| blob | a939785246c5f0b616a34ccfb30b84b5bbf5668f |
5 <head>
11 <link rel=stylesheet href="freeshell2.css" type="text/css"> <style type="text/css"> .tag { color: red; font-weight:bold}</style>
12 </head><body>
15 <ul>
17 <ul>
19 </ul>
20 </li>
22 <ul>
32 </ul>
33 </li>
35 <ul>
37 </ul>
38 </li>
39 </ul>
44 <p>
45 This is a loose collection of ideas and TODO items for the future
46 development of Org-mode. These ideas come from various sources,
47 mostly from emacs-orgmode@gnu.org, from direct emails to me, or from
48 my own day-dreaming. I don't always mention the source of an idea,
49 out of laziness. However, when I implement a good idea, I try to
52 something.
53 </p>
57 <p>On this page, I am using TODO keywords in the following way:
60 <thead>
62 </thead>
63 <tbody>
64 <tr><td><b>TODO</b></td><td>A development that is going to happen, once I have time or once I figure out how to do it.</td></tr>
65 <tr><td><b>IDEA</b></td><td>A new idea, I have not yet decided what if anything I will do about it.</td></tr>
66 <tr><td><b>WISH</b></td><td>A wish, probably voiced by someone on emacs-orgmode@gnu.org. This is less than a new idea, more a change in existing behavior.</td></tr>
67 <tr><td><b>QUESTION</b></td><td>A question someone asked, that needs some thinking before it can be answered</td></tr>
68 <tr><td><b>DECLINED</b></td><td>I have decided not to implement this feature, but I am keeping it in the list so that people can see it, complain, or still try to convince me.</td></tr>
69 <tr><td><b>INCONSISTENCY</b></td><td>Some behavior in Org-mode that is not as clean and consistent as I would like it to be.</td></tr>
73 </tbody>
74 </table>
84 <ul>
86 </li>
87 <li><span class="todo">TODO</span> Get rid of all the \r instances, which were used only for XEmacs.<br/>
88 </li>
89 <li><span class="todo">IDEA</span> Should we allow #+TODO as an alias for #+SEQ<sub>TODO</sub>?<br/>
90 </li>
92 Make them not hide the text after the final list item.
93 This is not trivial, we cannot usenormal outline stuff,
94 needs a separate implementaiton.
95 </li>
97 </li>
99 Is there a problem with the overriding map?
101 </li>
103 A way to put a TODO entry without starting a new section.
104 </li>
105 </ul>
109 <ul>
111 is this safe, or could things be messed up with this?
112 </li>
114 Right now they change the time stamps. This could be
115 M-left/right, but this could also cause inconsistencies, because
116 when on a time stamp, the same keys do exactly this.
117 </li>
119 These could be made available for toggling, just like
120 follow-mode. Examples:
121 <ul>
122 <li>
123 org-agenda-todo-list-sublevels
124 </li>
125 <li>
126 org-tags-match-list-sublevels
127 </li>
128 <li>
129 org-agenda-todo-ignore-scheduled
130 </li>
131 </ul></li>
133 This would apply for the list collection, not in the day entries.
134 However, I could also have a TODO keyword criterion that could be
135 used in the day entries, as one of the minor criteria.
137 </li>
139 This feature is not going to come, because block agendas can
140 achieve the same thing.
141 </li>
142 </ul>
146 <ul>
148 Not sure why this is needed.
149 Idea from William Henney.
150 </li>
152 Or something like that, to make standard links fast.
153 </li>
155 Would require a display property also for the label part of
156 links.
157 </li>
159 </li>
161 Info links of course only work inside Emacs. However, many info
162 documents are on the web, so the HTML exporter could try to be
163 smart and convert an Info link into the corresponding link on the
164 web. For example, we could use the GNU software site then
165 Name.HTML. Here is the link to be used:
166 <a href="http://www.gnu.org/software/emacs/manual/html_node/">http://www.gnu.org/software/emacs/manual/html_node/</a> Another
167 question is, is this URL going to be stable so that it makes sense
168 to actually put this into org.el?
170 </li>
172 I can use Google groups with a message id to find a USENET message,
173 even if the original link points to gnus.
175 </li>
176 <li><span class="todo">TODO</span> Remove irretrievable links from <i>published</i> HTML output<br/>
177 This is on David's table, and he will hopefully integrate a
178 mechanism for this into org-publish.el. The discussion about this
181 </li>
183 <a href="http://thread.gmane.org/gmane.emacs.orgmode/139">Tim Callaghan</a> started the discussion on this one.
184 Declined because this would depend on local variables and the
185 agenda commands are global. Also, this can actually be done by
186 specifying the file list in a custom agenda command.
188 </li>
189 <li><span class="todo">DECLINED</span> Make CamelCase words link to corresponding org-mode files.<br/>
190 Files in the same directory. Or some other way to easy do that.
191 Would be useful for publishing projects. This would be an
192 incompatible change, but I don't think anyone is using CamelCase
193 anyway?
194 This is marked declined, because link abbreviations now provide an
195 easy way to make links to other files.
197 </li>
199 I don't think this is really covered anywhere.
200 Maybe we also should protect characters in the visible part, to
201 make sure thing will never be on two lines...?
202 </li>
204 I guess each org file could write a .orgtargets.filename file, if
205 it has any radio targets.
206 </li>
207 </ul>
211 <p>Could I use a dynamic block for this?
212 </p>
216 <ul>
218 @4=.....
219 </li>
221 Demonstrate running averages.
223 </li>
224 <li><span class="todo">WISH</span> Make a variable that current line should be recomputed always<br/>
225 in each table, skipping headers of course.
227 </li>
229 This is for using constants.el.
230 Well, it is now possible to do this for the file, is this enough???
231 </li>
233 </li>
235 </li>
237 Computing changes fields and does not yet see correctly if the column
238 width has changed, in the case of a narrowed column.
240 </li>
241 <li><span class="todo">DECLINED</span> Alternative for the implementation of orgtbl-minor-mode:<br/>
242 I could use post-command-hook to set the variable orgtbl-mode.
243 I will not do this now and only consider it if problems show up.
245 </li>
246 <li><span class="todo">DECLINED</span> Table info lines are not necessarily comments in orgtbl-mode<br/>
247 Should I generalize this? No, because the table itself will not be
248 in the correct syntax for whatever mode.
250 </li>
251 <li><span class="todo">DECLINED</span> S-cursor motion to select part of a table, with proper highlighting.<br/>
252 Similar to CUA
254 </li>
255 </ul>
259 <ul>
261 This could/should be part of org-mouse.el.
262 <ul>
263 <li>
265 </li>
266 <li>
268 </li>
269 <li>
271 </li>
272 <li>
274 </li>
275 <li>
277 </li>
278 <li>
280 </li>
281 </ul></li>
282 <li><span class="todo">BUG</span> We have no clear view on what to do with properties upon export.<br/>
284 </li>
285 </ul>
289 <ul>
291 This is being phased out. Almost everything works under Emacs 21,
292 but in the future I wwill make little effort to support it.
293 <ul>
295 Declined, because I don't know how to do this. Too many
296 problems.
297 </li>
298 </ul>
299 </li>
301 <ul>
302 <li><span class="todo">QUESTION</span> Is there an issue with the coding system of HTML exported files?<br/>
303 In the code I used to have a comment saying that
304 <pre>
305 (and (fboundp 'coding-system-get)
306 (boundp 'buffer-file-coding-system)
307 buffer-file-coding-system))
308 </pre>
309 always returns nil, implicating that setting the coding system for
310 the export buffer would not work correctly. however, I have never
311 followed up on this and never had a bug report - so I am wondering
312 if there is an issue at all.
314 </li>
316 Declined, because I don't know how to do this. Too many
317 problems.
318 </li>
320 To make narrowing work under XEmacs, I would need to write a
322 would have to rework narrowing entirely. Unlikely that this will
323 happen, mainly because it is working in Emacs and so does not
324 bother me personally so much. Anyway, I don't know if people are
325 actually using narrowing very much at all.
328 </li>
329 </ul>
330 </li>
331 </ul>
335 <ul>
337 </li>
339 Declined because I don't really see the need for this.
340 </li>
342 Nic Ferrier has been pushing this agenda for a long time, for
343 example with <a href="http://thread.gmane.org/gmane.emacs.orgmode/140/">this email</a>. I am not deep enough into the semantic
344 web that I could do this myself. Maybe we can do this by
345 modifying the html exporter step-by-step?
346 </li>
348 Right now this is being done as part of `org-html-expand', which
349 does not seem logically correct.
350 </li>
352 The line-by-line processing may not be the fastest way to do
353 this. It certainly uses more resources. Right now the exporters
354 do work though, so it is unlikely that I am going to change this.
356 </li>
358 I remember this coming up several times, for example in emails
359 from <a href="http://thread.gmane.org/gmane.emacs.orgmode/156">Niels Giesen</a> and <a href="http://thread.gmane.org/gmane.emacs.orgmode/139">Tim Callaghan</a>. Org-mode does have
360 Tab-separated export of tables, so right now I don't really see
361 the benefit of adding CSV export. Are there applications that do
362 not understand tab-separated files, but do understand
363 comma-separated ones?
365 </li>
367 Time stamps have their own class now which allows to highlight
368 them etc. But I was wondering if something more useful could be
369 done with them, like a link to some kind of calendar...
373 </li>
374 </ul>
378 <ul>
380 </li>
382 </li>
384 The CLOCK lines could be there in addition, or could even be
385 removed, so that only one line is kept.
386 </li>
388 <ul>
389 <li>
390 Compute time interval on time range
391 </li>
392 <li>
393 Update CLOCK interval
394 </li>
395 </ul></li>
397 Org-mode re-binds C-a to make this command go to the beginning of
398 a visible line. There are other keys which might invoke C-a.
399 Should these keys be changed as well? one could use
400 `substitute-key-definition' on the global map to find them all.
401 </li>
403 </li>
405 </li>
407 Roland Winkler was kind enough to implement a new argument to the
408 `bibtex-url' command that allows me to retrieve the corresponding
409 URL, whether it is taken from a URL field or constructed in some
410 clever way. Currently I am not using this, because too many
411 people use an old Emacs version which does not have this.
412 however, eventually I will implement this.
413 </li>
414 <li><span class="todo">WISH</span> Get people to write articles about how to do GTD with Org-mode.<br/>
415 There is now one by Charles Cave, read it <a href="http://members.optusnet.com.au/~charles57/GTD/orgmode.html">here</a>
417 </li>
419 A key that can be applied to a tree, showing statistics:
420 <ul>
421 <li>
422 how many headlines
423 </li>
424 <li>
425 how many TODO
426 </li>
427 <li>
428 how many DONE
429 </li>
430 <li>
431 Checkboxes
432 </li>
433 <li>
434 etc....
436 </li>
437 </ul></li>
439 That could easily be done in an org-mode file. But then, maybe
440 this should really be a paper thing.
442 </li>
444 Here is some information about priorities, which is not yet
445 documented.
446 <ul>
450 i.e. it goes above top todo stuff 7 days
451 before due
453 i.e. it goes above top todo on the due day
454 TIMESTAMP is 0 i.e. always at bottom
455 but as a deadline it is 100
456 but if scheduled it is 99
457 TIMERANGE is 0 i.e. always at bottom
458 DIARY is 0 i.e. always at bottom
460 Priority * 1000
462 </pre>
463 </li>
464 </ul>
465 </li>
467 In plain lists, multiple demote/promote commands executed directly
468 after each other don't change the scope of the command - the
469 initially selected text continues to be selected. This is
470 inconsistent with the behavior of outline sections, were the subtree
471 for promotion/demotion is newly defined after each command. Which
472 convention is better? Should this be consistent between trees and
473 plain lists?
475 </li>
478 </li>
480 I am actually not sure, I might have addressed this already, but
481 my memory is failing me. Needs some checking.
483 </li>
484 <li><span class="todo">DECLINED</span> Think about Piotr's idea of treating TODO like a TAG.<br/>
485 The answer is probably NO because the simple part of TODO must
486 still be there and I like the keyword at the beginning of the line,
487 just like a checkbox. Of course you can make a TODO tag yourself
488 at any time.
490 </li>
493 </li>
495 Problem: If the keyword disappears, it will suddenly start to fold....
496 So my feeling right now is that the answer should be NO.
498 <p>
499 No, because S-left/right does this already pretty well
500 </p>
501 </li>
503 That counter shows what faction of entries has been marked DONE. I
504 am not yet sure how useful such a thing would be, because of the
505 huge number of entries that can be in a file, and the different
506 weight of such entries.
507 Declined because we do now have counters for checkboxes, and the
508 feeling is that this is not so useful for TODOs.
510 </li>
511 </ul>
519 <ul>
520 <li><span class="done">DONE</span>( <span class="timestamp">2006-06-01 Thu 09:15</span>) Commands to move through an item list<br/>
521 next item, previous item. What should happen at the boundaries of
522 the current list?
524 </li>
525 <li><span class="done">DONE</span>( <span class="timestamp">2006-06-01 Thu 09:16</span>) Force relative links, would this be useful?<br/>
528 </li>
529 <li><span class="done">DONE</span>( <span class="timestamp">2006-06-01 Thu 09:16</span>) Fix empty search string.<br/>
531 </li>
532 <li><span class="done">DONE</span>( <span class="timestamp">2006-06-01 Thu 09:16</span>) STARTUP options for logging<br/>
534 </li>
535 <li><span class="done">DONE</span>( <span class="timestamp">2006-06-01 Thu 09:16</span>) Make a variable for Pete, to have SCHEDULED at point<br/>
537 </li>
538 <li><span class="done">DONE</span>( <span class="timestamp">2006-06-01 Thu 09:16</span>) Fix ASCII export of narrowed columns<br/>
540 </li>
541 <li><span class="done">DONE</span>( <span class="timestamp">2006-06-01 Thu 09:16</span>) org-file-apps should allow regular expressions<br/>
543 </li>
544 <li><span class="done">DONE</span>( <span class="timestamp">2006-06-01 Thu 09:16</span>) fix problem with w32- versus mswindows-....<br/>
546 </li>
547 <li><span class="done">DONE</span>( <span class="timestamp">2006-06-01 Thu 09:17</span>) Document that font-lock-mode is needed<br/>
549 </li>
550 <li><span class="done">DONE</span>( <span class="timestamp">2006-06-01 Thu 09:17</span>) Add keyboard access and menu commands for org-project.el<br/>
551 C-c C-x C-p or something like that.
552 In the menu, it goes under export, or even its own group.
554 </li>
555 <li><span class="done">DONE</span>( <span class="timestamp">2006-06-01 Thu 09:17</span>) Document those new keys in orgcard.tex<br/>
556 C-c C-x C-p or something like that.
557 In the menu, it goes under export, or even its own group.
559 </li>
560 <li><span class="done">DONE</span>( <span class="timestamp">2006-06-01 Thu 09:17</span>) Remove SCHEDULED keyword when marking it done.<br/>
561 in particular when adding a CLOSED timestamp.
562 Only problem is that when unclosing it, it should be scheduled
563 again
565 </li>
566 <li><span class="done">DONE</span>( <span class="timestamp">2006-06-01 Thu 09:17</span>) Don't show tasks already scheduled for the future, only stuff which<br/>
567 is not yet scheduled.
569 </li>
570 <li><span class="done">DONE</span>( <span class="timestamp">2006-06-01 Thu 09:17</span>) Put title and tags in separate classes<br/>
572 </li>
573 <li><span class="done">DONE</span>( <span class="timestamp">2006-06-01 Thu 09:17</span>) Option to leave out TAGS from export<br/>
575 </li>
576 <li><span class="done">DONE</span>( <span class="timestamp">2006-06-01 Thu 09:17</span>) Option to leave out Timestamps from export<br/>
578 </li>
579 <li><span class="done">DONE</span>( <span class="timestamp">2006-06-01 Thu 09:17</span>) Checkable items like Frank Ruell's proposal<br/>
581 </li>
582 <li><span class="done">DONE</span>( <span class="timestamp">2006-06-01 Thu 09:17</span>) Classify Keywords and time stamps<br/>
586 </li>
587 <li><span class="done">DONE</span>( <span class="timestamp">2006-06-01 Thu 09:17</span>) Change default for include-all-todo to nil<br/>
589 </li>
590 <li><span class="done">DONE</span>( <span class="timestamp">2006-06-01 Thu 09:17</span>) Specify TAGS in a special line instead of collecting them dynamically<br/>
591 The could help to avoid typos and could be faster for very large files.
593 </li>
594 <li><span class="done">DONE</span>( <span class="timestamp">2006-06-01 Thu 09:17</span>) Some much faster interface for setting tags.<br/>
595 Basically, use single keys to add/remove tags from the list. Also
596 for this the idea to define tags in a special line would be good,
597 in order to have a way to define the shortcuts.
599 </li>
600 <li><span class="done">DONE</span>( <span class="timestamp">2006-06-01 Thu 09:17</span>) Remove date stamps from headline if they are not needed.<br/>
601 Bug report from Scott.
604 </li>
605 <li><span class="done">DONE</span>( <span class="timestamp">2006-06-01 Thu 09:17</span>) : Maybe s-up/down should change priorities only in headlines<br/>
606 Right now this works wherever the cursor is. If this is changed,
607 should S-up or S-down do something else?
609 </li>
610 <li><span class="done">DONE</span>( <span class="timestamp">2006-06-01 Thu 09:21</span>) Need a command to remove tabulators from a line.<br/>
611 Use this in ASCII export, to make sure we get the indentation
612 right.
614 </li>
615 <li><span class="done">DONE</span>( <span class="timestamp">2006-06-01 Thu 09:24</span>) Improve tab in org-cycle:<br/>
616 <ul>
617 <li>
618 there is a bug when the setting is 'white: It requires a
619 non-empty white line!
620 </li>
621 <li>
622 There should be another setting to have tab be the tabulator in
623 the white space at the beginning of the line.
624 </li>
625 <li>
626 Finally, when TAB closes an entry, maybe it should go beck to the
627 beginning of the entry??? On the other hand, it is good to be
628 able to go back to the old place with just another tab.
630 </li>
631 </ul></li>
632 <li><span class="done">DONE</span>( <span class="timestamp">2006-06-01 Thu 11:48</span>) Include TAGS into sorting.<br/>
633 But what strategy should be used when there are several tags?
635 </li>
636 <li><span class="done">DONE</span>( <span class="timestamp">2006-06-06 Tue 10:15</span>) Archiving an entry in current buffer goes to end of file<br/>
637 Should go to end of subtree.
638 This is done now, and I also control the amount of empty lines
639 created by the archiving process in general.
641 </li>
642 <li><span class="done">DONE</span>( <span class="timestamp">2006-06-06 Tue 10:16</span>) Narrowing agenda to current files does not work as advertized.<br/>
643 C-c a 1 t should make the TODO list for the current file, but does
644 not.
646 </li>
647 <li><span class="done">DONE</span>( <span class="timestamp">2006-06-06 Tue 10:17</span>) Radio TAGS for the fast tag interface.<br/>
648 Make lists of tags that are mutually exclusive. So when I turn on
649 @HOME, @WORK would be turned off automagically. A good syntax for
650 defining these groups in the #+TAGS line would be:
651 <pre>
652 #+TAGS: [@WORK(w) @HOME(h) @CLUB(c)] Laptop(l) PC(p) Car(r)
653 </pre>
654 This could indicate that @WORK, @HOME, @CLUB are mutually exclusive.
656 <p>
657 If I do this, I will also need a syntax for the global alist to
658 indicate the same thing.
659 </p>
660 <p>
661 It seems to me that with such grouping of tags, sorting would be
662 useful as it would improve the overview over the current tags. I
663 might even be able to support mutually exclusive tags with
664 fontification in the interface. Or arrange things such that the
665 mutually exclusive ones are all in the same row or column, to
666 optimize the visual feedback.
667 </p>
668 <p>
669 For the internal format, I think best would be something like
670 </p>
671 <p>
672 <pre>
675 </pre>
676 This setup makes sure that assoc and rassoc still do work as
677 expected.
678 </p>
679 </li>
680 <li><span class="done">DONE</span>( <span class="timestamp">2006-06-06 Tue 10:17</span>) Should tags be sorted in a certain way?<br/>
681 Right now, you can either set the tag sequence yourself through
682 the completion interface. If you use the fast tag selection
683 interface, the tag sequence depends on the order in which tags are
684 selected and deselected. maybe a specific roder would be useful
685 for example the same order as the one given in the configuration?
687 </li>
688 <li><span class="done">DONE</span>( <span class="timestamp">2006-06-06 Tue 10:17</span>) Remove narrowing cookies for export<br/>
689 Exported tables should not show narrowing cookies. In fact, if a
690 table line does contain nothing but narrowing cookies, the entire
691 line should be removed.
693 </li>
694 <li><span class="done">DONE</span>( <span class="timestamp">2006-06-11 Sun 07:32</span>) Track working times similar to time-clock.el<br/>
695 This was proposed by David O'Toole in an
696 <a href="http://thread.gmane.org/gmane.emacs.orgmode/387">email to emacs-orgmode@gnu.org</a>. He wants to be able to know the
697 times when he worked at a particular project. Some reporting
698 possibility would be needed to make this useful. Maybe sparse
699 trees with broken-down working times?
701 </li>
702 <li><span class="done">DONE</span>( <span class="timestamp">2006-06-15 Thu 13:35</span>) Make org-store-link do better in image-mode buffers<br/>
703 Just link to the file.
705 </li>
706 <li><span class="done">DONE</span>( <span class="timestamp">2006-06-15 Thu 13:35</span>) Use a ported version of noutline.el<br/>
707 Programming the old outline-mode is really a nightmare - the new
708 outline mode that uses invisibility properties instead of
709 selective display works really much much better. There is now
710 an <a href="http://thread.gmane.org/gmane.emacs.orgmode/214/focus=218">experimental port</a> by Greg Chernov, so when I find time I will
711 try if this works well. If yes, this port should become part of
712 XEmacs. Once that happens, I could remove a large amount of
713 ballast from org.el
715 </li>
716 <li><span class="done">DONE</span>( <span class="timestamp">2006-06-15 Thu 13:35</span>) Verify links during export<br/>
718 </li>
719 <li><span class="done">DONE</span>( <span class="timestamp">2006-09-25 Mon 16:11</span>) Allow counting checkboxes<br/>
720 Then some statistics should be displayed. One idea is for
721 checkboxes, to use a special box [%] which will be updated with
722 percent of done. I could have an alternative [/] that will be
724 two numbers are equal, I could highlight then in DONE face, just
725 like [100%]. All smaller numbers could be highlighted even in
726 TODO face if I wanted. Hmmm, I am beginning to like this.
728 <p>
729 Then: how to update this? Each time a checkbox is added or
730 toggled, go up and update all the counts, right up to something
731 which is not an item, or up to the previous header line.
732 </p>
733 <p>
734 Maybe I should also make an option for turning this on, to avoid
735 slowdown. I guess for long lists this could be slow.
736 </p>
737 <p>
738 Also would need a command for global update.
739 </p>
740 <p>
741 An maybe plain list and checkboxes should get their own chapter?
742 In Structure they seem to be a bit hidden.....
743 </p>
744 <p>
745 Also a menu sublist...
746 </p>
747 </li>
748 <li><span class="done">DONE</span>( <span class="timestamp">2006-09-25 Mon 16:11</span>) Allow stacking calls to org-occur<br/>
749 This was a proposal from Piotr. It is now possible, the second
750 and subsequent calls to org-occur need a prefix argument to keep
751 the previous highlights.
753 </li>
754 <li><span class="done">DONE</span>( <span class="timestamp">2006-09-25 Mon 16:12</span>) Create a hook to be run after an agenda buffer has been finalized.<br/>
756 </li>
757 <li><span class="done">DONE</span>( <span class="timestamp">2006-09-25 Mon 16:12</span>) Include TAGS into sorting.<br/>
758 But what strategy should be used when there are several tags?
760 </li>
761 <li><span class="done">DONE</span>( <span class="timestamp">2006-09-25 Mon 16:12</span>) Fixup tag display in agenda buffer.<br/>
762 The tabs mess up everything. Either remove the tabs, or realign
763 the tags to some useful column.
765 </li>
766 <li><span class="done">DONE</span>( <span class="timestamp">2006-09-25 Mon 16:12</span>) Blocks in agenda:<br/>
767 Make a single agenda buffer that contains several sets of items,
768 like the TODO list, a tags list, and a day view. Not a bad idea.
771 </li>
772 <li><span class="done">DONE</span>( <span class="timestamp">2006-09-25 Mon 16:12</span>) Make org-store-link do the right thing in dired-mode<br/>
774 </li>
775 <li><span class="done">DONE</span>( <span class="timestamp">2006-11-02 Thu 10:24</span>) Control over windows being used for agenda and related buffers <span class="tag">NEW</span><br/>
777 </li>
778 <li><span class="done">DONE</span>( <span class="timestamp">2007-01-24 Wed 12:59</span>) Extend access to table fields<br/>
779 Find a general way to get any rectangular region into a calc
780 vector
782 </li>
783 <li><span class="done">DONE</span>( <span class="timestamp">2007-01-24 Wed 12:59</span>) Allow links to remote images to be inlined.<br/>
785 </li>
786 <li><span class="done">DONE</span>( <span class="timestamp">2007-01-24 Wed 12:59</span>) M-TAB on an already complete tag should add the ":"<br/>
788 </li>
789 <li><span class="done">DONE</span>( <span class="timestamp">2007-01-24 Wed 13:07</span>) Need a command to do show hierarchy after isearch.<br/>
790 Maybe automatically, using isearch-mode-end-hook. But this does
791 not seem to work, because to is called before the overlays are
792 re-installed. I can create a new hook, isearch-exit-hook, run in
793 `isearch-exit'.
795 </li>
796 <li><span class="done">DONE</span>( <span class="timestamp">2007-02-18 Sun 07:16</span>) hook for users to do their own agenda lists<br/>
797 New function dumping headline into the agenda buffer, based on regexp
798 and callback, to allow general stuff. For the special tags, we can
799 search for one and check for the others using the callback.
801 </li>
802 <li><span class="done">DONE</span>( <span class="timestamp">2007-02-18 Sun 07:37</span>) Make C-c C-l also work on a plain link, converting it into bracket.<br/>
804 </li>
805 <li><span class="done">DONE</span>( <span class="timestamp">2007-02-18 Sun 07:37</span>) GNUS <i><gnus:mail.general#123></i> stuff should be "Email from:..."<br/>
807 </li>
808 <li><span class="done">DONE</span>( <span class="timestamp">2007-02-18 Sun 07:38</span>) Email/message context: What format?<br/>
809 I can think of two good formats for this:
810 <pre>
811 Email from John Smith (if I an the receiver)
812 Email to John Smith (if I am the author)
813 </pre>
814 or
815 <pre>
816 John Smith on: Some subject (this is the current Org-mode default.)
817 </pre>
818 The first format requires better parsing of the messages (to get
819 both author and receiver names), and also
820 some way to detect if I am the author of this email or not.
822 </li>
823 <li><span class="done">DONE</span>( <span class="timestamp">2007-02-21 Wed 12:04</span>) Allow internal links to match inside <b>other</b> link<br/>
824 Only the link itself that actually triggered the search.
826 </li>
827 <li><span class="done">DONE</span>( <span class="timestamp">2007-02-21 Wed 12:07</span>) Allow fully protected HTML code that will be exported as is<br/>
828 Currently, <...> works only for simple tags, and I think we
831 each match in the run preparing the export.
832 I guess I could require these at the beginning of the line, like
836 <p>
837 or something like that.......
838 </p>
839 </li>
840 <li><span class="done">DONE</span>( <span class="timestamp">2007-03-13 Tue 05:19</span>) Make it possible to set TOC levels independently of headline levels<br/>
841 The easiest way to do this would obviously be to allow a number
842 setting for the tco switch, not only a simple t/nil switch. This
843 can be nice for webpages: One could have only top-level or level
845 more complex structure below.
847 </li>
848 <li><span class="done">DONE</span>( <span class="timestamp">2007-03-13 Tue 05:20</span>) Filling is not done in a compatible way.<br/>
849 The current setup used the Emacs variables and machine to get the
850 filling of plain lists and everything lese right. XEmacs uses the
851 filladapt package, which has different ways to do the same
852 things. org.el should be smart about this and setup filling under
853 XEmacs correctly.
855 </li>
856 <li><span class="done">DONE</span>( <span class="timestamp">2007-06-06 Wed 17:44</span>) Repeating Schedules?<br/>
857 <a href="http://thread.gmane.org/gmane.emacs.orgmode/149">Dieter Grollman</a> requested this feature, but for now I have decided
858 not to do anything about it. I don't see a good way to implement
859 this, and I believe that cyclic diary entries are good enough for
860 such tasks.
862 </li>
863 <li><span class="done">DONE</span>( <span class="timestamp">2007-06-14 Thu 13:16</span>) improve item indentation when promoting and demoting<br/>
865 </li>
866 <li><span class="done">DONE</span>( <span class="timestamp">2007-06-14 Thu 13:17</span>) The TODO buffer lists possible TODO keywords<br/>
867 These are taken from the current buffer, so when multiple buffers
868 are involved, this list may not be correct.....
870 </li>
871 <li><span class="done">DONE</span>( <span class="timestamp">2007-06-14 Thu 13:20</span>) Conflict between bold emphasis and headlines<br/>
873 headline or an emphasis. This will not be easy to fix. The way
874 to do ii is to require a space after the star in headlines. This
875 requires to change outline-regexp, outline-level, the many many
876 times when outline regexp is used in the program, and also the
877 countless times when I am simply matching th stars directly.
879 </li>
881 <span class="timestamp-kwd">ARCHIVED: </span> <span class="timestamp">2007-07-02 Mon</span><br/>
883 </li>
885 <span class="timestamp-kwd">ARCHIVED: </span> <span class="timestamp">2007-07-02 Mon</span><br/>
886 same or additional indentation????
888 </li>
889 <li><span class="done">DONE</span> Make it easier to do monthly agendas, and agendas for ranges<br/>
890 <span class="timestamp-kwd">ARCHIVED: </span> <span class="timestamp">2007-07-02 Mon</span><br/>
892 </li>
894 <span class="timestamp-kwd">ARCHIVED: </span> <span class="timestamp">2007-07-02 Mon</span><br/>
895 similar to DEADLINE and SCHEDULED
896 Idea from Bastien
897 This could be done now using properties, I don't think we need a
898 special thing for this anymore.
900 </li>
902 <span class="timestamp-kwd">ARCHIVED: </span> <span class="timestamp">2007-07-02 Mon</span><br/>
903 Even when org-agenda-remove-times-when-in-prefix is set, there
904 seem to be cases where this does not work. I forgot what the
905 example was - does anyone remember?
907 </li>
908 <li><span class="done">DONE</span> Links are still highlighted in fixed-width environments.<br/>
909 <span class="timestamp-kwd">ARCHIVED: </span> <span class="timestamp">2007-07-02 Mon</span><br/>
910 This ties in with the problem that there is no way to have
911 protected HTML code in the file.
913 </li>
914 </ul>
917 </p>
919 </body>
920 </html>