two different communities: the Quality of Experience (QoE) community and the User Experience (UX) community, and span different disciplines. While the QoE community has primarily focused on technical aspects and subjective perception of waiting times, the UX community has been working on issues of acceptance, experience, and crucial design factors extensively for a long time. This paper aims to provide a survey of literature related to QoE modelling for Web browsing by addressing studies that deal with the impact of a wide set of system, context, and human influence factors. The survey shows that the QoE community has for the most part neglected relevant aspects studied by the UX community, which are needed for a more holistic understanding of Web QoE. On the other hand, UX studies may benefit from insights into research conducted in the QoE domain in terms of the impact of more technical factors on UX. Thus, by bridging these findings we argue the need for future multidisciplinary and multidimensional studies on Web QoE modelling, whose product, that is, multidimensional Web QoE models, are of interest to multiple stakeholders involved in the service delivery chain. Readers of this paper will benefit from a systematic analysis of surveyed papers, summary of key findings, and a discussion of open research topics that contribute to setting a research agenda in this domain.

show abstract

The Web, the Users, and the MOS: Influence of HTTP/2 on User Experience

Cited by 27 publications

References 13 publications

Narrowing the Gap Between QoS Metrics and Web QoE Using Above-the-fold Metrics

Narrowing the Gap Between QoS Metrics and Web QoE Using Above-the-fold Metrics

Perceiving QUIC

Survey of research on Quality of Experience modelling for web browsing

Contact Info

Product

Resources

About