Dilbert - gathering product requirements from 'the generic guy'.

See, I'm not sure who the guy in this Dilbert cartoon is supposed to be. Is he the project marketing manager, who hasn't a clue what he's doing? Is he one of the project stakeholders who is similarly clueless? I'm pretty convinced he's not the user, 'cause I've never heard a user refer to the software they use as "my software". There are often sufficient usability flaws for users to want to distance themselves from anyone who had anything to do with designining the product they have to grapple with every day to get their work done.

Still, this depiction is spot on in characterizing some of the chicken-and-egg type discussions that can take place at project inception. Ted, isn't much of a visionary though is he? ... guess that's why he's called 'the generic guy'. The question just remains, is that 'generic marketing','generic engineering', or 'generic executive'?

More blogs about Dilbert, requirements, users, stakeholders.

Convention, consistency, familiarity ... and the Gmail delete button.

Well, here's a case of when the guideline of 'make interfaces that are consistent with users' existing experiences' won the day. Now Gmail has a 'delete' button. And, I have to admit I was using the feature, even when it was buried under the dropdown. So I've been saved a few clicks, and my mental model of how to deal with a ton of data feels complete again. But it's a shame the adventure of email without a delete button had to end.

More blogs about customers, google, design, heuristics.

"Who says customers can't tell you how to innovate?"

A colleague from the field of interaction design recently commented "Who says customers can't tell you how to innovate?" and provided an illustration of a customers' design idea collected through some user research a number of years ago, and details of a recently shipped product that now (these few years later) uses the latest technologies to create the design that the customer envisioned.

My reaction is that if the research and design community is making such statements, that's rather arrogant, and probably going a bit far.

I think the statement that our community more frequently makes is that users are the experts at 'real life use'; but not necessarily the best people to design the next version. That customers can't tell us how to innovate, is a much more extreme view. Of course, this statement, made in its milder form, still does not mean that customers' design ideas are irrelevant, ... as was aptly shown by the example my colleague gave.

Designers can be expert at design if those designers have access to sufficient details about different types of user requirements, contexts of use, and task analyzes. If we couple that with appropriate training and design experience, then designers can make expert design decisions. Despite the 'research data - design' link that my colleague gave, most user researchers and interaction designers have been in situations where they have had to listen to customers explain what they think and believe they would like to see designed (in contrast to concerns and pain points). Often times the designer knows, either through previous usability tests, or through other design experience, that those design ideas would just not work, or would provide a worse experience for the user.

If we, as designers, haven't built up a good understanding of our users and their environments of work, ... the likelihood of customers creating better innovations than designers is presumably higher than it would otherwise be. Whether or not we have a good understanding of our users and their environment of work, it is surely the case that there are people in those work environments that we are designing products for, that may well have design suggestions should could be the next best thing. Of course this will happen, because these customers are living with the issues and frustrations (whether large or small) everyday. There are innovative thinkers in every field. And if our customers' design ideas are flawed, as user researchers we should listen, question, and analyze, so that we can uncover the frustrations and formalize the requirements that our customers are hoping their designs will address.

More blogs about customers, HCI, design, user research.

Girls' Middle School Inaugral Scholarship Breakfast - with Robin Jeffries.

I sometimes get to attend events that Sun Microsystems decides are worthy causes to sponsor - and to be seen at. Over the past few months I've been a guest at the Anita Borg Institute for Women and Technology Women of Vision awards where Sun's Radia Perlman won the Women of Vision 2005 award for Innovation, and the Women in Technology International San Jose conference where Sun's Barbara Bauer was inducted into the WITI Hall of Fame. Attending these formal dinner events, has been extremely interesting, and it's been wonderful to be able to learn about and celebrate the achievements of these wonderful women who work in technology as well as meet them in person.

This week the event I was invited to was the Inaugural Girls' Middle School of Mountain View Scholarship Breakfast. This event was especially appealing to attend, because a former Sun colleague, Robin Jeffries, was giving the keynote and receiving an award for her services to technology. Robin's keynote speech was really wonderful. There were two stories in particular in her talk that I cherished hearing about, and Robin has endorsed my re-telling them in my own words:

The first story is about the challenges of being a working mother. By the time Robin was just about to have her second child, she saw a job ad with 'her name written all over it.' She had recently identified her interest in the area of HCI, and was very eager to apply for the job. However, this was difficult timing. The hiring manager at the company she was applying to was extremely keen to have Robin work for them, and she's imagined over the subsequent years how he'd had to ask HR things like, "Is it ok for someone to go on maternity leave as soon as they join the company?" In the end she flew out for an interview taking her baby when he was just 6 weeks old. She arranged for baby-sitting by an agency and early in the morning the baby-sitter called to cancel. She was unable to arrange for a new sitter, since the agency were not yet open, and called the company she was interviewing at only to find her interview couldn't be delayed as she was making a presentation at 9:00am. So, she had no option but to take her baby with her. She set him down, and cross her fingers, .... ... ... and he was as quiet as anything throughout the presentation, cooing, and quietly gurgling in all the appropriate points of the presentation! Robin had thought that with all the complexities that she'd completely blown the interview, that didn't turn out to be the case at all, and they happily hired her.

And how can I relate? Well my story isn't as dramatic; but I did once do a telephone interview at home with my then 9 month old grabbing hold of my legs, to get my attention. As he got louder and louder in his demands, I had to resort to putting him in his crib, shutting the door, and doing the interview in the yard! I felt bad, but he calmed down afterwards with lots of love and cuddles, ... and I also landed the job!

The second story explains Robin's "ah-ha" moment that sparked her curiosity in HCI. Robin had been creating some software down in the basement of the company that she was working at (where programmers were always put) and was visiting her work colleagues up on the 3rd floor, where she was able to notice someone using her software. The software was HR related and the administrator using the software at that point in time was supposed to type in an employee ID but instead had started to type in a name. As soon as the error message started appearing she'd realized her mistake; but had to wait through the gradual appearance of the very clear and long explanatory error message that Robin had written to cope with this very situation. Back then the text appeared one character at a time on the screen, and it took a rather a long time before the administrator could begin working again. Robin noticed her sitting there tapping her fingers while the text arrived. This lead Robin to wonder how she could correct this - it was obvious that the error message needed to be shorter as the administrator had quickly figured out what was wrong. Robin thought about what the shortest form that error message could be. She thought through what an actual conversation would be like around this error, and that probably all one person would need to say to another for them to notice such an error would be "huh?". So she duly went back to her development environment and implemented this. The very next day, while she was sitting in the basement programming, she heard a commotion and someone knocking on the door and bursting in, in floods of tears proclaiming, "your computer just said "huh?" at me!". And this is when Robin realized that learning to develop systems that would appropriately support users in their tasks, was a much more complex and interesting subject to pursue, and hence this became her lifelong career.

Many thanks to Robin for letting me recount these delightful stories here - she told me these were her two favorite stories of the presentation too.

More blogs about Robin Jeffries, HCI, WITI, Anita Borg.

Jan 10th Sun - Oracle Announcement

Through working at Sun I was fortunate to attend the Sun Oracle Town Hall / Press Conference held at the Oracle Headquarters in Redwood Shores on Tuesday. It made for a fun (given Scott and Larry's humor) and happy (given the news and initiative announcements) afternoon. Scott and Larry in witty repose joked around to open the event. Scott began by asking if Oracle was going to buy Sun, "A simple yes or not will do". Larry replied to the tune of "You'll see it in the newspapers, ... Oracle's strong preference is to do everything hostilely." The joking and teasing continued through talking about the issue of making deals and doing business with each others rivals. Scott summarized the situation by saying "So, you aren't totally faithful, and we aren't totally faithful either, (pause and reflection) ...Maybe you less so than us." To which Larry responded, "We are talking about IT, yes?", and Scott replied "My wife is in the front row. I don't know where yours is."

The announcements made at the Press Conference are great news for Sun:

• Oracle is signing a second 10 year license to use Java
• Sun and Oracle announced that the middleware stacks from both companies would work well together.
• Sun will be packaging Oracle's Enterprise Edition Database Software with four-processor Sun Fire V490 and with servers that use the UltraSparc IV or IV+ processors. This means that using Sun Servers will Oracle database will be 25% cheaper than equivalent IBM or HP systems.

You can find out more by browsing through the Sun Oracle Kick-Off 2006 Press Kit on Sun.com

More blogs about Java, Sun, Oracle, News.

Google Pack Installation Experience.

Google released Google Pack this last week. There has been plenty of blogging about this, so my look at Google Pack will have a specific focus, on the install experience. I was particularly curious about this because Google are taking a number of products, not just their own and creating one install experience with one license agreement. Despite the critics, I think having arranged for this to happen, is itself, quite a feat. So I downloaded, and am equally impressed with the ease of the download experience - although I do note a couple of usability issues that it would have been nice of them to have dealt with before launching. I opted to download all the products, and the installer recognized I was already up-to-date with a couple of the products. The installation experience basically takes about 3 steps: license, install instructions, installation panel; but this can be lengthened if you decide you want to get more software as part of the pack. Wish all installations were this easy.

What follows is the procedure that I went through including the couple of usability issues I came up against. You can open any of the screenshots in a new tab or window to see them full-size:

Step 1: Screenshot of license page (web-page):
















Step 2: Simple 3-step install instructions (web page):

















Step 3: Install starts and a pop-up window displaying a 3-tab feedback panel is there to guide you through the process:
















Part way through the installation process you can see that it is clearly telling me which software has been newely installed, which software is already up-to-date, for which software I already have a premium version installed, and which need to be downloaded. The two progress indicators that are provided, deliver very clear feedback.















This screenshot shows the installed software:
















The preferences panel is suitably clear and simple:
















Here's my installation, pretty much completed:















Upon completion a new window is opened to tell me that Norton Anti-virus could not be installed. This panel provides a poor user experience. As long as I agree it plans to change my premium version for some lesser version, and provides a default to 'fix this'. I don't consider the down-grading of my software to be a fix! This is a confusing message to the user, and not a constructive or helpful user experience:
















Step 3: My installation is complete!















Moving to the 'Installed Software' tab I clicked on the link to the right to 'Get More Software',















and ...















... it was only then that I discovered that RealPlayer had not been included as default as part of the Google Pack. While I already had RealPlayer installed on my machine, this was annoying to me because I wanted it to be included in the Pack so that I could have it updated through the update mechanism. From this UI it's not clear whether there's a mechanism for adding a single product to the installed pack, and I didn't want to go through the software detection process again for all of the products I had just installed. I didn't see an option to select the desired software for myself earlier in the process. Perhaps Google don't want to give this option on the first installation. For me finding out at the end that Real Player not been included as part of my download, was a user experience issue. (Brad Hill - the unofficial google weblog - also comments on this issue)

My summary of the installation process for Google Pack is that aside from the two usability issues I experienced, the process is pretty slick. It takes just a couple of clicks through a couple of webpages, and a well-designed pop-up to detect and install software components. There were two Google releases this week announced at CES 2006, Google Pack and Google Video, while both are fantastic concepts, only this one could have been said to have been executed well. The final word goes to naming the missing components... why isn't StarOffice included in this bundle?

Tagging and Social Bookmarking, SDForum SIG

Attended an insightful evening 2 hours ago with founders from:

• digg.com - Kevin Rose, founder
• kaboodle.com - Manish Chandra
• del.icio.us.com - Joshua Schachter
• wink.com - Michael Tanne

at the SDForum Search SIG.

For those new to these sites, here's a one-line summary of each one:

• digg: Allows posting of stories and provides access to stories that fellow users think are great. Go digg something!

• kaboodle: Allows you to bring together your shopping considerations and mine those shopping considerations of others.
  

• del.icio.us: A site of favorites - you know the rest

• wink: uses algorithms that rank tags to determine search results
  

Google weren't presenting at this event, and while Yahoo! was represented by del.icio.us, it would have been nice to see search guru's from these two big mainstream companies in public conversation with these smaller firms who are finding new innovative methodologies to search using tagging and social bookmarking, rather than traditional page rank. But this evening was about showcasing those having something new and innovative in the arena of tagging and social bookmarking.

Can we trust tagging? - is the most basic question that should be asked, and the panel felt that the judgment is in the users. Given the appropriate tools, the users will get rid of spag (spamming tags) and the like. Joshua Schachter reminded us that no one taxonomy is correct, and currently tagging is the newer, and seems the better way to go to collate quality search results. The biggest challenge for search engines really is relevance - for example, is it possible to do a full text search that uses tagging to bring the things of relevance to the top? The real intent of searching though tagging is relevance and quality of hits. Take a look at Yahoo! Mindset for another interesting way to provide results of relevance.

The panelists agreed that regarding supporting appropriate behaviors and not forcing people into socially awkward corners was something that designers of search software using tagging and social bookmarking need to be very careful about. A quick vote was taken around the room for the numbers of people had the problem with LinkedIn, that they had a load of unanswered invites sitting in their 'inbox', because they felt these people were not close friends; but they socially felt bad about saying "no - I don't want to connect". ... That was most of us then. The panelists felt that the better direction was to avoid the "will you be my friend?" question; but to emphasize the connections between interests and content. This may mean that your best friends might not be your best connections because they may not share particular interests with you; but someone who is a stranger to you now, may be one of your best connections based upon interests and topics of focus.

What about setting standards for tagging? Take a look at TagCommons, for one early approach.

Charlene Li, from Forrester Research closed the question and answer session with 3 predictions for the coming year:

1. An ecosystem will develop, of being able to track what is good
2. There will be an inevitable couple of social disasters
3. There will be a different cast of characters in this space, not just the innovative crowd (present and featured at this event) but to include the traditional search space, and media.
   

At the end of the session 3 minute demos are encouraged from the audience, and as a sneak pre-view we saw:

• loomia.com, which will be launching in 3 days time. For now see:
  conversationsnetwork.org/recommendations and recommendations.loomia.com
• opinmind.com, that allows you to mine blogs for opinions. Note: you can put things like google vs yahoo into this - hmm, it's fun!
  

After attending this search SIG I'm going to be looking out for the next one. Just hope that they don't choose the second Tuesday of the Month every time, since that clashes with BayCHI.

More blogs about
searchsig.