search

thelfer / MFrontGenericInterfaceSupport

This project aims at providing support for MFront generic behaviours. This project can be embedded in open-source and propriary sofware
37 stars 35 forks source link

Issues with paper #35

Closed srmnitc closed 4 years ago

srmnitc commented 4 years ago

This is part of the review of JOSS paper openjournals/joss-reviews#2003

thelfer commented 4 years ago

Hi @srmnitc,

Statements in the paper such as This is error-prone and may lead to spurious or even worse inexact results and While this causes memory fragmentation, this is the most frequent choice, would be reinforced with suitable citations.

This is indeed a difficult point coming most notably from our own experiences of building complex simulations. I don't know if I can find any citations on that matter.

Would it be ok if we introduce proper word of cautions, such as: "According to the authors experience...", "One drawback of this approach is that it may lead to memory fragmentation"

srmnitc commented 4 years ago

I think you could also specify some reason as to why it is error prone and so on.

thelfer commented 4 years ago

For the third point, here is a small paragraph which highlighs how traditional approaches are error-prone:

In the authors' experience, this is error-prone in particular for inexperienced users and may lead to spurious or even worse inexact results. For example, when material properties such as the Young Modulus or the Poission ratio, are required by the behaviour, those are generally defined in the solver input file by an array of values and potentiel checks are limited to the size of the array. An user may thus invert two material properties. If those two material properties have the same order of magnitudes, computations might lead to a physically consistent result, altough a wrong result.

thelfer commented 4 years ago

@srmnitc

  • The JOSS paper has some typos which need to fixed.

I tried to fix that with the help of the co-authors and a colleague who is a native english speaker. English is hopefuly acceptable now.

  • The first page of the paper has numerous short paragraphs, which in my opinion affects the readability.

Apart from the first line, the other paragraphs have more than 7 lines. Each develops a specific idea, which is a common convention in French.

  • Another point is about citations. Statements in the paper such as This is error-prone and may lead to spurious or even worse inexact results and While this causes memory fragmentation, this is the most frequent choice, would be reinforced with suitable citations.

The statement "This is error-prone" has been discussed more in details. The statement about memory fragmentation has been removed because detailled discussion of this point is out of the scope of the paper.

srmnitc commented 4 years ago

Thanks @thelfer , everything looks okay now. One last point - maybe Mohr-Coulumb yield criterion needs a citation?

thelfer commented 4 years ago

Done ! Thanks again for the review.