Friday, June 22, 2007
Thursday, June 7, 2007
David Platt (http://www.amazon.com/Why-Software-Sucks-What-About/dp/0321466756 ) has a good and fortunately very popular talk here at Teched on "Why your software sucks" aimed at the very developers that write the software. IT is basic stuff like "your users are not you" but David Platt's examples and instance on "Just make it work" makes me reconsider some of what I have been thinking about for our Group Policy designs. Are we really making it just work?
So making GP just work is one thing. Hard to do because you can easily be overwhelmed by all the things you also make the software do and all the complexity you could expose. But the notion talks well to my corner stone thought that you have to build in knowledge about what people should do.
But how do you accomplish "Just make it work" and "you are always building a platform"? The latter is just as important as it states that the role of a design team is to make a UI that lets you, the person who paid us to work for you, able to recompose, reconfigure, and mash together a new use for our software that we did not anticipate. How do we anticipate your next move and at the same time let you redefine what your next moves are going to be in the future.
I can't say I have an answer yet. I think we have to separate the focus. You always have access to the data. But we make a purposeful UI based on one understanding of you. Maybe that is too old fashioned, too old school. Hmmm, still thinking.
Tuesday, June 5, 2007
Steve’s focus on economics and on rephrasing all security questions as economic questions reminds me a whole lot about usability discussions over the past decade or so. Security, Performance, and Usability share the characteristic of being basically not-interesting. They are assumed and only considered when they are absent. No one buys a product because it is secure, performs well, or is usable UNLESS that is not the standard in the category and then that is just an indication of an immature product category. The trio of Security, Performance, and Usability plays the role of the base-guitar of any product. It lays the foundation for a great user experience, but it does not make up a product.
In his 1994 book “Usability Engineering” Jakob Nielsen spends a long time on ROI, talking the language of the C-level folks and so on. Same path as Steve is walking with security now. And I believe the future for at least Security and Usability will also share many characteristics. Usability went from after the fact quality control to a design discipline and is making its inroads into product definition and strategic asset. In other words, Usability as a discipline has moved from cost center to primary value generator.
Some of the talks I have seen here at TechEd leads me to believe that Security could take a similar role. Security is not about fiddling with settings, but about making IT behave in a way that gives the business a competitive edge. If for instance the IT department can demolish the stupid wall between inside the firewall vs. outside the firewall and give all employees access to company resources regardless of location or computer, then a major hurdle will have been removed for when & where to work and will ultimately make the company more competitive and productive. If security can enable this scenario, then that is a strategic business initiative, not some after the fact patch update or firefighting.
All the WPF stuff including the VS support, Blend, Design, and latest Silverlight is just pure plain awesome.
Server 2008 is great. System Center Essentials is mega cool. heck, loads of good stuff.
And we also have stuff that just suck or products with really exiting technology wraped in a horrible UI. But I think our new products have decent to great UI and there are fewer and fewer of the "arrg, that sucks" products. All goodness.