IJCK reviewer 1 comments

[kyleniemeyer]

Currently four database formats are offered for storing combustion experimental data. These are CloudFlame (Excel/CSV format), PrIMe (XML-based format), ReSpecTh (also XML-based format), and ChemKED (YAML-based format). Elaboration of the last format is reported in the submitted manuscript. The motivation for the elaboration of this new format was based on the statement that YAML is more human readable than XML. I think that this is simply not true. Even a first-time user can read the raw datafiles of both formats with equal efficiency. Also, currently the suggested new format is much more limited (e.g. it is able to encode ignition delay time experiments only). The presently planned features of ChemKED are already available in the other three formats.

However, I suggest the publication of an improved and shortened version of the manuscript in IJCK. The reason is that a permanent and laboratory-independent storage of experimental data is an important issue and I agree that a public discussion promotes the acceptance of a widely used common format.

Specific comments:

• I think that the paper contains too much information science details. Still, as the authors note, the description is not comprehensive and they refer to the user manual. I suggest the creation of a significantly abridged version that contains the basic information technology points only and does not discuss the details. For example, Chapter 4 could be deleted and Chapters 2 and 3 may also be shortened.

• p. 1 "Competing standards such as the Cantera [2] CTI format or FlameMaster [3] lag behind considerably, although use of the former continually grows due to Cantera’s open availability."
  Note that FlameMaster is also freely available from the GitLab repository of the developers.

• p. 2 "Second, the XML format is intended to be a machine-readable markup language rather than a data format, and its lack of human readibility presents a barrier to creating and working with database files."

XML is a text format and it is fully human readable, similarly to YMAL. Even a first-time user is able to read the chemical content of a PrIMe XML file and an experienced user is able to write such a file.

I suggest adding to following block to this para:
"The original PrIMe was on an Amazon cloud server. Recently it was shifted to the computer center of DLR Stuttgart. At the new location it is called "PrIMe 2.0" [https://prime.cki-know.org/]. This is identical to the original PrIMe apart from technical changes due to the different computation background. Professor Frenklach plans to develop further PrIMe (using name PrIMe 3.0) and this version will be availabe at [http://primekinetics.org/].

• "For example, where PrIMe uses internal bibliographic references, ReSpecTh adds a field for typical bibliographic data." I suggest the following extension:

"For example, where PrIMe uses internal bibliographic references, ReSpecTh adds a field for typical bibliographic data, including the DOI of the cited arecicle. Each datafile has a unique own DOI and unique URL address. [http://respecth.chem.elte.hu/respecth/reac/CombustionData.php]". Also species can be identified by InCHI or SMILES.

• "The ReSpecTh standard also provides machine-readable formats for describing ignition experiments, including a field for the definition of autoignition and the ability to specify facility-specific effects." Modified text:

"The ReSpecTh standard provides formats that are both human and machine-readable for describing ignition experiments, including a field for the definition of autoignition and the ability to specify facility-specific effects. Also, it describes PSR and flow reactor speciation data, burner stabilized flame speciation data and burning velocity data."

• "However, ReSpecTh experimental files are another XML-based format, and, as such, suffer from the same usability issues as PrIMe."

I think that the limits of an XML database is set by the actual set of tags and not the format.

• "Moreover, the standard remains closed, requiring potential users to register with the standard’s authors to access the specification and the growing database."

I suggest instead:
"Registration is open for everyone who provides a valid institutional e-mail address."

• "... hydrogen/syngas [11] and ethanol [12]. In all four cases, they converted numerous experimental datasets from the literature into the ReSpecTh standard."

Extended verion:
"... hydrogen/syngas [11], methanol and formaldehyde [C. Olm, T. Varga, É. Valkó, H. J. Curran, T. Turányi: Uncertainty quantification of a newly optimized methanol and formaldehyde combustion mechanism, Combustion and Flame, in press] and ethanol [12]. In all five cases, they converted numerous experimental datasets from the literature into the ReSpecTh standard. The database currently contains 93240 experimental data points in 1376 XML-format data files [http://respecth.chem.elte.hu/respecth/reac/CombustionData.php]".

• p. 5 "type (required, string): How ignition delay was measured;"

The options described here is a very limited subset of the ignition delay time definitions used in experimental articles. For example, backward or forward interpolated values of a measured quantity are further possibilities.

• p. 8 "In the latter case, some information may be lost, as ChemKED files both encode more details and specify more granularity of some information than ReSpecTh files."

This was true for ReSpecTh format 1.0, but no longer true for ReSpecTh Kinetic Dataformat 2.0. The latter covers all info of the ChemKED files (and more).

• Missing data in the references: Year is missing is refs. [5], [6], [9], [10], [11], [12], [21],[25],[27]

• [3] Please add reference: https://www.itv.rwth-aachen.de/index.php?id=13 Also, the latest version is from 2017.

• [7] Further info: Paper P1-04, ISBN 978-963-12-1257-0

• [8] Please refer to ReSpecTh Kinetics Data Format Specification 2.0 instead.

[kyleniemeyer]

I think that the paper contains too much information science details. Still, as the authors note, the description is not comprehensive and they refer to the user manual. I suggest the creation of a significantly abridged version that contains the basic information technology points only and does not discuss the details. For example, Chapter 4 could be deleted and Chapters 2 and 3 may also be shortened.

Yeah, not gonna do this. Reviewer 2 wants more detail in many places, and a thorough explanation of all this and examples are important.

[kyleniemeyer]

Note that FlameMaster is also freely available from the GitLab repository of the developers.

Yeah, but not openly—I just tried to view it, and only people with an account can see it... so it's effectively a private repo.

[kyleniemeyer]

XML is a text format and it is fully human readable, similarly to YMAL. Even a first-time user is able to read the chemical content of a PrIMe XML file and an experienced user is able to write such a file.

I suggest adding to following block to this para:
"The original PrIMe was on an Amazon cloud server. Recently it was shifted to the computer center of DLR Stuttgart. At the new location it is called "PrIMe 2.0" [https://prime.cki-know.org/]. This is identical to the original PrIMe apart from technical changes due to the different computation background. Professor Frenklach plans to develop further PrIMe (using name PrIMe 3.0) and this version will be availabe at [http://primekinetics.org/].

Same comments as for #3 ; yes, technically XML is readable because it is written in text, but it is much less understandable. This is widely agreed upon in the tech community... I wonder if there are any citations we can use here.

We can mention the new location of PrIMe, but the new features are not documented or explained anywhere. And we certainly can't discuss any plans Frenklach may or may not have in our paper...

[kyleniemeyer]

"For example, where PrIMe uses internal bibliographic references, ReSpecTh adds a field for typical bibliographic data, including the DOI of the cited arecicle. Each datafile has a unique own DOI and unique URL address. [http://respecth.chem.elte.hu/respecth/reac/CombustionData.php]". Also species can be identified by InCHI or SMILES.

Oh this is new—you can see some information publicly now. I don't think we should adopt this text verbatim, but we can update this slightly.

[kyleniemeyer]

"The ReSpecTh standard provides formats that are both human and machine-readable for describing ignition experiments, including a field for the definition of autoignition and the ability to specify facility-specific effects. Also, it describes PSR and flow reactor speciation data, burner stabilized flame speciation data and burning velocity data."

Sure, we can add some of this, since ReSpecTh does go beyond just autoignition.

[kyleniemeyer]

I think that the limits of an XML database is set by the actual set of tags and not the format.

To some extent, yes, but the format plays a big role in getting people to use and work with the format. Other than saying that, I'm not sure how to support that statement, though.

[kyleniemeyer]

I suggest instead:
"Registration is open for everyone who provides a valid institutional e-mail address."

Really? We already say you have to register. And there's nothing stopping them from changing this in the future; openness isn't baked in.

[kyleniemeyer]

The options described here is a very limited subset of the ignition delay time definitions used in experimental articles. For example, backward or forward interpolated values of a measured quantity are further possibilities.

Have you seen anything like this? I'm not entirely sure what this is referring to.

[kyleniemeyer]

This was true for ReSpecTh format 1.0, but no longer true for ReSpecTh Kinetic Dataformat 2.0. The latter covers all info of the ChemKED files (and more).

We'll have to look at v2.0 of their format and update. However, a quick look shows that their bibliographyLink still doesn't include granular details about a reference, just DOI and "description". I'm not sure if we were referring to more than that, though.

[kyleniemeyer]

Reviewer 1 all done!