On Mon, May 12, 2008 at 6:43 PM, Bo Peng <[EMAIL PROTECTED]> wrote: > >> In that sense, the > > > reversibility problem is yours, not mine. > > > > > I'm perplexed about this, since I don't have any such problem.
Let me try it the last time. My design goal is to 'create a lyx format with embedded files that can be opened directly by anyone on any OS, and produce identical outputs'. My proposed approach achieves it. I do not have to provide an 'unbundle', namely 'extract all embedded files' feature. As a matter of fact, because it is far less useful than inset-level features such as 'un-embed', I have no plan to add such a feature before 1.6.0 is released. The reason why I can do this is because my approach is not intrusive. Thanks to 'individual embedding', users can continue to use external (possibly out of tree) files, or embed these files. With help from inset-level operations such as 'un-embed' or 'edit', there is no strong need to extract all embedded files. Your case is quite different. Because there are major inconveniences of your bundled mode, such as no support for external files, you have to provide both bundled and unbundled modes, and have to face the reversibility problem. I have demonstrated the non-reversibility problem (inconvenience or feature, however you want to call it) of your approach using a few working scenarios. You can shrug them off (quote: "I do not have any such problem"), but as long as we have both presented our cases, this thread has served its purpose so we do not have to continue this debate. Cheers, Bo
