Obviously improve our existing importers and exporters. This is especially true for the OpenOffice and RTF filters.

Add support for more filters. This is especially important for the MS Office 11 XML format, as it stands to be a major player in the market.

Better clipboard support. We want every imp/exp to be clipboard capable.

Honesty parameter. Basically, a rating of how true the imported format represents/translates to ABW

Specify a list of mime types during construction, query them later

Pass arguments to imp/exp using some mechanism (CSS like, maybe?)

Use LibGSF exclusively in all Imp/Exp, so that all files and transports look equal; so that we handle more transport types, and so that we can handle MSWord, OpenOffice, and WordPerfect files on all of our platforms. Transport types include: Memory, Disk IO, VFS, WebDav, IStreams, Bonobo Components, ...

See also: Import/Export

Edit->Paste Special, and then choose one of the available formats

Remove Edit->Paste Unformatted after Paste Special is implemented

Order clipboard formats when pasting according to honesty (unless otherwise overridden by Paste Special)

More sophisticated (and more XP) clipboard handling, including a better API for retrieving data types, and a way to lazy-populate clipboard contents (while copying/cutting)

Fault tolerant clipboard. As we support more file formats (like HTML), the odds of our encountering a problem while pasting go up. Ideally, we'd like to just suck down the next available format and try again.

Enable multi doc-range clipboard - mainly for table columns and non-contiguous selections

We'll need the things above from Clipboard and Import/Export. We'll want to be able to drag+drop any format that we can import. We'll want to use GSF so that we don't have to hit a temporary file here. This will essentially be 100% complete when the GSF and mime work above is done.

Use Enchant exclusively on all platforms. Link here.

Clean up all known bugs

Correct speed regressions

Enable Paragraph borders and shading, stealing code from Gnumeric where possible

Images as backgrounds for things like [cells, tables, paragraphs, sections]

Text frames

Clean up mess left behind by 2.0

Feature polish; continued improvement and development

Remove all encoding-less strings, except for input/output translation.

Deprecate UT_String. Replace with UTF8 strings.

All internal encodings are UCS4 or UTF8

Refactor ev, xap, util as necessary - especially menu and toolbar code (wrt icons)

Cut down on code duplication

Remove duplicate map/hash classes.

Make things easier for Criawips to use

Further improvement; bring all dialog documentation up to 100% coverage

Better integration with native help browsers

Come up with a sane C API/ABI for a lot of this stuff. Incrementally make plugins use it. Plan on using a lot of opaque structures and C "namespaces". Mainly want to abstract away the more useful PD_Document, FV_View, XAP_App, and XAP_Frame bits

Enable on-demand loading of plugins. - Possibly look at the framework Gnumeric is using.

Use above C API/ABI to expose ourselves to the outside world. Build C API as needed, not all at once. Project: AbiCapi

PERL, Python, Scheme, ... using SWIG. This will be complete once the AbiCapi work has been done.

Re-organize how menu and toolbar icons are done

Unicows/NT UCS2 string support, including keyboard/input and output (i.e. dialogs)

Finish removing the last of the GTK+ deprecated stuff

More HIG work

Use LibEGG stuff where applicable (Menus, Toolbars, File dialog)

Icon theme integration (export our toolbar icons' names)

Comply with the recent files Freedesktop.org standard

Will have a working OSX port by 2.2

Maintain feature parity with Win32 and UNIX ports

Unfortunately, a lost cause. Unless Jun makes progress in leaps-and-bounds, remove this code.

Remove dead and unused modules

Restructure how plugins, docu, abidistfiles, etc... are done

Remove screenshots and other worthless binary garbage from CVS