Dear SharePoint Community,
Chris Riley of Cloudshare and I both love SharePoint ECM. I think. But Chris and I can’t agree on the relative role of taxonomy and folksonomy in SharePoint Managed Metadata Service.
Please take a look and weigh in on the debate. Hint -- I like folksonomy. And he’s wrong! Please share your opinions so long as it matches mine. Thanks!
From: Chris McNulty To: 'Chris Riley'Subject: RE: Taxonomy vs. Folksonomy
I just don’t understand it. Term variation is fine – you have the tools to promote them into your precious taxonomy – but good folksonomy is a leading indicator for good taxonomy. Not the other way around.
How can we resolve this? Moderated debate with Christian Buckley? Dance off? Interview at Weehawken? Can we ever agree on anything? If not, well, go tag yourself!
Chris McNultySharePoint Strategic Product Manager | Quest Software Simplicity at Work | firstname.lastname@example.org
From: Chris Riley [mailto:email@example.com]To: Chris McNultySubject: RE: Taxonomy vs. Folksonomy
Not building and planning a taxonomy upfront is equivalent to setting a day for technology nuclear fallout. The deployment will fail. Users will create every variation of every term possible. They will not even understand their own classification, and basically create a modern version of a very bad thing we all know. Called shared drives. Yes your system will effectively work to make the proliferation and messiness of these modern shared drives even more rapid.
The reality is that planning a head, and building the system, forces users to do the right thing, and create an understanding of the content they contribute. It will allow them to be more effective, and efficient. Not only that the process of building the taxonomy will unite the organization at large with a greater understanding of the content they own.
Want flexibility fine. As I said synonyms are a great way to keep the right amount of flexibility.
Sure it’s time required upfront that can sometimes seem like a never needing battle, but if you do it, you won’t fail.
Chris Riley | Product Manager & EvangelistMobile 925.640.4361
CloudShare, Inc.CloudShare.com/blog | Twitter: @HoardingInfo | LinkedIn
From:Chris McNulty [mailto:Chris.McNulty@quest.com] Sent: Friday, November 11, 2011 6:06 AMTo: Chris RileySubject: RE: Taxonomy vs. Folksonomy
Wow. I was so stunned I needed a month to compose myself (and change jobs!)
I just don’t agree. Strict taxonomy, especially at the outset, is too alienating and limiting for most users. You made a reference to the taxonomy being “done”. Taxonomy is never “done”. A proper information architect is always on the lookout for new terms and synonyms, new sets. We learn this through watching usage patterns, search terms. And, of course, user generated keywords!
It’s very nice to think that we can build a high-in-the-sky taxonomy before users arrive with their messy documents and personal information maps. It sounds really great, up there, 14,000 feet up in the Rockies. But down here, at sea level, in the big city, we need to get things done. We can’t wait for your ivory tower information theory to complete a needs analysis.
But I think you’re absolutely right about choosing the friendly departments, like engineering first. If you pick the right users their folksonomy will BE the taxonomy.
From:Chris Riley [mailto:firstname.lastname@example.org]To: Chris McNultySubject: RE: Taxonomy vs. Folksonomy
First off. The type of content that needs a taxonomy is not rich media. I agree. But documents, a must. Second, you are not waiting for someone to add terms. If they are not all added up front you are doing something wrong. I suspect what your response on this will be. Taxonomy’s are set upfront before a single document is loaded, if the company has a retention schedule, this is an initial guide. They should be done per division, not organization wide. In my experience they take about 6-12 hours per division. Starting with the cleaner departments like engineering is always a good start. Once the taxonomy is done it should have no “other” buckets, use the proper synonym’s and be no deeper them 4 terms per parent term. If a taxonomy needs to be updated this should be a gentle process.
I agree doing taxonomy retroactively and waiting for terms will never work, but this is not the proper way. Having folksonomy with taxonomy can overcome the flexibility limitations. But to truly be organized and compliant, a strict taxonomy is mandatory.
And no I’m originally from CO J
Chris Riley, ECMp, IOAp, CAPTUREp
Product Management & Evangelism, CloudShare
Twitter | LinkedIn
From:Chris McNulty [mailto:email@example.com] Sent: Sunday, October 02, 2011 5:48 AMTo: Chris RileySubject: RE: Taxonomy vs. Folksonomy
I think you’re flat wrong here, sorry.
Synonyms are great if other people’s actions predict another users thought patterns. It never works out that way. If I want to tag the photos from my SPTechCon session as “ChrisPix”, or maybe “Sunset” because that’s what time it was and that’s how I remember it, why should I:
a. Wait for an information architect to add it to a term store
b. Mess up a highly organized term store with undisciplined, highly personalized tags?
And, anyway, aren’t you from the Bay Area? California, personal empowerment, all that? Really, you’re going to argue against individualized folksonomy? Is this the *real* RileyBeebs? Or has HoardingInfo taken a new approach?
Hell yeah. Taxonomy kicks folksonomy ass.
I think we should record this debate.
I love giving users power and control, and feel there is an opportunity for that even in a controlled taxonomy. But I also know that when it comes to ediscovery, compliance, trusting users to use the right terms, in the right way, without a stick guide will never fly in court. I think there is a way to have both with a taxonomy. And often support taxonomy with folksonomy, but never support folksonomy alone. Your right could just be my legacy ECM talking, but that legacy ECM made legacy laws like HIPPA and SOAX. And these laws mandate good taxonomy. So it’s better to plan for the worse. And training your users to accept a taxonomy, with great synonym’s that satisfy their needs seems obvious.
Chris@CloudShare.com | 925-640-4361
From:Chris McNulty [mailto:firstname.lastname@example.org] To: email@example.comSubject: Taxonomy vs. Folksonomy
Really looking forward to seeing you at SharePoint Conference in Anaheim, and catching up with your new role at CloudShare.
Although, I must confess, I’m somewhat baffled at your comments the last time we were in the same room. Did you really claim that the taxonomy services in SharePoint 2010 Managed Metadata Services were superior to folksonomy and keywords? Really?
I mean, how Enterprise 1.0 of you. Seriously. If the past few years have taught us anything, it’s this – trust your users. Keywords are easy to set up, easy to maintain, easy to use. If you’re an end user, just type away. Want to define tags to show up in a search? Done. Want to create a topic that people can use and follow? Done.
What about central control? What about predefined hierarchies? Who needs them??? Why wait for the high priests and priestesses of IT and metadata to get around to it?
And regardless of how central planners *think* the world should be organized, users will lead us the way to how the information actually, really works. Trust users, don’t fear them.
Tell me where I’m wrong.
Christopher F. McNulty