The opensource model became very popular in the last 2-3 decades of software development and you can find great examples of open-source software used everywhere – operation systems, mobile platforms, graphic libraries, databases, and even business applications. But we don’t see it much successful in PLM.
Last week’s discussion about PLM and open-source caught my attention. Check it here – Is Open Source PLM Doomed. I was following open source in PLM for a very long time. Check some of my articles here. My last commentary – Closed Open Source PLM, Open Source Conundrum and Open Source in PLM and Manufacturing.
Getting back to Yoann’s conclusion why open source and PLM don’t go along? Here is a passage from his article.
First of all, if we look at all the open source libraries. Very few libraries actually get massively maintained. It requires a lot of developers and users to maintain some sort of hype on a project to get people really interested in it. That’s a first issue. PLM is a niche, there aren’t enough tech dev and users to keep the hype.
The other PLM open source category, is the marketing one. Companies develop their product and suddenly declare themselves open source by creating an Open Source version called “Community”. It usually allows companies to download the software and play with it, but you don’t get many people to actually look at the sources and collaborate on development.
Open source often talks to tech people. PLM is not a real tech environment. Software sales are not much based on the product technicality. It is more functional requirements and politics.
Jos Voskuil provided a great answer and detailed explanation of why different players in PLM aren’t interested in open source. Check this out.
I somehow have the feeling (not done any research) that Open Source PLM has two reasons to fail – I will try to explain them here: 1. If Open Source PLM is a development platform without business logic (just data model and a set of predefined behaviors – like SmarTeam/Aras – the value is all depending on the implementers – as each company has a specific legacy and need. The business logic per company is killing the open-source value. 2. PLM has always been an infrastructure connecting to discipline-specific tools (MCAD, ECAD, Simulations, ERP, …) that all require potential custom-made interfaces. As long as there are no integration standard formats, the integration becomes a licensed approach for the connecting solution. Therefore as PLM can be implemented in so many different domains, different industries, different authoring tools, I believe to have a generic open-source PLM will always be too simple or too basic for the real needs.
Let’s get back to the question – why we cannot see open source in PLM? To give you an answer, let me step back to very basic things for a minute. Let’s talk about source code first.
Source code is a part of the software that most of the users don’t see. It is actually the code that software engineers wrote to make it work – database storage, models, data extraction, saving data, visualization, managing accounts, data access, and many other things. Software engineers that have access to software code can improve the code and give it back to all others.
In many cases, software code is proprietary and belongs to the person or company who made it. But open source is different. In the open-source model, a person or a company will give source code to everyone to use and improve.
Consider anyone who is interested to get source code of the product and to do something with this code to re-use it for his own development or for distribution as a customer of open source. And it only makes sense to make something if you have enough customers. Make an analysis of all successful open source solutions and you will see how each of them found his customer niche – developer components, operation system, database, mobile platform. It didn’t happen to PLM, because PLM developers and customers were not interested in open source. Whoa… you can ask me. How come?
In my view, everything the PLM industry wanted (and still wants) is a different business model, which can make PLM more affordable. Most mature PLM development is owned by vendors that selling these solutions to large enterprises. PLM is a foundation of platforms that used to manage data and lifecycle, require long education and implementation cycle, consulting and support. PLM vendors own enterprise market, but the mid-size manufacturing enterprise market is completely open. Guess what… it is hard to take. Remember – customers first. And enterprise customers and mid-size manufacturing companies are so different. Therefore many attempts to sell the same product to both markets in PLM failed.
For a very long period of time, the PLM industry is looking at how to democratize PLM and to bring great ideas of PLM to small and medium-sized businesses. Until now, it didn’t happen, but some time ago, it was a hope that open source can do it. Well… it didn’t. Everything opensource PLM was able to do (and I’m talking mostly about Aras PLM) is to provide free download and community. It helped many companies and developers to start using PLM and make it more popular, but not so much business. Especially for Aras, which departed to focus on large companies to sustain their growth and provide returns to Aras investors.
The last thing to talk about is timing. As you know, timing is everything. Bad timing can fail any good idea and technology. A decade ago, the timing was good for open source. Not anymore. These days the timing is good for SaaS and cloud solutions. Small and medium-sized companies are going well along with the idea of SaaS applications. SaaS creates a new business model that can provide a good foundation to distribute the PLM solution. Look on all high performing PLM vendors according to best PLM software G2 PLM grid and you will see that all top high performers are SaaS solutions – Arena, OpenBOM (disclaimer – I’m CEO and co-founder) and Propel.
What is my conclusion? In my view, open source is a history for PLM. The important thing to learn from open source is about timing, customers and market. Open source PLM at its best time didn’t match these three elements together in a successful way. It was a great attempt, but it failed. It had its bright moment and helped companies like Aras to become more visible. PLM software is using tons of open source components now. Moving forward the timing is for SaaS PLM now. But it requires the right business models, technology, solutions. And the most important for SaaS PLM is not to miss its customers and what they actually need. So, open-source PLM is dead and SaaS PLM is coming. Just my thoughts…
Disclaimer: I’m co-founder and CEO of OpenBOM developing cloud-based bill of materials and inventory management tool for manufacturing companies, hardware startups, and supply chain. My opinion can be unintentionally biased.