.....................................................................
Assigned : READINGS - This week, read Chapters 5 - 6 from the "polar bear" text.
Book: Information Architecture for the World Wide Web 3rd Edition, by Peter Morville and Louis Rosenfeld, ISBN 100596527349
TAKE AWAYS
Chapter 5
ORGANIZED SYSTEMS
(page 53)
As IA we organize info so that people can find the answers to their questions and strive to support casual browsing, direct searching, avoid scrolling = aim is to label system to give it sense to user.
The Web provides environment to create this atmosphere.
IA are more and more becoming like librarians to centralize information and organize it for users.
(page 54)
Blogs, Corporate websites, freedoms to publish, content growth, information overload,
exabyte=billion gigabytes=wow!
Classification systems built on foundation of language and words can be understood in more than one way=ambiguity.
(page 56)
Heterogeneity=collection of objects with unlike parts.
Homeogenous=similar or identical elements=allows structure to classification systems.
Some websites=heterogeneous provide access to documents which in turn articles, journals, link outs to pages or other websites=granularity.
(page 57)
The way some people organize their files on their computer can be maddening.
Labeling and organizing systems are affected by their creators perspectives.
Website are designed for multiple users and all users have a different way of understanding info and how to organize it.
(page 58)
The choice of labeling and organizing IA for websites has huge impace on how users of the site perceive the company, products, service.
IA's must be sensitive to ogr's polictical environment and company image, products and services.
(page 59)
Schemes=telephone books, bookstores, supermarkets, tv programs guides=help us access the information.
Telephone book exact(alphabetical) vs supermarket=ambiguous at best.
(page 60-67)
-alphabetical-index, dictionaries
-chronological-archives, announcements
-geographical-traveling,maps
-topic-yellow book pages, consumer reports
-task-applications, things to perform
-audience-repeat visitors/bookmark the site, clutter free, schematic, self identity
-metaphor-trash can =recycling bin, desktop, folders, intuitive
-hybrids-mixed together of various things but laid out so easy mental-model to user
(page 69)
Top Down approach
Hierchy/taxonomy=organized info into hierchies: family trees, kingdoms, classes, species, organizational charts, books, chapters
These are simple familiar ways to organize.
(page 70)
Taxonomies that allow cross-linking=polyhierchal
(page 71)
Number of links safely include is constrained by users abilities to visually scan the page rather than by their short term memories.
INSTEAD:
-recognize the danger of overload options
-group and structure at page level
-subject your designs to user testing
(page 73)
Bottom up Approach
Data arranged for ease of speed and search and retrieval=example Rolodex
(page 74)
Relational Database Model(page 75)=data stored within set of relations or tables.
Rows in tables represent records, columns represent fields, data may be linked via keys.
Tag documents = controlled vocabularies metadata = powerful search, browse, filtering, linking occurs. CMS Content Management System software manages your metadata and controlled vocabularies.
(page 76)
Diagram showing structural approach to defining a metadata schema
(page 77)
Free Tagging=collaborative categorization, mob indexing, ethno classification
Tags are public=serve as social navigation
(page 80)
Break down site into components, tackle one question at a time, decompose content into narrow domains=identify opps for effective org systems.
____________________________________________________________
TAKE AWAYS
Chapter 6
LABELING SYSTEMS
(page 82)
Labeling=form of representation
Works as a short cut to chunks of information
If done poorly, users suffer, can't find what they need
(page 83)
Need feedback from users/sometimes response to feedback too delayed
(page 86)
Labels- Two formats
Textural
Iconic
Contextural links=hyperlinks to chunks of information on other pages or to another location on the same page
Headings=labels that simply describe the content that follows them, just as print headings do
Navigation system choices=labels representing the options in navigation systems
Index terms=keywords, tags, subject headings that represent content for searching or browsing
(page 98)
General Guidelines-Designing Labels
-Content users and context affect all aspects of the IA
-Ambiguity makes is hard to assign labels to describe content/don't make it difficult for the users
(page 99)
Narrow scope wherever possible.
Reduce the number of possible perspectives on what a label means.
Labels, like organizations and navigation systems, are systems in their own right.
(page 99)
Consistency
-Style:(Strunk and White)
-Presentation: fonts, sizes, colors, whitespace, grouping
(page 100)
-Syntax:verb based labels for instance, single syntactical approach
-Granularity:equal in specificity,i.e. restaurants, burger king, fast food franchises
-Comprehensiveness: reduce noticeable gaps, mistakes, keep things consistent
-Audience:know your audience, consider the language you use, terminology consistencies
(page 101)
Sections Labeled & Navigation
Top of page types of labels=i.e. buy tickets, hotels, travel info
Body navigation types of labels=i.e. track your flight, what to expect in airport
Bottom of page types of labels=i.e.home, sitemap, FAQs, Contact us
(page 102)
Controlled Vocabularies and Thesauri
Usually created by prof's like librarians alike.
Accurate representation and consistency.
Seek out narrowly focused vocabularies that help specific audiences to access specific types of content.
(page 103)
Unfortunately there are not controlled voc's for every domain. But seek out as much as you can before creating label system.
-Taxonomy Warehouse
-ThesauriOnline
-Controlled vocabularies
-Web Thesaurus Compendium
(page 104)
Creating new label systems
Labels can be derived from content in the site.
auto extraction=of meaningful terms
Take terms output by software (auto extraction) and use them as candidates for controlled vocabulary.
(page 105)
Ask content authors to suggest labels for their own content.
Use advocates and subject matter experts=librarians, switchboard operators, subject matter experts.
(page 106)
Card sorting=exercises are good way to learn how your users would use information.
open=cluster labels for existing content/closed=provide subjects with existing categories and sort content into those categories
(page 108)
free listing=select an iten and have subjects brainstorm terms to describe it
(page 109)
Indirectly from users=check search queries used
Search-log analysis=search analytics least intrusive data on labels your site's audiences actually use.
**Labels represent relationship between users and content-analyze and adjust regularly
.....................................................................
Subscribe to:
Post Comments (Atom)
No comments:
Post a Comment