The problems found with such a setup may not have been caused by a single design decision, such as choosing very narrow lines of text. These problems could actually result from a combined effect of several design decisions where they would by themselves not produce a noticeable negative effect. Thus, the possible advantage of the integral approach is illustrated.
The disappointment, mentioned earlier, with the absence of clarity about how
scientific efforts can add to the craft of the typographer specifically pertains to one step of the
typographic design process; choosing a suitable typeface. One would expect to find more
systematic ways of selecting type for a particular application or a particular reader audience.
In fact, only one method has been found which will quickly give the typographer an
impression of how a particular typeface is perceived by readers. Figure 20 displays the table of ratings
used in this method, proposed by Bartram (1982). Perhaps more methods have been developed, but none seem to be
well known or seem to have proven their value. Surely, such evaluation procedures for
typefaces could be useful. This is another example of how a practical approach may benifit
the typographer.
![]()
Figure 20:
Typeface Personality Evaluation; an example of reader-scoring of typefaces on known dimensions (after Bartram, 1982)
If the sum of the ratings is R and the numbers of scales is n, then the Score is equal to (R/n - 4)/1.5
![]()
| Scales | Ratings | ||
| Beautiful | 5 | Evaluation | |
| Pleasant | 6 | Score: E=(26/5 - 4)/1.5 = + 0.8) | |
| Good | 5 | ||
| Expensive | 4 | ||
| Interesting | 6 | ||
| ---- + | |||
| Total: R= | 26 | ||
| Bold | 3 | Potency | |
| Strong | 1 | Score: P = (7/3 - 4)/1.5 = - 1.11 | |
| Heavy | 3 | ||
| ---- + | |||
| Total: R= | 7 | ||
| Happy | 6 | Mood | |
| Hot | 5 | Score: M = (18/3 - 4)/1.5 = + 1.33 | |
| Relaxed | 7 | ||
| ---- + | |||
| Total: R= | 18 | ||
| Fast | 4 | Activity | |
| Active | 3 | Score: A = (7/2 - 4)/1.5 = - 0.33 | |
| ---- + | |||
| Total: R= | 7 | ||
![]()
In the example of the instruction manual, the user trial is equivalent to a courtroom situation; the design as a whole is put to trial and judged, using the reader as a witness as well as a judge. The role of the jury is reserved for both researchers and designers, be it that the two parties serve under different systems of law. When the reader finds the design to be ineffective in some way, the jury will look at the case and considers in which ways the design could be rehabilitated. The design could then be altered and return to court until it is acquitted and ready for use. Of course, it is quite impossible to treat every design so formally for lack of time and money. It is possible, however, to bring up for trial a number of designs that exemplify certain conditions of use. For each of these conditions, different design-decisions are suitable. Three ways come to mind in which conditions can generally differ; between applications, between the ways of dividing the different elements on a ‘page’, and between languages.