We’ve had a lot of questions lately about whether SharePoint constitutes content management for technical publications and dynamic information delivery. There was also a paper published by the Aberdeen Group about Securing Unstructured Data that made the point, none too gently:

Companies face a significant challenge with respect to safeguarding critical data in unstructured formats (i.e., in documents and file systems), a problem made worse by the ubiquity of web-based portals and collaborative tools such as Microsoft SharePoint.

They tracked:

Number of actual security-related incidents related to unstructured data (e.g., data loss or exposure)

Number of non-compliance incidents related to unstructured data (e.g., audit deficiencies)

Human error related to unstructured data (e.g., policy violations)

This tracks with what we find when we see companies trying to use SharePoint as content management rather than what it’s good at: you get basic for search capability, some audit trail information, and ad-hoc collaboration.  If all you’ve got is a file system, then SharePoint is a massive improvement. Truly, getting any of this where you previously had none is a huge win.

However, if you’re talking about true content management, then it’s critical to define what you mean by that and what you’re actually trying to achieve with it.

Most SharePoint strategies fall short because SharePoint itself is not truly content management. SharePoint cannot handle any complex associativity or relationships between compound document structures. It can’t enable life cycle management or process automation. SharePoint is good as a delivery vehicle for intranet publishing sites — for sharing published content for review and retrieval across enterprise functions. SharePoint is more of a portal view into corporate file servers.

SharePoint is simply a better way to access the files you use every day. But it isn’t content management: SharePoint is document management.

Content management systems manage information modules or chunks — parts — of documents. Document management systems focus on asset management and, as such, manage complete documents — finished information products. The input to a document management system is the output of a content management system.

A content management has structure built in — structure around the content, structure around access, workflow, monitoring and filtering, encryption, and management. A true content management system can pass the same validations that a class three medical device can pass.

A CMS manages information not files. A CMS manages everything required to deliver information to consumers from it’s creation through to publication, distribution and delivery. A fully-functional content management system has the ability to initiate workflow, to make decisions on document construction based on metadata stored inside the information components as well as the system itself. It tracks more than who last looked at the document or who last modified the document: it tracks functional activities related to the document. It associates content reviews to information components so that information quality improves over time. A content management system can answer the question “Where is this used?” and “How many documents are affected if I change this drawing, legal boilerplate, or feature description?”

In DITA environments, these kinds of questions are critical and having the ability to manage topic-level information components is essential. It is not uncommon for even a small DITA deployment to have literally 1000s of information components that are combined, recombined, and reconfigured into multiple documents and deliver those documents into a multiplicity of formats — PDF, HTML, eBook, Text, online help, Eclipse Help, etc.

For technical publications, it is this end product — the PDF, the set of HTML files — created from mounds of other, smaller, content components that would be shared with the enterprise, not the components themselves. A content management system would automate the composition, publication, and delivery of these final formats into a SharePoint environment for collaboration and review across other parts of the enterprise.

PTC’s Arbortext Content Manager integrates with SharePoint to provide this level of enterprise sharing and collaboration. Companies get the benefits of SharePoint for internal collaboration for social product development without hindering the activities required to get to that point in the product development process. Arbortext Content Manager doesn’t replace SharePoint, rather it delivers final information products to SharePoint for enterprise collaboration.

Quotes taken from the Aberdeen Group paper: “Securing Unstructured Content: How Best-in-Class Companies Manage to Serve and Protect” published June 2009 by Derek Brink.

This week, I participated in a stimulating and lively discussion on LinkedIn. Here’s the question:

What’s the best toolkit or set of tools for use by a team of content creators to automate the process of getting content from differing sources and file types (text docs, media, graphics) online?

The first thing I do when I see a question like this is find out more about what’s behind the question. There’s a lot of implications in a question like this. I want to know whether the asker is talking about multi-channel publishing or automating the hand-work of posting content to a web or portal location.

These are two very different goals.

Here’s what I asked in return:

Are you looking for “lights out” publishing that can be completely automated to more than just PDF output? If that’s what you’re looking for then a true XML publishing system (not afterthought, bolted-on one) is your best bet here.

We work with Arbortext and have customers who are completely automated, achieving 90% reuse, and can guarantee the quality of their information because they’ve certified their publishing system so that they can pass FDA audits.

In addition, they’ve been tracking translation costs the last seven years across multiple business units that have both XML and traditional publishing systems and have data to prove that translation costs can be an order of magnitude lower once you go to a true XML publishing system like Arbortext.

Not everyone needs that level of quality guarantee, but being able to reduce cycle time, increase quality, and save money all at the same time is good for everyone.

Happy to follow up with you offline.

Sure, I’ll mention my product line. That way the questioner has somewhere to go to get information about whether what I’m talking about applies to their situation and goals.  But I’ll also push to move the conversation about vendor-specific features, tools, etc offline and out of the public discussion in a place like LinkedIn. I’m not there to advertise, I’m there to lend expertise if it’s appropriate.

Many of the suggestions that followed encouraged the asker to do some automated script handling via XSLT in order to bring the content into a manual, desktop publishing application like InDesign.

It’s a suggestion that still boggles me.

If the idea is to reduce the manual effort and shorten cycle times, then you really want to avoid introducing the manual steps required to flow XML through InDesign.

Arbortext provides the full system of products available to produce multiple output formats without having to build the tools to bridge applications from multiple vendors. All products are tested and delivered together, so the burden isn’t on your team to validate upgrade paths of individual tools or individual products. Arbortext does all of that for you.

Arbortext has a history of success in the publishing industry. We have several magazine customers who automatically produce pages that are derived from very complicated layout rules.

For example, Arbortext can auto-fill advertising content. it’s very common for us to see customers implementing rules like these:

• if one side of the page has dark print, then we want a similar dark ink image on the reverse side of the page because the pages are thin
• If Advertiser A is on the left hand page, then Advertiser B cannot be on the right hand page; and/or it must be N # of pages away.

Also, if an editor wants to hand-tweak the page before the deliverable is finalized, Arbortext the capability to do this: editors can move graphics and reflow text if they want to.

All of this is available from within the publishing engine. There’s no need to step into yet another design tool to make changes.

And do I need to mention that there’s an Arbortext Content Manager that includes workflow capability — completing the full single-vendor solution picture ?

You might say that Arbortext Content Manager is a late comer to the publishing industry. It wasn’t originally targeted and publishing. However, we’re starting to see it unseat traditional partners like Documentum because of performance issues and globally distributed teams.

It was a great conversation. Everyone on the thread contributed real technical information rather than just product mentions designed to confuse the issue — and add more work — for the questioner.

This week I had a great conversation on LinkedIn. Here’s the question that got asked:

What’s the best toolkit or set of tools for use by a team of content creators to automate the process of getting content from differing sources and file types (text docs, media, graphics) online?

My next post will address the question itself, but the experience of answering questions on LinkedIn has been interesting — and a lot of fun. This particular question generated a lot of conversation that was stimulating to everyone involved.

For now, I want to talk about what I’m finding on LinkedIn.

As you might expect, there were a lot of opinions and a lot of suggestions. But there are always two things that surprise me when people answer questions on LinkedIn:

1. Almost no one asks the question asker questions to find out exactly what their situation is.Context is meaningful to vendors. It’s just as important to find out when you don’t fit as when you do.  On LinkedIn, you always get vendors saying, “Try mine!” with no qualification whatsoever. How can a question asker know if the product being suggested applies to them at all when a vendor doesn’t take the time to ask about their business, their goals, or their requirements?
2. The impression that “open source” solutions are free.Open source is not as free as it sounds. You’re responsible for independently validating the products you’re stringing together when it’s time to upgrade (any one of them). You’ve got to build the tools to join systems together (if you do it the wrong way, it’ll be harder to upgrade than you think). You’re at the mercy of other people to implement that one feature that you really need (unless your company is sponsoring someone as a developer on that open source project — which nearly no one does).It often requires a software developer in house who can manage the open source products you’re using. Luckily, these guys can typically be found in your IT department and they’re really, really good at what they do. These guys are a shining light in the darkness, but all of these other costs need to be factored in before you just jump in.

Our podcast schedule has increased. We’re delivering one every two weeks now.  The podcasts are our attempt to get things that are inside our heads and out into a searchable, learnable, location. It’s an attempt to take that tribal knowledge and make it less tribal.

Three podcasts were published in the last two months:

1. Berry Braster, TedoPress — his experiences, the expertise and business of Tedopress, and his advice for preparing content for translation.
2. Gareth Oakes, Global Publishing Solutions — the expertise and business of GPSL and Gareth’s advice for working with Arbortext Advanced Print Publisher (formerly 3B2) and stylesheets.
3. Keith McGinnis, Review Publishing — the innovative projects that they’ve been doing to revitalize and transform their company from a newspaper to true media company.

You can find the PubWright Podcast series here and on iTunes here.

Updated!: Read more about Arbortext business unit and the leadership heading it up.

We just launched our new website.  It’s got a whole new redesign. It’s been a while coming, but we’re glad to have this done especially with the launch of the new Arbortext business unit last month. More information on that coming soon..

In the meantime, we’re excited to have the new site. We’re focused on Arbortext software more than ever and we’re continuing our dedication to the community and to growing the resources available to our customers so they get the right help at the right time in the right way to best serves their budgets and their needs.

We’ve got several new and interesting pages that are permanent new additions:

1. Single-Sourcing Solutions community efforts page:  I knew we gave information away like candy. I’m still amazed by how much we actually do. This page lists the things that we at Single-Sourcing Solutions do for the community because it is simply part of what we do every day. http://www.single-sourcing.com/company/subscribe.html
2. Resources from the entire Arbortext community: There are lots of resources out there from all kinds of sources. This page has a list of those resources. http://www.single-sourcing.com/company/community.html
3. What New with Arbortext page:  We’re keeping a page that talks about what’s new with the latest release of Arbortext products that’s a handy reference page for anyone considering upgrading to the current release  http://www.single-sourcing.com/products/new.html
4. Our meta-partnership network.  Single-Sourcing Solutions, Inc. believes strongly in creating healthy alliances with other partners so that we can leverage with each other in creating the best solution for our customers. http://www.single-sourcing.com/company/partners.html (Read more about how our Meta-Partnerships create the best solution for our customers.)

We’ve got all the standard pages — about us, products, solutions, etc – but these four are my favorites. They really demonstrate who we are, what we do, and why we do it.

As we post news and new pages, we’ll be making announcements to our Twitter stream: You should follow us on Twitter here.