|
In a way, the they have done that.
If you look at the first epub 3 specifications that added lots of crud like "epub:switch", and "multiple refines on refines", they have now deprecated all of that and have decided to instead focus on things supported by modern webkit/blink/browser engines. And therefore they have stopped creating differences from what the normal html (whatwg spec) allows.
That means more support not less in modern readers that typically use browser engines. They also now look at usage levels by current e-readers and publishers in their decision to deprecate little used features.
If you look at the whatwg spec, you see a living spec that is evolving but browsers engines still support all of those old features (back to html 3.2 and in some cases before such as font tags, etc).
So as long as your epub content can be read in a browser engine, it will not ever be fully lost.
Proper document archiving is a different beast entirely. That requires archiving both the operating system and application software for reading it for the long term and keeping up virtual machines to host it all. But since xhtml/html is just marked up text for the most part, almost anything will be able to extract content from it.
Everyone is looking for an ebook format that will run on all reading devices forever but that is just not going to happen. Many e-readers have never updated their firmware so there is just a hodge page of epubs with workaround kludges for those devices. And Kindle pushing its own proprietary format just makes things worse, not better.
So creating a decent epub3 version epub (without kludges for Kindle and broken devices) as the base version and archiving that is probably the best bet for archival purposes and future proofing given the ease that xhtml can be manipulated with simple software.
|