[dlahaye]

I am citing OpenFoam by referring to the Weller-98 paper. Is this accepted practice?

Are these better ways to cite particular OpenFoam solvers such as rho(S/P)impleFoam, buoyantFoam and reactingFoam? That is, can the inclusion of these solver to the OpenFoam distribution be traced back to particular references?

Many thanks, Domenico.

[Santiago]

Originally Posted by dlahaye

I am citing OpenFoam by referring to the Weller-98 paper. Is this accepted practice?

Are these better ways to cite particular OpenFoam solvers such as rho(S/P)impleFoam, buoyantFoam and reactingFoam? That is, can the inclusion of these solver to the OpenFoam distribution be traced back to particular references?

Many thanks, Domenico.

Well, the older versions of buoyantFoam were based on Oliveira & Issa's work HERE , PISO, SIMPLE and, by implication, PIMPLE, can be cited to Issa's work, and Spalding & Pantakar. When you refer to the OpenFOAM library itself (its facilities, classes, NOT ITS SOLVERS) then your citation is correct.

There is a rather toxic way of citing within the OF community (not to say everyone does it), and that is to cite derivative work made using OF as if said work were the original for the underlying algorithms. You might need to dig a bit, also because sometimes the nomenclature between OF and the rest of the academic world is different. One noteworthy example: On the turbulence models you have to be specially careful, because some of the impemented models are either 'non-conformant' variations (i.e.: dynamic smagorinsky-lilly) or particular flavors (see S-A, KOmega, oneEqnEddy) of a general class of models, usually based on conference works. My suggestion is, for the sake of being rigurous, to modify your codes to make them attain to the standards, or to pray that someone has undug the papyre on which they have documented the particular solver/model.

[dlahaye]

Thank you so much for sharing your thoughts and the reference. Kind wishes.

[Tobermory]

Another suggestion is to read the header files of the different submodels in the solver, since these sometimes list the papers where the models have been drawn from - eg check out https://cpp.openfoam.org/v8/kEpsilon_8H_source.html.

[dlahaye]

Thank you for getting in touch and sharing your insights.

The references that you point to are indeed valuable to see.

However, these references to not convey the information of who went through the effort of coding and testing parts of OpenFoam. I do imagine that for early versions of OpenFoam, the Weller-98 covers pretty much the complete library.

For more recent versions of OpenFoam, the situation for me is much less clear.

Can the development of rho(S/P)impleFoam be referenced with Weller-98? Who should I cite to acknowledge for the new developments (e.g. in handling turbulence, thermodynamics or time-stepping in these solvers).

Same question for reactingFoam.

[Tobermory]

Oh wow, this is a bit of a minefield as I understand it. The answer is that many people have contributed to the opensource project, and a few have had a quite considerable impact ... but maybe felt that they have not always had the recognition that they thought they were due. With strong personalities, this has led to some friction between parties, and you will see aspects of this in the forum occasionally.

My advice is to avoid the above tarpit, and to not try and attribute the code/applications to specific people (bearing in mind that you can still reference the underlying models, by using the references in the header files, as per my earlier suggestion), but to embrace the opensource aspect and reference www.openfoam.org, as per the advice in https://openfoamwiki.net/index.php/FAQ/General.

Good luck!

[Edited to clarify my point, following Santiago's later post]

[Santiago]

Originally Posted by Tobermory

Oh wow, this is a bit of a minefield as I understand it. The answer is that many people have contributed to the opensource project, and a few have had a quite considerable impact ... but maybe felt that they have not always had the recognition that they thought they were due. With strong personalities, this has led to some friction between parties, and you will see aspects of this in the forum occasionally.

My advice is to avoid the above tarpit, and to not try and attribute the code/models to specific people, but to embrace the opensource aspect and reference www.openfoam.org, as per the advice in https://openfoamwiki.net/index.php/FAQ/General.

Good luck!

This has nothing to do with the Weller vs. Jasak feud. This is about authorship of algorithms within OpenFOAM...

and to not try and attribute the code/models to specific people

What kind of advise is that? That's most unprofessional, if not unethical, particularly if you are in academia. So, according to you, we should acknowledge PISO to Weller, Jasak, etc, and not to Issa? Or k-w not to Menter? What about the tens of articles produced every year, contributing new algorithms using OpenFOAM? Is OpenFOAM a religion now, "If you ask Me anything in My name, attribute it to me"?

You'll never cite MATLAB whenever you implement/use a, say, FEM algorithm there, in place of the Author of the algorithm itself... Whenever you write a article in Word/LaTeX, unrelated to Word/LaTeX, do you cite Leslie Lamport & Bill Gates for it? What these examples have in common with OpenFOAM is that THEY ARE MEANS TO AN END, and are incidental to the task at hand.

Particularly, OpenFOAM is a library for solving PDEs using Finite Volumes, and is in that context that you must cite it, otherwise you must respect the Authorship of the particular algorithm you are using. One clear example that such can be done has been with IsoAdvector. Clear references and documentation were put in place within the code itself and in subsequent literature. But this is the exception, not the rule.

There is an effort to create an OpenFOAM Journal, in an attempt to solve this problem... Let's see where that goes.

[Santiago]

Originally Posted by dlahaye

Thank you for getting in touch and sharing your insights.

The references that you point to are indeed valuable to see.

However, these references to not convey the information of who went through the effort of coding and testing parts of OpenFoam. I do imagine that for early versions of OpenFoam, the Weller-98 covers pretty much the complete library.

For more recent versions of OpenFoam, the situation for me is much less clear.

Can the development of rho(S/P)impleFoam be referenced with Weller-98? Who should I cite to acknowledge for the new developments (e.g. in handling turbulence, thermodynamics or time-stepping in these solvers).

Same question for reactingFoam.

Some thoughts:

(1) You don't get any credit because you reproduce someone else's work. Verification, and implementation is important, but is not original, so you don't deserve acknowledgement for that. Most algorithms in OpenFoam come from authors outside the developers of FOAM, even the programming model in OpenFOAM is not original from OpenFOAM. On the other hand you have the MULES algorithm, which was conceived by one of the developers of OF...

(2) The time integration schemes. Some of them date even before the conception of the Neumann computer model. The more recent time schemes, such as local time stepping, come from transonic flow simulation research, even before OpenFOAM... One Important reference is Venkatakrishnan V. et al.

(3) What do you mean by "Testing parts"?? If you mean to verify the different spatio-temporal schemes, Jasak did a pretty exhaustive job about that in his thesis...

(4) There is still a confusion on what OpenFOAM is: OpenFOAM is a library for solving PDEs. Is not a solver, it is not a collection of solvers and turbulence models. So citing OpenFOAM should be done in the contexts of its facilities, not the solvers that come within.

(5) The algorithms for rhoPimpleFoam, pimpleFoam, pisoFoam, you can find them in any standard book for CFD (Peric, Blazek, etc). Obviously, they will not refer to them as pisoFoam, etc. Besides, these algoritms pre-date weller-98...

[Tobermory]

Originally Posted by Santiago

What kind of advise is that? That's most unprofessional, if not unethical, particularly if you are in academia. So, according to you, we should acknowledge PISO to Weller, Jasak, etc, and not to Issa? Or k-w not to Menter? What about the tens of articles produced every year, contributing new algorithms using OpenFOAM? Is OpenFOAM a religion now, "If you ask Me anything in My name, attribute it to me"?

No - that's not what I am suggesting - and I didn't mean to push your buttons, although it seems I have - apologies for that.

In fact, if you read the trail, in my first response I suggested that the OP reads the header files of the models he is using and quotes directly the source of the models - as you would expect in any academic paper. But that doesn't help when the OP is trying to reference the source of simpleFoam for example ... for that, I am suggesting that it is difficult (impossible) for any of us to attribute the true credit for these parts of the OF library ... hence my suggestion to reference the project rather than to try and pick some individuals ... unless you know for sure that it was their work. Probably confused because I clumsily used the word model instead of solver/application - let me fix that.