Letters / response

American Journal of Pharmaceutical Education,  Spring 1998  by Gilkey, John C,  Tong, Theodore G,  Shepherd, John

• E-mail
• Print
• Link

Multimedia Instruction and Costs

Dear Sir:

We are writing to respond to the article "Development of Multimedia Computer Applications for Clinical Pharmacy Training," by John Schlicht and Bruce Livengood, that appeared in the Fall 1997 issue of AJPE. The authors, discussion of the benefits of CAI is salient and to the point, and their description of the process of creating multimedia instructional modules is mostly in line with our own experience. There are some aspects of their discussion, however, that are not commensurate with our experience, and that should be brought to the attention of those who are intending to create their own instructional modules. These are as follows:

The authors state that is no advantage to scanning images that are to be incorporated into a module at a resolution higher than 75 dpi (approximately the screen resolution of a monitor, 72-75 dpi). This would be true if: (i) there were no blemishes or dust/dirt on the surface of the original or on the glass of a flatbed scanner, (ii) the images being scanned were precisely what was needed in the final product: and (iii) the scanned image will never need to be enlarged to accommodate a higher resolution display. However, the first of these conditions is virtually never met-the images must be edited to remove blemishes (or, in the case of scanned slides, a nonuniform background, even if the background was uniform in the original slide). Very often, the second condition is also not metthe images must be altered to fit the current need. Finally, prudence dictates that images should be scanned at a resolution high enough to accommodate future needs, so that they need not be scanned and edited again.

Editing and modification are much easier and give much better results, and acceptable results can only be obtained when resizing, with images scanned at a resolution much higher than screen resolution. Twice this resolution in both directions would be a good start, or about 150 dpi minimum. Furthermore, to take full advantage of the capabilities of the scanner, images should be scanned at the optical resolution (not the interpolated resolution, which can be much higher but contains no more information) of the scanning device, which will be about 300 dpi for a $300 scanner and 600 dpi or more for better scanners. Images should also be scanned, edited and resized in 24-bit color (or higher, if available). Since 24bit color is effectively a continuous representation of color space, color replacements, spatial transformations such as rotating and resizing, and various types of filtering can be applied with fidelity to 24-bit images. Eight-bit indexed color (256 colors) is not a continuous representation of color space, so these same transformations applied to 256-color images give rise to objectionable artifacts. The image-editing software mentioned by the authors can be used to reduce the images to the desired size. followed by reduction to 256 colors, when editing is finished. This result will also be better than that obtained by reducing the image to screen resolution during scanning, even if no editing needs to be done, since the image resizing algorithms of the editing software are generally better than those of the scanner hardware. The original high-resolution edited images will then be available for future editing and resizing to accommodate future higher-resolution displays.

Images scanned as suggested above (and video and sound files, as the authors point out) can be very large (about 85 MB for an 8" x 10" image scanned at 600 dpi and 24-bit color). The best solution for short-term storage ( less than six months) of these files are Zip (100 MB) and Jaz (1 GB) drives, which are probably the most common type of large capacity, removable medium drives found on the machines of collaborators and the target audience.

Longer-term storage is most economical with tape, although CDRs (writable CD-ROM) disks or CD-RW (rewritable CDROM) disks are more reliable and allow much faster access to the information. CD-ROMs (in this case, CD-Rs) are also the only viable alternative to the distribution of multimedia presentations, which if they contain video and sound will be tens of megabytes in size: virtually everyone has a CD-ROM drive on his or her computer (although many of these drives cannot read CD-RWs hence the recommendation for CD-Rs. which most older drives can read). Digital Video Disks (DVD) will not be an alternative for some time. DVD drives are now very uncommon, and while the "older" DVD-I drives read existing CD-ROMs, DVD- I machines that can read CD-Rs appeared only recently. This summer, rewritable DVD disks should appear, but DVD-I and DVD-II drives will not be able to read them. In any case, it will probably be at least another year from next summer before a significant number of people own any type of DVD drive.

The authors used Multimedia Toolbook to create their presentations. Scripted applications are indispensable for providing the user the degree of control over the flow of information that is needed for an optimal instructional environment (e.g., one's perusal of a textbook is not strictly linear-why should a multimedia instructional module be less flexible?). The authors. however, describe Toolbook as an application that is easy to use. and requires little or no prior programming experience. Toolbook, as any other scripted application, is not particularly easy to use. The syntax of its scripting language is indeed much like English, but a new grammar must be learned, and the syntax is totally unforgiving. Be prepared to spend some time, with frequent references to third-party manuals, learning the scripting nuances required to get all but the simplest routines to work.