Open Source PLM Factoids

June 30, 2010

I’d like to continue discussion of Openness and PLM (see my yesterday post – Closed Thoughts about PLM Openness) with the discussion about Open Source. Open Source is one of the questions that always raised last time when it comes to the discussion of Openness. I got to read an interesting article – Being acquired is the best thing for a FOSS project from Network World Blog. They presented an interesting perspective on a particular niche of Open Source. The following numbers caught my interest – Sourceforge hosts about 7,000 security projects. Daniel says, “From these 7K only 10% will survive; they seem to die quickly.”  I made me think what is the right formula for Open Source in PLM?  Yoann from Prodeos wrote in his post (in French)  PLM=PLM = BPM + ECM + PM + CMII + … (if standards == true). His idea of various component syndication seems to me interesting (I just hope my French was good enough to understand it with the help of Google Translate). All these together made me think about the following aspects of PLM and Open Source:

Open Source Openness
This one is an important. Open Source imposes openness and democracy. I think, a portion of openness will be a good addition to today’s PLM business. Open Source can provide a different flavor of PLM implementations. However, the discussion about what means Open Source and how it will go beyond just providing free licenses.

Open Source Communities
This element is a integral portion of any open source projects. This is what I call “a life indicator” for the open source. If you have a working community, an open source project will have a future. Therefore, wiliness of people to be involved is important. Multiple projects can bring even more interest into development of communities. More projects in this space can create a viral effect on the future development of Open Source options for PLM.

Founding Companies
I think, commercial companies play a significant role in the development of Open Source projects. We all know about the role of IBM, HP and others. I think, role of founding companies is important. However, Open Source needs to be a “development community” project first and only after to become part of the commercial company. Aras, as a company associated with PLM and Open Source, can play an interesting role in this space.

Open Source Enterprise
This is not a simple question. In my view, Open Source was established mostly as “development projects”. Enterprise is different from many aspects- business, support, customer orientation. If a community of people involved into the open source PLM projects will grow, I can see two potential routes for PLM Open Source to enterprises: 1/IT development project; 2/Commercialization by a particular software vendor.

What is my conclusion? Open Source and PLM are still “Terra incognita“. We can see and hear lots of opinions, numbers, prospects. The key question today is how to develop open source community projects in PLM and how to make them connected to other open source projects related to enterprise software. A potenal candidate can be, for example, Drupal. This can be an interesting path for Open Source in the enterprise. Just my thoughts…

Best, Oleg

Closed Thoughts About PLM Openness

June 29, 2010

My post Open vs. Closed PLM Debates last week and related Fortune CNN Money Blog article by Jon Fortt – Chrysler’s Engineering Software Shift created some very interesting experience for me. The level of interest, especially the amount of comments, is one of the things that drive my analyzes after actually post published. The special experience I’ve got with this blog post was that I got most of the comments by email and not on the blog website. It made me think, may be about additional issues related to the “openness and PLM” that I didn’t discover before. So, I decided to pull some of my thoughts about PLM and Openness here and see if we all can generate a reasonable discussion about that.

Are We Open?
I think, CAD/PLM lives in the world of competition on openness. The question “Are We Open?” sounds as a wrong question to me. In my view, there is no Black & White in openness. The issue of openness needs to be related to the specific characteristics of software that can be measured. The examples of such characteristics are – Open API, support for existing standards, availability of software for customization and extension, ability to publish or exchange information about data models and file formats, etc. I think, we can collectively come and find more characteristics. A special characteristic of openness is the open source, and it needs to be analyzed separately, in my view.  So, in order to get an answer on the question “Are We Open?”, we need to come and analyze various aspects of software.

PLM and Integrated Software
I’d like to emphasize the topic of “integrated software” in the context of discussion about openness. It is a tricky one. Enterprise Software, in general, as well as CAD and PLM software specifically is growing and the question of integrating different pieces of software becomes more and more important. Customer demands are to have a better integrated software and software vendors (especially a big ones) are focusing on the questions how to make it happen. Sometimes it comes to the point where pieces of software that before had a weak connection becomes tightly integrated and dependent. Is it a good thing? I think, it depends… The latest debates about DS V6 platforms and tight connection between CATIA and ENOVIA are actually coming to this point, in my view. I think, the intent of DS is to provide “best integrated software”. Does one size fit all? I think, DS engineering wizards definitely had in their mind a question how to create a next level of CAD and Data Management functions in a single box. I can see similar trends also coming from other CAD/PLM vendors.  The best non-PLM association I’d like to come with is Apple platform. When it comes to the unique experience customers are having with Apple product, you can make a compromise on openness of the platform. Will it continue for the long run, I don’t know? However, I see these strategies work for Apple these days.

PLM Openness and Customers
I think, this is a real measure of your openness. Whatever you do is for customers and not for competitors. Customers need to have an ability to define what are their openness needs. The reality I see on a customer side is that nobody is using CAD and PLM software coming from a singe vendor. The maturity of industry will be measured by the ability of vendors to come to the compromise of how to serve customers with the best performing software. How vendors can achieve it? It is a very good question… However, it shows the overall maturity of the industry.

What is my conclusion? I think, openness is ready hard. To play this game right, you need to see both worlds at the same time – customers and competitors. And this is the exact order how to see it. My bet is that openness wins for the long run. I think, we will see more software that will be measured by how it performs for customers. One of the performance characteristics will be how information managed by the software can be open and available for Pull (I’m going to post more about the “Pull” in the future) . It will be the end of “openness competition”. Just my thoughts…

Best, Oleg


3 Steps To Improve PLM Collaboration

June 28, 2010

In the landscape of PLM, collaboration is overused word. You can hear “collaborative” a lot, and you’d be thinking all problems of collaboration are already solved. However, I think, we are only in the beginning of starting to have a real collaborative solution. The most important thing for a collaborative solution is a context. To have a right context is important, otherwise you cannot efficiently collaborate. Think about emails or files. You cannot be sure your co-worker or manager is looking on the same piece of information.

I had chance to read Google Enterprise blog about New Sharing Options in Google Docs. Take a look on the following video.

It made me think about some fundamental things we are missing when creating PLM collaborative solutions.

Unique names for resources
PLM manages different pieces of product data, documents and many other resources. Depending on what is the scope of PLM implementation, the amount of data can vary. What need to be done to collaborate efficiently is to have an ability to provide a unique name for data (or resources) we want to collaborate on. By doing that, we will be able to have a constant context to use for collaboration. The unique resource name can remind you web (URI) – everything on the web has their own unique naming. We can use the same to identify your product data in the organization.

An obvious, but very important addition to unique naming/identification. If there are resources, we need to apply security rules to ensure you have appropriated rights to collaborate on the specific data set. Security functions can be global for the organization or local to facilitate a collaboration need for a specific person who may be outside of the organization.

Share Options
The third component of a successful collaboration strategy. You need to be able to invite a person and to provide a particular piece of data as a context. If you use unique naming and appropriated security model, you cannot go wrong. Your context will be always well identified and access by multiple people at the same time. You can use various nice options for share such as an ability to transfer, share, ability to redline, edit or just view. You need also to provide information to users who is sharing this context, for the moment.

What is my conclusion? I think, PLM is using word collaboration very intensively. However, there is one thing, many of the collaborative PLM software is missing – unique name of the resources to collaborate on. Unique resource names can solve lots of problems PLM has today. Just my thoughts…

Best, Oleg


PLM and A Single Point Of Disagreement

June 25, 2010

When you talk to a sales person from one of the PLM companies, you for sure will be exposed to a “Single Point of Truth” vision. On the surface, you can see it as a very powerful message. What can be wrong with having a single point of truth on all design, requirements, engineering bills, manufacturing plans, support materials and customer calls? Sound a great opportunity finally to organize all you have related to your product development. However, is it really true?

View of the World from 9th Avenue

There is a legendary New Yorker magazine cover by Saul Steinberg called “View of the World from 9th Avenue.”It comprises a “map” of the world from a “New Yorker’s” point of view. Looking west from 9th Avenue in Manhattan is the Hudson River. Beyond that is  a flat view of the rest of United States. Then you see the Pacific, Japan, China and Russia. If you think about manufacturing and product development, you can find a very similar picture, depends on who is the person you are talking to. My conclusion is there is NO single point of truth. Everybody sees the problem or product data differently.

Single Point Of Truth Process

So, what happens when PLM implementation comes to the company? In the nutshell, every PLM implementation is trying to create a single point of truth for the organization. It means to go all the way from a data mess to the agreement about how to manage product data in the organization. The most typical process is when a company is taking PLM vendor’s blueprint of a data management schema and starting to customize it. This is a main reason why the process of PLM implementation is long and painful. You need to have different people in the organization to agree about data management principles. This is a very painful process. People in company departments have different goals and priorities. This is similar to New Yorker’s view from 9th avenue. There are multiple PLM methodology to deal with this called “Role-based views”, but technologically they based on the assumption to have a single model of everything.

The New Goal: Single Point Of Disagreement?

One of the possible ways to start doing PLM differently is to stop applying this painful “agree on a single model” process. People need to have a way to work in the world where their views are different, but their views can be synchronized and integrated. This is a not trivial task. It seems to me as a more appropriated way to solve this problem in comparison to what we have today. What need to be done is to find what are differences between people view on data in the organization. It can help to create an integrated product development data landscape.

What is my conclusion? To create an integrated and balanced way to manage product development is not a simple task. PLM is missing this point and assumed the actually data model integration will be created during PLM implementation and will be driven by customers. I can see it as a mistake that makes an implementation process lengthy and implementation costly. To resolve this problem will help to bring a desired simplification into PLM world. Just my thoughts…

Best, Oleg


Open vs. Closed PLM Debates

June 24, 2010

I read Fortune CNN Money Blog article by Jon Fortt – Chrysler’s Engineering Software Shift. In the competitive world of PLM software it raises again the question about what is the better choice – Open or Closed? The context of this article is leaked information about Chrysler’s movement from CATIA to NX or, maybe more from DS PLM product lines to Siemens PLM. However, author made a nice association between the engineering software story and bigger story related to the strategy of closed platforms such as Apple, Oracle and Cisco. It made me think how I see the future of Open vs. Closed routes in PLM.

CAD Openness
The debates about openness of CAD and later PDM/PLM software is not a big news in the industry. From the early beginning, CAD applications tried to protect themselves, by creating a proprietary format to store geometrical models and drawings. For the long period of time and until now, practically all leading CAD vendors are using closed file formats. It created a separate industry of companies working on translators and supporting so called “interoperability”. The discussion about CAD openness and interoperability is probably the longest one I can remember in the history. I’m not sure we’ll be able to see the end of this story. The current situation reflects clearly the conflict of vendor’s business models and user interests. CAD industry veterans outline the future of CAD (MCAD) will remove this barrier and make CAD product more open. You can take a look on my blog post – CAD Future: How To Liberate Data.

PDM/PLM and CAD Integration
In the beginning, PDM was about just managing meta data about CAD File. It started from revision management and release control. Most of PDM system in the market managed to have multi-CAD integration strategy by supporting multiple vendors. However, customers were interested in more integrated products. Evolution of PDM product into PLM, including their ability to manage a diverse set of product data and processes, just added more fuel into development of future PLM Platform strategies. It was a time, when vendors started to think seriously about how to create completely integrated product suites. Dassault V6 is a first kind of system that introducing CAD/PLM system bundle.

CAD vs. PLM Openness
What is the difference between CAD and PLM Openness? In my view, it is an interesting turning point in the overall story of engineering software. In the real world, customers are working with a diverse set of tools. In the world of pure CAD, their decision to work with multiple CAD products was hard, but doable. Many of the customers (especially big ones) worked historically with multiple CAD products. PLM is adding new flavors in the old story about CAD openness and interoperability. This is the place where the world of CAD files ends and companies are starting to think more about how to manage all engineering and enterprise data assets.

The Future Is Open?
The most important question is how we can move into the future where data will be more open. I think, many of the companies, are thinking how to solve this problem. It looks like a very promising future to make data open. However, the business reality is different and companies are continuing to make profits from closed platforms. The following quote from Fortune article is the most important, in my view:

When the closed strategy works, it can yield outsize profits — Apple’s recent financial results being a prime example. But it can backfire, too. If a company’s bundle of products doesn’t work together well enough to justify the added cost, customers can get turned off. That’s the danger for Dassault. Joe Barkai, analyst at IDC Manufacturing Insights, says that in this age of consolidation, automakers are more likely to be looking for flexible design systems that can easily share data with a new partner or supplier.

What is my conclusion? I think, we are going to see more and more stories related to development of Open startegies. There are clearly two possible options: 1/To create excellent integrated product suites and sell them to customers (i.e. Apple story) or 2/To develop open strategies. My take – I think Open game is hard. However, the prize can be big. Just my thoughts…

Best, Oleg


PLM and The Power Of Pull

June 23, 2010

I had chance to have a talk with David Siegel, entrepreneur, speaker and the author of a new book “Pull“. We met during the Semantic Technology Conference in San-Francisco yesterday. The sub title of this book state – The Power Of Semantic Web To Transform Your Business. You can take a look on David’s web site – The Power Of Pull. My first question to David was about the idea of name “Pull”. His answer was very interesting. The working name of the book was Business 3.0. However, people didn’t get this name. So, he tried to get to the bottom line of what his ideas are about and came to the definition of “Pull”. In the nutshell of Pull, we are going to move to the world where information will be available at the time we need it. I found it interesting, since it solves, in my view, one of the most important problems we have today with enterprise software in general and PLM specifically- how to make systems more intuitive?

Pull Concept: Product Lifecycle Management

Here is my simple explanation about how to shift to Pull in PLM. On the very basic level PLM is about how to track Products and Information related to products from the early beginning (actually interaction with potential customer about what product they want) through all phases of product design, engineering, manufacturing and disposal. The interesting thing – I can see this chain does exist today. However, when we want to touch it, get information about it, analyze it, we need to push a lot of stuff around us. It related to all aspects of a product life cycle. We need to work on so many sources of information to get actually what we need. Existing systems are trying to organize everything around products. The so called “single point of truth”. So, it puts “the product” is in the middle. The shift to Pull may happen when we will put people in the middle of the circle. Then all information about a product will become findable and available for pull.

How to Pull Product Lifecycle Data

Product Lifecycle is represented by data. There are lots of data around us representing ideas, information about potential customers, requirements, product design, manufacturing, supply chain, retail, physical products. The most interesting question for me is how we can make this data available and retrievable at the time we need it? The naive answer – just search across all web sites, intranet and enterprise software products to find all what I need. However, here is the problem – I don’t know what I need to get. I’d prefer somebody will take care to organize information around me. This is “aha moment” to think about “pull” as a different concept. Product Lifecycle Data need to have an ability to be organized to become available. I can see initial shifts into this direction across multiple spaces. One of them is personalization of search and organizing of social and real-time information. I think, we will see more products going into this direction.

PLM, Pull  and Openness

Openness will play a significant role in PLM shift to Pull. Today, PLM is the kingdom of proprietary information. Software vendors take a significant care about how to make data created in their software products available only in their own products. There is a simple and obvious explanation to that- business. They are making money by selling their software products. Just think about, what if tomorrow they will make money from making data created using their software product available? This is a major shift to “Pull” concept. It will shift industry from a closed and proprietary world we live today into the open world of tomorrow.

What is my short take on this? Important. Take your time to read this book. It contains lots of ideas in diverse set of business fields. Find your problems. Focus on low hanging fruits. Get it done.

Best, Oleg


PUI: Not PLM UI. Future User Experience.

June 22, 2010

I’d suggest to have some fun in the beginning. What is PUI? Do you think this is a new PLM UI? No, marketing didn’t get it yet… I hope, they will not do so. I’ve got this PUI TLA reading Enterprise Irregulars Blog – Three Types of GUIs: Past Present and Future. Thanks ArnoldIT blog for referencing me to this article. The author explains three different type of GUI – Task UI, Desktop UI and Process UI.  I found it very interesting. It made me think about the evolution of user experience in engineering applications and PLM.

TUI, DUI, PUI in Engineering applications
If you think about CAD applications, they look very task oriented. You have a 2D drawing (lately transformed into 3D model) and everything you do around this drawing or model, can be considered as tasks. You can create a new model or modify an existing one. Another big group of applications, CAE is also focusing on different tasks (i.e. calculations, etc.) around 2D/3D models.

The initial development of data management application created an idea of “Desktop”. In the beginning, data management was focused on “meta-data” management. Meta-data was considered as everything surrounding CAD files. The need to present data in various forms created lots of modifications of data forms. You can find it in most PDM/PLM applications during last 10-15 years.

Process was one of the top ideas that PLM focused on for the past 5-6 years. Organization can understand a “process language”. Software vendors in PLM space learned it from their big ERP brothers. I had chance to see lots of PUI modifications focusing on different processes.

PLM and PUI problem – Scalability
In my view, process orientation was a big excitement for PLM and other engineering software companies. It created a scalable model to grow the amount of applications, they can provide to companies. Since everything is a process, imaging how many process-oriented application you can create using this approach. Licensing wasn’t the last reason in this decision. Looks like PUIs were the way to go. PLM companies are tried to implement PUI approach. Take a look on the quote from Enterprise Irregulars Blog:

Now imagine business or enterprise apps that are process based, not single task items knit together by DIY process: An app that can pick up an idea, issue or request and run it through an unpredictable process that might look like a ball of yarn all the way to an implemented idea, a solved issue or a happy customer.

Where is the problem? In my view, the biggest problem is a scalability of this process. Every organization is different. Process oriented applications became very complex. This is what happens to PUI in PLM. Customers cannot implement an existing process oriented application. However, to tailor it will skyrocket cost of services.

Future User Experience
It is clear to everybody that the level of the complication kills. So, what will be the UI in the future? Many of companies are looking for the next big thing in UI. I can identify two main trends in PLM UI simplification – 1/ Embedding into the mainstream UI (i.e. Email, Excel, SharePoint, etc.), 2/Imitating successful applications in consumer space (i.e. Facebook). What will be the next big thing in UI? Do you think it will come from augmented reality? Games? Voice? I can see SIRI acquisition by Apple as the sign of possible direction in the future. I had chance to write about how I see the future user experience in my blog too.

What is my conclusion? User Interface is a hard problem in PLM apps. I think, we are in the mist of UI discussions. I can see many people are innovating in this space and trying to find a solution to this problem. We are in the beginning of mass migration of application from consumer space to the enterprise software. It will be very interesting to get back to this discussion in a couple of years.

Best, Oleg



Get every new post delivered to your Inbox.

Join 288 other followers