Reading Reflection – Week 14

With all we have learned this semester about Web Information Architecture, without the buy-in of the business, nothing can be gained.  As information architecture has not yet risen in the minds of most business leaders as something they need to strive for, Information Architects need to not only implement solutions but also “sell” them.  Understanding what information architecture is and how it can benefit the customer is a key step in the process.

Our text book describes two kinds of people that may need to be “sold” on how information architecture can make a huge impact on their business.  The first is the “buy the numbers” people that need the hard facts, like ROI, investment costs, and other metrics that have a “direct and quantifiable impact.”  The second type is the “gut reactionaries” that trust their instincts and experience.  They need more scenarios and case studies to feel a real problem that existed and see how proper information architecture aligned with their business practices to fixed most (if not all) their issues.  Some other methods of educating potential clients are by using boot camps, expert site evaluations, and competitive and comparative analysis.

As businesses understand the need to go deeper into the realm of information architecture, the two should develop a symbiotic relationship.  As they align, so will the structure of their website with the goals and strategy of the business.  Our book refers to this as “the feedback loop of business strategy and information architecture.”  See figure 18-1 below.



Reading Reflection: Week 13

This was a very short chapter that covered some tools that IAs use in a variety of scenarios.  Tool use can be quite different in each organization depending on software that is already in use, the familiarity that the staff has with new and existing tools, and the needs of the site.  The book divides the tools into the following categories:

  • Automated Categorization
  • Search Engines
  • Thesaurus Management Tools
  • Portal or Enterprise Knowledge Platform
  • Content Management Systems
  • Web Analytics / Tracking
  • Diagramming Software
  • Prototyping Tools
  • User Research and Testing


Each category explained what the software tool did, listed some example tools, and listed URLs for more research.  One takeaway from the chapter was dealing with search.  Last weeks I talked about search systems and some questions to ask when and if it is needed.  During our reading I came across this quote – “As content volume grows, search will become the heart of most web sites and intranets.”  I think this is true for most sites, but the challenge comes in getting everyone involved onboard with a set of tools that is easy to use and affordable.

As anyone that has had interactions with sale people know, you have to ask the right questions before diving into a new software tool or platform.  A lot of times the sales promises just don’t work out the way they said it would.  It is the job of the IA’s to conduct their own research – vendor customer calls, online research, software testing, etc to make sure the product is the right fit.


Reading Reflection: Week 12

Continuing on from last weeks post, this weeks focus is on search systems.  IA’s need to adequately determine if their website needs to add search capabilities and if so, what is the best approach for the users.  There are many factors that need to be addressed to start to answer the question of making a site searchable. A few questions that may help to put this into perspective are:

  • Is there sufficient content and/or recourses?
  • Are their better alternatives?
  • Will users bother with it?
  • Is the site dynamic?
  • What are the user expectation?

Search helps with larger sites that contain too much information to browse.  It is helpful for users that know exactly what they are looking for (typing correctly spelled words or phrases) and for those that don’t (can correct spelling, use controlled vocabularies, etc).  Narrowing the vast amount of information down to a smaller stack to be dealt with is the goal.  With any luck, the information that one is trying to find is one of the first few choices.  But getting to that point can be rather complex.

Thinking back to the dawn of the Internet and when I first got connected, finding information has improved dramatically.  As search system have evolved, they have become “smarter” at interpreting a user’s query to show results that are closer to what they were expecting.  The early search systems mostly relied on boolean searching using operators such as AND, OR, and NOT.  Users had to adapt to a machine-like mindset to do any advanced searching.  As the user base of the Internet began to grow, more data was gathered on how users actually searched for information.

Besides just indexing web pages, documents, and media, there are other aspects that aid sites that contain a lot of content.  One is what data should not be indexed (such as header information, navigation, etc).  Depending upon the audience, splitting up information into multiple zones (or categories), adding query builders, and adjusting between recall and precisions can all have an impact on the search results.  Only through knowing your audience, testing, and making iterative changes, will you finally get to a balance that gives the most benefit to your users. 

Reading Reflection: Week 11

Our ability to search through massive amounts of information has only continued with the dawning of the information age – vastly aided by the World Wide Web.  Everything from books and journals to how to video’s to medical conditions can be queried to gain a better understanding of a user’s interests.  “In 2012, we created 2.5 quintillion (2,500,000,000,000,000,000) bytes of data every day (Reference).”  With all the data that on the Web, the question has mostly switched from “Is the data available?,” to “How do I find the data I am looking for?”

To that end, search has been an evolving topic and a big business.  When a normal user thinks of searching the Internet, they typically thing of Google, Yahoo, or other search engines.  The thought that “I’ll Google it” is now they way we think.  Looking behind the scenes shows us how complex search actually is.  As IA’s we must obtain the right balance between correct and concise answers.  Too much data, although relevant, can overwhelm the user.  Too little, and they will go look elsewhere.

Thesauri, controlled vocabularies, and metadata have been developed to aid users with their queries.  Metadata “is definitional data that provides information about or documentation of other data managed within an application or environment.”  It can add simple keywords (or tags) to existing data that describes it, so it can be found later.  Controlled vocabularies use preferred terms for searchable data.  One example is a synonym ring, which defines “equivalent” items for search to help with retrieving the correct data.  An example synonym ring may contain the words coffee maker, Keurig, Cuisinart, Mr. Coffee, and Hamilton Beach.  An authority file is similar in that it lists preferred terms and their acceptable values.  It does not use broader synonyms or other variants.  The most robust (and expensive) search system we reviewed this week is that of a Thesauri.  Our book says it is “A controlled vocabulary in which equivalence, hierarchical, and associative relationships are identified for purposes of improved retrieval.”  It uses many layers and associations to construct the best answers to user queries.  There are also a few variants that can be implemented depending on the need and budget.

Search has come a long way since the introduction of the Web, which has helped refine the information age into something normal users can access. 

Reading Reflection: Week 10

Usability evaluation allows IA’s to interact closely with the users to see how each design iteration impacts their experience.  This is accomplished by assessing how users are impacted through different interfaces and improving its usage during the design phase.  The five quality components of usability are:

  • Learnability
  • Efficiency
  • Memorability
  • Errors
  • Satisfaction

Using the components above, the end result should be a useful site – one that is both usable and provides features the user needs.  The basis for usability testing is a user-centric design where the user base is incorporated through a series of tests to collect responses and feedback to better the experience.  These often take the form of informal or formal testing using controlled experiments to record user interactions.  From previous studies, only a small sampling from the users is needed for accurate representation.  There has been no evidence that a larger test group obtains better results.  In some situations, larger audiences can be beneficial, such as when rolling a new interface out to thousands or millions of users.  Bucket testing allows for larger audiences to experience the new interface at one time to either gain feedback or to easy the user population into the new design over time.  A few other methods include encouraging participant motivation, long-term studies, avoiding experimenter bias, and timed responses.

Mobile design deals with all aspect of displaying content on a host of mobile devices, all of which contain multiple browsers, connection speeds, screen sizes, and processing power.  Older web designs were constrained to one resolution, no matter the device’s screen size.  Today’s designs can adapt automatically if they are coded correctly – detecting the devices size and other requirements.  Specific mobile sites are also popular.  For smaller devices, IA’s need to take into count other aspects as well.  They are simplicity, image sizes, proprietary plugins (such as flash), non-persistent connections, location, and too excessive scrolling.  Meshing design and mobile adds a new dimension to an IA’s requirements.

Reading Reflection: Week 9

This section had a variety of topics including strategy, design and documentation, and content analysis heuristics.  Strategy from chapter 11 (it was split between the last two readings), finishes with highlights of the project plan and presentations.  The project plan is the formation of a structured schedule for the overall site construction.  Both short-term and long-term plans can be created to flesh out the remaining architecture.  Presentations should always be geared towards the audience.  IA’s need to get the prominent parties on their side and help them understand the recommendations.

Design and documentation was covered in chapter 12.  Proper diagram creation helps the viewer grasp the multi-dementions that a site can fill.  Multiple views are necessary as sites take on more complexity.  Viewers also need audible guidance in addition to diagrams to minimize any misunderstandings.  Blueprints are used (often as sitemaps) to show page and content relationships using label and proper organization.  Wireframes are where they visual pages start to take shape.  They are the templates of how the site’s pages will look and feel.   Typically created for the main and other important pages, they also help keep the site’s consistency.

Depending on the size of the site, content can be dealt with multiple ways.  Content can be split into chunks that are referenced on multiple pages.  That way, it can be modified in one place, and then automatically updated in multiple places.  It can be handled from a simple text document to more complex databases.   It is a way to reuse data, by separating the content (data) with the container (site objects or pages) to promote consistency and efficiency.  When inventorying a site’s content, content models can be developed to allow automatic “extras” to be displayed (such as related items) during a user’s search.  Controlled vocabularies add metadata (data about data), definitions, and other information to a site’s content that allow for a thesaurus and other data linkages to further exploit a site’s possibilities. 

Some of the next steps are creating a website prototype and documenting everything.  An important aspect of documentation is why certain decisions were made and the “How to” of things so future iterations keep to the same guidelines for consistency, manageability, and findability.

Lastly was an online reading pulled from the cross-section of users, content, and business context that provides qualitative results and general trends while organizing the report and helping to identify signification issues that may not be obvious otherwise.  The former was a result of the set of 11 heuristics for analyzing website content (below).

Information scent
Bounded horizons
Multiple access paths
Appropriate structure

Reading Reflection – Week 7

My team is also responsible for the group discussion this week.

Chapter 10

Research is avery important part of developing any quality website.  It begins be gaining an high-level understanding of the organization’s goals, reviewing existing materials, and meeting with the strategy team.  The entire IA process includes periods of research, strategy, design, implementation, and administration.

The research framework consists of the following:

Context – Includes “business goals, funding, politics, culture, technology, and human resources”

Content – Includes “Document/data types, content objects, metadata, volume, existing structure”

Users – Includes “Audiences, tasks, needs, information seeking behavior, experience, vocabularies”

It is incorporates meeting with the various teams to ensure the project is a success.  Stakeholder and User Interviews are also extremely important.

From Chapter 11

The strategy report is often the largest, hardest, most detailed, and important deliverable for the IA (Information Architecture) team.  It combines all the acquired research thus far (including previous results, analysis, and ideas) into a clearly communicated document.  Organizing this document is also a challenge.  The strategy report should include high-level visuals and be paired with verbal explanations to explain ideas and answer any questions.  The main ideas need to be broken down into consistent and cohesive goals.

Breaking down the strategy report into pieces, the textbook’s example (pages 279-283) included:

Executive Summary – The executive summary should present a high overview of the major problems and the team’s recommendations.  The goals and methodology should clear, which helps to set the tone for the entire document.  The summary should generate interest in the readers so they continue reading.

Audiences & Mission/Vision for the Site – The mission for the website should be restated so it is fresh in everyone’s minds.  The context for the audience and goals should be conveyed as they fit into the larger design.  Also define any vocabulary for users roles and the intended audiences.

Lessons Learned from Benchmarking, User Interviews and Content Analysis – This is where everything that has been learned is referenced to make your recommendations.  The book describes it as “the bridge between your research and analysis.”  The results should backup your decisions and be clear for the reader to see.