Kevin Kelly — Help Wanted: who rejects technology?
Dave Winer: “I make shitty software and I write shitty specs, but for all that shittyness, they’re amazingly popular and somewhat useful.”
If you are on a technical writing project, are there any systems that let you centralize and organize preferred terms, telephone numbers and such so that you can manage these things?
Solving Mobile Challenges with Psychology-driven IA: I was going to say: “bullshit!” (that title!), but then came this and I was hooked: “Instead of telling the deleted user, ‘You were deleted by Sharon,’ its message reads: ‘Don’t bother with Sharon.’ (These messages are, of course, adapted to the local culture.)”
ongoing · Google Adsense (if the page doesn’t show up, change the encoding to Cyrillic (Windows). To change the encoding, go to View > Encoding in IE6, Win)
“RFC 1, “Host Software”, issued thirty years ago on April 7, 1969 outlined some thoughts and initial experiments. It was a modest and entirely forgettable memo, but it has significance because it was part of a broad initiative whose impact is still with us today.” (RFC stands for request for comments. I was wondering about that.)
– Pervasive Environments and Ubiquitous Computing
– Shared Interaction Spaces
– Playing and Learning
– Tangibles & Virtual Environments
– Collaborative Learning
– Theory & Conceptual Frameworks
– Technology Mediated Communication
– Interactive Art
Stack up on printer paper.
I am shopping for a digital videocamera. Can I efficiently edit digital video on my laptop of 350 Megs RAM and 500 Mhz? Should I buy a videocamera with less pixels instead of more? I don’t really want to buy a new computer.
Ben Hammersley.com – Sporting Gent’s guide to the Semantic Web Kinda cool presentation.
Ben Hammersley.com: “The answer to the problem of false metadata is, in fact, more metadata.”
The new Google ads on Easy Topic Maps are showing up now. Have a look and see for yourself what you think about their relevancy.
So I got in at work (9.00), filled my 1.5 liter bottle of water, it’s 10.30 and it’s empty. I’ve found that filling it up completely works better: if I fill it only halfway I drink half as much (I don’t take more refill trips). I am also getting off the train one station earlier and then walking to work (20 mins). Same in the evenings.
Me and the Credit Card Company Call Center guy:
Me: I have this 29$ charge on my bill for not paying on time. But I never received the previous bill.
CCCCC guy: Just a moment please.
12 minutes of happy music. I hang in there.
CCCCC guy: This is what happened: long explanation of how it’s really my fault.
Me: OK. I do think $29 is a bit steep. If you could remove that charge that would be great. If not, I’d like to cancel this card now.
CCCCC guy: Well, we can do that as a one time courtesy. Long explanation of how it really was my fault.
Jon’s Radio: “What open source projects often lack is not the engine, but the steering wheel.” So true.
Even though I have a computer with the basic system requirements, Six Degrees slows my system down to an unusable crawl after I install it. And there’s no way to turn it off: if I tell it to be inactive it still eats up memory. This is the second time this happens – I don’t think it’s a coincidence. And I’m pissed off: I want this software. I need it. But it wastes my time every time I try to use it (and that for a program that calls itself “timefreeing technology”). Arg.
I wait for the day when bloggers stop linking to articles about blogging on the NYTimes, and the NYTimes starts linking to blog posts.
Information architects have long felt left out by the software development
tool makers. Companies like Rational
(recently purchased by IBM) base their tools on a process that is optimized
for complex software development but doesn’t really address the needs of information
architects. We end up using a variety of tools like Word, Excell or Visio that
are just generic enough to be frustrating. Keeping large and complex deliverables
up to date is a challenge: make a change in the content inventory and you also
need to update the sitemap. And why can’t the sitemap be automatically generated
from the content inventory? And exported as, say, a requirements document?
The search for the all encompassing tool is nicely illustrated by this
case study. It describes how an information architect team at Razorfish
used GoLive to
manage their deliverables. From the article: "[…] documents tend to end
up separate and independent from one another. They are often owned by different
people, reside in different locations, and are created in different formats.
It is not uncommon that, by the end of a project, updating something as simple
as a navigation label requires updating half a dozen documents or more".
This dream goes beyond providing a central tool to manage information architecture
deliverables. IBM has been working on a modelling methodology and toolset called
will let you model everything (from audience analysis to use cases to interface
widgets) in a UML-like visual language. When you are done modelling, click a
button and a prototype of the software gets automatically generated – or that’s
the dream, anyway. It is the ultimate abstraction, bringing with it all the
abstractions bring to software development. Still, the dream is hard to
resist, so we keep looking for tools focussed on information architects’ needs.
Enter Ubiquity RP. I interviewed
Victor Hsu at Axure software, who are developing a tool that provides a central
modelling and documentation environment for information architects called Ubiquity
RP. Ubiquity lets you document requirements from user needs to wireframes, and
automatically exports documentation and prototypes.
Ubiquity is based around a 5 step process.
In the last step you can generate a specification document and a working prototype.
Q: Who are you targeting with this product?
Victor Hsu: Our target users are people who are responsible for the
functional design of web applications. Their responsibiities includes understanding
user and stakeholder needs, defining how a system will be used, designing the
visual interfaces, and communicating the design to clients, users, stakeholders,
and development. To put it in terms of work product, we are targeting users
who create prototypes, functional specifications, or proposals for web applications.
Q: Small companies?
We are finding that both small and large companies are interested in Ubiquity
RP. The types of companies we are targeting include companies with internal
IT organizations developing intranets and extranets, web design and development
consulting firms, and web application product companies.
Q: Information architects?
Many of our target users have Information Architect titles, but others are
Business Analaysts, Project Managers, Product Managers, Developers, and Consultants.
Q: Why did you develop it?
From a very grand perspective… We are after the Holy Grail of software development,
the ability to create complete applications through code reuse and code automation
without programming. We believe that for this to be achieved, it has to start
with strong information architecture.
From a more practical perspective… We began developing the solution to help
resolve many of the inefficiencies and frustrations that development faces when
functional requirements are poorly defined or communicated. We quickly realized
that there was an equivalently urgent need to better elicit and communicate
requirements to users and stakeholders. Since we were unable to find a good
solution to these problems, we saw an opportunity to create a tool that would
make it possible to elicit and communicate quality functional requirements more
efficiently and to make that process repeatable.
Q: What is the development roadmap for the product like?
Our first product is Ubiquity RP, “real-time prototypes, specifications, and
proposals.” The next release of the Standard Edition of Ubiquity RP is scheduled
for Q2 2003. It will include enhancements to the diagramming and drawing functionality.
The tools available in the Standard Edition will be Site Flow (currently Sitemap),
Wireframe (currently Page Design), and Generate Specification and Prototype.
This is intended for users who are primarily interested in prototyping.
The Professional Edition will be released shortly after in Q3 2003. This edition
is designed for users who will be creating specifications or proposals in addition
to prototypes. It will include all the tools in the Standard Edition plus Vision,
Needs, potentially Use Cases, and enhanced features in Site Flow and Wireframe.
We intend to do one release of each edition per year.
In addition, an Enterprise Edition is expected to be introduced in 2005 which
will provide a more collaborative environment for teams to work together on
a centralized design. Future product lines that we are considering include a
product for managing the requirements that are produced from Ubiquity RP, and
a product that leverages the designs from Ubiquity RP to lay the groundwork
Q: What is your background? You mention information architects on your site
– are you involved in that community?
My background has been primarily along the lines of project management and
software development. In each role, I have assumed some information architect
responsibilites and have worked closely with product managers and designers
to produce specifications and prototypes. I am not currently as active in the
IA community as I would like to be.
My involvement with the IA community thus far includes joining and beginning
to contribute to several mailing lists for IA professionals. I have also been
in contact with several thought leaders in the community who have been very
supportive of our efforts. We are investigating opportunities for Axure to become
more involved with the IA community. I was particulary interested in the concept
of building a knowledge base of information architecture resources maintained
by the community that is suggested on the Asilomar Institute for Information
Architecture website (www.aifia.org).
Q: I tried the product, and I like the general idea. But I am worried I
won’t be able to also keep using my current tools. What are the plans for making
the product work nice within my toolset? Or is it meant to be the only product
We chose to develop our own suite of tools rather than try to integrate several
existing tools for many reasons. Primarily, the tools on the market were not
treating the designs (site flow diagrams and wireframes) as web application
design data. For example, a rectangle in a Visio diagram labeled “Page 1” is
a rectangle with a label. In Ubiquity, a page element in a diagram is a page
object with specific data associated with it including a wireframe representation
and user-defined page attributes.
In addition, it was particulary important to us to tightly integrate needs,
siteflow, wireframes, etc. such that we could implement traceability in the
future. I’m excited to see the day when a user can see an automatically generated
diagram tracing needs to use cases to site flow to ui elements and eventually
to code! I don’t think use of our tools will replace the use of the Visio’s
and Dreamweaver’s of the world for designing web applications. Each has it’s
strengths and weaknesses. But we are going to do our best to consolidate the
strengths into our tools.
Down the road, we see potential integrations with diagramming tools like Visio,
HTML tools like Dreamweaver, analyst tools like Rose, requirements management
tools like Requisite Pro and DOORS, and development tools like Visual Studio
and WebSphere Studio Application Developer.
Q: I always like tools like this to be customizable as much as possible.
Do you plan some kind of scripting environment, where I can script importing
or exporting and such? How about the stencils (‘libraries’)? Can I create my
own? Can I share them? Are there plans for functionality like that? Any other
plans for letting users customize the app to their needs?
The document format for Ubiquity RP is entirely XML based so a user with an
understanding of the schema would be able to script pretty much anything they
wanted to manipulate the document. We are planning some changes to the format
so we are not recommending that any users do this at this time, but it will
certainly be an option in the future.We would like to develop the framework
such that custom tools built by 3rd parties can be added to the suite. A person
could build their own Needs Manager and plug it into Ubiquity RP for example.
Customizable stencils are slated for the next release. You will be able to
create your own. There are also plans to allow users to share stencils as well
as elements in what we are calling the repository, which contains reusable diagram
or wireframe elements.
For example, if you have designed a stock ticker that you would like to reuse
across multiple pages or designs you will be able to do so.
Along the lines of a central knowledge base of IA resources, I would love to
see a central repository of Ubiquity RP stencils and design elements that our
users could share.
Q: Are there examples of real life projects (not just showcases) being done
with Ubiquity? If so, what are the lessons learnt that might guide future development?
Several companies have already used or are using Ubiquity RP for real life
projects. We have learned that the application can be used for several different
purposes and that users value different aspects of the product.
For example, one IT organization primarily uses the prototyping aspects to
quickly validate the requirements for the numerous development requests that
they receive from other departments. On the other hand, a consulting company
that is using Ubiquity RP values the documentation produced for the purpose
of presenting high level proposals to potential clients.
This has impacted our plan for packaging and developing the product. The Standard
Edition will be focused on prototyping. A common request for enhancing the prototyping
capabilities has been to enhance the diagramming functionality for both the
Sitemap and Workflow tools which we are addressing in the next release. The
Professional Edition will be focused on creating proposals and specifications
in addition to the prototypes. This places a greater emphasis on the types of
documents and the formatting of the documents that Ubiquity RP can produce.
Q: Why ‘Ubiquity’?
Ubiquity is a play on the word "ubiquitous", meaning existing everywhere.
We want our solutions to be accessible, usable, and effective for anyone that
is responsible for designing a web application whether they are amateurs or
experts, technical or non-technical, in small or in large organizations.
I know you didn’t ask, but RP stands for Real-time Prototyping. We very much
want to accelerate the process of prototyping to make multiple iterations of
prototyping prior to development more feasible for time and budget constrained
Q: On the website it says "If you are interested in guiding future
development, please email us at email@example.com." – how does that work?
When people contact us expressing interest in guiding development, it opens
up a dialogue to begin discussing their needs, our plans, and how we can make
Some inquirers have asked for specific features to be added to the next release
and others have a greater interest in discussing our long-term goals and mission.
Having that on our website has given us the opportunity to speak with people
who see the potential in our solution and are excited to see it grow. We’ve
been very happy with the response we’ve received.
Q: So, did you develop the axure site using ubiquity?
Yes we did! The tricky part was trying to design Ubiquity with Ubiquity :)
If you want to try out Ubiquity, you can download
a 30 day trial version for free.
A Russian IA glossary: in4arch
From email conversations with various people, it seems that getting mod_rewrite to work on Apache on a Windows box is near impossible if you are not endowed with special sysadmin powers. Which means that if you want to develop rewrite code locally, you need a linux box or something. Can someone please tell me this is not true? I refuse to believe. I am not such a techie that I want to get linux box in my little NYC appartment – let alone learn to administer it. I really just want to do rewrite code, and I don’t particularly want to do it on my live server.
There is a common information architecture myth that says: �Within a labeling system, labels should be grammatically consistent�. If one label uses a verb or is phrased as a question, all the labels should do the same.
Like in a game of whisper telephone, subtle guidelines easily get misunderstood. In this review of the first bearbook (1), the author writes: �Labels should be grammatically consistent; if some labels are nouns and others are verbs, readers will be confused.� Next we get people like Marcia Yudkin writing (PDF): �The words and phrases [for labels] should be grammatically parallel�, and gives a rather bad example: �What We Do, Who We Are, Our Portfolio, Contact Us� should, according to her, become: �What We Do, How We Do It, Who We Are, How to Reach Us�.
Recently, I was on a project where we had decided, during the analysis phase, to use labels with verbs. The entire IA was based on activities, so it seemed to make sense.
We soon ran into trouble though: using verbs throughout became harder and harder (sometimes it just didn�t make sense), and during usability tests, people had a hard time finding the right links. Making matters worse, convincing people that using verbs throughout in the labels wasn�t the right approach had become politically difficult.
I asked the SIGIA-L list for help, and soon learned I wasn�t the first person to struggle with this. Cynthia Ramlo wrote: �I worked on an application where this approach, during the analysis phase, had been agreed upon. It was very difficult to implement and it was not clear what verb would be associated with common actions (view, see, look, peruse).�
Donna had experienced this too: �I too have in the past tried the consistent grammatical structure – starting with verbs for tasks/actions. However, I always get to the point where it looks silly or the verbs start repeating, and I drop back to a combination of nouns & verbs.�
Peter Merholz also chipped in: �I’ve dealt with this on a number of projects in the past. My original inclination was that all labels (particular if there kin labels, such as items available in global navigation) should share grammatical structure — if one is an action label, all should be action labels.
For me, the lesson was, verb, noun, action, thing, whatever, it kind of doesn’t matter. What matters more is that you utilize words that are likely rattling around in your visitors’ heads, words that resonate with them.� Amen.
Livia Labate expanded on that with some thoughts on labeling for international users: �One problem I can see depends on whether your target audience is a native speaker of English or if English is a second/third language. I just finished a course on the Semantics of the English language (so I could create better labels!) and my conclusion is (as Cynthia mentioned) that nouns (and adverbs) are more easily learned than verbs [�]�.
Going back to our earlier example, is �How we do it� truly a better label than �Portfolio�? Of course not! Keeping an eye on grammatical consistency can help when developing labels. But people look for certain key words, and if they don�t see them in your label, you are in trouble. Using labels that work for users is much more important than grammatical consistency in the labeling system.
(1) I couldn�t find my copy of the first bearbook, so I�m not sure what is said about labels in there. Anyone?
I recently came along Travel UCD, a usability consultancy specialized in a very specific market (travel and hospitality web design). I was intrigued: it makes sense to specialize in a market like this because you can reuse a lot of specific knowledge, but I didn’t know of other usability consultants specializing like this. I wondered how it was working out for Alex, and whether it could be a model for others. Alex has an academic background is in user interface design & development, a degree in Applied Computing and has worked within the travel, IT or web industry since 1995.
Q: Hi Alex, thanks for doing this interview. When did you start the company,
and how did you come up with the idea of limiting your offerings to such a specific
Alex: I started in April 2002. At that time I had just been employed
as head of web design & development for a leading consortium of hotel chains
in Europe. With my expertise in creating hotel reservation websites for both
B2C & B2B marketplaces, and prior experience as a Managing Director of a
UK based tour operator, it seemed natural to continue within this sector with
Travel UCD. A large amount of the design work that is undertaken on travel websites
involves having an understanding of travel & reservation technology – not
the normal realm of a web designer or usability consultant. It seemed like there
was a market for this.
Q: You are offering some reports with best practices for download; some
free, others for payment, similar to what the NNGroup does. What is the thinking
behind offering these? Do they sell well? Do they lead to potential clients
The NNGroup model was an inspiration, but our reports address specific design
questions common to many online travel websites – much more niche than the NNGroup.
A useful part of the reports is that we include all tasks and observations from
our usability testing. Designers working on similar websites can take the same
tasks and repeat the usability testing on their sites and compare results.
We started the reports project for three main reasons. First, it demonstrates
to clients and potential clients the kind of knowledge that a specialist usability
consultant can bring to a project. Usability is still an alien term for many
project and product managers and I wanted to have something tangible that I
could demonstrate. Much of my work (prototypes, designs etc) is for future projects
– or projects that are not public – so I can’t show this work to potential clients
for contractual reasons.
Second, and this is important, it defines our intellectual property in a public
location. As I work in one specific sector I may be working for clients who
compete with each other in some way. By publishing the reports I can clearly
demonstrate the key intellectual property that Travel UCD know prior to a project
Third, it provides some revenue from sales. The free report has had about 6000
downloads since July 2002. The 2 recent reports, available for US$150 each,
haven’t quite sold that many, but enough to cover the research and promotion
costs. Most of the household names in online travel from all corners of the
globe have purchased their copies as well as a few travel technology providers.
Q: How do your organize your research? Do you wait for client engagements
to pay for testing? Or do you organize tests outside of client work? How do
you store your data? (video, …?) Do you revisit earlier tests for more insights
The research for the reports is conducted proactively. I choose topics that
I know are of keen interest to travel website designers & product managers.
These questions may have come up at industry conferences or from previously
unanswered design questions from earlier projects.
I begin by looking at the main design approaches to a particular issue. I then
evaluate statistically how many travel websites apply the design in the alternative
ways. Following that I then take a sample of 4 or 5 websites who all use different
approaches and then conduct usability testing with 12 users on these 4 or 5
sites. Although not statistically significant, I believe that by tying my observations
from the testing with the statistical analysis and my own experience, I can
devise guidelines that define best industry practice.
At this stage I have not revisited earlier tests as there are so many areas
left to study!
Q: So what do you think is the best site in this category you have seen,
from a usability point of view?
I tend to look at travel websites on a product by product basis rather than
as a whole site. For example I may say that a certain site has a good hotel
reservation booking process whereas another site may have a well designed flight
section. At this time I don’t have an all round firm favourite. In my opinion
even the major online travel websites that everyone knows and uses demonstrate
usability flaws that impact on the user experience.
This leaves some space in the market for a well designed website to acquire
greater market share using usability as the core of their strategy much like
Google has for search engines. In the travel industry this would probably be
an existing company not a new entrant as the success of travel websites is not
solely defined by design; underlying technology and having a large range of
well priced products and inventory are equally important.
Q: What would your advice be for someone wanting to set up a usability consultancy?
Ensure you have a potential client list before you set up as finding them afterwards
is much harder.
Thanks Alex! I am looking forward to seeing how your company flourishes. I
also wonder wether this could be a model for other people wanting to start a
usability consultancy. I can imagine a usability consultant specialized in,
say, human resource intranets. Or gambling sites. Lots of possibilities there.