Tuesday, October 31, 2006

Docusign.net heuristic evaluation

https://www.docusign.net is the subject of this heuristic evaluation.

This application uses the technical convention of an authenticated user on a website printing a document to a local service which has been previously installed on the local box to obtain a legal signature from a remote party or to send or respond to such a request. This “printed” document then appears in a registry in the Web interface on which the user will perform actions which the Web services provides. The document with its various suggestions for use, such as SIGN HERE, grants the ability to have legal digital signatures.

From the point of view of a user the convention is sending a letter, with the contents, envelope, and stamp of approval, proof of receipt and similar mail conventions.

While I found it stunningly easy to use, there arose what I refer to as “Omissions of the 3rd Kind” which means they are not intuitive and require more in depth experience with the system to understand, or some simple reason prevents the enduser from understanding the interface such as fear of use.

The first one was encountered after successfully following the steps to install the application locally, and printing the document I wanted to send it to someone to have it digitally signed. As I went through the crabwise steps appearing and clicking “Next”



Addressing the header in the Envelope.

After passing through all the steps, it became apparent that the application was going to allow me to send a document to be signed but not to open and digitally sign a document myself, and then forward it. I stepped through the entire process, taking screenshots of the process. I located the help files and played them and found that the interface did support signing and sending. But first the document needed to be prepared with the locations for signing embedded in the file.

So I stepped through the application again. I felt that instead of “New” in the interface it might have a option to Request Signature, and one for Send Signature as a scenario based design could offer. But as I used it the second time it seemed more intuitive.

I took it at face value, I trusted it. Signers IS Signers – it is anyone including me. So I addressed the “Envelope” to myself.



Adding myself as the addressee for sending and signing

Then I simply followed each step as it was encountered. Forwarded the document to myself through the application and voila, it was done.

The only real limitation I could immediately discover considering the purposes I wanted to use it, is if the new window is open, any documents you have not “printed” will not appear in the open window. This application detests the back button which is not a familiar scenario for some kinds of web uses.

Eureka! This application works from the heuristic point of view and it works well, with three simple caveats
1. View the basic instruction presentation
2. Trust the application to lead you, don’t overthink it
3. Read the interface.

For most users I believe they could get this to work the first time.

Empathy in Well-considered Human Centered Design

Reviewing three basic value propositions to well-considered human centered design by activating the humanistic standard ‘empathy’ in real ways presented by Leonard & Rayport, Dr. Friedman and Holtzblatt & Beyer (see details at bottom of post) I feel that as humans make products it only seems practical for design concinnity to include human-involved practices in the field, resulting in recursive feedback into existing and new systems.

Each author offers a slightly different view of what effective empathy means, but all see it as helping individual people in particular and the masses in general, by listening and asking questions through a variety of methods. All mentioned work-arounds which people employ unconsciously to combat commonplace non-functional technical environments.

Empathy is the ability to feel affection for other living beings, like in a mirror image, to understand other’s motivation, from the dictionary –
“Identification with and understanding of another's situation, feelings, and motives.”
Or in less romantic terms from Holtzblatt and Beyer – “We want our feet to be sore where their shoes pinch ”

Leonard and Rayport see empathy as an exciting way to “delight the customer ” through a process of market research with observers capturing data, analyzing the brainstorming to create prototypes. They also consider it as culture shift to ensure sustainable successful companies. The AOL section seemed misplaced – who cares if a company is technical or not – nobody wants their private contact information sold without consent.

Friedman uses a business based phrase to capture the attention of managers with “Value Centered Design ”. Friedman championed people strongly, with her powerful view of empathy, she framed the wording and examples to convince businesspeople to attempt these solutions. Her crowning materials were the moral views on informed consent particularly where web based user tracking was outlined, illustrated with “how a user might benefit or be harmed by it’s use. ”

Holtzblatt and Beyer use empathy because people just want get their work done on computers, not focus on the tool itself. They elucidated the processes of contextual inquiry. They want users as “strong participants in the design process.”

There are nothing but advantages to the methods outlined with only a few disclaimers:
1. These adaptations consume cycles (resources) and increase the “time to market”; a clear short term disadvantage: as engineers say:
“What do you want for your product? Good quality? Inexpensive? Quick to get to the market? Good, cheap, quick: pick any two.”

2. Change to refocus on the real needs of consumers is impossible for some well established producers (dinosaurs). As Donald A. Norman states in “Want Human-Centered Development? Reorganize the Company”
“It is difficult for a company to make the transition to a consumer-driven marketplace.”

These megalithic monstrosities drive themselves out of business -- with top down management they will not allow employees to have input, much less customers.


3. From experience how could anyone think for a New York second that the practice of tracking end users through cookies does not affect their work? ‘Sniffers’ and ‘click trackers’ grind robust browsers down to a snail’s speed – to read these articles online SpyBot, defragger, and “anti-virus” software had to run to restore the browser’s speed to normal. Less knowledgeable users have to pay for support.

4. Democracy as a human value; living in a failing democracy one wonders if the highly prized human values stated in these article will linger long enough to overcome greed.

empathy. (n.d.). The American Heritage® Dictionary of the English Language, Fourth Edition. Retrieved October 12, 2006, from Dictionary.com website: http://dictionary.reference.com/browse/empathy

Holtzblatt, K., & Beyer, H. (1993). Making customer-centered design work for teams.Communications of the ACM, 36(10), 92-103. Retrieved 1 Dec 2004 from http://doi.acm.org/10.1145/163430.164050.

Leonard, D., & Rayport, J. F. (1997). Spark innovation through emphatic design. Harvard Business Review, 75(6), 102-113.

Friedman, B. (2004). Value Sensitive Design. Encyclopedia of human-computer interaction. (pp. 769-774). Great Barrington, MA: Berkshire Publishing Group.

Wicked Problems in User Interface Design

The wicked problem examples from urban planning that Rittle & Webber (1973) present ring true for information systems design because when addressing and serving more and more people’s diverse needs naturally wicked problems arise, hopefully to be solved, in large part with monkey-like mischievous solutions.

Working with Apple Computers, Mountford (1990) expressed those wicked problems in computer information systems outlining on the first page:
“The world is rich with data, rapidly becoming more and more varied in media type…
The challenge is how to best present and represent such data within the interface, to transform it into useful information.”

Making this suggestion for resolving these problems:
Future interfaces need to incorporate new information types and to accommodate new types of users with additional customized real-world interface metaphors that make information easy to find and use.”

Suggesting merging the efforts of two groups not generally thought of working together, scientists and artists. Comparisons from design traditions, animation, theater, architecture, industrial design, and information display (using Tufte’s famed cartography example of Napoleon’s March to Moscow: The War of 1812”). Suggesting looking at real world tasks and environments to find new ideas which in the technology and music fields we call “mashups” today, Mountford writes:
“Some people believe that new ideas are almost always the result of collisions –juxtapositions or recombinations of ideas” (Koestler, 1964)

These are outlined based on Adam’s (1986) brainstorming steps, as new uses for the object, adapt the object to be like something else, modify, magnify, minimize, substitute, rearrange, reverse or transpose, combine the data into an ensemble. This advice seems virtually the same as Jasper Johns’ esthetic theorem, now considered basic advice to artists --
"Take an object. Do something to it. Do something else to it."

Role playing is suggested as a creative exercise to delve more deeply into the change the idea or object metaphor, and outlining four personality and functional roles, the Explorer, the Artist, the Judge, the Warrior, that could be worn like a costume on Halloween, but accurately providing a warning that if interacting all at once, perhaps the role of the judge should wait to sort the ideas presented instead of eliminating them.

In a way Johns is a good example of the polar differences between scientists and artists, with information science providing solutions to needs looking for something one doesn’t know or have and Johns ‘something’ becoming something else based on “things the mind already knows. ”

Mountford ends advising interface designers to live in the present and future, anticipating the challenges for designers in new ways to use computers having suggested prototyping models, and these several techniques that are in wide use today.

Life too far in the future isn’t always economically feasible, but combining with role playing should help software projects remain practical and satisfying for those of us who enjoy building successful information tools and toys, or just imagineering we are. I feel the analysis is especially accurate as far as customizable user interface design goes.

Jasper Johns sketchbook, page 42 from 1963–1964 http://www.thenation.com/doc/20001218/danto/2

Quote obtained Oct 31, 2006 from http://www.metmuseum.org/toah/hd/john/hd_john.htm

Rittle & Webber / Mountford on Wicked Problems

Rittle & Webber (1973) present the concept that due to the complex nature of human beings scientific answers fail because of the nature of the problems (in the context of Urban planning), such as hard to measure or soft things - societal concerns - which as stated:
“do not have a solution in terms of definitive and objective answer. ”

Mountford (1990) collects ideas from a variety of sources, and advises In effect doing the same thing for information system design - advising finding ways for artistic and scientific people to collaborate, approaching things from a variety of levels of human experience, stating:

“some of our new interface ideas will come from people who study thought, language, entertainment and communication, as well as from people who study hardware, algorithms, data structures. ”

However some wicked problems can only be resolved in granting end users rights and freedoms when it comes to information. Some problems are not served by direct end user participation because it won't help anything and it would not make sense for the many.

Take a WAP (wireless) phone application which has a GPS function (locator that knows where the end user is all the time), and a Search function, so if an end user wants to find a coffee shop within a short distance, they run a Search and the 5 closest coffee shops are returned.

What users may be interested in doing is setting their own preferences to Starbucks Only coffee houses, or only hotwired coffee houses which have computers available for rent. They may be interested in knowing how other people who went there rated the coffee, etc. They may also be planning to meet someone there for reasons other than coffee (but that is for an adult audience).

Google weighs query responses using a variety of ways, but one of them, perhaps the strongest, is by how many times someone clicks on it. This is just like voting - there is a good argument that personalization is the greatest contribution that the American post war baby boomers, with their large purchasing power have made and continue to make today. Companies, and other ensconced or older established organizations still experience enormous trouble facing these kinds of 1 to many, 1 to 1 individualization or personalization changes.

During today’s and yesterday’s job interviews I advised allowing end users to set their preferences on a Web site when they are at a full sized computer to help them find whatever they want on the smaller screen, even if some “father knows best” defaults must be present.

While there are things that the end user may wish to configure, the typical end user is not interested in exactly how the algorithm returned the Search sort to the database, and has no interest in writing it, or contributing to writing it, or even ever hearing the word "algorithm" -- users just want the results and fast.

One of the questions I was asked in a recent interview was a standard sort question for an algorithm, my response was true to that of a designer (or an end user) – “Find a database engineer, sorry that’s not how I think about or consider these problems.”

This remains a problem that inexperienced or less well educated technical or scientific approaches believe that individuals should know everything – most end user/customers can not function in this manner and should not be expected to think about how the machine runs to obtain the results they want, even if they have some kind of mental model.

Imagine if Google functioned only as well as the user’s understanding of search engines permitted – they wouldn’t find anything – they have to rely on Google’s code poets to figure out how to take Search from the functional status of a hammer and make it into a starship. Same goes for Live Search, in my humble option, but that’s a completely different subject.

The wicked quality of being human means that solving ill-defined and sticky problems should not be and can not be the domain of just one facet of thinking about a problem or a system, but as my recent interview demonstrates simplistic thinking about designers roles and customers needs still dominates the software industry.

One solution is everytime a designer uses the term BRAINSTORM and wants to use this technique they should define it so that all present understand that some wild and crazy ideas are likely to form, and just as likely they will go away later, leaving just the good stuff, it's all part of the process. If critics make their presence known immediately, the uber creative will find someplace friendlier to be.

Holtzblatt, K., & Beyer, H. (1993). Making customer-centered design work for teams.Communications of the ACM, 36(10), 92-103. Retrieved 1 Dec 2004 from http://doi.acm.org/10.1145/163430.164050.

Leonard, D., & Rayport, J. F. (1997). Spark innovation through emphatic design. Harvard Business Review, 75(6), 102-113.

Friedman, B. (2004). Value Sensitive Design. Encyclopedia of human-computer interaction. (pp. 769-774). Great Barrington, MA: Berkshire Publishing Group.