I found out that this week, the first full week of January, is International Blog Delurking Week. Go ahead, Google it. This is a week where followers of a blog are asked to say “Hello” and comment on what they hope to see from the blog in the new year.
Besides a GUI that does everything, what do you hope to see from OpenSees in 2020? More Python, more documentation, more verification? Do you have ideas for blog posts? Do we need an OpenSees Delurking Week where we select obscure models and see if they can pass a patch test?
Let me know your thoughts in the Comments section, or just say “Hi”. Thank you for following the blog.
Portwood Digital 
Hi Positive Definite,
Thanks for sharing so much information on the development of OpenSees.
I’d like to see more theory manual (documentation) of OpenSees. I tried to read the codes of corotational transformation, fiber section, displacement-based and force-based element. But it’s hard to understand only from the codes. Sometimes I had to transform the codes to formulations. Sometimes I tried to find the developers’ dissertations, which may be outdated compared with the codes. I think the documentation showing the function and formulations of different classes and methods will be very helpful for OpenSees users.
Best,
Xinlong
LikeLike
Hello Xinlong,
Thank you for the comment. The code is definitely difficult to read, even for those familiar with the formulations. Because the models are rarely documented, we end up with many duplicate formulations, especially with material models.
PD
LikeLike
Hi PD,
Thanks for developing and maintaining OpenSees over the years. I would like to see more examples of geotechnical earthquake engineering problems.
best wishes,
Gopal
LikeLike
Hi PD,
Thanks for developing and maintaining OpenSees over the past decade or so. The best thing I like is its openness and free to everyone.
best wishes,
gopal
LikeLike
Hello gopal,
Thank you for the comment.
PD
LikeLike
Two years ago, before I get to know the OpenSeesPy, I thought OpenSees should be written from scratch since the world of programming has been changed a lot over the last 15 years. From my point of view, Delurking Week is a good idea if you can do it.
LikeLike
Hello MSB,
Integrating OpenSees with Python definitely extended OpenSees useful lifespan by several years. Based on the other comments, delurking periods for solid elements and geotechnical examples would be good ideas.
PD
LikeLike
Hi,
As a huge fan of OpenSees, I have been contemplating on this matter for some time now. I think we should have two goals in mind : “user-friendliness” to increase OpenSees outreach and “technical supremacy”.
Now regarding the first objective: I do not think a GUI is the ultimate answer. We are in an era that people are comfortable coding. The problem lies in the lack of user-friendly documentation, exciting examples that are easy to implement and “double-checked”. Although it is nice to be able to draw and see the model, it is absolutely crucial to have clear documentations on, let’s say, elements with their theory background, source code development, multi-tier examples and accessible codes. I aknowledge that there are several examples and documentation online and honestly they are the really (and may be ) only useful resources, but I think a lot can be done [ Btw, if you ever wanted people helping with documentation, I’m down]. In addition, we have disproportionate documentation as different materials are developed by different people: OpenSees should ask people to provide standardized documentation before adding their element/mat to source code. I can go on and on about what we can do for technical transparency but I stop myself. Lastly, I want to note that shifting to OpenSeesPy is a major game changer and can increase OpenSees appeal drastically.
Regarding technical supremacy, oftentimes OpenSees is acknowledged as only a macro finite element code. It will be nice if its awesome capabilities are extended to other realms of finite elements. I acknowledge there are several initiatives in this direction, but it will be nice to expose these initiatives and have a larger initiative to encourage people to contribute to such cause [funding opportunities, workshops, etc.]
LikeLike
Hello stostruct,
Thank you for the comment. I agree 100% that OpenSees sorely lacks documentation. Many examples online are out of date and in some cases just plain bad examples. Pretty soon, the main OpenSees will move away from the wiki documentation and toward the RST format used for OpenSeesPy at readthedocs.io. Being run through GitHub should make it easier for contributors to add documentation.
PD
LikeLiked by 1 person
Hola PD,
Thanks for sharing your knowledge through your very interesting posts.
OpenSees and OpenSeesPy are just awesome! I think a great plus would be adding Energy Recorders to avoid computing it as a post-process analysis.
Saludos desde Ecuador,
Bryam.
LikeLike
Hola Bryam!
Muchas gracias por la respuesta. Some other people have been interested in energy recorders as well, and they may have been added to a few elements and materials.
PD
LikeLike
Hello PD,
Firstly, many thanks for all of your work and support. It means a lot to us – OpenSees users, to see all these new changes: this blog, github, documentation etc. Hats off!
Secondly, I have been struggling for some time with the development of a new joint element. We are already using it locally, and hopefully we will submit it. However, I’m not so confident on my programming skills (C++ and Visual Studio stuff). I’m new to both C++ & OpenSees. Therefore, some tutorials/support (tips/tricks) on how to implement/add new elements/materials to the library, without getting into conflict with the already existing ones, would be highly appreciated.
Best regards,
Cristian
LikeLike
Cristian,
Thank you for the feedback! I’m glad you have found the posts useful.
I haven’t written a material or element “from scratch” in a few years, but it would be a good thing for me to try again, especially with recent versions of Visual Studio, and document the steps and post here along with screenshots. I’ll think of a relatively simple uniaxial material model I can implement.
PD
LikeLike
Dear P.D.
I want to thank you for taking the time to provide the posts of blog.
I found that you implemented some algorithms into OpenSees such as Secant Newton Algorithm, Krylov-Newton Algorithm,….
It would be very valuable if you compare and explain algorithms.
Best,
LikeLike
Hello MSB,
Thanks for the comment! The Newton algorithms are definitely something on my to-do list for blog posts.
PD
LikeLike