By Richard Hamilton, RL Hamilton and Associates

image Humans are wired to put things in buckets.  We have an innate need to create categories and sort things into them. While the average human leans this way, technical communicators dive off the cliff. From the DITA Maturity Model to The Content Wrangler’s article categories, if it’s not tied down, it will get dropped in a bucket.

This is not a bad thing; buckets help us organize and retain information. If you are shopping for a television, having a set of criteria in your mind—screen size, plasma vs. LCD, screen resolution, price—lets you bucket the choices and narrow them down to a manageable number.

Given both the ubiquity and potential usefulness of categories, I was surprised when a web search didn’t turn up a set of categories for classifying web content delivery strategies. So, of course, I had to create a set.

This article proposes a set of buckets and draws a few conclusions about content delivery from the exercise. I’m just scratching the surface here, and would be glad to hear from others about the usefulness and appropriateness of this way of looking at web content delivery.

Web Content Delivery Categories

This categorization focuses on presentation of technical information, though it could be adapted to other uses. The buckets suggest a rough hierarchy, moving from No-ware to Active-ware, but as you will see, they do not follow a strict progression.


No web presence at all. Unless you work for a part of the government that even the NSA does not know about, you are probably not in this category. These days even buggy whip companies have web sites (I’m not kidding, check out And of course, the NSA has a website, too.


Content is shoveled onto the web in whatever form is convenient. This may mean that entire books end up on the web in a single PDF or Word file. While clearly not optimal for all content, Shovel-ware is not always a bad thing. For example, many companies post PDFs of their product manuals that are identical to what was shipped with the product. That way, a customer who has lost the manual can get a printable version on the web.

An example of a good use of Shovel-ware is IKEA, which posts nearly all of its assembly instructions on the web in PDF form.


Content is developed and displayed as books, usually in HTML or PDF, with an index and search capabilities. With good search and well-designed books, this can be a reasonable way to go, though it does not take full advantage of the power of the web.

Hewlett-Packard’s documentation web site is a good example of Book-ware, with some Shovel-ware for documents like product pamphlets and installation guides.


Content is designed, developed, and assembled for delivery on the web using a modular methodology.  This is the first bucket that targets content for web delivery and brings the unique considerations of displaying information on the web into play.

Technologies like the Darwin Information Typing Architecture (DITA), and the modular methodologies that most people use with DITA, are examples of Lego-ware. Interestingly, the closest I could find to a general set of content strategy buckets was the Dita Maturity Model mentioned above. That model is a nice way of categorizing Lego-ware, and to a lesser extent, Custom-ware.


Content is designed and developed like Lego-ware, but the display of content can be customized by users, either directly by allowing them to select what they want to see or indirectly by having the interface select what they see based on information about the specific product or service being used.

An example of Custom-ware is the Boeing Aircraft MyBoeingFleet service, which gives customers access to customized information for aircraft they own or lease. Another example is FLOSS Manuals, which was featured recently on The Content Wrangler.


Wikis, forums, on-line chats, webinars, and so forth. Anything that lets you interact with users on-line falls into this category. This category is orthogonal to the others; you can have interactive features in any environment except No-ware.

There is a continuum within Active-ware, based on the level of interactivity, with Wikis at one end, mailing lists and forums a bit further along, twitter next, and webinars/live chats at the other end. There are many good examples of Active-ware, the best known being Wikipedia. In the technical communication world, both DocBook and DITA have wikis, and the DocBook community hosts a freenode chat at irc://freenode/docbook.

So What?

One of the good things about bucketing is that it gives you a new perspective on something you may have been looking at for a while. Creating this set of buckets led me to a few interesting ideas that I might not have considered before.

New World vs. Old World

There is a clear divide between Old World uses of the Internet (Shovel-ware and Book-ware) and New World uses (Lego-ware, Custom-ware, and Active-ware).

The Old World contains clear evidence of carry-overs from previous technology. Shovel-ware shows this most directly, since it is a direct transfer.  Book-ware shows it a bit more indirectly, through structural elements like tables of contents, chapters, and so forth, but the influence is still discernible.

The New World, while often having recognizable elements of the past, at least attempts to forge a new paradigm. In the case of Lego-ware and Custom-ware, the paradigm is modular structure and a separation between content development and deliverable structure.

Re-considering Active-ware

Active-ware, while clearly New World, builds from a different set of paradigms, modeling collaborative structures like conferences or symposiums in the case of Wikis and face-to-face meetings in the case of chats and webinars. This explains why Active-ware is orthogonal to the other categories and suggests that successful approaches to Active-ware will draw from a different set of models than the other categories.

This also implies that some of the things we value most in technical communication, like good writing and complete and accurate solutions, may have less importance, if any, in Active-ware. To some degree, this is happening already in a chat room, no one thinks twice about bad writing unless the meaning is completely obscured but it is likely to expand as Active-ware matures.

To come back to the buckets themselves, I am putting them out as a straw man; I think they represent a reasonable way to look at various web content delivery strategies, but I don’t claim that they should be set in stone.  In fact, I invite comments, questions, and brickbats.

About the Author

Richard Hamilton is principal consultant with R.L. Hamilton & Associates, specializing in documentation management and the application of XML technology to documentation. He is the author of the forthcoming book, Managing Writers: A Real World Guide to Managing Technical Communication People, Projects, and Technology, which will be published by XML Press later this year.

Copyright © 2008 Richard Hamilton