Open source PLM conundrum
Oleg Shilovitsky
CEO @ OpenBOM | Innovator, Leader, Industry Pioneer | Transforming CAD, PLM, Engineering & Manufacturing | Advisor @ BeyondPLM
The original article was published here. More articles on Beyond PLM
Once open source software was a no-go solution in enterprise software. I remember debates and discussions about open-source code with legal department in a large PLM development company back in 2000s. They asked me to bring good reasons why to use open source software... Fast forward in 2016 - open source software goes mainstream these days. It is not limited to web. There are some interesting examples of open source software usage for enterprise.
The question about open source and PLM is entertaining engineering and software industry for the last few years. It is not a simple task to understand all nuts and bolts of open source software - licenses, distribution, business models, ownership, etc.
My attention was caught by TechCrunch article - The Money in Open-Source Software. It gives a detailed perspective and examples of companies associated with open-source software. I captured few interesting notes.
VC companies poured a significant amount of money into open-source businesses for the last 5 years.
According to industry estimates, more than 180 young companies that give away their software raised roughly $3.2 billion in financing from 2011 to 2014. Even major enterprise-IT vendors are relying on open-source for critical business functions today.
Open source is a great delivery model. Just give away software as open source is not creating viable business. The business model behind is a differentiation factor that can make it financially successful.
Despite the growing popularity of open-source software, though, many open-source companies are not financially healthy. Just like eyeballs didn’t translate into actual online purchases during the first dot-com era in the late 1990s, millions of free-software downloads do not always lead to sustainable revenue streams. Make no mistake, open-source software is a brilliant delivery model to drive user adoption, and it’s poised to drive increasing market value in the coming years. But it’s not a business model on its own.
There are very few examples of financially successful open-source businesses
Just how challenging is it to build a big, profitable open-source business? Consider this: Besides the ongoing success of Red Hat — the company now boasts a $14 billion valuation, built over 20 years — and MySQL’s acquisition by Sun for $1 billion in 2008, there are few other landmark exits in the history of open source.
PLM open source lessons
For the last few years, I've been asked about open source PLM projects and vendors. Some of these projects went dark. Here are just few with live websites - OmniaPLM, beCPG, OpenPLM, DocDokuPLM. However, the only company with strong traction and visibility is Aras Corp developing Aras Innovator.
Aras claimed something unique called "Enterprise Open Source PLM". How open is Aras PLM source code? I've been trying to find the answer on this question on Aras PLM website.
Our approach was to combine multiple software formats, OSI-compliant open source,community source and commercial platforms, in a mixed source structure to provide the assurance necessary for business-critical solutions while delivering the flexibility for collaborative innovation. We call it enterprise open source.
Complicated? Not much. In a nutshell, Aras open-source is a combination of Aras commercial platform (read- traditional PLM software), free distribution and community solutions developed on top of Aras core platform. Most probably, Aras won't give you to fork their code, but you might check it with Aras. Here is an old comment from 2011 I captured from Aras forum with request to get Aras open source. According to the comment, you don't need source code to customize Aras.
But, make no mistake, Aras enterprise-open source is a brilliant business that put Aras in a unique position to challenge top 3 major PLM providers. The jury is still out.
Open source vs Free
Without go too much into legal terms, you would like to differentiate between open source and free software. I shared some of my thoughts about it back in 2010 in my article - PLM: Open Source vs Free. I think what we've seen in the past few years in PLM and more specifically with Aras Innovator project. It was mostly around building a subscription business around free software distribution. In a nutshell, Aras Corp. gives away Aras Innovator that allows you to do anything you want- use, implement, build solutions. But devil is in details and there are many use cases of Aras' implementations that will force you to pay for Aras subscription. I guess, it is a good news for Aras and it drives Aras revenues up.
What is my conclusion? Manufacturing companies are looking for alternatives to expensive PLM solutions. This is a main driving force behind supporting open source PLM projects. At the same time, all manufacturing companies are highly sensitive to potential risks related to vendors without solid financial base and unproven business models. It can potentially drive their decisions against open source PLM projects and vendors with unproven business models. Just my thoughts...
Best, Oleg
Photo credit - Mirage in the Western Desert in Egypt. © Damien Lovegrove
CEO Håkan Kårdén AB, Advisor, Co-founder Eurostep
9yRemember that a big thing with PLM is the life time of data - product data last for 10, 25, 50 or even more than 100 years. Regardless of open source, free SW etc the huge investments in data need to be protected. Vendors and SW will come and go. Standards are important.
Needs and requirements are connecting all product development actors. Are your Requirements Engineering process and tools designed for the best interactions between them?
9yIt is true that you don't need the Aras source code to develop very advanced configurations (i.e. PLM dedicated applications) thanks to Aras' generic and highly'extendable data model but also thanks to client and/server programming capabilities, including UI customisation. You can also stay in the free side if you are very careful in the way you develop. recPLM has been developed in this mindset.
Founder & CEO : Tijori.ai ^ Family Support System for your loved ones
9yHi Oleg , Your observations are right in light of open source with comparison to Licenses deployments -but future of PLM is more sustainable and beneficial to end customer in Cloud solutions rather than opensource or licensed deployments. ARAS has surely made major PLM players rethink of their strategy and such kind of challenges are good for keeping industry competitive and on the edge. Best Regards, Prashant Dhonde
Business Transformation Executive | SaaS Operations | Driving Profits through Innovation & Leadership | Solving the Ugly Problems
9yThanks for the article Oleg Shilovitsky. Your closing comments regarding financial security of a provider are crucial. When you're a young startup, open source is attractive if only because you have precious little capital available. You scrape together everything you can. Everything is constantly at-risk, including daily survival. Established businesses rely on established processes to support them. The processes need to be as close to 100% reliable (i.e. zero risk) as possible. This means not only the tool itself, but the ability to support it and obtain timely assistance when 'something' happens. When making buy/build decisions in the past, the topic of open source components would come up at times. I was always open to the discussion for non-mission critical areas. Financial viability of the provider must always be a consideration. No one wants to explain to their exec team how a Million dollar implementation stalled--or failed--because the company supplying a 'free' component is no longer there to help fix it. That, is perhaps the biggest bane of Open Source. Cheers!