LinuxWorld 2008 Prediction #8: Industry applications

Print Friendly

Podcast of LinuxWorld 2008 Prediction #8

Although I’ve previously published the slides for the talk I gave at LinuxWorld 2008 in San Francisco, I thought it might be useful to add some additional comments in the blog about each of the eight predictions I made. This is not the full text of what I said nor a full discussion of the slide, but just some ideas that flesh out what I meant. The full one hour video of the keynote talk is now available at the conference website.

Slide made of prediction at LinuxWorld 2008

If we consider a simplified software stack, right above the hardware is the operating system. Clearly with Linux, FreeBSD, OpenSolaris, and others, free and open source have had great success.

Above this we have what IBM and others call middleware: http servers, Java web servers, databases, email servers, groupware, and other types of servers. Apache rules the http server space, both open source and proprietary, while the other categories have strong proprietary and open source entries. Apache Geronimo, Red Hat JBoss, MySQL, Postgres, Apache Derby, Yahoo Zimbra, Scalix, and many other open source projects have been successful in getting developers and users to adopt them.

Above this we have the cross-industry application areas like ERP and CRM. Compiere is well known in the open source ERP space, though there are other contenders and people seem to be looking seriously at what open source providers are developing in this area. In the CRM area there are many projects, though SugarCRM is probably the best known.

As an alternative to “CRM on my server,” there is also the software-as-a-service offering. Of course, in the proprietary space, Oracle and SAP aren’t doing so badly.

Just about every industry has customers that need to be managed, even if you call them “clients” or “donors.” Every industry deals with payroll. Every industry has supply chain, even if it’s to order beverages for the vending machine down the hall. I call this category “generic businessware.”

The big systems provide common modules and then allow you to extend the system in various ways. So if you have an “enterprise application suite” then it might have the usual software but also custom developed applications or user-designed reports, for example.

Above either the operating system, middleware, or the generic businessware levels it is possible to build applications that only work in one specific industry. It is these industry-specific applications that I am addressing in this prediction.

Sakai logo

So far we have seen very few open source industry specific applications or frameworks outside the public sector. In Education, both Sakai and Moodle are doing well and compete vigorously against each other.

Over at Eclipse there is the Open Healthcare Framework (OHF) Project which is described as

… a project within Eclipse formed for the purpose of expediting healthcare informatics technology. The project is composed of extensible frameworks and tools which emphasize the use of existing and emerging standards in order to encourage interoperable open source infrastructure, thereby lowering integration barriers. We currently provide tools and Frameworks for HL7, IHE, Terminology, Devices, and Public Healthcare Maintenance.

Outside the public sector, I’m aware of the OpenQuote project for the insurance industry. Beyond that … not so much.

Open Healthcare Framework (OHF) Project

I’ve been quoted in a couple of articles because during the talk I said that I’m tired of waiting for more open source industry applications. That is, if more are coming, they are taking their time!

Given the open source industry applications that already exist, we know that the future is not one that only has proprietary industry applications. Will all such applications eventually be open source? If not, what will be the balance between open source and proprietary applications in this area? Right now it is overwhelmingly in favor of proprietary.

Maybe that’s the way it will stay. There’s no guarantee we will see a huge movement to open source industry applications. Will the wait for the “Year of Desktop Linux” be replaced by that for the “Year of Open Source Industry Applications”?

I don’t think we need to wait forever to find out. Open source in general has huge momentum but there may end up being market categories that are resistant to the movement. At some point you have to look at what is going on and say “you know, I really don’t think it’s going to happen.”

I’m offering this observation as someone who has looked at open source for some time, not as someone who is particularly bemoaning the absence of these applications. That is, I don’t personally need the open source apps just as I don’t personally even need the proprietary ones.

The public sector really seems different here and I think it’s important not to extrapolate to other industries. That is, don’t say that just because open source is so strong in Education that means it is just a matter of time before it becomes dominant in Chemical/Petroleum.

Now if you disagree with me here, just as you might disagree with what I’ve said in the other predictions, you have a great opportunity to prove me wrong.

Start coding!

Get a community together. Figure out your business model. Evangelize. Build to real requirements of industry. Do it better than the proprietary guys. Don’t just solve the problems that were figured out before, offer solutions to the really thorny issues that affect us now and are growing worse.

Of course, the “proprietary guys” won’t be sitting still either. Competition will drive innovation and value for the customer.

Unless something changes drastically, I don’t see open source growing rapidly and soon in the broad industry application space. Maybe it will just take more than ten years, but as I said in the talk, “either it will happen or it won’t.”

If you want it to happen, make it so. That’s the power of the model and the community for important software that solves real problems.

Previous: LinuxWorld 2008 Prediction #7: Standards


  1. There is a significant amount of open source for education, for example SkoleLinux . However, there isn’t a significant deployment of open source software in my children’s schools; nor in other schools I know about.

    The US Government seems to think US to back 21st century learning there’s a gap that needs addressing between ‘Business’ deployment of information technology, and ‘school’ deployment of information technology; and is proposing to invest $50M in attempting to bridge the gap in the USA.

    25 cents per head. A drop in the ocean. Better than nothing, though.

    I think it’s related to the way that schools acquire software. It drives them towards ‘renting’ deeply-discounted proprietary software, rather than grabbing for themselves the profusion of open-source software that is scattered all over the public Internet and building on it.

    OS/2 and SmartSuite are history, we know that. Of more interest is what the future holds, and we need these schoolchildren to build it.

  2. Hi Bob,

    Firstly, can I say thank you for mentioning OpenQuote, I am part of the project and it is something the whole team dedicates a lot of energy towards, so any help raising awareness is really appreciated.

    As you say, open source solutions in vertical markets are few and far between. I believe that a major factor is the large amounts of money traditional software houses spend, both developing and delivering solutions that have huge feature scopes (some may say monolithic) covering the entire vertical. This not only makes it hard for open source projects to compete directly, but also for new software houses using the traditional charged for/commission based licence approach.

    A problem with traditional software in vertical markets and its licensing is that it alienates large sections of the vertical being delivered to. Two examples:
    1. Organisations who specialise in a very specific process of the vertical (i.e. a small cog in the machine), these don’t require most of the functionality delivered by these large commercial solutions, and they are reluctant (or cannot afford) to pay large licence fees for software that they will only partially use.
    2. Larger organisations who need all of the functionality but may already have software in place for most of it, in these cases finding software to fit the gaps may prove a more cost effective solution than replacing the lot.

    For OpenQuote we hope to take advantage these groups, this means small to mid-sized insurance brokers/agents looking to take their quotation process on-line, and insurers with in-house developed solutions who need to add on-line quotation functionality. We don’t pretend to offer an end to end insurance solution, we have deliberately specialised in an area where we see a real need. Where will this take us? Well perhaps as the software becomes more popular it will grow to deliver more areas of functionality, but that will really be something for the community that grows around OpenQuote to decide.


    Matthew Tomlinson
    Applied Industrial Logic

  3. Hi Bob,

    Check out TriSano …

    Just open sourced yesterday.

    “TriSano™ is an open source, citizen-centric infectious disease surveillance system, the result of a collaborative effort between county health advocates, state health officials and Collaborative Software Initiative.”


Comments are closed