Scott wrote:
Bowerbird wrote:
you've got some slippery thinking here about the "id" markers. do they already exist in the text? how did they get there? how did the database, which links id markers and annotations, come into existence? how are annotations shared with others? can annotations be made on annotations? what about things like graphics and movies -- how can they be utilized as annotations?
I don't have time to debate Bowerbird on these points, but if the original poster or anyone else who is actually going to work on annotation has serious questions, I'll be happy to throw in my $0.02 if I notice the thread.
Most of the time one doesn't have to author an actual implementation to determine whether it will be hard or not. Most experienced and even inexperienced programmers instinctively know the difficulty of most proposed applications. It's like saying "if I go to the roof of a tall building and toss a bowling ball off the side, it will begin accelerating to the ground." It's obvious what will happen --there's no need to even waste the time and run the experiment. Some aspects of the current discussion about external annotation of digital texts is similar. If one wants to implement some system, planning and forethought are needed *before* writing lines of code. For example, does XML confer benefits in the annotation system over plain text, or vice-versa? Obviously, Bowerbird wants everyone in the digital text arena to embrace regularized digital text, and he is (apparently) building a set of working applications (he calls them "tools") to prove his point. All the power to him. But we certainly have the right to bring up the requirements issues and call into question whether regularized plain text is sufficient for all uses and needs of the digital text universe. There are those among us, including yours truly, who believe that XML should form the core of digital publishing processes and formats. Even if Bowerbird implements his system, we will ask: "will it do this and do that?" We *know* that XML and its many associated W3C and IETF specifications and RFCs confers a powerful and sufficient foundation to do all the myriad things proposed for the digital publishing universe (that I know of at least, and I've looked at a *lot* of advanced uses.) And we don't have to write code to *know* this as true (see bowling ball discussion above.) Those who listen to Garrison Keillor's "Prairie Home Companion" are familiar with the fictional "Ralph's Pretty Good Grocery" (RPGG) in Lake Wobegon. Their advertising slogan is "If you can't get it at Ralph's, you can probably get along without it." Bowerbird's system is clearly a RPGG since I know it will NOT do everything that has been discussed for digital texts. Whether it will hit the sweet spot and win the hearts and minds of the ebook masses (which must include all the important stakeholders in the digital publishing universe), who will overlook its deficiencies, remains to be seen. I'm skeptical, but will wait to see what arrives. Jon