To determine the extent to which the links on this site do their jobs.
To make links work well for users, though, the authors must follow a dozen strategies, and within those broad approaches, a number of tactics.
These strategies and tactics for links appear here as a set of guidelines based on research and actual experience, documented in books, journal articles, and online styleguides. (References appear at the end of each strategy). Each guideline, then, provides a method for a writer to follow, or a heuristic.
In this evaluation we test the text against these guidelines. This, then, is a heuristic evaluation.
Here’s how to perform a Heuristic Online Text (HOT) evaluation of links on your target site.
1. Save this file with a name that includes
q The site you are analyzing
q The aspect you are evaluating (brevity, in this case)
q Initials
q A period
q A suffix indicating the file type (doc for Word files, htm for HTML files)
Examples: ibmbrevityjp.doc,
yahoobrevityds.htm
2. Go to the site, and locate a fairly typical page
that has several links, anywhere on the page, outside of menus.
3. In this file, type the subject of the page,
under Sample #1, below.
The subject appears in the title bar of the window (not including ads for your browser) or in the major heading at the top of the page. Use whichever best articulates what the page is about.
4. Copy the page and paste it into this file after
the subject.
5. Return to the page and copy the URL for that
page, then paste that into this file, in the line right after the paragraph.
The URL is the address of the page.
6. Type today’s date on the next line, to show when
you collected the sample.
7. Repeat this process, collecting at least three (3)
pages.
If possible, find pages with different kinds of content.
Tip: You may want to print out your samples, so you can look at their text on paper as you write onscreen.
8. Apply the HOT Evaluation to the samples you have
collected, filling out the evaluation form.
If a strategy or tactic seems irrelevant, omit it from your evaluation. Note that this will change the total possible points.
Subject:
Page:
URL to the page:
Date investigated:
Subject:
Page:
URL to the page:
Date investigated:
Subject:
Page:
URL to the page:
Date investigated:
Test
In each page, I found only one or two links that
failed to describe the target, or none.
YES=1, NO=0.
Impression
Overall, did the links themselves seem
descriptive?
YES=1, NO=0.
Example
Which link seemed particularly repetitious (if
any)?
Comments
If this guideline does not seem to apply, indicate
that here.
If you have further thoughts, or reflections, about
the way the text follows, or ignores, this guideline, please put those
observations here.
Test
Next to each link, I found extra information about the
target, in ordinary text, or a label.
YES=1, NO=0.
Impression
Overall, the site seemed to offer a lot of information
about the link targets, so I could tell where I would be going, before I
clicked.
YES=1, NO=0.
Example
Which link seemed particularly bad because it lacked
an accompanying description?
Comments
If this guideline does not seem to apply, indicate
that here.
If you have further thoughts, or reflections, about the
way the text follows, or ignores, this guideline, please put those observations
here.
Test
With each link, I found that the text of the link
contained words or phrasing similar to the text in the title or major heading
on the target page.
YES=1, NO=0.
Impression
Overall, the links seemed to echo the titles or major
headings.
YES=1, NO=0.
Example
Which link seemed particularly different from the
title or headings on the target page?
(Quote the linktext and the title or headings).
Comments
If this guideline does not seem to apply, indicate
that here.
If you have further thoughts, or reflections, about
the way the text follows, or ignores, this guideline, please put those
observations here.
See: America Online (2001), Apple (1997), Berners-Lee (1998), Bieber et al (1997), Borges et al (1998), Bricklin (1998), IBM (1999), Levine (1997), Microsoft (2000), Nielsen (1995, 1998, 1999e, 1999f), Nielsen & Morkes (1997), Rosenfeld & Morville (1998), Spool (1997), Spyridakiis (2000), Sun (1998), Tchong (1998b), W3C (1999) in the bibliography at http://www.webwritingthatworks.com/HTres2cbiblio.pdf.pdf
Test
When a link appears within a sentence, it goes at the
end. .
YES=1, NO=0.
Impression
Overall, the linktext acted as the emphatic element in
the sentence or paragraph.
YES=1, NO=0.
Example
Which link appeared, distractingly, at the beginning
or in the middle of the sentence? (Quote the whole sentence).
Comments
If this guideline does not seem to apply, indicate
that here.
If you have further thoughts, or reflections, about
the way the text follows, or ignores, this guideline, please put those
observations here.
See: Bricklin (1998), IBM (1999), Levine (1997), Nielsen (1997b), Sun (2000) in the bibliography at http://www.webwritingthatworks.com/HTres2cbiblio.pdf.pdf
Test
In each link within running text, I found the text
made sense.
YES=1, NO=0.
Impression
Overall, the linktext seemed to make a meaningful
contribution to its sentence.
YES=1, NO=0.
Example
Which link seemed particularly out of touch with its
surrounding sentence?
Comments
If this guideline does not seem to apply, indicate
that here.
If you have further thoughts, or reflections, about
the way the text follows, or ignores, this guideline, please put those
observations here.
See: Arthur (2000), Berners-Lee (1995), Levine (1997), Nielsen (1997b) in the bibliography at http://www.webwritingthatworks.com/HTres2cbiblio.pdf.pdf
Test
There is a section of related links, with descriptions
of the target info.
YES=1, NO=0.
Impression
Overall, the pages do not refer very often to other
pages within the site.
YES=1, NO=0.
Example
Which page seemed to need links to related info? (If any).
Comments
If this guideline does not seem to apply, indicate
that here.
If you have further thoughts, or reflections, about
the way the text follows, or ignores, this guideline, please put those
observations here.
Test
The running text contains pointers to related pages on
the site.
YES=1, NO=0.
Impression
Overall, the running text seems thickly related to
other content on the site.
YES=1, NO=0.
Example
Which paragraph seemed to need links to related
info? (If any).
Comments
If this guideline does not seem to apply, indicate
that here.
If you have further thoughts, or reflections, about the
way the text follows, or ignores, this guideline, please put those observations
here.
See: Ameritech (1997), Apple (1997), Berners-Lee (1998), Bricklin (1998), Farkas and Farkas (2000), Gagne & Briggs (1979), Horton (1990), Levine (1997), Microsoft (2000), Nielsen & Morkes (1997), Nielsen (1999f, 2000a), Reigeluth et al (1980), Robinson & Knirk (1984), Sheetz et al (1988), Shneiderman (1998), Sun (2000) in the bibliography at http://www.webwritingthatworks.com/HTres2cbiblio.pdf.pdf
Test
There are links to other sites on every page, or most
pages.
YES=1, NO=0.
Impression
Overall, the pages do link to relevant info on other
sites.
YES=1, NO=0.
Example
Which page seemed to need links to outside sites, but
did not offer them? (If any).
Comments
If this guideline does not seem to apply, indicate
that here.
If you have further thoughts, or reflections, about
the way the text follows, or ignores, this guideline, please put those
observations here.
See: America Online (2001), Berners-Lee (1995), Levine et al (1999), Lynch (2000), NCSA (1996), Nielsen (1999a), Nielsen & Morkes (1997), Spyridakis (2000), Sun (2000) in the bibliography at http://www.webwritingthatworks.com/HTres2cbiblio.pdf.pdf
Test
There are breadcrumbs, highlighting, graphics, or text
to show where this page is within the larger structure. The page has a logo or
site identifier on it.
YES=1, NO=0.
Impression
Overall, I knew where I was.
YES=1, NO=0.
Example
Which page seemed to need some indication of its
location within the hierarchy? (If
any).
Comments
If this guideline does not seem to apply, indicate
that here.
If you have further thoughts, or reflections, about
the way the text follows, or ignores, this guideline, please put those
observations here.
See: Apple (1997), Bricklin (1998), Conklin (1987), Nielsen (1995, 1996, 1997b, 1997d, 2000a), Utting & Yankelovich (1989), Zimmerman (1997) in the bibliography at http://www.webwritingthatworks.com/HTres2cbiblio.pdf.pdf
Test
The page contains at least three of the following:
status information, date, phone number, comment link, author name.
YES=1, NO=0.
Impression
Overall, I found out what I wanted to know about the
page.
YES=1, NO=0.
Example
Which page seemed to lack meta information? (What was missing?).
Comments
If this guideline does not seem to apply, indicate
that here.
If you have further thoughts, or reflections, about
the way the text follows, or ignores, this guideline, please put those
observations here.
See: Apple (1997), Berners-Lee (1995), Cook (1997), IBM (1999), Levine (1997), Lohse & Spiller (1998), Lynch & Horton (1997), Nielsen (1999d), Spyridakis (2000), W3C (1999) in the bibliography at http://www.webwritingthatworks.com/HTres2cbiblio.pdf.pdf
Test
The URL contains words or text that I can parse, to
understand where I am, or what I am looking at.
YES=1, NO=0.
Impression
Overall, I found the URLs made sense.
YES=1, NO=0.
Example
Which URL seemed made up of gibberish, if any?
Comments
If this guideline does not seem to apply, indicate
that here.
If you have further thoughts, or reflections, about
the way the text follows, or ignores, this guideline, please put those
observations here.
See: IBM (1999), Nielsen (1999f) in the bibliography at http://www.webwritingthatworks.com/HTres2cbiblio.pdf.pdf
Test
When I choose to View the Source of the page, I see
that there is alternate text for every image, animation, sound, or video. The
headings are conventional H1, H2, H3.
The page uses stylesheets, relative font sizes, few or no frames, a
single table, if any.
YES=1, NO=0.
Impression
Overall, I found the site made an effort to be accessible.
YES=1, NO=0.
Example
Which page, if any, seemed least friendly to people
with special needs?
Comments
If this guideline does not seem to apply, indicate
that here.
If you have further thoughts, or reflections, about
the way the text follows, or ignores, this guideline, please put those
observations here.
See: Nielsen (1999f), W3C (1999) in the bibliography at http://www.webwritingthatworks.com/HTres2cbiblio.pdf.pdf
Test
If the page offers a file for downloading, they tell
me how large the file is, how long it will take to download at a particular
speed, what format it is, and what software I can use to view it.
YES=1, NO=0.
Impression
Overall, I found the site made an effort to help me
decide whether or not to download the files.
YES=1, NO=0.
Example
Which page, if any, seemed least friendly to people
with special needs?
Comments
If this guideline does not seem to apply, indicate
that here.
If you have further thoughts, or reflections, about
the way the text follows, or ignores, this guideline, please put those
observations here.
See: Nielsen (1999f), W3C (1999) in the bibliography at http://www.webwritingthatworks.com/HTres2cbiblio.pdf.pdf
Test
Content that is new, or changed, has some identifying
information, to alert me to the fact that the info is new or updated.
YES=1, NO=0.
Impression
Overall, I found the site made clear what was new, and
what was old.
YES=1, NO=0.
Example
Which page, if any, seemed to need a little more
identification of the new or changed items?
Comments
If this guideline does not seem to apply, indicate
that here.
If you have further thoughts, or reflections, about
the way the text follows, or ignores, this guideline, please put those
observations here.
See: IBM (1999), Nielsen (1999d), Microsoft (2000), Sun (2000) in the bibliography at http://www.webwritingthatworks.com/HTres2cbiblio.pdf.pdf
Test
When I choose to View the Source, I see that the meta
tags for Description and Keywords contain words and phrases that reappear in at
least two of the following: the title, the major heading, the first sentence on
the page, the running text, and the linktexts on the page.
YES=1, NO=0.
Impression
Overall, I found that each page could convince a
spider that its content is relevant to the search terms someone might use to
find the page.
YES=1, NO=0.
Example
Which page, if any, might seem to be fooling a spider,
by having keywords and descriptions that do not match the rest of the page
content?
Comments
If this guideline does not seem to apply, indicate
that here.
If you have further thoughts, or reflections, about
the way the text follows, or ignores, this guideline, please put those
observations here.
See: Belew (2000), IBM (1999), Kilian (1999), Lloyd-Martin and Whalen (2001), Nielsen (1999f), Sun (2000) in the bibliography at http://www.webwritingthatworks.com/HTres2cbiblio.pdf.pdf
Assigning a grade to text is always a bit arbitrary. But counting up the points for these sample pages, we reach this diagnosis:
Total Points:
Total Possible:
Percentage:
90-100%: Excellent links.
75-89%: Pretty good links, but occasionally confusing.
60-74%: Could use some work.
45-59%: Clumsy.
25-44%: Impenetrable.
0-24: Comic.
In a few paragraphs, summarize your most important observations—both
positive and negative.
List the top five problems with the links, and in a sentence or two,
summarize what you would recommend as solutions.