I was looking around at Apple’s KB, and noticed that they’ve recently redone their kb numbering scheme from the old 6 digit system to a new two digit identifier + new article number. I was wondering what their thinking was behind that. They’re still in transition with a ton of redirects from article with the old naming convention pointing to the new article numbers. Other products haven’t been migrated yet to the new system.
For example, an article previously known as 306876 is now known as TS1278. It looks like they have the following delineation:
They’ve also gotten rid of exposing the antiquated keyword system like Microsoft has in some places, where Apple used to have keywords like ktech and kmosx that mean nothing to anyone that doesn’t know what the glossary of terms are like.
In an age of internet search, does it make a lot of sense to do this type of mass renaming and categorization? Surely there are competing articles that might be troubleshooting and how to. Why worry about categorization if search is driving the majority of your traffic? Maybe their content isn't complex enough to have this type of struggle. With our content at Microsoft, we often have content that can mix some how to information along with information on how to fix error messages, etc. Anyway - back to the categorization problem…
It's unlikely that customers sit around and browse the Apple KB, just like it's unlikely that customers sit around and browse our KB. The knowledge base is a place you go because you have a question that needs to be answered, not to get random bits of knowledge as you browse the web.
If the above premise is true, then how would renaming the article numbers to the new nomenclature doing them any good? Perhaps they're trying to get a better handle on the content they have, or maybe they have reporting issues that makes it easier to just rename these articles. The amount of stuff you can store in internal metadata still just makes it an odd choice for them.
Do we need to have our content re-branded? For the broader company, we have a series of "fw links" - these are the static small links that point to larger, dynamic addresses. For example, http://go.microsoft.com/fwlink/?LinkID=90854 points to the huge URL of http://www.microsoft.com/downloads/details.aspx?familyid=173E6E9B-4D3E-4FD4-A2CF-73684FA46B60&displaylang=en. As the large URL moves around our internet pages, all we have to do is update the fowarding link that the LinkID=90854 passes and we don't have to communicate to customers that a site has changed. It's worked out pretty well for our product groups and some of our content, but I think we could leverage it even more. I suppose the downside is that the URLs aren't entirely friendly since no one knows from the address what LinkID=blah could even point to.
Have you gone through a content re-numbering/re-naming process? What was the driving factor that made you do it? How did it work? Was it successful? Let me know by leaving a comment in this post.
Hunter DonaldProgram Manager - Knowledge Management StrategyCommercial Technical Support