Studies and multimedia

File formats in "barrier freedom" test

The following file formats were tested:

1. Text-based file formats

• ASCII (.txt)
• HTML (.html)
• RichText (.rtf)
• LaTex (.tex)
• Microsoft Word (.doc)
• Excel (.xls)
• XML (.xml), sub-language .mathml

2. Graphic-based file formats

• Corel Draw (.cdr)
• Scanned formats saved as graphics (.jpg, .gif, .bmp, .tif)

3. Semi-graphic file formats

• Adobe Acrobat (.pdf)
• Postscript (.ps)
• PowerPoint (.ppt)
• Flash (.fla)

---

Mode of procedure in file-format tests (methodology)

File format selection:

The selection of the file formats to be tested reflected the use of the formats in courses offered at the Universität Karlsruhe (TH). In addition, formats were selected which are currently being used by our blind and partially-sighted students. For greater clarity we have separated the file formats into three groups: text-based, graphic-based and semi-graphic file formats.

Layout of the questionnaire :

The questionnaire was split up into three blocks. The first block contained the technical prerequisites necessary to test the selected format. The entries requested were the operating systems used, programs and screenreader . In addition to definition of the type of visual impairment the student was asked to assess his/her own skills with reference to these elements. The second block inquired about the legibility of texts, tables, graphics, frames , mathematical expressions, images, hyperlinks, imagemaps , navigation aids etc. Questions in the third block referred to the creation of files and texts, where the study aids implemented and their accessibility were under investigation.
The students were given great scope to answer open questions concerning specific factors, problems, personal experiences and recommendations.
The tests were conducted in October and November 2002 and were subsequently discussed and evaluated during the course of several group meetings.

Result:

The tests clearly demonstrated the widely differing usability levels of the file formats depending on the students visual impairment.
Partially-sighted students submitted differing evaluations depending on the type and severity of the visual impairment concerned. Some students were able to use file formats which other students found to be inaccessible. In the case of the partially-sighted students in particular, a relationship was evident between the use of a file format and the individual students' previously-acquired skills. Some students know tricks enabling them to use formats which are difficult to access.

---

File-format test summaries

The summaries contain a brief assessment of the file formats we tested. They describe whether and how visually impaired students are able to work with the respective formats. The evaluations reflect the experiences of the blind and partially-sighted students and are to serve as recommendations for and against the use of a format within the scope of a course with respect to freedom from barriers.

1. Text-based file formats

• ASCII (.txt):
  ASCII is a pure text representation form and does not involve any other elements, such as graphics. The tools used to create and read the format are easily accessible.
  During the text creation process care should be taken that line length is constant. We recommend a line length with a maximum of 80 characters as Braille displays can represent either 20, 40 or 80 characters.
• HTML (.html):
  HTML pages must be designed in accordance with W3C standards. Appropriate labelling is important for graphics, imagemaps and frames . Layout creation and recognition requires support from a sighted person.
  Unnecessary barriers arise through the "incorrect" use of HTML elements. Example: An empty, invisible frame (table or graphic) is used as a spacer. The blind user receives a false impression of the layout of the homepage.
• Rich-Text (.rtf):
  Continuous text with simple formatting elements (bold, italics) is easy to create. Tables are legible and can be created if this function is supported by the screenreader. Mathematical expressions and graphics cannot be read by blind and most partially-sighted students, rendering scientific work in the fields of natural science and engineering science impossible.
  Hence, the appearance of a document depends both on the application with which it is created and the application with which it is read. Consequently, text and other elements, such as graphics, may be displaced. This can have an adverse effect on the legibility of RTF documents.
  Example: The section switch is represented differently in the English language than in the German language.
• LaTeX (.TeX):
  LaTex is the only file format which enables the reading and writing of mathematical expressions through the direct legibility of the source file . The text structure stands in the foreground in the source text, whilst the layout of the document is defined by the document type specified in the source text. Visually impaired students do not benefit from these advantages as they use LaTex in the graphic version as do their sighted fellow-students.
  Processing tables and diagrams is possible with limitations.
  Reading is made more difficult through additional formatting commands (e.g. change of font, centring).
• Microsoft Word (.doc):
  Continuous text with simple formatting (bold, italics) is easy to create. Tables are legible if this function is supported by the screenreader . Visual feedback is required from a sighted person to create tables. In addition, the application requires good spatial imagination.
  Blind and most partially-sighted students are unable to read mathematical expressions and graphics. Hence, scientific work in the field of natural science and engineering science is not possible.
• Excel (.xls):
  Tables are easily legible and easily navigable using Jaws . It is very difficult to exploit the mathematical potential offered by this format as this requires very extensive Excel skills.
• XML (.xml):
  XML has not yet been extensively tested for blind and partially-sighted students. A corresponding tool is not yet available. This is primarily due to the multiplicity of opportunities this format could create. Accessibility cannot be ruled out at the present time, but the format has hardly been used for text-processing programs so far. This will change in the future and will open up great opportunities with regard to the barrier freedom of XML.
  It is very promising as a starting format for the conversion of mathematical scripts, for instance, into languages accessible for the blind.

2. Graphic-based file formats

• Corel Draw (.cdr):
  The format is illegible for blind people. The fact that objects can be freely positioned requires visual feedback for the processing and generation of documents, the support of a sighted person is necessary.
  The format is accessible for partially-sighted students who are able to work with graphic elements as almost all the features are available without having to use a mouse.
• Scanned documents saved as graphics (jpg, gif, bmp, tif):
  These file formats are unsuitable as text formats as they only contain image information. These formats are used to scan books.
  Blind students are able to process the text involved by means of OCR (Optical Character Recognition) software and can recognise it with limitations. At the beginning, care must be taken that people without visual impairment can read the text on the screen after it has been scanned for processing. During the scanning process a resolution of at least 300 dpi should be used. Text information can be accessed. Hand-written texts and graphic information remain inaccessible here. The best results are achieved with the TIFF format.

3. Semi-graphic file formats

• Adobe Acrobat (.pdf):
  PDF can only be read using a screenreader able to use the range of functions of MSAA (Microsoft Active Accessibility). Coded documents cannot be read. There is an additional program for Acrobat 4.5 which enables the export of PDF files to HTML. Legibility is restricted to text and hyperlinks. Graphics, tables, diagrams and other elements are inaccessible.
  Some partially-sighted students are able to read PDF documents. Sans serif fonts, such as Arial, should be used however. Attention should also be paid to font size and colour contrast.
• Postscript (.ps):
  The file format is not legible. Continuous text can be extracted to make it accessible. To do this, fonts must be used for text representation. All other information (tables, mathematical expressions, graphics etc.) are lost.
• Powerpoint (.ppt):
  PowerPoint is legible in principle. This depends on the creator of the text having observed the following criteria:
  Graphics, diagrams and imagemaps must feature an alternative text (e.g. through a note function).
  The layout of a foil must be systematic. Jumping from the left bottom edge to the top upper right edge within a foil should be avoided.
  Arrows and animated elements cannot be perceived. Alternative descriptive comments (e.g. in the note section) are required.
  Colour contrast is important for partially-sighted students. Dark scripts on a dark background and light scripts on a light background are to be avoided.
  PowerPoint foils must be made available to the students concerned before the respective lectures so that they can prepare.
• Flash (.fla):
  In this format there is a difference between static and dynamic elements. Static elements such as text and navigation units are legible using special tools. Animated elements cannot be recognised. Animation sequences must be described in the form of text for blind and partially-sighted students.

next >