Linear or non-linear modeling that’s the question?

Recent discussion got my attention and i would like to add my 2 cents to the pot:

http://www.deelip.com/2009/06/siemens-holds-on-to-history.html

http://www.deelip.com/2009/06/is-direct-modeling-honeymoon-over.html

Compare to Dan Staples, Deelip Meneze or Paul Hamilton, I do not have inside knowledge of what’s going on under the hood of a 3D modeler, my knowledge is some how limited to how to use those tools and generic informations about how things run under the hood.

In the blogsphere of CAD information, my belly button is still wet and I do not have the weight in media of Evan Yares, Ralph Grabowski or any other, this is use has a friendly term, old-timer.

I do believe in my point of view, even if I may look bias due to the fact that I blog about a specific product or technology.

Where I want to jump in the conversation is when we talk about the hypothetic death of history base modeling.

Peoples who try to resolve this question with a linear minding approach will conclude about the death of the history base modeling, espacially  in regard of the direction the industry is taking.

History base modeling will never die or disappear, what will change is how we navigate inside of it. This is why i usually talk about linear modeling and non linear modeling in parallel of history and non history, because i believe both can exist in the same file structure.

And i believe we need to make the distinction to help the debate on the subject.

 

Little history and analogy

What the hell is he talking about linear???? non Linear?????

Ok give me few seconds to  expose my point of view…..or for quick answer visit the Lexis section…..

For the last 15 years we have been educate in a linear way to create solid model and before this we had a 15 years of 2D drawing that shape the base of linear modeling. Changing two generations of CAD designers will take much more then 10 years and after those 30 years, some technologies still rely on basic 2D DWG.

Iwould like to make a analogy to a TV program (big, bigger, biggest) that list the ten big leaps that make sky scrappers possible. Until the 19th century, buildings of over six stories were rare.No one could imagine a building taller, having great numbers of stairs to climb was impractical for inhabitants.

The first big leap that lead to taller buildings……… The invention of the elevator change everything. Peoples where able to move from floor one to floor ten then back to floor five. Elevator give the possibility to navigate in a non linear way inside a building. This open the door to a new generation of Supertall towers.

Same here in the CAD business. The introduction of Synchronous Technology allow users to navigate in a non linear way inside the history of a part.

 

Break from the box design concept

Looking at how we sort information’s  inside a pc ……a folders structure with no communication between folders, this reflect our way of design.

 We have assemblies and sub assemblies that contain parts. Those parts contain features that need to be in a specific order. Each feature has  their own steps that need to be follow, the plane step, the geometry step etc…..

Making edition force users to open box after box after box to access data.

This box conception have advantage but also it’s disadvantage.

Our brain work more like a relationship database than a drawer. And this is what non linear modeling try achieve, build a relational design or organic design.

From my point of view it is not if history will continue to exist or not, but how we will navigate inside the history of the part. Everything seem to point to non linear navigation inside the history of a part.  Forget about the boxy way of doing things in a linear modeling.

Same for features, they will not disappear, under Syncrhonous Technology we will have  procedural features. They will have behavior store in their core that will allow us to  define priority base on the context.

At PLM world conference, those who paid real attention, ST2 introduce round re-order to have a specific result. In the same presentation a slide show how we can set priority in the model. I do not have the slide or access to the presentation and hope this one pass the beta testing.  I build my own slide out of ST1 to expose the concept  base on what i can remember from the presentation.

 

The procedural feature that add material will have the priority in the left side. On the right side the cut will be prioritize. This could be done no matter the order they where create. The location of the procedural feature in the Pathfinder has no influence allowing users to sort those in a way that make sense to help understand the model.  What it is show in the Pathfinder are “Collection of surface” either system collection or user collection.

Conclusion

I believe the history modeling will survive but how we will navigate inside that history will change to a non linear modeling way, allowing designers breaking from the box design of linear modeling. At the same time i am a strong believer that Synchronous Technology will introduce at some point thru  procedural feature the concept of relational database to help prioritize action inside the model or the design.