CCK08: How to Profit off of Open Source, Or at least pay the Bills

I have been struggling with Stephen Downes views of Open Source and how it should be the wave of the future to ensure the most effective learning environment possible. My part time job teaching for RMC would have not problems with this model since what I teach is not designed around cost recovery.

However, my day job with Innovatia as an Information architect left me wondering how I could maximize the user and learner experience using open source (and not locked behind a LMS or username/password) while still enabling my company to earn a living producing telecommunications user documentation, and offering telecommunications training with our own LMS hosting service. I was wondering how I could take what I had learned form this course (which my employer was paying for) and implement it in the corporate training world. There appeared to be limited use of this “open source” model for “profit” organizations.

I had just read Disrupting Class by Christensen (Review here) which largely advocated a similar open source concept (very disruptive!) which I thought would work in the public K-12 government funded environment, but might not work as well in post-secondary or corporate training world.

Then serendipitously, Stephen provided the course with “Models for Sustainable Open Educational Resources” as an optional reading for Week 10. Then the light bulb came on for me 🙂 I could now discuss open source publication with my Innovatia sponsors when I brief them about the course and provide business model options.

To summarize Stephens’s views on funding for Open source authoring, he discusses:

1. Endowments model
2. Membership Model
3. Donations Model
4. Conversion Model
5. Contributor -Pay Model
6. Author-side Payments
7. Sponsorship Model
8. Institutional Model
9. Government Model
10. Partnerships and Exchanges

Now I will move into some specific exampels aroud authoring in the telecommunciations world.

In the telecommunications world, typically user manuals are published under the institutional model where the manufacture absorbs the cost of developing the manuals into the products list price. So there is always pressure to reduce authoring and distribution costs. This has led to XML authoring tools to allow maximum reuse of the content into multiple formats (PDF, web html, on product html help, CD-ROM, and Print). Some manufactures provide limited copies or access and then charge for extra access or copies at an attempt at cost recovery.

Training is a different story. Typically, training is based on a cost recovery basis. Instructional designers figure out how the content needs to be broken up and for what target audiences need addressing, and then build courses around that. It is not unusual in telecommunications to see 5 day classroom courses costing in the $3-5,000 range plus the cost of travel and accommodations to the manufactures training facility. There is a real move afoot to make training online either by eLearning or virtual classrooms with virtual labs (connectivity to working telecommunications equipment). Going online reduces the course delivery cost, reduces the customers travel cost to almost 0, and allows the frequency of the course offering to increase to customer demand levels since there is no physical classroom limitations (in other words, moving to the MOOC model of almost unlimited registrations).

In an effort to further reduce training costs, organizations are looking to DITA-XML and the OASIS eLearning specialization so that the technical publications content can be reused for eLearning content development. This works fine if you are a big player like IBM, Sun, Cisco, or Xerox. But, what if you are a small player and can’t afford these expensive authoring tools and complicated information models?

A solution gaining increasing popularity is where the manufactures “Sell” the rights to the training to a specialized training provider, who then uses these XML authoring tools to a large scale to produce the content, and the manufacturer receives a royalty. Sometimes there is partnership where the manufacture pays a portion of the training provider for the content conversion. The benefits for the manufacturer include reduced training costs for the customer, increased training availability, and usually better training outcomes, thus reducing the need for customer support staff. Oh yeah, the customer is so happy, they buy more training, and come back and buy more product instead of going to the competitor. The specialized training provider gets the benefit of an increased library, and access to customer list of potential learners.

Now how do you make this open source and still pay the bills. One way would be to make the training content truly open like MIT. To recover costs, the manufacture or the training provider could charge for certification exam, access to mentors, discussion groups, and access the training equipment. So if certification credentials are import to the customer, then this model works.

Now Stephen Downes suggests why people even need to pay for training. If the user manual is well written, people blog and do customer evaluations about products, then it becomes easier for John Q Public to never have to pay for training. However, some products are complicated like in telecommunications. I do not know if most people have seen the inside of a telephone company Central Office, but it can be a little bit complicated (See picture). Handing a technician a user manual does not cut it. There is a movement afoot to make technical documentation simpler. A manufacturer has a profile for each customer and the equipment it operates. Using an information model identifying what software features the customer is initially using and DITA XML topic based authoring, each customer can dynamically build their own user manuals based on a query to the manuafacturer’s information repository.

This sounds a bit nebulous so I will give an example: If the manufacturer has 8 ways of implement a feature, and the customer has chosen just one way, then the customer gets the instructions for the their chosen way, and does not receive the extra non-essential material.

This technique can also work for training. Customers identify themselves, what level of training they need, and what job function, and a course can be built dynamically. LMS have been oeprating in this fashion for almost a decade to the point of ven pre-testing to allow the learner to opt out of trianing they already know (wouldn’t it be nicer if the LMS already knew your trianing profile though? But that is an enterpise systems integration dilemma suitable for another blog post).

This works for the initial implementation, but can you imagine the manufacture maintenance of the back end customer information. What if the customer changes features without telling the manufacture. In this case, a simple questionnaire can gather and update the customer’s features list before generating the unique manual or training course. The cost of paying for this method would still be born by the manufacture or training provider, although in the case of the training provider, there would be cost recovery. Some models have proposed a charge for these spontaneously built user manuals that would be cheaper than having to pay for the whole, unneeded suite of manuals (that just gather dust if print, clog up hard drives and become out dated in both cases).

To make the system truly user-friendly, there is some thought about having the manufacture making the initial information available to the customers, and then have the customers customize the information repository to their specific needs. For example, I have already explained how the customer can build dynamic content around their features, but a customer could also using Wiki-like features, go in and upload their system schematics, photos, maps, or IP addresses and then have the content repository publish a unique document for the requestor. The automotive industry is already moving in this direction creating unique user manuals for each customer based on the features selected at the time of purchase. The arguments is why publish 3-400 page car manual that no one will read because there are too many radio options listed so people never learn how to operate their radio (or car jack, or lighting system etc).

A final thought is using Wiki-like editing features would be to allow users to spontaneously generate examples for procedures etc that the manufacture may not have time to develop while trying into get the product to market. The Wiki type editing would also allow users to point out errors exactly at the point of confusion instead of having to use elaborate change request feedback mechanisms. Current updating of publications is a long and expensive process sometimes taking months to a year before the documentation is updated. This means customers become turned off and may switch to other manufacturers who are willing to invest in having up-to-date and correct information. Some manufacturers are reluctant to go to this user community extreme for fear of lawsuits. In this case, the manufacture will have to invest some time in getting staff to validate and then approve this form of feedback. However, the benefit is a happy customer that keeps returning.

If you have made it this far, thanks for reading the online ramblings as I try to sort out the user/learner centric philosophy, telecommunications training and manuals, and reconcile it with a technique of making it pay for itself.

Thanks Stephen!

Tags: CCK08