Spaced-repetition flashcards for Org Mode https://git.cjennings.net/org-drill/atom?h=fork 2011-05-20T12:10:34+00:00 2011-05-20T12:10:34+00:00 eeeickythump devnull@localhost 2011-05-20T12:10:34+00:00 urn:sha1:1c77d6cf3ace2de1e9c6b7b6f53ec2a93935ee6a 2011-05-20T11:22:18+00:00 eeeickythump devnull@localhost 2011-05-20T11:22:18+00:00 urn:sha1:b17bf579bd1ab9b3efc9c19a26e3efc59845fce3 - 'conjugate' card type can now show a hint (VERB_INFINITIVE_HINT) - New command: org-drill-again: run another drill session using leftover items from the last session, i.e. don't rescan the item collection. - If org-drill-resume is called and there is nothing to resume, offer to call org-drill-again - Automatically prompt to save all modified buffers when a drill session finishes (turn off with 'org-drill-save-buffers-after-drill-sessions-p') - Politely skip cards with unknown card types - When suspending a drill session, print a message describing the key that runs org-drill-resume (it it's bound to a key) - org-drill-merge-buffers: can now optionally be prevented from copying unmatched items from SRC into DEST - org-drill-merge-buffers: fixed a bug where it prompted for certain property values if they were not set in the item being processed - org-drill-merge-buffers: ensure that *all* items in DEST are stripped of their scheduling data 2011-05-11T22:36:28+00:00 eeeickythump devnull@localhost 2011-05-11T22:36:28+00:00 urn:sha1:65e12900233c3665f952653df892cb003b11091a (ie where items are sourced from). The variable can take the value 'directory', meaning items will be sourced from all org files in the same directory as the current file. This is useful for splitting up a monolithic item collection into smaller files, while still including all those files in drill sessions. The values of the SCOPE argument to org-drill have changed slightly. Possible values are the same as those for 'org-drill-scope'. - 'file' now means current file/buffer, respecting restrictions (default, same as old 'nil') - 'file-no-restriction' means current file, ignoring (overriding) restrictions (same as old 'file') - new value 'directory' (see above) - other values have the same meaning as before. 2011-05-10T04:52:23+00:00 eeeickythump devnull@localhost 2011-05-10T04:52:23+00:00 urn:sha1:0aeff8516d30ce8b29865db8ce4c40803157d75d various clever tricks such as 'synching' the item collections of two people. At the beginning of a drill session, IDs are assigned automatically to all drill items that do not possess them. This is slow if you have a large collection, but it only happens once. - New command 'org-drill-merge-buffers'. Called from buffer A, and given buffer B, imports all the user-specific scheduling data from B into A, overwriting any such information in A. Matching items are identified by their ID. Any items in B that do not exist in A are copied to A. A scenario where this could be useful: * Tim decides to learn Swedish using an item collection (org file) made publically available by Jane. (Before publishing it Jane used 'org-drill-strip-all-data' to remove her personal scheduling data from the collection.) A few weeks later, Jane updates her collection, adding new items and revising some old ones. Tim downloads the new collection and imports his progress from his copy of the old collection, using 'org-drill-merge-buffers'. He can then discard his old copy. Any items HE added to HIS copy of the old collection will not be lost -- they will be appended to his copy of the new collection. - Instead of overdue items being reviewed in a completely random order, they are now ordered by the number of days overdue, so that the most overdue items are seen first. When two items are the same number of days overdue, then the order is random. - slightly adjusted how 'random noise' is applied to intervals, to give wider spread - we now use the port of the Common Lisp random number generator, in cl.el, instead of emacs' builtin RNG - Random number generator is now reseeded using system time at the beginning of each drill session. - Hints inside clozed text areas are now invisible during drill sessions if the clozed text is not itself being hidden, ie if your card contains [Moscow|Russian city] you will only see [Moscow] in the answer. - The '...' is now shown after the hint text rather than before it, i.e. '[Russian city...]'. You can override this by actually including '...' in the hint itself. - The minibuffer prompt now displays the card 'type' for testing purposes, as a single letter: N=new, Y=young, o=old, !=overdue, F=failed - New card type: hide2cloze (hides exactly 2 randomly chosen areas of clozed text) 2011-04-30T04:47:42+00:00 eeeickythump devnull@localhost 2011-04-30T04:47:42+00:00 urn:sha1:c27d9035ef5ba5120157649cbfc8e0587ab4a08c - 'conjugate': retrieves properties VERB_INFINITIVE and VERB_TRANSLATION from parent item, and uses its own property VERB_TENSE to prompt the student 'Translate the verb INFINITIVE and conjugate for the TENSE tense' or 'Give the verb meaning TRANSLATION and conjugate for the TENSE tense' - 'translate_number': using third party library spell-number.el, prompt the student to translate a random number to or from a non-English language (the library can handle numerous languages) - examples of both in spanish.org - org-drill-card-type-alist can now take a second function name, for controlling how the ANSWER is displayed - items can have weights (DRILL_CARD_WEIGHT). The interval is divided by the weight when scheduling, so eg an item with a weight of 2.0 will be tested twice as often as a normal item. - New command: org-drill-tree. Same as org-drill using 'tree' argument. - New command: org-drill-strip-data: deletes all scheduling data from every item in scope. Intended for use if you wish to share your item library with someone else. - Fixed bug in simple8 algorithm where items failed on their first review were not having intervals reset - Ensure all markers are freed before starting a new drill session. 2011-04-22T03:41:08+00:00 eeeickythump devnull@localhost 2011-04-22T03:41:08+00:00 urn:sha1:97aad887284ac4d9590e31f843d4f4f11393e9be still recognised as a synonym, for backwards compatibility) - new card type 'show1cloze' -- like hide1cloze, but only reveals one of the areas of clozed text in the item. - when 'revealing' an item during a drill session, do not show contents of any subheadings which are themselves tagged as drill items. This means you can have drill items inside drill items, without giving their answers away when you review the parent item. - better ensure visibility state of the org buffer is restored after the drill session. - syntax highlighting of [clozed text] now works if other faces are applied inside the clozed text area, eg [the *largest* city] 2011-04-14T20:46:35+00:00 eeeickythump devnull@localhost 2011-04-14T20:46:35+00:00 urn:sha1:e698616493d45db983fc1d3054790b71536450a4 some items have been moved or deleted. We now check for this and discard offending markers, rather than failing outright. - Final report now shows number of items that will become due tomorrow. - Fixed a rounding bug which affected deciding whether items are 'overdue'. 2011-04-12T23:31:30+00:00 eeeickythump devnull@localhost 2011-04-12T23:31:30+00:00 urn:sha1:8b72dfbded08af1444a795c379d4bd588b2a8a3a Use 'M-x org-drill-resume'. - Finessed the prioritisation of items during drill sessions. - We now present failed and OVERDUE items first. Overdue status is defined in days and has some leeway, defined by a new variable 'org-drill-overdue-interval-factor'. - Next we present YOUNG items (learned or relearned recently, so have short intervals). The definition of 'young' is via the variable 'org-drill-days-before-old'. - Finally we present a random mix of OLD items and NEW items. - Final report gives info about counts of pending overdue, young, and old items. - Timer shows '++:++' if > 1 hour passes while reviewing an item. 2011-04-08T04:22:13+00:00 eeeickythump devnull@localhost 2011-04-08T04:22:13+00:00 urn:sha1:a222b4dfb486991960e9c12e3df90d518c81801a - Item drill data is stored in a clearer format, each value has its own property, the old 'org-learn' LEARN_DATA property has been dropped. Items that contain LEARN_DATA are silently rolled over to the new format. - For SM5, matrix of optimal factors is now correctly *global* rather than per-item, and is saved between drill sessions as an emacs customization variable. - We store interval data as floating point numbers rather than rounding to the nearest integer. - New experimental algorithm: 'simple8', based on SuperMemo SM8. - New options: - org-drill-forgetting-index: define the percentage of items that it is "acceptable" to fail. A warning message is shown after the drill session if more items than this were failed. - org-drill-learn-fraction: governs how fast time intervals increase with each successful repetition of an item. - org-drill-adjust-intervals-for-early-and-late-repetitions-p: if true, alter next scheduled date if an item is reviewed earlier or later than it was supposed to be. - Almost all variables have been declared safe for file-local reassignment. - We now ignore org-learn-always-reschedule, regarding it as always true. - Post-session report now shows a count of dormant items (those not yet due for review). 2011-04-03T22:21:12+00:00 eeeickythump devnull@localhost 2011-04-03T22:21:12+00:00 urn:sha1:9ab44310d5d5e4d07e07bb83df3c6c994201afe1 invisible while each item is being tested. - OF matrix entries are rounded to 3 decimal places. - Item E-factors are not modified if the item is "failed" (in accordance with Dr Wozniak's description of the SM5 algorithm).