MindReader: How to best read your mind?
MindReader turns one year old Sunday. The initial version ("Let GyroQ read your mind") only recognized 13 keywords. Its vocabulary has grown quite a bit over time. As MindReader grows up, there are a few keywords that need to be revised or added.
Projects and Sub-Projects
Earlier versions were configured to recognize "project" and "subproject" and assign ResultsManager custom icons for these, but I later realized that it would see both strings as "project". We've also subsequently seen that you often use the word "project" in a "next action" that you don't want turned into a project. The current version now uses "rmproject" and "rmresult" to avoid that issue and some have customized this instead to "isproject" and "isresult". Another possibility might be "make project" and "make result". Any recommendations or suggestions for the default configuration?
Notes and Links
The program currently uses the clipboard text as a note or hyperlink if the entered text has "note" or "link" in it. I've noticed that this sometimes leads to problems if I use either string (e.g. "buy notebook") when I'm not actually intending to add a note or link. Should the default triggers be "see note" / "see link"? Those phrases could act as reminders from dashboard that the tasks show has supporting information.
Info Only
I'm finding that while using OutLinker to transfer message links into MindManager, that any given "next action" usually has 5-10 additional messages that provide supporting reference for it . While you can send it information in as "next actions" and then consolidate later, I'm thinking that it would be better to have the option of telling MindReader not add the action-icon. What syntax would you prefer to use for that? "Info:"? "noaction"? "reference"?
Complete
Some folks may want the option sending information into maps to document actions they have completed. How would you want to convey that? "Done:"?
Due Date
This isn't really a question but an announcement. Previously you needed to add due dates within # # delimiters in order for them to be recognized (e.g. #12/25/08#) but now MindReader will look for the "/ /" and recognize 12/25/08 by itself (or 25/12/08 if that is your system's format). You can still use #12/25# or #Dec 25# but I find it easier to just type it out. Don't overuse this feature or your next action list will turn red and your NAA score will suffer :-).
MindReader is fully customizable via the mindreader.mmap (see Mapping the MindReader Configuration Map), so ultimately these are yours to choose, but I'd like to pick defaults that make sense.
Comment below with your thoughts.
Entries 
Jose Miguel Bolivar said,
January 18, 2008 @ 4:16 am
Happy Birthday, MindReader!
It is hard for me to believe MR is just one year old. I really have the feeling I have been using it since years.
Re the keywords, I have been using "isproject" and "isresult" for a while already and found them quite convenient.
I see some risk with "normal" expressions such as "see note" or "reference" since you may sometimes want to write "see notes from…" or "file X for reference…".
Why do not standardizing the syntax and use "isnote", "islink", "isinfo" or "isdone" (alternatively "iscomplete")?
As additional benefit, that common syntax will make those keywords probably easier to remember.
ActivityOwner said,
January 18, 2008 @ 7:32 am
With way too much "info" streaming in from outlinker, I wsa fairly desperate for the "info only" feature so I went ahead and implemented/uploaded some things (see below). I continue to be interested in feedback on questions above.
One nice thing about MindReader is that keywords are not mutually exclusive. For example, if you want "Call Sam" and "phone Sam" to trigger your @phone context, you just add both keywords to the "Context" branch. The only time problems can arise is when keywords share a substring (i.e. you want to avoid having "project" and "subproject" or "monthly" and "bimonthly").
In this case, there is no reason not to add all the "is" keywords you recommend so i went ahead and added them. In general I try to have MindReader keywords be as intuitive/automatic as possible ("Buy Eggs tomorrow!") and then supplement them with explicit settings (e.g. "Eggs [@errand 1/19/08!]")
I have provisionally added the following:
project:
result:
info:
info only:
complete:
done:
half done
not done
see note
see link
I also changed the upgrade routine so that each change is only made once. This will save time and avoid need for user to keep reversing an "upgrade" keyword that you don't want (e.g. if you want to avoid "see link). Deletion of keywords now includes user confirmation. If someone was using "link" and "note" I didn't want it suddenly to disappear on them as they would loose info. Adding that confirmation to added keywords is a someday/maybe as I need to figure out how to not swamp a new user. Ultimately we'll have mindreader.mmap entirely created by the upgrade routine, which will make that easier to manage.
Magnus said,
January 18, 2008 @ 4:42 pm
I totally agree with you regarding need for "info only:".
Have installed latest update but canĀ“t get it to work so far. The action icon is still there….
ActivityOwner said,
January 18, 2008 @ 7:29 pm
Can you try selecting a topic and type "m info:" to see if it removes the icon?
Did you make sure you put mindreaderNLP.mmbas in my maps and did it prompt you that it needed to upgrade mindreader.mmap?
Ron said,
January 18, 2008 @ 7:30 pm
I like Jose Miguel's suggestions - isproject, isresult, isnote, islink, isinfo, isdone - they are consistent and therefore easier to remember (for me anyway). Your suggestions project:, result:, etc. with the trailing colon would work too but I find the "is" keywords easier and faster to type than colons.
I find that where I have departed from your default verbs is most often to add a shortcut - "eow" instead of "end of week", "eom" instead of "end of month", "!" instead of "deadline", etc. - I look to minimize the number of keystrokes required as much as possible and find some of the defaults to be too wordy.
One other thought to ponder - I often find that I am repeatedly entering the same set of keywords over and over again. It would be nice to have some sort of Keyword Set or Group feature.
I really like MindReader and Outlinker - together they are an incredibly powerful combination that makes the potential of RM and MM reality.
BTW - do you ever sleep?
BillS said,
January 18, 2008 @ 9:34 pm
Like Ron I tend to like the unix approach:
1. rmp - project
2. rmr - result
As for the colon, I typically use it in the task like this:
1. Research:
2. Review:
Perhaps a hybrid concept like:
1. isp: project
2. isr: result
3. isl: islink
4. isd: isdone
etc
My 2 Euros worth,
Bill
P.S. AO sleeps, he just has 3 clones of himself :^()
ActivityOwner said,
January 18, 2008 @ 10:00 pm
Great suggestions — now we just need a scrabble pro to spot our false positivie risks! Don't want to have any d*isinfo*mation in our dashboards:-).
One downside of some of these is that they won't look as natural in the final task so we'll tend to want to put them in [], which adds keystrokes, but perhaps that's true of "info:" etc. as well. Remember the pros can always customize your config file, so the question is what defaults will be most intuitive to new users.
I do sleep, but with the power users of the site mostly several time zones away, I wake up early each morning to see the new ideas (and assignments) that have arrived overnight.
BillS said,
January 19, 2008 @ 6:49 am
OK, new users are a different audience. In that case, then AO's provisional keywords make sense & if we use Jose's stanardization, we can add
1. is project:
2. is result:
3. is stalled[or snagged]: (85% completed)
etc
I like the idea of using 'is' & the colon. I think this would look most natural to new users who may or may not know about or use the [ ]. FYI, when I first started using MR I didn't know about the [ ] & my tasks looked weird. Now, my tasks still look weird, but for different reasons.
ActivityOwner said,
January 19, 2008 @ 12:14 pm
One context that MindReader can be used is in marking up task during an on-screen group branstorming session. That's one place where I see some of the "keyword:" approaches feeling more natural.
ActivityOwner said,
January 20, 2008 @ 10:18 am
The discussion above (which is great) caused me to wonder if we should have the option of having certain keywords and classes of keywords automatically deleted from the string without the need for [brackets].
For example, How often is it necessary to have someday, isproject, isinfo !, tomorrow, @web, name@, P1, 1/'1/08, 15m remain in the string once it has been mindread into markup?
Could be a nice NAA prize feature request :-).
ActivityOwner said,
January 20, 2008 @ 11:12 am
re: "Could be a nice prize feature request for NAA"
If only I had some self control and didn't just implement and upload it. :-)
Say goodbye to most bracket requirements….
BillS said,
January 21, 2008 @ 11:11 am
AO,
Not to be a contrarian, but one way to avoid finding substrings would be to require that parsing will only occur within the [ ]. My vote would be for this option.
Your personal devil's advocate,
Bill
ActivityOwner said,
January 21, 2008 @ 9:52 pm
Bill — It could be done relatively easily, but its not high on my next action list (work on that NAA score :-). I'm aspiring to go the other way and find ways to avoid all need for brackets.
Desitination maps seem to be the final frontier for me on the bracket front, particularly at work. At home most things flow to the right map based on kids' names, finances, bills, saw, outlinker, mindreader, blog, etc.
Perhaps we could invent some syntax for situations where the destination map is extra and needs to be deleted (e.g. map:finances) and perhaps extend that to search for files by name when not found in links branch in that situation.
Important MindReader Update for Outlinker Users » ActivityOwner.Com said,
January 22, 2008 @ 7:50 pm
[...] I mentioned in "MindReader: How to Best Read your Mind", the original "link" and "note" keywords used to tell MindReader to use [...]
Ron said,
January 24, 2008 @ 9:00 am
I think I have to agree with Bill on this one. Typing the [] is very quick and easy and puts me in "I am now entering keywords" mode. Only parsing within the brackets would also eliminate most false positives which are a problem using my minimalist (or unix) approach to keywords.
One minor thing I have noticed - when using "isinfo" the engine is applying the blob icon. Not a bid deal, but I would not expect that to happen.
Ron
ActivityOwner said,
January 24, 2008 @ 10:06 pm
Ron — So you would want to type: "Call Nick about Acme Project [@phone Nick@ Acme Friday]" and have MindReader ignore the text outside the brackets? It certainly can be implemented as a non-default option, but will take a focused hour to avoid introducing bugs so have to put it on the someday/maybe list for now.
On the topic of keywords, I put up a skeleton list of all the MindReader keywords (extracted from mindreader.mmap). R.S. is interested in generating some improved documentation so I put this up to get him started. Feel free to jump in.
http://wiki.activityowner.com/index.php?title=MindReader_Keywords
re: blob
Thanks for the catch. I was running with this option off. I changed the code and will upload with next round. It is a one liner to fix it in the "Addtheblob" sub routine by inserting the middle line here:
Or ActiveTopic.Icons.ContainsStockIcon(somedayicon)) _
Or ActiveTopic.Task.Complete=-1 _
And AddBlob Then ActiveTopic.Icons.AddStockIcon(mmStockIconMarker7)
Ron said,
January 25, 2008 @ 1:00 am
Well, no - and yes. I guess I really fall somewhere in between. I would expect MR to recognize "Call Nick" and apply @phone and Nick@ without the need for brackets. However, your example assumes I have a map that exactly shares the name of the project or subject of the call (Acme). In my setup that is not generally the case and I often find substrings in the topic equaling one of my map keywords resulting in my tasks being shunted off to strange, unexpected and unintended destinations. So, maybe what I am saying is there are two classes of keywords - those that work regardless of where they are (inside or outside brackets) and those that only work within brackets. Taken a step further this would be configurable on a keyword by keyword basis. Does that make sense?
ActivityOwner said,
January 25, 2008 @ 8:15 am
Yes — In theory we could add keyword by keyword customizations by using icons on the configuration map.
Map keywords are a key area to think about from a false-positive and "easier positive" point of view. Note that the links branch can be two layers deep now and it searches the second layer of each subtopic before each subtopic.
I find this feature very powerful. For example, my "blog" branch has "blog" as the 1st layer topic and subtopics for "mindreader", "outlinker", "mark task complete", etc. That way all I need to do is mention outlinker in a task and it goes to the right map. I could further enhance the macro to go a third layer deep and have "mindreader ideas" go to a different map than "mindreader bugs" and the rest go to "mindreader" or "blog". I have another branch of "finance" with subtopics "quicken" "download statement" "2007 taxes".
I think a third layer might be useful so I'll add it. Question: Should all the third layers be search before the 2nd and the 1st? Might be a little slower but be more false-positive resistant.
ActivityOwner said,
January 25, 2008 @ 8:36 am
Continuing on this theme, note that you can have redundant entries, so if your kids names are Tom and Jane, you might have
"Tom" - "Tom and Jane"
"Jane" - "Tom and Jane"
"kids"
so that a task related to both "Tom and Jane" are intercepted and go to "kids" map rather than one of their individual maps.
I'm definitely going to take this a step further and have recurse to allow as many levels deep as user chooses.
Ron said,
January 25, 2008 @ 6:36 pm
Hmmm… Wow!
I am going to have to play with this a bit. I don't know if it completely answers the need, but it sounds like it goes a long way toward it - and - introduces compelling new capabilities.
The possible problem I see is if I have a Map for one of my staff members - Jane - and a map for a project she is working on - let's use the Acme Project again - if I enter "Discuss Acme Project Status with Jane", where does MR send the task - to my Acme Project map or to my Jane map? Further, in one case I may want the task to go to the project map and in another I may want it on my Jane map (or both?).
If you could share your mindreader map (or a likeness of it) to show how you are addressing these issues it probably would help.
Ron said,
January 25, 2008 @ 8:08 pm
FYI - The code link on the Wiki is pointing to MindreaderNLP-dev.mmbas.
ActivityOwner said,
January 25, 2008 @ 11:35 pm
Hi Ron — I just uploaded a new version of mindreaderopen.mmbas that allows you to organize your links in as deep a tree as you like.
I have noticed the problem you mention with the staff-member/project map conflict on my work maps recently. I got around it by changing the keywords for the staff maps to firstnamenamemap instead of firstname and then I just bracket the [firstnamemap] when I want it to go to one of them.
The advantage of the multilayer map is that it lets you organize your a large number of maps in groups and to intercept the more specific destinations and fall back to the more general maps at the root.
I need to see how I can fine tune my approach with this new feature onboard.
the "-dev" on the wiki branch is an artifact of the early days when I envisioned I'd have a stable "download file" and more dynamic text code. As it turns out, I prefer to just keep wiki current with what I'm using. The downside is that some bugs creep in, but the upside is that if there is a bug in the online version I experience it as well.
ActivityOwner said,
January 26, 2008 @ 12:51 am
I just made one additional change. If you download the new version of "fq" and mindreaderopen.mmbas, you can now eliminate the need for isnote/islink keywords and also avoid bloating your queue with random clipboard text.
Now if you want to grab a link or note, you just type control-shift-G, and then toggle to fq and type in your action. The link or note will automatically come along.
To get this to work automatically without need for keywords, you need to use the m tag to "setoption:usetextgrab:1".
If this makes sense to folks, I'll change over "q" as well and set the option above by default.
Note that latest version of mindreaderopen.mmbas also allows you to organize your "links" branch as deep as you like. it will search from the bottom up and use the first match it finds.
MindReader Update: Capture notes and links without using keywords » ActivityOwner.Com said,
January 27, 2008 @ 11:48 am
[...] could get false positives if you used "link" or "note" in your [...]
BillS said,
January 27, 2008 @ 6:34 pm
I think recursing bottom up in the link branch helps, but I still like the parsing only in the [ ]. I also use abbreviated names too in my mapmap.
So, my entry typically looks like [waiting for wn ! p1 mymap #03/01#]Promotion is successfully implemented
This would give me a task of:
Promotion is successfully implemented
And icons for: Priority 1, deadline
And state date: today
And end date: 03/01
And Resource: Bill Nestle@
And sent to: mymap
I don't look for the 'Call Bill' in the task, rather I enable task information & task markers and view it from the resource. The only downside to this is the extra real estate of the task info under the task.
Chuck Brown said,
March 7, 2008 @ 3:23 pm
Ok, at the risk of sounding like a complete idiot, I'm going to just say it… I've been 'messing' with MM for almost a year. The best thing I've found about MM is the Mindreader stuff. I see the beauty of the system, I get what it can do for me, I just don't get how to get there. It seems as though there are endless tags that are being constantly updated and improved with features that are really useful, and I'm still struggling with what it means if an "@" is before a word or after it, with how I turn this stack of papers on my desk into meaningful tasks in maps. At one point, I just dove in and created a map with everything I could think of, really not doing the work of assigning them to projects and dates and todo's, and I generated a massive graphical todo list that wasn't much better than what I had already. Then, I split that into multiple maps, each with its own 'focus', and that resulted in more unmanageability.
So, I've come to the conclusion that somewhere, I've shortchanged myself in the learning process, and really am missing something vital here. So, I went to the MindJet site, thinking what I was missing was a basic understanding of MM, but could really find nothing useful there to change that. I went back to the MindReader stuff, but here I find dozens of people happily using and improving the MR code, none seeming to have the comprehension problems I am. Can someone point me in the direction I need to go with this? I'm jealous! I see what this can do for me, and really want to get there, but just can't seem to find the path from here to there!
BillS said,
March 7, 2008 @ 5:27 pm
Chuck Brown,
I will refer you to the 2 additional tools that we're using:
ResultsManager & GyroQ by Gyronix
These are located at http://www.gyronix.com
MindReader is an add-on to GyroQ
I would also start from the bottom up on this blog to get a history. It has taken ~year to get where we are. Understanding this background will get you up to speed more quickly. Also, look for the ResultsManager Project Map as this is a good starting point for organizing tasks around projects.
Also, while not completely necessary to starting the endeavor understanding Getting Things Done gets you alot farther in achieving your goals.
http://davidco.com — Getting Things Done
Good Luck
ActivityOwner said,
March 9, 2008 @ 9:50 pm
Hi Chuck — You raise some important and diverse issues.
The "AO tools" are relatively stable lately and I've been thinking that most value added "next action" in near term is focusing on documentation and examples and less on new code. I enjoy these tools a lot, but even I forget to use all the options, tags, and syntax available and I created them!
I understand your confusion on MindManager and ResultsManager. It takes awhile to grasp where one starts and one ends. MindManager is an environment and promotes creativity and brainstorming. ResultsManager helps pick up the pieces and sort them into dashboards. GyroQ feed the monster quickly with new 'stuff" and MindReader helps facciliate marking the "stuff" up with meaning.
Not sure if you have RM yet. It can feel like a big investment on top of MM price tag, but its worth every penny once you get rolling with it.
As Bill recommends, reading Getting Things Done or listening to the audio book in your daily commute is a good way to understand the principles and mindset behind the stuff here.
If you are running ResultsManager (which is key to all this), running the "next action analysis" script on your daily action dashboard can help you figure out what to do next with your current setup. It is by far my favorite tool and is helping me climb out from under in ways I haven't been able to for a while.
Re: @context and person@
GTD is all about translating your "stuff" into "physical next actions". That can mean being in a place, a state of mind, or with someone. ResultsManager uses person@ rather than @person as it likes to treat people (who can owe you things or be owed things) differently than @contexts.
What's next…
If I was serious about answering your question, I'd pull up a fresh map and set a goal/project of "Chuck understand how AO stuff works" and then start enumerating sub-projects of "contexts understood", "tutorial/example case generated on blog", and then figure out a simple next action for each, but having a full plate (and a declining NAA score), I'll toss the next action back to you :-).
How can we best help?
Chuck Brown said,
April 1, 2008 @ 4:00 pm
What SHOULD be happening when I select an entry on a map, then enter "m isinfo:" in gyroq? Sometimes I seem to get a white task start marker, sometimes no change at all. I understand that this is actually the percent complete in mr, but how do we go from percent complete to understanding that this is an information only item?
Am I missing a step in the configuration (i.e. setting up a custom marker), or missing something bigger?
Thanks!
Chuck Brown said,
April 1, 2008 @ 4:10 pm
I highlighted multiple emails, then hit the outlinker button (using latest code on the site for outlinker). Just testing the options, I hit "c" for compose. Not exactly sure what this is supposed to do, other than the obvious. So, when I hit c, Outlook came up sorry with a blank email, and that was it! The composed message didn't have any reference to or from the original message, and outlinker seemed to exit at that point. I think either something isn't working, or I'm missing something…
ActivityOwner said,
April 1, 2008 @ 10:25 pm
Hi Chuck — "m isinfo" should remove the task icon from the topic. Its primary purpose is to allow info to come into mindmanager via gyroq and outlinker and not create an action item. When you use the keyword from the gyroq m command the white square should flash onto the topic and then be taken off.
The "c compose" command doesn't really have that much use in outlinker. It is just a command to open a new blank message not associated with the one you are currently viewing. I added it early on thinking it would be a more convenient way to initiate a new message while outlinker was running, but I don't tend to use it and it seems to generate some confusion.
Chuck Brown said,
April 3, 2008 @ 10:28 am
I've been using the CTL-SHIFT-G/fq method for links, but am trying to understand how the [islink] tag works. My first assumption was that it would take what was on the clipboard and use it as a link, but this doesn't happen for me. The sequence:
Highlight link, copy to clipboard
GyroQ "Q Keep this link
Simply creates a task with a label "Keep this link", and wipes out the clipboard.
Is this not the correct use of "islink"?
ActivityOwner said,
April 3, 2008 @ 7:10 pm
Hi Chuck — islink is obsolete. MindReader originally put the current clipboard into the queue for every command just in case the user might have wanted to use it as a note or link. That just created too much risk/bloat for the queue file so I changed it to only grab links/notes when using the control-shift-g.
See:
http://www.activityowner.com/2008/01/27/mindreader-update-capture-notes-and-links-without-using-keywords/
Chuck Brown said,
April 4, 2008 @ 11:34 am
Off topic here, but this is as good a place as any… I have a desktop in the office, where the suite is working great (MM, MR, GQ, RM). I have 'My Maps' in 'My Documents', which is a network based folder that my workstation (Vista) sees as the real 'My Documents'. I synchronize My Documents on a notebook PC as well, as offline folders. I had assumed that was it to keep everything in sync, but I'm not so sure. Are there any other files or settings that I need to make sure get sync'd between them? I believe the gyroq tags are in gyroq.ini, which is in my maps, so it should be fine.
Other than My Maps, what should I need to make sure stays the same between the two computers, or is that really all that needs to happen?
ActivityOwner said,
April 4, 2008 @ 5:11 pm
All the GyroQ and MindReader configuration lives in "My Maps". If you have custom ResultsManager dashboards, you need to open them and select "install template" from both computers. I don't know if it is possible to instead sync a configuration file somewhere else. Alterhatively, you can export/import your ResultsManager configuration from the mindmanager menu.
Ken Florian said,
April 9, 2008 @ 4:21 pm
Just restarted using the various "set date" functionality of mindreader. I have the latest macros and the 1.9.4.7 of gyroq.
I am unable to set a "start date" using the "m" markup tag. I have a topic with a due date of 4/18/2008. I try to set a start date with "m" by typing 2 days before.
The task information marker in MindManager actually flashes to 4/16/2008 but then sets itself back to 4/18/2008.
I don't want to set durations or anything like that…just absolute start and end dates using the markup tag. I am overlooking something.
Thanks for your help.
Ken
ActivityOwner said,
April 9, 2008 @ 8:49 pm
Hi Ken — I'm puzzled by this one. Just to clarify, you have a due date of 4/18/08 and no start date set right? You then type "m 2 days before" and the due date toggles to 4/16/08 and then back to 4/18/08? (the "m " toggles to m tag).
I have no idea what might be causing this. I've tested on my system and don't see any issues. If nothing else, the "2 days before" keyword shouldn't be touching the due date; only the start date.
What happens if you type "m 4/19/08"? Does that move the due date successfully? How about "m start in 2 days"
Chuck Brown said,
April 10, 2008 @ 9:45 am
Should changes made in a dashboard be pushed back to the source map? For example, I found that there were a few items showing up on my dailyactions/Committed Next Actions that were really info items. So, I highlighted them, did a 'm isinfo', and the task icon cleared. However, when I went back to the source map (Bits & Pieces), the items were still flagged as tasks, so of course, when I rerun the dashboard update, they are tasks again.
I was pretty sure the update should have pushed through, because when I selected another task, and did 'm next wednesday', that DID update the source.
Chuck Brown said,
April 10, 2008 @ 10:18 am
Should 'm {link}' move an existing task to a different map? It doesn't seem to; if not, is there another 'mindreader' way to do this, other than cut/paste?
ActivityOwner said,
April 10, 2008 @ 5:40 pm
If you want to move a topic from one map to another, highlight it and hit "s keyword" in gyroq. The "s" toggles to the s tag and they destination keyword dictates its destination. You have to have configured the destination keyword using the "k" tag.
Chuck Brown said,
April 11, 2008 @ 8:45 am
How do the mr resources, contexts, categories, and areas relate to the 'master' context, owners, categories, and areas in Results Manager? Do they flow to each other, or are they independant?
Chuck Brown said,
April 11, 2008 @ 9:00 am
Sorry for the disjointed posts… trying to collate the GTD book, MR documentation, and RM documentation in my head! :) RM says to add people to whom you speak with frequently to the contexts list. In MR, we have resource verbs that seem to do this, but also have contexts. So, if I have 'Joe' working for me, that I assign things to and discuss things with frequently, should Joe be a resource, a contact, or a context, or all of the above?
Ken Florian said,
April 11, 2008 @ 9:15 am
Here is a screen recording of the behavior.
http://www.kenflorian.com/sample/tabid/454/Default.aspx
ActivityOwner said,
April 12, 2008 @ 9:17 am
Thanks Ken — That screencast was great — Can we sign you up to do some tutorials and demos! Thanks for taking the time to help get to the bottom of this.
I love this hobby. No need for Saturday morning crossword puzzle when you have mysteries like this one to solve.
I think I figured out exactly what is happening and replicated. I suspect you have a keyword entry on your mindmanager.mmap in the "starting" branch that is blank and has a "0" in the notes. Specifically you might have blanked out the "same day" keyword. That would cause anything you "read" to be effective considered as having the "same day" keyword.
I have updated mindreadernlp.mmbas and mindreaderopen.mmbas to avoid this issue generally for any entry in mindreader.mmap. You can download or just look at the mindreader.mmap and fix there.
Let me know if this doesn't address your problem.
ActivityOwner said,
April 12, 2008 @ 11:05 am
Hi Chuck —
ResultsManager has the capability to learn/remember your contexts/areas/categories and present them as drop-down menus in the edit activity dialog (and in GyroQ if you prgram it that way). I personally don't make use of that feature as I usually just prefer to type it out quickly or (even better) to have them set by inheritance from project above or by MindReader keywords. So, to answer your question, there isn't a programatic connection between the the MIndReader configuration and the RM lists.
re: People/Contexts
This can be a little confusion at first. If you wanted you could have a context (e.g. @Joe) for things you want to talk to Joe about and that would work fine from a context list point of view.
The power of the ResultsManager "relationship central" branch is that it combines a "Contact Joe About" list with a "I owe Joe" and a "Joe owes me" lists, so when you have Joe in front of you or on the phone, you have a lot to talk about..
In order for ResultsManager to "know" that "Joe" is a resource and not a place, it uses the convention of Joe@ rather than @Joe and puts then entry in the resource field rather than the context field.
MindReader automatically sees the word after some key "resource verbs" as a resource (e.g. waiting for Joe, Talk to Joe, I owe Joe, Ask Joe, Remind Joe, etc).
The "ResourceList" branch is mindreader is optional. It is just for cases when you want to use a short name in the text but need a formal resource name (e.g. "Chuck" for "Chuck Brown"). This applies when you are working in a group settting where everyone needs to stick to one name for each person.
I think I'm caught up on your questions — if any have slipped by let me know.
Ken Florian said,
April 14, 2008 @ 2:01 pm
the "same day" topic was broken in my mindreader.mmap.
Thanks for helping me figure it out.
Do you have a specific suggestion or two on a tutorial to record? I will try to do a couple in the next month.
Ken