As a test, ask a friend of yours for their favorite PowerPoint presentation. Now grab your favorite PowerPoint presentation. Now, open them both in PowerPoint (but not in the same file). Present one. Immediately when you finish the first, show the second. Now, do they look like 2 presentations that were pasted together, or do they look like one continuous course?
Instead, take the content from the two presentations, put them into a single PPT file, and make sure that they are consistent in:
1. Look and feel: backgrounds, fonts, navigation, structure
2. Style: Text style, location of test questions, parallel structure of notes
3. Navigation text: This doesn't really apply to PPT, but with web sites/web courses you need additional labels and texts ("next page", "previous page", "grade the test", etc.) These need to be consistent throughout a single course, or the student will be very distracted.
Once you make sure these are all consistent, you can serve the content, and it will look smoother.
The difference here is whether you do run-time reusability or you do With eLearning if all you do is modify the style sheets at assembly time, you are not really using the SCOs unchanged.
I am a proponent of design-time or assembly-time reusability, and as you can tell, I don't like run-time assembly. More than that, I am a proponent of being able to re-use components of a SCO such as graphics, multimedia, pages, page groupings, test pages, etc. Some of these are what SCORM 2004 calls "assets".
A good tool will let you copy/paste content modules from various sources so that you can create a unified course. For example, suppose you want to number the SCOs so that the user knows at all times where they are. If you number them before assembly, and then you change the assembly order, the numbering will be incorrect. If you use content from different authors, it is important to make sure all the "stock" texts, navigation images, tables-of-contents, etc. are consistent. This is nearly impossible if you limit yourself to run-time re-usability.
In the web course process, there is the publish step. This is comparable to the print step with documents. During the publication step, you assemble your course for delivery. The way SCORM is designed, the publish step is more like grabbing multiple documents that have already been printed, and staple them together in a new order just prior to delivery. As opposed to printed documents that are linear, web courses are multidimensional with a "web" of interconnections. SCORM forbids the interconnections between SCOs. However, this does not prevent the author from giving the student clues about the interconnections between SCOs. For example, with the ReadyGo tool, if you choose to make each chapter of your course a SCO, the tool still builds a table of contents for the entire course that shows the other chapters but doesn't let you jump to them through the course. Within the chapter, the student is allowed to navigate anywhere.
If you design your SCOs so that the primary objective is re-usability, then you'll find that you're making them smaller and smaller, and more generic. You can entirely avoid the navigation issues by making each SCO a single-page object. You can also make tests so that each test is a separate SCO. (With many LMSs you must do this if you want to find out if a student understood a specific data item because they only store the score for each SCO.) However, I feel that these steps are bad for the end-user. When going from SCO to SCO, the LMS must close the current session and open a new session. The best turn-around I've seen for this is in the 5-10 second range. Typical delays are around 20 seconds. Any delay between pages of content will cause the learner to drift off the subject (and often they go check their e-mail or instant messages). So, the more granular your courses are (for re-usability) the slower the delivery and the more disjointed it will be for the learner. Now, if there are any visual disparities between the SCOs, these further disrupt the learning process.
So, yes, you can re-use SCOs at run-time. You will probably win some awards for your ability to use the SCORM, and you'll feel proud of yourself for creative re-use of content. But what about the learner's experience. Shouldn't that be the highest priority? If you alienate enough employees (with learning experiences that they dread), what's the point of re-usability?