My colleague Anthony Bradley posted a thoughtful response to my post Web 2.0: Now with Fewer Features. First he wonders why I didn’t mention Gall’s Law. Well, If I were to cite a law it would be Sowa’s Law, which I sometimes refer to as Sowa’s Corollary to Gall’s Law:
Whenever a major organization develops a new system as an official standard for X, the primary result is the widespread adoption of some simpler system as a de facto standard for X.
Sowa’s law is closer in spirit to what I am trying to highlight — that the key to success is simplifying the complex.
Second, Anthony poses a far more difficult challenge: “I don’t see social applications as immune to the innovators dilemma.” In a nutshell, Clay Christensen predicts that an innovator’s product starts simple, but then the innovator adds functionality over time to satisfy the needs of the upper end of the market until the product overshoot the needs of the mainstream market. So why wouldn’t this happen to social software products?
Well, I’m going out on a limb here because I really do admire Christensen’s work. But I think a certain class of products, or at least a certain aspect of products, resists creeping featuritus — and even occasionally exhibits diminishing featuritis: User Interfaces. User interfaces tend to increase in complexity much more slowly than the technology behind the interface. For a paradigm example, look at the dashboard of an automobile (just the controls used to drive, not the stereo or climate control). One could argue that the dashboard has gotten simpler since the early days when you needed meters for oil pressure, voltage, water temperature, and other “engine monitoring” indicators. Even the tachometer (RPMs) is mostly there for show.
Another example is the QWERTY keyboard, which hasn’t gotten too much more complex — despite Microsoft’s attempts to add Window and Menu buttons, etc. And look at the mouse — Apple actually decreased the number of buttons from two to one.
What Clay Sharky highlighted in his talk but I neglected to mention in my post, is that the reason social interfaces may tend toward simplification (contrary to Christensen’s Law) is that the most important thing about some social interfaces is the speed and ease of personal interactions they enable. Any UI complexity that gets in the way of the immediacy (and intimacy) of personal interactions, tends to get whittled away over time.
Let me give a couple more examples of things that have gotten simpler instead of more complex:
- A twitter post is simpler than a blog post
- An IM/SMS message is simpler than an email
- An iPod’s interface is simpler than the MP3 players it displaced
- An iPhone’s interface (only one button) is simpler than a typical cell phone’s
- Google’s search interface is simpler than the search portals it displaced
- A WII controller is simpler than previous game controllers
I could go on. But I think I’ve established my point that it is an oversimplification to say that every product or every aspect of a product inevitably gets more complex. I do think there is something special about User Interfaces that tends to keep them simpler longer and that when they do get overly complex, they tend to refactor into a simpler replacement faster.
People’s need to connect with one another is so deep and so powerful that they are constantly seeking simpler ways to do it.
14 responses so far ↓
1 Chris Townsend // Oct 2, 2008 at 10:25 am
To me, this sounds like a mis-application of “Christensen’s Law.” Your writing seems to assume that “moving up-market” necessarily implies increased complexity and/or additional features. I’m not sure where this assumption comes from, but it is certainly a critical piece of leverage in the crux of your argument.
In fact, I think Clayton Christensen himself would be eager to point out that this assumption is not a clear inference that is possible from what he wrote in his book. In _The Innovator’s Dilemma_, the line of reasoning is that disruptive new products move up-market because they become *better* over time, which does not necessarily translate into increased complexity and/or additional features.
2 Bruce Robertson // Oct 2, 2008 at 11:52 am
I agree with Chris: better is not necessarily more complex.
3 Nick Gall // Oct 2, 2008 at 12:50 pm
I’m flabbergasted.
Christensen is quite clear on the relationship between increasing performance (the northeast pointing arrow and complexity. First, I just did a search of Innovator’s Dilemma for the word “simpler” ( http://books.google.com/books?ct=result&output=html&id=SIexi_qgq2gC&dq=dilemma+christensen+complexity&ots=AhtNkEC9Hj&pg=PP1&lpg=PP1&sig=ACfU3U2vf1jk4pak6O3_0g6Yyd3A_192WQ&q=simpler ). And page after page reveals statements such as this one: “Generally disruptive innovations were technologically straightfoward, consisting of off-the-shelf components put together in a product architecture that was often simpler than prior approaches.” (p.15) Read the search results yourselves: there is no question that Christensen believes that disruptive technologies beginning to satisfy mainstream market needs are simpler (which I equate with “less complex”) than the incumbent technologies, which are on target to go beyond mainstream needs.
Second, he is even clearer in Innovator’s Solution. Read pages 127-128 and look at the chart ( http://books.google.com/books?id=ZUsn9uIgkAUC&pg=PA167&lpg=PA167&dq=christensen+“attractive+profits”+complex&source=web&ots=M4MSTyg4vI&sig=4Giais66-ECWsHB3kMJWSqNaG0E&hl=en&sa=X&oi=book_result&resnum=3&ct=result#PPA127,M1 ). He clearly labels the line representing the top end of the market “Interdependent Architecture”. In other words, as technologies move to the upper end of market needs, their architectures become increasingly interdependent, ie complex.
Bruce and Chris, Please cite something to support your very unorthodox view that in Christensen’s model moving up-market “does not necessarily translate into increased complexity!”
4 Bruce Robertson // Oct 2, 2008 at 1:08 pm
I wasn’t quoting Christensen’s work — I was stating my belief. I’ll be clearer next time.
5 Chris Townsend // Oct 2, 2008 at 1:22 pm
Hi Nick,
Okay. First, let me say that I have *not* read the Innovator’s Solution, so my comments are restricted merely to that book. And before I posted my previous comment, it’s true that I did not go back through the Innovator’s Dilemma line-by-line. Perhaps I should have made this more clear.
Nonetheless, from what I can see, the evidence you’ve provided doesn’t really address my comment (although I don’t think you’re “wrong” either). My point from the beginning was intended to highlight that increased complexity *does not necessarily* manifest in the UI or user experience in particular. Instead, this complexity may appear in other ways, which may in fact contribute to their true value, even if the superficial look-and-feel of the product-in-question is, in fact, still quite simple (or even simpler).
One of the passages you quoted is a good illustration of this distinction, and the crux of my argument: “…put together in a product architecture that was often simpler than prior approaches.” In this quote, the operative word for this discussion is “architecture” — which certainly is not synonymous with, or even fully inclusive of, the “social interface” that you are talking about in your post. I think it’s important to compare like with like, and therefore Clayton Christensen’s work doesn’t seem to apply directly to the kind of arguments you’re developing in your post.
With that said, if you take away the Clayton Christensen straw man, I’m actually in complete agreement with your post, and I found your insight quite interesting. But I do object to the peceived attempt to discredit Dr. Christensen, whose work I respect very much.
6 Chris Townsend // Oct 2, 2008 at 1:24 pm
woops sorry, in the first sentence of my previous comment, there’s a typo…I meant to say “…are restricted merely to the other book.”
7 Nick Gall // Oct 2, 2008 at 2:00 pm
Chris, I think you and I are in agreement now. I didn’t intend to contradict Christensen, whose work I too respect greatly. Think of my observations on UI simplicity as more of a needed clarification of Christensen’s work.
There is no doubt that Christensen believes that moving up market entails increasing architectural interdependency (ie complexity) within a technology. But he is relatively silent on WHERE such increased complexity manifests itself in said technology.
I hope I have established at least the beginnings of an argument that as a technology moves up market, the UI of a technology is less likely to manifest this increased complexity vs other parts of the technology’s architecture. As I said in the original post: “User interfaces tend to increase in complexity much more slowly than the technology behind the interface.”
Agreed?
8 Frank Hood // Oct 2, 2008 at 11:54 pm
Unfortunately I couldn’t find a link on Eric Burke’s site (stuffthathappens.com), but I think his cartoon says it all.
http://swissmiss.typepad.com/photos/uncategorized/2008/03/10/imagesapple20google20and20you.png
9 Nick Gall // Oct 3, 2008 at 9:47 am
Frank, Thanks for the hilarious comic. I loved it so much I hunted down the link. Here it is: http://stuffthathappens.com/blog/2008/03/05/simplicity/ . I’m going to post it to my blog as well.
10 Chris Townsend // Oct 6, 2008 at 11:12 am
Nick,
Yeah, totally agreed, thanks for the clarification and for your diligence in hearing me out. Generally, I find it most rewarding to reach this kind of mutual agreement as an end-state…provided that we’ve sufficiently worked through the superficial semantic differences in order to establish a shared lexicon for constructive dialogue going forward.
Three cheers for reasoned disputation on interesting and important topics!
Cheers
Chris
11 Nick Gall // Oct 6, 2008 at 11:29 am
Thanks Chris! It was good for me too.
12 Less is More - The Comic // Oct 15, 2008 at 6:18 pm
[...] a comment by Frank Hood on my less is more post, here is the perfect comic to illustrate my point. Be sure to read the comments to the [...]
13 Twitter Trackbacks for Less is More — Especially in Social Interactions [gartner.com] on Topsy.com // Aug 29, 2009 at 9:09 pm
[...] Less is More — Especially in Social Interactions blogs.gartner.com/nick_gall/2008/10/02/less-is-more-especially-in-social-interactions – view page – cached [...]
14 sH22ALEXA // Jan 28, 2010 at 4:58 am
With the aim to prevent the mistakes in some history dissertation, this would be much better to order the interesting information associated with this post from the dissertation service in web. That is not very hard to have the good grades taking that great way.
Leave a Comment