I suspect there is room in this world for both open-ended and close-ended learning systems depending on individual circumstances.
I want to raise an issue related to the construction of OELEs and tools that support learning. I take it that if one is to decide one's own goals and methods, one still does not want to progress any more slowly than necessary. Thus learning tools become important. At my university a group of researchers have been working on a tool for learning the Japanese writing system. Here, gentle reader, I must beg your indulgence for using Japanese as an example yet once again.
What they are building is a camera attached to a computer which you can point at some piece of writing that you can't read. (If you don't know, Japanese is very hard to read and you can't just look things up in a dictionary very easily.) You are walking down the street and you see a movie poster which attracts your attention. You wonder what it says. You point the camera at the poster and capture the image. The image is digitized and an OCR program converts it to text. The text is rendered into English (or whatever) and you try to make sense of the machine translation. The words are stored for future reference. The words are retrieved from a concordance and displayed in various contexts, etc. A reasonable little open-ended learning tool.
The issue I want to raise is that most of the work involved in producing such a tool is beyond the education of most people in (so-called) IT. They simply don't know enough about computers, and digital images, and OCR, and so on. I suspect that the really (intellectually) interesting parts of developing OELEs is going to involve building tools that support learning in ways similar to the above.
Problem is, the real work will be done outside of IT. The WWW and Mosaic really should have been designed by IT people but instead they were done by some nerdy computer guys, not cool IT folks like us.