Standardization Procedure for Data Exchange

Round 1

Reviewer 1 Report

This paper is not well-integrated.  The ideas do not follow logically from the initial setup of the arguments through to the case study and then the conclusions.  For example, while AI is discussed and used as a framing device early in the paper, it is never mentioned again.  This is a problem because we do not need to know how HTML5 came to be standard so much as we need to know how the lessons offered by its development should guide our understanding of data standards and interoperability in the current environment.  This is the source of the paper's promise, but it does not deliver.

Too much space is given to describing the changes to the W3C's internal process.  The reader is left unsure which details matter to the overall thesis (how interoperability can emerge among competitors) and which are background.  More attention needs to be given to the logical flow of the arguments and tying the discussion back to the state of web technologies now.  Why are these new hypotheses valuable?  How can they be applied to the present situation?  These are not things the reader should need to guess.

The figures also need to be improved.  For example, in Figure 2, no explanation is given for what "revenue stream" means in the context of the diagram.  On the left side, for instance, it is implied that Apple predominantly relied on hardware sales for revenue whereas Microsoft was focused on software.  That's fine, but then on the right, only Google Drive is highlighted as a revenue stream.  Are we to understand that Office Online is not a revenue stream for Microsoft?  The push toward getting everyone to sign up for Office 365 suggests otherwise.  Also, is Chrome not a revenue stream for Google?  The dominant market share of their browser has certainly affected their ad business.  Likewise, Figure 4 is difficult to read.  It took me a moment to understand that Figure 3 and Figure 4 are the same except for the milestone markers, but they are visually cluttered and do not effectively show how these changes relate to the formation of the hypotheses.

I do think the subject matter of this paper is important, which is why I would like to see an improved focus on the ultimate significance of the case study.  The adoption of HTML5 is positioned as a singular event that played a key role in modern web architecture.  If we grant that, why are the four aspects of the process that the author molds into hypotheses the ones most likely to impact the formation of standards today?  I was waiting for this argument to be made convincingly, but after getting to the end of the paper I believed I was left with a list of the hypotheses and no compelling reason to accept them as important to present battles over data standards.  I was also hoping to see a greater discussion of the tension between standardization and innovation, given that it was mentioned several times but not really explored.  Who are the innovators in the modern data landscape?  How might standardization hold their progress back?  The paper doesn't have to answer that specific question, but those are the kinds of basic questions I think a project like this should address in order to be a valuable contribution to the literature.

Author Response

Prease see the attachment

Author Response File: Author Response.pdf

Reviewer 2 Report

Dear Author,

 I have appreciated reading your paper. You have posed an interesting question and potential four-hypothesis solution for enabling a standardization of data sharing.  I would suggest you to:

1) make more clear the role of ontology (e.g., OWL, knowledge graph, etc..) and how do they can support your solution

2) to highlight that your solution might enable a cross-stakeholder data traceability. Traceability is required for both the verification and validation (V&V) of AI systems, to allow possible investigations of failures, to provide additional information that might be used by XAI modules.

Author Response

Please see the attachment

Reviewer 3 Report

The paper is much at the policy and strategic level rather than at the technical level. The whole work centers around a case study with HTML5.

Because HTML5 is a developed technology, the significance of this work should be articulated and highlighted.

The novel contribution of this work is somehow unclear. It is hard to see how it will benefit the standardization of WWW.

Author Response

Please see the attachment

Author Response File: Author Response.pdf

Round 2

Reviewer 1 Report

The author made an honest attempt to incorporate the previous round of comments. I still believe some of the earlier critiques apply, but the paper is closer to publication quality. Some figures are still not readable and do not convey the intended information. Finally the paper needs to be edited for flow and clarity.

Author Response

I added the last sentence to the abstract and updated figure 1 and 5 to highlight the role of the standard specification, the main argument of this paper. This paper is not about standardization but convergence of specification for bigdata usage. I tried to use the word convergence and interoperability instead of standard and standardization especially in sections of background and discussion.

Reviewer 3 Report

The paper is good as a review or comment on the standardization issues and policies, esp., regarding HTML, WWW.

It is not a typical technical paper but more like a magazine paper.

Author Response

The analysis was conducted according to case study method. I added the explanation of qualitative case study to 3. Materials and Methods section. The variable extraction in this paper follows qualitative research methods in social sciences. The paper is written according to the cross-disciplinary concept of this special issue.

Round 3

Reviewer 1 Report

Recommend a final edit by a native English speaker.

Author Response

I have made the manuscript be checked by a native speaker, and changed some expressions.