Paper Revision

Questions

• 1. Overall Rating Provide your overall rating of the paper. (counts as 1...5 in overall review score)

  • Probably accept: I would argue for accepting this paper.

• 2. The Review Write your review of the paper here. Please identify positive and negatives of the work and the paper, plus any suggestions for revisions. Provide constructive criticism to help the authors to improve their work.

  • The paper presents the development of Soundcool Online, which is a re-implementation of Soundcool in web language, allowing its use with a browser from any platform. The motivation of the work is clear and of great interest. The document is well structured and several aspects are addressed, which makes it interesting for future research and development on similar tools.

The Related work section could be extended, as there exist many online music creation tools.

About how the modules are interconnected, no "wires" indicate how they are connected. The user has to hover the mouse over a box to highlight the modules. How is this done exactly, and what is the motivation for not always showing wires?
  • DONE - Kristin



Regarding Section 5.2, the authors propose as a future work to allow multiple remote users to access, edit and play with the same project. This however, would require some additional implementation work, as well as designing new features in order to help users to efficiently collaborate on a project. The idea of "performance" proposed from the authors is not very clear. How would this be different from a project? Would this just correspond to a "virtual room" that a performer can give access to, or does it imply additional features? * Done - Kristin



Some aspects of usability are not very clear. How exactly can we add new modules. These modules are "stacked in column". Is the order of the modules important? Can we re-order them, in order to organize the project in a logical way? * DONE - Kristin

 I would love to be able to try the application!


Reviewer #4 Questions

• 1. Overall Rating Provide your overall rating of the paper. (counts as 1...5 in overall review score)

  • Probably reject: I would argue for rejecting this paper.

• 2. The Review Write your review of the paper here. Please identify positive and negatives of the work and the paper, plus any suggestions for revisions. Provide constructive criticism to help the authors to improve their work.

  • The paper presents Soundcool Online, a Web Audio re-implementation of the original Max/MSP implementation of Soundcool, a system for collaborative music creation used mainly for educational purposes. Authors say that their tool stands out from the others mainly by its collaborative capabilities and large set of modules (it’s a modular system for sound synthesis), that can be compared to what we find in the Eurorack standard.



Section 3 paragraph 2 could be omitted as the motivation for having an online version has already been exposed in Introduction.


The collaborative aspects could be better categorized, when we are talking about a collaborative online tool, we imagine something synchronous (a la Google Doc, where multiple users are connected at the same time and interact on the same applications that updates all users’ GUI), or asynchronous (you work on a project and share the link, someone else works with it at a different time, or with some parts “locked” by other users, or work with a copy of the project until they share a new link). Here you say “they can share screen using zoom”, huh. Then, ALL applications are collaborative! Ok, you talk about controlling the screen with mobile devices, this is a sort of local collaboration. Even if the scenario is zoom + multiple set of people who interact by phone, the way it is described in the related work and introduction is a bit misleading. Clarification would be welcomed and a table with other collaborative tools to compare would help seeing what is different from other solutions.



In Chapter 4.1 you talk about the now obsolete ScriptProcessor Node and you raise questions about AudioWorklet Node that is the new standard. Except in Safari it is now available in all major browsers, and people from the WebAudio community know since a long time the flaws of Script Processor Node. Making the choice to rely on SPN is questionable. Many tools such as CSound, Gibbers, and 2 online commercial DAWs rely on AudioWorklet nodes, and this technology is quite mature now.



Section 4.3 is too technical and the style is not suited for a scientific research paper. Same remark for section 5.1.The discussion about latency is interesting but the style is more like “telling a story” than the one of a scientific paper…



There have been several works about WebAudio plugins (see papers presented at WAC 2018 for example), and the paper you cite by Letz and al. present how the FAUST IDE can be used to generate plugins that follow the WAP standard, that is a superset of the WebAudioModules standard (this latter targeting more C++ developers who want to port their native plugins to the Web platform). WAP supports many different technologies, not only FAUST as it seems to be when one reads this section.



About webaudiomodules see :

Kleimola, Jari, and Oliver Larkin. "Web audio modules." Proc. 12th Sound and Music Computing Conference. 2015.



About the WAP standard:

Buffa, M., Lebrun, J., Kleimola, J., Larkin, O., & Letz, S. (2018, April). Towards an open Web Audio plugin standard. In Companion Proceedings of the The Web Conference 2018 (pp. 759-766).




To conclude : this tool seems interesting but the paper could be improved by removing redundancy (the advantages of a web application vs native appears multiple times), and use a more synthetic / scientific style. The article is pleasant to read but the presentation lacks scientific rigor. There is no evaluation of the quality of this tool, a small user evaluation campaign would have been welcomed. The details about the implementation could be omitted except if they have an important role in the final result. The collaborative aspects are interesting but not clearly described. Do you have any scenario/use case that could explain why it is really a great tool for education?


Reviewer #5 Questions

• 1. Overall Rating Provide your overall rating of the paper. (counts as 1...5 in overall review score)
  • Definite accept: I would argue strongly for accepting this paper.
• 2. The Review Write your review of the paper here. Please identify positive and negatives of the work and the paper, plus any suggestions for revisions. Provide constructive criticism to help the authors to improve their work.
  • Interesting paper that proposes a collaborative web alternative to an existing tool. The main concerns and motivations are well exposed and the resources/technologies are explained. Specially, I have found very smart and helpful the solution made for the limited bandwidth in schools, which can be a real problem in the development of such tools in this kind of contexts.

I have made a set of minor comments/suggestions that could be added to make the paper a bit clearer:

- Maybe it would be good to add WebPd (https://puredata.info/downloads/webpd) on section 2.
  • DONE - Kristin



• In section 3, I haven't found clear enough how to save a session. Is it made by the JSON file? Should this be then stored locally?

  • DONE - Kristin



• In section 3.2, Freesound should be cited.

  • DONE - Kristin



• Section 4.3: It is mentioned that Amazon AWS ElasticBeanstalk has been used for the deployment of the application. Since this is an educational project that tries to promote the open source mentality, maybe you could try to think of an open source web app for this purpose.



- Section 5.1: It would be interesting to add some sentences about how the delay could be improved, at least just as future work.

- Section 5.2: Would there be limitations in terms of numbers of users at the same time for this solution?



• Section 7: This is an idea or suggestion: Since it's an application targeted to schools, it would be interesting to give more support to the figure of the teacher.
Maybe it would be nice to have some kind of hierarchy where the teacher can control the student's projects and make some comments/corrections.

Meta-Reviewer #1 Meta-Review Questions 1. Overall Rating Provide your overall rating of the paper. (counts as 1...5 in overall review score)

• Probably accept: I would argue for accepting this paper. 2. The Meta-Review Please summarize all of the reviews and discussion for this paper. Focus on the major strengths and weaknesses of the paper. What is your overall recommendation about whether this paper should be accepted? Provide constructive criticism to help the authors to improve their work.
• Most of the reviewers agree to accept this paper, and I support them. However, authors should address multiple points that have been pointed out by reviewers before submitting a camera-ready version.