Dryopithecus
It would be very nice if we could set a keyboard accelerator for "Unlink highlighted thought from active thought".

Currently, DELETE key is assigned to "Forget/Delete highlighted thought/link", and this accelerator cannot be changed.

I personally never use DELETE key to forget/delete thoughts/links, because I found it too easy to accidentally forget thoughts. Every few weeks, I found some thoughts mysteriously disappeared from the Plex, and eventually found them in the Forgotten Thoughts list, I must have accidentally hit the DELETE key when I was trying to delete text in the search field or notes.

I personally prefer to use DELETE key to unlink highlighted thought and Ctrl+Shift+Delete to forget highlighted thought.


Quote
Darkstar

Seconded.

-Darkstar
Quote
Steeph
Dryopithecus wrote: It would be very nice if we could set a keyboard accelerator for "Unlink highlighted thought from active thought".

Currently, DELETE key is assigned to "Forget/Delete highlighted thought/link", and this accelerator cannot be changed.

I personally never use DELETE key to forget/delete thoughts/links, because I found it too easy to accidentally forget thoughts. Every few weeks, I found some thoughts mysteriously disappeared from the Plex, and eventually found them in the Forgotten Thoughts list, I must have accidentally hit the DELETE key when I was trying to delete text in the search field or notes.

I personally prefer to use DELETE key to unlink highlighted thought and Ctrl+Shift+Delete to forget highlight thought.


Nice idea, the more customisation, the better.

There might be a downside to this though. If you accidentally press DELETE you just forget the thought but won't lose anything. You just remember it and voila.

If this change was implemented and you accidentaly press DELETE the link is lost. Forever. No 2nd change. Unless you happen to notice it very shortly after it happened so you could Undo it. But that is the whole problem, you never knew you deleted a link so you probably will not notice it.
PB user since 1998

Mind over matter?
I don't mind and it doesn't matter.
TB 8.0.2.1 Pro on Win10.1 Pro 64bit JVM 1.8.0-112
Quote
dyslucksia
I like Dryopithecus's intention, after all anything's better than not being able to use the keyboard at all to delete links!

But I don't see why we need a keyboard accelerator at all. It would be far simpler to have a Thought context menu option, saying "Delete link with Active Thought". PB must be able to tell if any thought visible in the Plex is linked to the Active Thought or not.

If we had to use a keyboard accelerator, you'd have to use both the keyboard and the mouse, just to delete a link! I also disagree with Steeph, there's no risk of losing a link if you accidentally delete it; Undo will get it back for you.

This issue doesn't bother me right now as I like my little macro; all I have to do is move the mouse cursor over the link to be cut, press Win+X, and - snip! This is a link-oriented, not a Thought-oriented, solution.

I thought, we have Ctrl+Click to add a Thought to selection, so why not a key-enhanced click to cut a link? Haven't looked back since. Well, you know where to find my macros if you want it.

What does annoy me is 5.5's new feature of not making highlighted links stand out from the rest by making them the same color as unhighlighted links. Let's hope they change their minds on that one.



PB 5.5.2.1 on Windows XP, J-1.6.0_17
Quote
Steeph
dyslucksia wrote: If we had to use a keyboard accelerator, you'd have to use both the keyboard and the mouse, just to delete a link! I also disagree with Steeph, there's no risk of losing a link if you accidentally delete it; Undo will get it back for you.

Well, yes and no. Perhaps I've should have explained it better.

Indeed the undo function will restore your link, but what if you've accidentally deleted it without realising it? The next session or a few days later the undo list will be empty and your link forever lost.

Compare this to accidentally forgetting a thought. It will wind up in the forgotten thougth list where it will stay forever until you restore it or delete it. Even after a kazillion years you would be able to restore it. Not so with a link.

Which brings me to the idea that links should be forgettable as well, similar to thoughts. Bit difficult to implement though I think. Perhaps the list should only show forgotten links for the current thought orso.
PB user since 1998

Mind over matter?
I don't mind and it doesn't matter.
TB 8.0.2.1 Pro on Win10.1 Pro 64bit JVM 1.8.0-112
Quote
dyslucksia
Hmm. I understand now what you're suggesting, but can't see it happening for a long time, if ever. Links are the poor relatives of Thoughts in PB for reasons that have been posted many times here. Until link labels are divorced from link types, and preferably made multi-line, they won't be considered to be capable of holding enough information to make broken links worth retrieving.

PB already keeps a record of deleted links in its output.log file. All you need do, if you wish, is write a small script, macro or batch file to save these logs for x number of days after closing PB, or before restarting it.

Here's an extract from my last session in a test Brain, with irrelevant stuff deleted:
Code:

~Create Thought: 2,41 [aaachild]
~Delete Link: 199 [1] [AE099301-3460-9BD0-8B58-271E3836A603]
~Delete Link: 200 [1] [E72CAD26-235F-9059-7CD1-2EA754B6EDBE]
...
~[Undo]Delete Link: 199 [1] [AE099301-3460-9BD0-8B58-271E3836A603]
~[Undo]Delete Link: 200 [1] [E72CAD26-235F-9059-7CD1-2EA754B6EDBE]
...
About to shutdown due to window close...

PB 5.5.2.1 on Windows XP, J-1.6.0_17
Quote
Darkstar
It would be trivial, from a data point of view, to keep links around as "forgotten". It really is just one more column for the link table. Fir example, call it "IsForgotten", and with it being a bool type, now you now what is active link and what is forgotten. You could review "forgotten links" the same as you do with thoughts. They'd just show the two thoughts the links connected. Put the exact gate indicator on the outer side of two thought names, and you would end up with something like:

Forgotten Links
(c) Test ---- Thought 39 (p)
(c) Test ---- Thought 41 (p)
(j) Larry ---- Shemp, Replacement (j)
(p) Moe ---- Relatives (c)
(p) Shemp ---- Relatives (c)

Selecting a forgotten link would act like selecting a forgotten thought. The plex would display it, in "forgotten" colors. It would make the active thought whichever end of the link it found first. Then you could "remember" the link, or delete the link, or do something non-related.
-Darkstar
Quote
dyslucksia
Sounds like what Undo and Redo do do right now, only it's limited to about 8 do's.
PB 5.5.2.1 on Windows XP, J-1.6.0_17
Quote
Darkstar
Well, effectively, it would allow you to "undo" any "forget link" that hasn't been "deleted" (just as you can do now with thoughts).

Of course, you couldn't bring back a link from non-forgotten thought to a forgotten thought--- as those come back when the forgotten THOUGHT is "remembered". But other then that minor item, it would put links on a closer level to thoughts. Which they should be, as they are a VERY important element in PB--- in fact, they are often more important then thoughts, depending on the information being tracked.

In fact, lately I've been thinking that links should really just be directed thoughts--- that is, they would have all the major capacities of a thought, they are just a "connector" type of thought, so they'd start off a special class of thought, showing a kind of connection between two thoughts. In the beginning, PB didn't have any special thoughts, but with the inclusion of tags and events (both are actually "defined thought types), connector thoughts would fit into that newer design paradigm.

And then, we could have graphics on a link (think about this--- big Arrow graphic showing a direction for the link--- or a wedding ring to show a "marriage", etc), names (thoughts have names, after all), it would have types (so same functionality), color (same functionality), keywords/labels (which it has now, but a link TYPE connector would still be able to have its own individual set, as Thought Types can, so improved functionality). As a connector, it would just have an added attribute--- the two thoughts it connect, and how. But this extra attributes MIGHT be just as easily handled as attachments (internal references to the two thoughts), and THAT would set up for allowing future version of PB to do references to thoughts in other brains.

Just some thoughts on linkers of thoughts.
-Darkstar
Quote
Steeph
Darkstar wrote: It would be trivial, from a data point of view, to keep links around as "forgotten". It really is just one more column for the link table. Fir example, call it "IsForgotten", and with it being a bool type, now you now what is active link and what is forgotten. You could review "forgotten links" the same as you do with thoughts. They'd just show the two thoughts the links connected. Put the exact gate indicator on the outer side of two thought names, and you would end up with something like:

Forgotten Links
(c) Test ---- Thought 39 (p)
(c) Test ---- Thought 41 (p)
(j) Larry ---- Shemp, Replacement (j)
(p) Moe ---- Relatives (c)
(p) Shemp ---- Relatives (c)

Selecting a forgotten link would act like selecting a forgotten thought. The plex would display it, in "forgotten" colors. It would make the active thought whichever end of the link it found first. Then you could "remember" the link, or delete the link, or do something non-related.


Yep, that's the easy part.

I was thinking more of the display of the forgotten links. Say you have 200 forgotten links and you would display them similar to how forgotten thoughts are displayed. That would be meaningless.* Links are defined by what they connect so you would need to see the thoughts and other context to be really able to decide what to do with it.

An option would at least be to show forgotten links for the current thought and/or thoughts that are 1 click away. Then at least you immediately have the context of the link. That however would also imply that thought are displayed that are only connected by a forgotten link, so perhaps you would need to introduce another colour orso to indicate it is not current.


* And probably impossible, because links are nameless
PB user since 1998

Mind over matter?
I don't mind and it doesn't matter.
TB 8.0.2.1 Pro on Win10.1 Pro 64bit JVM 1.8.0-112
Quote
Darkstar
links have names. That's what you put in the label of a link. Most links don't get that field filled in, but it is where you'd put that sort of information.

Having all the "forgotten links" show for a particular active thought could be useful. So if they kep the forgotten links, then they could certainly do this.

As for color--- just default forgotten links to use the same color as forgotten thoughts. Add it in as a settable color in the preferences if it is needed.
-Darkstar
Quote
Steeph
Darkstar wrote: links have names. That's what you put in the label of a link. Most links don't get that field filled in, but it is where you'd put that sort of information.

Hmmm, yes. Hadn't thought about that. Actually, come to think of it, I've never even given a link a label. Occasionaly I assign a linktype, but that's it.

To phrase the problem better, by default links don't have names unless specifically assigned. And unlike thoughts, this is not mandatory. I even suspect most people never label links anyway. I wouldn't like/want mandatory link labels though.

I imagine that even having a label wouldn't help much without the thought context, but that's just speculation on my side.
PB user since 1998

Mind over matter?
I don't mind and it doesn't matter.
TB 8.0.2.1 Pro on Win10.1 Pro 64bit JVM 1.8.0-112
Quote
dyslucksia
If you recall, the original idea of this thread was Dryopithecus' request for a keyboard accelerator to unlink Thoughts (quite a reasonable suggestion IMO).

So what's the point of saving retrieving broken links, other than to restore them if they were inadvertently broken? My philosophy is that if I accidentally break some links, I'd realize this pretty soon and if I couldn't get them back using Undo, I'd always have PB's output log to fall back on. If this didn't work, all I'd need to do is bring up a list of orphan or parentless Thoughts, and I'd pretty soon find any broken links. About the only thing I'd lose might be their Types/Labels , but it wouldn't be too hard to guess what they might have been and to restore the type by resetting it to a type that still exists. If I couldn't do any of the above, too bad. It wasn't that important. Get over it and move on.

Note that like Thoughts, you can't create a link type without having a link there to assign it to, but even if you destroy the link later and there are no other links of that type, the link type still persists unless you specifically delete it via another link.

What if one or both Thoughts at each end of a link have been deleted? Would you still want to retrieve the link between them, and if so, how would it be displayed? This seems a bit futile to me. Note that at present, individual links have no substance, which is why their design has only one label per type, and the type's name is its label. Once you start differentiating between links of the same type by assigning different label content depending on their location, you then have to create a whole new database of link data objects (dependent on their connecting Thoughts) as opposed to link type objects.

What I feel is wrong is this desire to assign content to a link's label, because in reality the content depends on which side of the link you are viewing. For example, take a family tree with John Smith as the grandfather and Sue Smith as the granddaughter. Then if John Smith is offscreen, the link stretching from Sue Smith back to him might read "Sue's grandfather is John Smith". Likewise, the same link from John's Thought might read "John has a granddaughter named Sue."

I would implement this not by putting two lots of data in a link object but to incorporate each datum in the Properties of the relevant Thoughts. Then if the link is inadvertently cut, no data is lost.

I don't think Forgotten Links is the way to go, but we can certainly look at hiding some links by including an invisibility attribute that suppresses their display. I do this right now by assigning a type with the same link color as the background of my plex, so the link is hidden unless I hover over it, though the target Thought is displayed, and anyway, the gate will show if there's a hidden or visible attached link by the nature of its solid color.

For those who like using wallpapers, we need an invisibility checkbox that suppresses link display for any given link type. Now that would be really useful. Has anyone suggested this? We've had the option to suppress display of all links now for some time, though I have no use for this. When 5.5 is ready to use it will allow us to suppress display of sibling links, long overdue to reduce plex clutter. Reports filter also lets us suppress certain Thoughts and thus their links too.

PB 5.5.2.1 on Windows XP, J-1.6.0_17
Quote
Darkstar

Output.log is overwritten every time you start up PB. So it isn't a good repository for "link changes".

When a thought is deleted, its links are deleted. Links depend on both sides of what it connects to be in the database.

Undo will only work in session.

I've lost links. It isn't nice to realize that I'm missing several hours of my work and I will have to try and recreate it months after doing it. Don't forget--- PB lost links in version 4 AND in version 5. I expect it will continue to happen where it doesn't TAKE a link when you are mad linking or when the plex is busy--- and PB will OFTEN display that "missing" link is actually there (as an invisible link--- with the gates on and the thoughts "arranged" as if it was there--- but it isn't, as shown by a reloading of that brain will show).

So--- outright loss (by program), data corruption (by program), and pretending it made the link but it didn't (by program)--- all reasons for PB to offer us a "link history" table to bring back old forgotten links or at least let us review them. And that is not even talking about user actions (unlinking something and then finding you need it back 3 months later, accidentally getting overzealous in unlinking stuff on Friday and then realizing on the following monday or tuesday that you removed stuff, clicking on the wrong link while doing too much multitasking and not realizing your removed the wrong link for a few weeks, etc)

If a link is showing BOTH sides of what it connects, it is SHOWING its direct context.

-Darkstar
Quote
dyslucksia
Maybe the answer is for us to diligently create macros, scripts or batch files that on closing a Brain, append the output.log file to a backup master log.

Or perhaps PB developers, recognizing that data corruption still goes on, might be kind enough to add an optional feature that automatically does this when we close a Brain. I.e., the output log is not overwritten, only appended to. Also a warning to us to delete it when it reaches a certain user-defined size
PB 5.5.2.1 on Windows XP, J-1.6.0_17
Quote

Add a Website Forum to your website.

Newsletter Signup  Newsletter Signup        Visit TheBrain Blog   Visit TheBrain Blog       Follow us on Twitter   Follow Us       Like Us on Facebook   Like Us         Circle Us on Google+  Circle Us         Watch Us on Youtube  Watch Us       

TheBrain Mind Map & Mindmapping Software     Download TheBrain Mind Mapping Software