SLAC PEP-II
BABAR
SLAC<->RAL
Babar logo
HEPIC E,S & H Databases PDG HEP preprints
Organization Detector Computing Physics Documentation
Personnel Glossary Sitemap Search Hypernews
Unwrap page!
Comp. Search
Who's who?
Meetings
FAQ Homepage
Archive
Environment
Administration
New User Info.
Web Info/Tools
Monitoring
Training
Tools & Utils
Programming
C++ Standard
SRT, AFS, CVS
QA and QC
Remedy
Histogramming
Operations
PromptReco
Simulation Production
Online SW
Dataflow
Detector Control
Evt Processing
Run Control
Calibration
Databases
Offline
Workbook
Coding Standards
Simulation
Reconstruction
Prompt Reco.
BaBar Grid
Data Distribution
Beta & BetaTools
Kanga & Root
Analysis Tools
RooFit Toolkit
Data Management
Data Quality
Event display
Event Browser
Code releases
Databases
Check this page for HTML 4.01 Transitional compliance with the
W3C Validator
(More checks...)

"Web sites that work - Designing with your eyes open"
A course given by User Interface Engineering
San Francisco 3-4 August

Introduction

The course was designed to present the results of UIE's research into the usability of web sites which was carried out by observing users who were employed to retrieve particular information from particular sites. The factors correlated with success were presented. The course taught us how to test our own sites and thereby see them through the users eyes.

Questions for web designers:

  • Why do people come to your site?
  • What are the most important features of the site?
Web designers need to understand the site's content and user requirements.

Good Usability Test tasks

  • Based on what people do in real life
  • Focus on high-risk areas of site
  • Do not direct users to answer
  • Avoid leading terminology
  • Are do-able

Search engines

Users of search engines in UIE studies:
  • use them half the time.
  • are less likely to succeed with them.
  • are less likely to give up with them.
  • usually try one word searches
  • tend not to use boolean logic
Search logs (if available) can be analysed to help in redesign of site and links. Key words files can be used to help search engine - e.g. in META tag.

UIE advise the removal or burying of search engines. Concentrate instead on links and user requirements.

Links

Definitions:
  • Content link leads to content
  • Category link leads to a list of more links
  • Keyword link is part of a search
  • Back(button) or home(page) links
Successful link is one that contributes to a successful task. UIE tests give likelihoods of success as 47%(Content); 31%(Category); 24%(Keyword); and 12%(Back or home). Likelihoods of users giving up were 3%(Content); 12%(Category); 4%(Keyword); and 6%(Back or home)

Users usually begin with category links and resort to Keywords after about 3 or 4 clicks.

Users lose the scent due to:

  • Dead links
  • Lying links
  • Missing words
  • Misinterpreted context
  • Links that go from narrow to general
  • Ambiguous links (link names must communicate information)
Therefore important to make effective links. UIE find that links are more important than graphics or search engines in successful tasks. Links should be descriptive and clearly different from one another on the same page.

UIE found optimum number of words per link to be around 9 to 10 words as they gave the highest likelihood of success. Link length was found to be particularly helpful for category links.

Brackets and bullets gave higher success likelihoods (about 42% each) than icons/symbols (33%); image links (29%); buttons (24%); and picture links (20%). Brackets and bullets help users in scanning and grouping links. Image links are time consuming to make and are no more succesful than any other type of link.

Page Layout

Here defined as grouping and positions of elements on a page; levels of information; and what types and style of elements.

UIE found that fewer, longer pages were better than more, shorter pages. The more groups of links the better but they found that left and bottom panels were not as successful as link groups. Probably because they contain short less descriptive category links.

More levels of information per page helped the user - i.e. more ways of getting to the information you want. Also the more information you give about what's contained in the whole site the better. Users were found to not mind scrolling but, since they knew that the information could be found, this may not be surprising.

Over-designed pages are not useful to users. Structured lists of links are much more useful. However content links should be kept near the top of the page.

Whitespace and readability

Whitespace is defined as non-information bearing space on a web page. UIE found that users preferred sites with less whitespace and hence greater information density. Whitespace means less in the user's browser and so prevents them from seeing the big picture.

Readability is defined as text which is neither related to a link nor anything to do with content. Users prefer less "readable" sites as they are looking for buzzwords not grammatical structures.

Graphics

Defined as images, pictures, backgrounds, animation, but nothing to do with links nor layout. UIE found that graphics don't affect the user's success nor the "fun" they had on the site. Users don't like to wait for useless graphics to download and they found animation distracting.

User's knowledge

Users become more critical of a site when they become more familiar with it. Those who were knowledgeable on the web-site's subject matter, or the internet itself, were no more successful at finding information than non-experts.

Thus site design and its usefulness to the differing requirements of users is all important.


BaBar WebMaster: miw@slac.stanford.edu