Secure DevOps for PLM

Digital thread in a shared dream

Once upon a time, Hollywood was a true Detroit for the souls. Movie marketing people of the 1960s figured that the audience enjoyed watching epic historical battles. Money poured in, and dozens of excellent movies were created, culminating in the majestic “Waterloo” in 1970. Movie marketing people of the 1980s discovered that the new generation of viewers desired supernatural and scary, and a Satanic rush followed, with some of the masterpieces like “The Prophecy” with Christopher Walken and Viggo Mortensen – unfairly forgotten since.

Then, at some invisible moment in time probably around 2008-2010, American movies collectively mutated into something totally different. Nowadays, a movie is never a decadent act of creativity or even a corporate greed: it always contains a pill designed to steer viewers into a certain way of political thinking. As a result, an overwhelming majority of these movies have become extremely bland and predictable, and I stopped considering them worthy of my time.

That’s how I missed Christopher Nolan’s “Inception” with Leonardo DiCaprio, and feel lucky for running into it much later. Besides bedazzling special effects, the movie is about “an idea, the most infectious thing in the world.” Once introduced into a human mind, it will grow and will “either define or destroy a person.” My own subconsciousness immediately pointed at the connection to PLM and the entire digital transformation debate.

A product starts with an idea. An idea (via requirements management) develops into design data stored in the PLM system. ERP keeps material information and MES is about a myriad of activities related to the shop floor. And so on.

The “Inception” miracles were attributed to a technology called “shared dreaming,” where several people can participate in the same virtual adventure inside their brains. An Architect designs the “maze” to simulate reality as perfectly as possible. Similar to The Matrix, dream worlds have rules. These rules can be bent or outright ignored only by truly inspired dreamers.

The movie is full of amazing concepts:

  • Inception vs Extraction – embedding the idea without leaving a trace vs. stealing information without leaving a trace either. Extraction is intellectual property theft, but with a twist. The victim must first explicitly hide it in a safe – and consider it secure.
  • Projections – totally imaginary participants in the dream, who at some point figure out the dream has been invaded by mischievous rule-benders and start attacking them. Sounds similar to entrenched corporate interests. I can only imagine what “militarized subconsciousness” and “armed projections” can mean in the engineering innovation context. Perhaps Nokia, Kodak and GE can tell us more about that?
  • Waking up with a “kick” – multi-layered dreams and being lost in a limbo. I cannot but think about so many failed “digital transformation” projects, that ended up precisely in this type of limbo. These projects’ members were probably enjoying their collective dreams for too long (time stretches longer in a dream), forgetting to validate themselves with reality. “Inception” dreamers could (usually) emergency-escape to reality by killing themselves, while in the PLM universe projects or entire companies are often simply shut down.

I remember a presentation circa 2005 about how next-gen PLM systems will transform the world of manufacturing. Not a single word about connected devices, digital threads or anything like that. A few loners were pushing for engineering ontology research. After IBM Rational object-oriented database and ENOVIA LCA exposed assembly fiascos, I personally considered relational databases and physical CAD files the only sane option for persistent data storage.

PLM circa 2021 is no longer focusing on engineering – it is about providing information/decision support along the entire lifecycle of products. The notion of digital thread itself is getting obsolete in favor of digital graph or digital web, while digital twins look more and more like triplets or, for some extremist dreamers, quadruplets.

There is a graph database based PLM solution on the market. SysML has evolved into a requirements management must-have, penetrating supply chains from top to bottom. OSLC, a decent dream of engineering data interoperability, showed new signs of life during the recent OSLCFest. The concept of DevOps in PLM, which I just started discussing with Senticore customers 5 years ago, is being rapidly brought to the fore.

Dreaming is indeed dangerous, especially when it is someone else’s subconsciousness. While the proprietary OEM data in the relational databases is always reasonably accessible for a trained SQL developer, going smarter and deeper can inadvertently make those OEMs brutally trapped in their PLM vendors’ dreams.

Will the data “ownership” be possible with the distributed graph databases? How will the semantic web play out in the world where it is still necessary to “trust but verify,” and actually, what are true intellectual property jewels and trade secrets, and what are not?

A big question on how to define “configuration” of the same hardware and multiple versions or variants of software, and then track all this reliably across the various dreams, hits the nail on the head. And as much as I am trying to keep my head cool, I start wondering if the graph database technologies integrated with public or private blockchains will indeed become the answer.

Further on in the DevOps for PLM domain, unleashing yourself into an unrestricted open-source code usage dream – without malicious code detection and licensing controls – looks to me like polishing brass on the Titanic. Fortunately, a case study of how this can be done properly is the USAF PlatformOne project, with their “cattle vs pets,” “everything as a code,” Iron Bank hardened containers repository, and “baked-in” security. If the government can do it, so can industrial OEMs.

Altogether, with all the dangers and challenges involved, not dreaming is even more dangerous for both humans and companies: the former go dull, the latter go extinct. So, both humans and companies must continue dreaming, while maintaining that precious link to reality. The future of engineering domain has never looked brighter.

Let’s go for a spin?