The Mentography of Rights



 

see dialogue
 


This old thought has been coopted for a greater purpose wink

What we call the top of our ontology is pretty arbritrary. Many logic systems call it a "thing". I choose to call it an "item". So in my system of logic here, everything is a member of the class called "items" Hense the tag on this item "type items" is redundant.

Tags

  1. ontology
  2. rights
  3. mentography
  4. quads

Comments


Mark de LA says
I think this got renamed thought 
PS I emptied the group logic into test. (2 items)
seems as if maybe it should be discussed whether I should be able to send seth’s thoughts in group logic into the sandbox (test) .

Holmes says
You ask “should be able to”. You should not be able to move thoughts in groups you are not a member of. If you are a member, you can do anything. That is the cart blanch way the system has always worked. Group members must play nice with each other.

Seth says
lol … this is a funny one … i’ll give it 3 laughlaughlaugh.

First of all, yes Mark, you had the right to do what you did … er, or you would not have been able to do it … the system functioned, as Nathan mentioned, flawlessly yes.

Now let’s get into the matter of “should” you have done it.   Well you knew that what the purpose of group logic.  But let’s suppose that you did not know its purpose … yet you had access to do whatever there because it is open.  You could choose to obstruct its unknown-to-you-purpose … or ignore the group.  Here, me thinks, is where the golden rule doth inform

Now obviously the owners of the group will need to defend themselves against this kind of malicious intent.  In this case i will just move the thoughts back and hope that you will play here ethically in the future.  After all, we are still humans acting socially here towards each for our mutual amazement and heart … are we not?

Seth says
MR of group mark 2016-02-23 07:49:59 [item 487#46499]
I also emptied the group recipes bad links on one item & mush in the other.  can delete it.
wizzard ok, i’ll delete the group now thumbs up

Seth says
Poe of group da 2016-02-23 08:26:22 [item 487#46504]
I think it may be time to do the planned reworking of how open groups function. I was about to work on full proliferation, but with the new relocate feature, it is even more important that there are true boundaries in groups that have “open thought” functionality. Releasing in this state is likely to cause trouble for everyone, unless we simply disable open group creation.
seth of group seth 2016-02-23 08:27:33 [item 487#46506]
?
MR of group mark 2016-02-23 08:32:09 [item 487#46507]
… & closed groups with lots of members 
?

Holmes says
MR of group mark 2016-02-23 07:42:52 [item 487#46497]
Notice I said discussed – not dictated – changes may be needed else we might all become monad groups of one. Perhaps a pending move to be reviewed by the author or a pre-approved right to reorganize to certain authors or members of a group.
Poe of group da 2016-02-23 07:54:41 [item 487#46500]
… and you still said “should be able to”, which is not asking of a human, but is asking if it is allowed, which implies “the system allows”. So I tell you what the system does. I have no interest in your conversation with Seth. I am replying to your clear question about system allowance … which is my area to respond right now.
seth of group seth 2016-02-23 08:07:31 [item 487#46502]
?
MR of group mark 2016-02-23 08:27:18 [item 487#46505]
notice also we never had the ability to move items before – so the situation as it was is no longer. smug
Look, make up you mind.

I read every word you write. You used language that had no other possible meaning, the way it was actually written, than to ask a question of me, pertaining to how the system operated.

So, either be more exact in your use of words, as you claim to believe is important. Or, don’t worry about what you write and use intuition to understand and your thoughts to pre pave what others will get.

I can adapt to either. And being Gemini, I will do both as I wish. I am asking that you make up your mind if you are going to continue to throw around the “Bablon” analogy every time someone else’s words are not to your liking. You can’t eat your words and spell them too!
 

Mark de LA says
seth of group seth 2016-02-23 08:37:00 [item 487#46511]
but i suggest discussing changes to rights in group fbi, not here in group logic.
MR of group mark 2016-02-23 08:44:49 [item 487#46513]
Seth – move them ! laughing no more exact than that!  When I make suggestions & discoveries I use the language appropriate to such.  When I make a recommendation I will use language appropriate to a recommendation. When I make a command or imperative …… I will use language appropriate for that . I rarely will do new agey munging unless I am talking to a munger-of-the-moment. 
seth of group seth 2016-02-23 08:51:29 [item 487#46516]
sigh … i was just talking about discussions of changes to the way rights work here.   seems to me you are munging that with the rwg you are haveing with nathan.
yep, it was hard to say who N was talking to so I bundled by thoughts & sent them back up – needed to be saif. it was a forking moment.

Seth says
seth of group seth 2016-02-23 08:37:00 [item 487#46511]
but i suggest discussing changes to rights in group fbi, not here in group logic.
Poe of group da 2016-02-23 08:39:40 [item 487#46512]
Not sure if you are saying that to me or not. I am not discussing. The work to do has already been discussed and is clear. I am only letting you know a change in the wind and which way it has blown me.
seth of group seth 2016-02-23 08:47:21 [item 487#46514]
i said that to anyone who wants to discuss the changes in rights … you .. me … mark. 

If it is already in your imagination, and you need no further discussion to match your intentions to others, then go for it, nathan.  I trust you will get it right … frequently you do … sometimes you do not and then we are faced with your incessant rationalizations.
Poe of group da 2016-02-23 08:50:37 [item 487#46515]
The quad editor is right. Just saying. You would come to that understanding too if you actually looked at what is in the table and how everything there interacts instead of just noticing that it takes more effort than you want to put into it to read it as it is.
seth of group seth 2016-02-23 08:58:26 [item 487#46518]
i woudn’t call it a “quads editor” laugh just saying.

I did look at the table, and understand how it can work. 

Hopefully you will make it work practically for thinking domains going into the future. 

Go for it !
Poe of group da 2016-02-23 09:04:30 [item 487#46520]
Well, no. I won’t make that editor work practical for thinking domains. That editor is for system developers only, not even for plugin authors. It is a window into the raw guts, but highly useful for that and as a tool to change things until specific editors can be made for particular areas.

One interesting way to make a thinking.domain would be to use one quad table for everything. Many sites do this. However, even our use shows that such a table would necessarily be a hybrid and would need to vary from strict quad semantics in order for the software to stay efficient. At best, one would have to filter out the ambiguous system rows before treating it like a properly structured quad table.
yep .. i even wrote a whole site using only quads … talked about here as tagtalking.net … i operated it here for years.   i ended up deleting it because it got corrupted by hackers … which had nothing to do with it being written in quads … and because others on the net were already doing it better than i could.  

Mark de LA says
Poe of group da 2016-02-23 09:15:56 [item 487#46522]
Your wallaby has always been to create one scheme that rules them all. One way that works for everything and is readable by humans. Cybermind was the first such. You have made many other attempts.

Personally, I don’t think that is needed. Nature doesn’t do that and yet nature functions quite amazingly fitting together in billions of diverse ways without underlying uniformity. I think the computer world would be fine that way too, and very nearly is exactly that.

But even if one did want a unified data theory for everything, modern software is not up to the challenge. To even get close to it would require a processor built to operate directly upon the underlying theory. Which I know you envisioned a Cybermind processor long ago.

In any case, it is your wallaby and I have realized that nature seems a better way than the one ring, so I don’t sweat it or search for it. I find the best natural pattern in the individual situation in front of me and meld with it.
Nature use the carbon atom which is tetravalent – that’s a big 4.  CH4 (methane) …… on up;  builds some nice things. thumbs up

Seth says
Poe of group da 2016-02-23 09:15:56 [item 487#46522]
Your wallaby has always been to create one scheme that rules them all. One way that works for everything and is readable by humans. Cybermind was the first such. You have made many other attempts.

Personally, I don’t think that is needed. Nature doesn’t do that and yet nature functions quite amazingly fitting together in billions of diverse ways without underlying uniformity. I think the computer world would be fine that way too, and very nearly is exactly that.

But even if one did want a unified data theory for everything, modern software is not up to the challenge. To even get close to it would require a processor built to operate directly upon the underlying theory. Which I know you envisioned a Cybermind processor long ago.

In any case, it is your wallaby and I have realized that nature seems a better way than the one ring, so I don’t sweat it or search for it. I find the best natural pattern in the individual situation in front of me and meld with it.
Well if you ever actually read what i have said about such a “scheme that rules them all” you would realize that has never been my intention.  In fact, my view is almost opposite to that.   I have written much about that here and on the larger web.  Your assumption in your first sentence is just plane false.   

But incidentally we are in the 21th century and not back in 1980 and things have progressed no just a little bit since then.   Now data is usefully shared from domain to domain … you might want to think about what is behind that actually happening … maybe google big data … or ask why search engines can provide such exact information ….

google san fernando
google nathan bawden
The natural process behind this happening is nothing less than just the type of semantics that i have been promoting.    Thinking Domains could participate in this evolution.   Or not … others are learning how to share data from diverse authors and sources … let them do it, not us. 

 

Seth says
incidentally, the rule is … if you cannot express a situation unambiguously in one quad in the context in which you are operating … add quads until you are modeling the situation sufficiently such that you can. 

Holmes says
Poe of group da 2016-02-23 09:29:44 [item 487#46523]
However. I did start all this by asking you to see where I screwed the pooch and see if you could look at the quads and find a better pattern. Not a better unified editor for the current ambiguous data, but a better pattern for storing the data. I would still be interested in that if it is not too late and if it does not slow things down much overall.

One major issue to look at is the fact that there are not enough columns in a quad to store a group/user/property/value relationship in a correctly semantic way. It would need to be a quint, not a quad.
MR of group mark 2016-02-23 09:36:06 [item 487#46525]
.. or LISP which is one cell with a back & forward pointer to the next in a chain. 
seth of group seth 2016-02-23 10:06:37 [item 487#46527]
… er … not enought columns in a quad to “stor a group/user/property/value relationship in a correctly semantic way” laugh

i am going to give that one 4 laughs laughlaughlaughlaugh

… mostly for you conceptions of “what a semantically correct way” requires.  

As far as i can tell, what you have there works just fine “semantically”  … where specifically is it hurting?
Editing for one. A <group id> is not a properly expressed context. Making it a name instead of an id would not change that. It is a random value being used as a context identifier. An editor must key on something to know how to represent the rest of the data. Usually that is context.

The same problem exists for any generalized routine that must deal with the data. Any such routine (or human) would not know what a group/user/propery/value quad is in any generalized way, because each of the columns could have virtually any value. It would be a noise row to a generalized quad routine.

Our system knows because our system does not deal with the data in a generalized semantic way. It knows how ask for collections that it deals with in specific ways. It knows what the ambiguities in the table are and how to ask for things that get around the ambiguities.

Holmes says
In simpler terms. I have a user and a group and I must find the intersection between them as well as what they have on their own in as efficient a way as I can. The group/user/property/value rows provide the intersection without extra table joins. And not only for rights, but potentially for any properties where there is an intersection.

Seth says
seth of group seth 2016-02-23 10:38:41 [item 487#46536]
Editing for one. A <group id> is not a properly expressed context. Making it a name instead of an id would not change that. It is a random value being used as a context identifier. An editor must key on something to know how to represent the rest of the data. Usually that is context.
nathan



Well yeah … http://www.fastblogit.com/60 is poorly chosen for a context.
 

i would have done it more like you did users …
 
context subject verb object  
group 60 rights JSON Data the user 1810 is really an object here inside of the json data
group 60 label “tiggerandhobbs” ← needed for humans only
group 60 permanence 1 ← matches perfectly with what you did here


or you could factor all three quads into the Json data of the first quad for efficiency … no difference really .. same semantics. 

or you could have drawn the arrow from the other direction … {user 1810 rights JsonData}
Poe of group da 2016-02-23 10:55:14 [item 487#46541]
Values only go to JSON objects when they are secondary and don’t need to be looked up on the fly. Or another way of saying it is, when they don’t need to be indexed as primary level rights should be for efficiency.

Not quite seeing a relationship relating a group to  a user to a value in your table there. I can think of a way to do it like that, but it would require considerably more code to compile out into the final rights. Since rights need to be compiled on every single page build and even every AJAX call I am choosing to keep their compile time down as much as possible over a semantically perfect way to store them. At least in the semantics I have been able to think of so far.

How would you show that full relationship? I won’t tell you mine in case you do see a better way so I won’t bias you.
Lots of ways to do it … and like you say, factor into the subject and verb exactly what the efficiency of your indexing requires.  

It might be best to allow the verb to contain the full verb-object paris, in which case the object in the quad would not be used.   Do it that way when the verb does not need to be indexed.   The if the verb needs to be is indexed in the quad, to keep your system consistent, the object could be only a list of objects. ←  idea new might be an idea that you  have yet to consider.  

But the graph is the same regardless of how the quad is factored into JSON.  That certainly can be programmed.

So i don’t need to specify that here.  I’ll just show you three mentographs, drawing the arrow one direction or the other. 



this is one way
 


here we draw the arrow in the opposite diretion.  note the change of context.
 


and finally if rights themselves are the most efficient context
 

Semantically speaking all of these arrows mean exactly the same things just as unambiguously.   You get lots of choices as to how to factor … choose the most efficient for the task at hand.

Seth says
Poe of group da 2016-02-23 11:30:45 [item 487#46550]
Not to get too far ahead, but this is a side path you brought up.

Compound property names is something I have played with. It does have the advantage of allowing the table to remain more semantically intact. But it also makes queries more difficult to build and read and understand, especially when there are lots of combinations. I usually stay away from compounding names because it seems more kludgy and can grow out of control easily. But it is efficient query time wise. That’s a tough call. I am not sure that making the table more semantic is a good trade for making the queries more complicated to build, especially when no one really has a true need to deal with the tables as a semantic whole here … it is only a theoretical “nice thing”.
 
seth of group seth 2016-02-23 12:25:47 [item 487#46559]
compound names of arrows sound really  great thumbs up.

if you look at these as just labeling arrows, it really doesn’t matter how you draw the arrow, in my way of thinking it means the same one way or the other.  

the only thing to consider in that regard is how people will understand it.  but with careful  labeling, you can make any model quite comprehensible by even the weirdest of person. 
Poe of group da 2016-02-23 12:31:43 [item 487#46560]
Well even you are saying that it is not getting rid of obscurity, just moving it from one place to another. I know from experience that compound arrow SLQ queries are harder to develop and maintain. I’ll think about it in reference to the issues we have so far … but I don’t see a real clear reason to bundle the complexity in the arrow than in the columns of the table so far. It’s just six or a half dozen.
?

Seth says
moving this thought to group fbi

Holmes says
With this stratigee then a generalizer could read the compound context name and know what is in the subject and verb columns from it. All author group relationships would have the same type of name and the first part would be the actual name of the property.

right-author-group/<uid>/<gid>/JSON

 

Holmes says
Okay. I was forgetting for a moment that all of these are text columns. Even when storing an integer, it is still indexed as text. So in this case, it doesn’t matter if you combine group and author id’s as long as they have a separator character.

Holmes says
p.s. Just to keep in mind. At the current rate of change 10 years is the expected lifetime of any software or system. Notice that fbi1 falls right in line with that, completely changing now at the 10 year mark. Data has a longer lifespan, but data can generally be transformed into any other format as needed.

Seth says
Poe of group da 2016-02-23 13:12:57 [item 487#46569]
So then, domain/<gid>:<uid>/rights/JSON would work and is equally fast on the sql server side, though more complex on the building of both the read and the write queries in PHP. It is semantic. It would require parsing of the subject field in order to be used in an editor ... no direct editing ability with a generic table editor like we are doing at the moment … but the editor could be taught to deal with all such values in a reasonably generic way.

I still don’t know. It is different, but just seems to shuffle complexities across the whole spectrum around, better one place, worse another. It’s not a clear difference. The only advantage is better semantic conformance … and I am not convinced that will ever make a difference in the expected lifespan of current thinking domain software.
seth of group seth 2016-02-23 13:24:13 [item 487#46572]
I don’t know what “editor” you are talking about … remember we don’t get a quads editor according to the developer.

The time it takes to process these rights when an administrator changes them is uninmpotant.  Because that only happens once when they are changed.   The only processing time that counts is building the pages and processing them.  I really don’t know what you are dealing with here.  If you are talking about your programming time … well shucks … do the best that you can do.   If it were me, i would make the quads themselves model a reasonable way for humans to think about it,  then build once and for all a semantic browser editor with filtering that could select out what we want administrators and teachers to be able to cognize, find, and change.   Tall challenge perhaps. 
Poe of group da 2016-02-23 13:31:12 [item 487#46574]
Yes. Time is only relevant to that required to serve a page and/or process an ajax request. When I speak of time, it is always in how long it will take to know current rights in that environment ... as the rights must be compiled according to current user and group on every page and request.

Editor is in respect to a generic editor, or any generic quad aware utility, and how extensively that utility would need customization to deal with the quads in any way beyond their raw values as I do with the raw editor right now.
yes agreed.

Like i have been shouting about, if it were me, i would write a good one once … put some real work into getting it right … and then use it wherver it is needed.   Being able to do a selection filter should be able to adapt it to any particular usage.   Don’t forget we really cannot totally predict what kinds of settings we will want to let people change in thinking domains two or three years from now.   But really, we do not want to have to customize a editor for each one when it comes up?

Seth says
Poe of group da 2016-02-23 13:38:36 [item 487#46578]
I’m not a graphing kind of person. I model everything in 3D in my head and travel around and through it … looks in my head a lot how mincraft looks on the computer screen.
doesn't matter … and i have heard that before … it represents the same thing wherever  and however  you model it.
 

Seth says
Poe of group da 2016-02-23 12:45:04 [item 487#46564]
Okay. In thinking about it there is confusion. Mainly because you are calling it a compound arrow. In order to maintain both the <group id> and the <author id> as distinct and indexable variables, the only column left is the one that tells that it is a right, or what it’s property name is. If I keep that in the verb (arrow) position then it leaves context as a variable which breaks semantics.

If I make the context something like “right-author-group” then it would be fairly clean.

right-author-group/<uid>/<gid>/JSON

That looks semantic, but then it is not an arrow, it is a context.
seth of group seth 2016-02-23 13:12:00 [item 487#46568]

well that appears to be this mentograph … also pictured above. 

now who’s idea was that, yours or mine, or does it matter wink

Note that we can label an arrow with any other node.  That is the way this modeling system works.   The context defines its meaning.  That is the way semantics works.
Poe of group da 2016-02-23 13:17:52 [item 487#46570]
You say that’s the same, but what does that look like as a quad? It doesn’t seem to look like anything we have talked about. For one thing, all the actual rights can be inside the JSON. They don’t need to be selected, only applied. Only the group and user need to be selected.
seth of group seth 2016-02-23 13:25:21 [item 487#46573]
did you read what i wrote above about how such a graph translated directly into a quads?   sheeeze ...
seth of group seth 2016-02-23 13:33:18 [item 487#46575]
insidentally the quads you scoped above,  right-author-group/<uid>/<gid>/JSON, just switches <uid> with <gid> from my graph.  No real difference there. 
Poe of group da 2016-02-23 13:34:59 [item 487#46576]
No. I guess I don’t know how to read your graphs. I can’t see how my quad matches the graph at all. The switching of the gid uid doesn’t help. LOL
hmmm … now who is being obstinate … unwilling to jump to another universe.   What, do i need to do … dig up a beginners course in mentography?  I think john sowa has one.   Your kidding though … playing stupid … trying to draw me out … or what … laugh

Seth says

Here it is from the original CyberMind manual. 

Note the graph is read off directly into some language.  Here i used CyEnglish … didn’t have Json yet.

Incidentally this is not just my Bozometry … this is the way knowledge is represented.   I can go find a refresher course in RDF graphing … it will give you no more … but take you a lot longer to read.

And here is another single page primer on mentography and translating them into linear language strings … http://robustai.net/mentography/Mentography.html  … with lots of examples underneath.


Of course the simplest grock is just to realize that the circles are subjects which represent things in the world or in some other context.  The arrows are the named relationship the subjects have to each other.

Seth says
Incidentally there is quite a bit of prior and implement art re converting graphs into JSON … you might start here What is the mentography of JSON ? ...and in particular from the comments on that item … this http://decentralyze.com/2010/06/04/from-json-to-rdf-in-six-easy-steps-with-jron/  .   I will bet that this eventually made it into a W3C note if not an outright recommendation. 

surprise oh it is still a draft … https://www.w3.org/TR/rdf-json/

Seth says
Poe of group da 2016-02-23 19:05:03 [item 487#46590]
I converted the database and here are the actual query changes. Not too different.

old

myQuery("SELECT object FROM quads WHERE context IN ({1},'domain') AND subject IN ({2},'all') AND verb = 'rights' ORDER BY subject DESC, context DESC", $uid, $gid)

new

myQuery("SELECT object FROM quads WHERE context = 'domain' AND subject IN ({1},{2},'all') AND verb = 'rights' ORDER BY subject ASC", $uid, json_encode(array($gid, $uid)))
?

Seth says
Poe of group da 2016-02-23 18:36:30 [item 487#46589]
Well, I don’t think I really need to understand the grafs. What I have found is that there is one way I can see is most likely better. That is:

domain/[<gid>,<uid>]/rights/JSON

The subject term is human readable as group id and author id and is also the JSON encoded format for an array … so it can be encoded and extracted without any other parsing than the standard JSON encoder. For you in your group it would be

domain / [77,1815] / rights / {json}

There are JSON parsers on server and browser side so there isn’t any place it can be dealt with as a key or as it’s parts with no special code and JSON parsers are compiled and very fast and it is an easy pattern for generalized utilities to recognize with a very simple regular expression test to know what it is and how to translate it.

i.e. /\[d+,\d+\]/.test(value) would determine that value was a group/author key and then JSON.parse(value) would turn it into an array for direct access of the values. Even the PHP side query value would simply be json_encode(array($groupId,$authorId))   and not some squirrely concatenation of numbers rebuffered to text.

With the efficiency of utilizing a JSON array as the compound key the overall simplicity and efficiency tops all the other methods including the one I am using now.
seth of group seth 2016-02-23 19:28:36 [item 487#46593]
that works too.  i suppose then you will need to write a special customized plugin for a person to edit it. 
Poe of group da 2016-02-23 19:35:38 [item 487#46595]
It will always need custom edit code. This technique is perhaps the least weighty of them. But as I said a couple of times, we don’t want any wizard directly editing quad tables. First we would have to train them in what all the different things in the table are for. Then they could still break the whole system if they make mistakes. Wizards of any quality and knowledge should have custom editors available for feature areas that only show them, and let them change, things that are safe and matter to that feature area. Especially as the quad table continues to grow in size and features, a general quad editor is pretty much useless here to anyone except a system developer, and the prime origin.
fine sure,  if those are your assumptions, then that hags together well for you to write the code.  

On the other hand, back in the 70s i wrote a app where even clerks directly edited selected settings in substantually quads formats and even added settings themselves as necessary … the system ran for about 15 years and after i wrote it, i never changed the code again.   So it is hard for me to believe your assumption that “It will always need custom edit code”. 

But hey, do what works … i have no axe to grind here.  Certainly custom edit code can make specific known settings more user friendly to change.   Thing is, when new settings come at you willy nilly, and even wizzards are inventing new games to play,  keeping your custom settings form updated is going to take a maintence  job.

Just giving you a valid view from an alternate universe heart … bounce & jump and go for whatever works in yours

Holmes says
seth of group seth 2016-02-23 14:03:06 [item 487#46582]


give me a moment and i’ll correct the graph where rights are the context.  i forgot to draw a circle around the subjects. 

← ok corrected. 

As to reading it off in our verson of JSON … you would need to tell me which way you are ziggin on the matter that i hghlighted above as a new idea. 

 
Poe of group da 2016-02-23 18:01:11 [item 487#46588]
I understand all of them except the variations of this one. I can’t figure out a way to map what I see here to
right-author-group/<uid>/<gid>/JSON
or really to anything like that or close to that.
If I am dense I am dense. I have looked at it a dozen times over.
seth of group seth 2016-02-23 19:20:27 [item 487#46591]
I think i mentioned up there somewhere that my graph was switched from yours.  
Here is  one matches your quad: 
{ right-author-group  <uid> <gid>  JSON }

Note that i am assuming that if there is a single object it can just be put in the object field,  yet if it is a list of objects that it could be in Json object.   Yet i see no reason that it could not just be a comma delimited list right there in the object field.  That way if you think about it the only time you would need a JSON object actually in the quad would be if you wanted to code the set of pairs right in the verb – see newidea highlighed above.
And of course there is no reason when the rights that are read from the quads table into something more efficient for page loading and processing, the whole thing could be created into a Json object.
 
Poe of group da 2016-02-23 19:27:34 [item 487#46592]
Okay. Happy you get it. I still don’t. But I am probably tired and really, I don’t need to.

The reason for keeping the actual rights as a JSON object even when not necessary logically is that JSON objects combine (merge hiarchially) very easily with built in functions. It’s just an array_merge in PHP and a jQuery extend in javascript. Hence I don’t have to do anything special to compile the rights other than to the merge or extend the chain of JSON objects in the correct order. Very efficient and well worth the extremely slight cost of JSON encoding.
Okay. It’s morning. One more try to see if I can get it. What I can’t seem to find in the diagram is how a group is associated to an author. In the quad right-author-group/<gid>/<uid>/JSON the group is in the subject column and the author is in the verb column so the relationship between that author and that group is contained in that single row of the table.

How/Where does the diagram show the relationship between a particular group and a particular author so that it matches the quad definition, even loosely? It also looks like you expanded out the JSON, but that’s fine, I can account for that. It’s not seeing the group → author relationship being shown that I am confused about. It seems to me the diagram only represents ½ of the quad association.

Don’t even know if this is correct mentography … but here is how my mind would want to diagram the quad.

Holmes says
So anyway, back to the program in progress …

The following applies only to the domain context.

Not sure if you call it an arrow or what, it is not clear if an arrow can only be a verb in the terminology, but the thing that makes the whole issue resolve nicely is utilizing compound subjects. Recognizing that the subject is not an id, but is a text label, I can compound it in lots of ways to encode additional relationships. I added one more unique compounding as [groupId]. So now we have.

 
context subject verb object  
domain #### <property> JSON an author property in the domain
domain [####] <property> JSON a group property in the domain
domain [####,####] <property> JSON a bound group-author related property
domain all <property> JSON a domain property default (applies to all of the property)
domain settings <property> <any> a setting of the domain (css, sign-in thought, etc)

So the “use rule” in my thinking.domain verse becomes:

Compound subject’s only to encode relationships. Until required otherwise, always seek to keep the compounding format unique to a particular use so that it can be identified out of context if needed. Until required otherwise, only compound domain properties.

This rule seems to maintain semantic purity and encompass all the variety of relationships we need.

And one further nice thing. The particular choice of authorId, [groupId], and [groupId,authorId] naturally sorts in SQL in the correct order to be applied so that group overrides group-author overrides author. That was not planned ahead of time, but really simplifies the compiling process.

Holmes says
seth of group seth 2016-02-24 07:01:59 [item 487#46601]
In the middle of the night it dawned on me what you were getting at with your “<gid>:<uid>” wierd looking coding in the subject of a quad.  It represents, as shown in your graph, the rights of a particular person in a particular group.  It is a subject …. a most useful one. 

Mentography has alway been a science and a art.  It has evolved to precisely represent the relationship between between thingeys.  The only complaint you would get from the school of mentography would be that you didn't bother to label the arrows.  

However  you expressed the arrow labels  in the quad.  
{ domain <gid:uid> <property> JSON }  where obviously  the <property> is what is labeling the relationship between the subject (here a arrow itself) and its objects in the JSON expression.   Frequently i have found it necessary to make an arrow itself the subject or even the object of another arrow.  In classical mentography there is no such thing as a two way arrow, that would be expressed as two different arrows pointing in the opposite direction.  So you have pushed the art by doing that here … kudos yes
Okay. Well now I understand better.

I guess it doesn’t show up well here due to the size of google drawing components, but those are actually dots, not arrows, on the ends of the line between gid and uid. There is only one arrow pointing down to the JSON from that line. But I was not sure about it anyway. Makes sense to label the arrows too, I guess I skipped it because the label was not an exact label, but could be any property.

Seth says
In the middle of the night it dawned on me what you were getting at with your “<gid>:<uid>” wierd looking coding in the subject of a quad.  It represents, as shown in your graph, the rights of a particular person in a particular group.  It is a subject …. a most useful one. 

Mentography has alway been a science and a art.  It has evolved to precisely represent the relationship between between thingeys.  The only complaint you would get from the school of mentography would be that you didn't bother to label the arrows. 

However  you expressed the arrow labels  in the quad.  
{ domain <gid:uid> <property> JSON }  where obviously  the <property> is what is labeling the relationship between the subject (here a arrow itself) and its objects in the JSON expression.   Then too, you labeled the context in the quad differently than in your graph … pickey pickey wink

Frequently i have found it necessary to make an arrow itself the subject or even the object of another arrow.  In classical mentography there is no such thing as a two way arrow because mathematically all arrows are single ordered pairs … google labeled directed graph.  So your diagram expresses two different arrows pointing in the opposite direction … but of course we don’t need to draw that distinction in the data.   So your graph works … the perfect way to represent two arrows in one figure … pushing the art by doing that here … kudos yes.

Seth says
Poe of group da 2016-02-24 06:15:53 [item 487#46600]
small note It is not a programming issue, only a human readability issue, but it is not clear if static subject names should be singular or plural. For instance settings, should it be “settings” or “setting”. On the one hand each row is a single setting, but on the other hand that row is a remember of a group of settings. I am not sure which way to go. Does the semantic web have a convention?
seth of group seth 2016-02-24 07:26:18 [item 487#46605]
i always looked at the plural as the name of the class of things and the singular as a single instance of the class … {whatever <setting> isa <settings>}  … just set theory.   when i tag i can’t seem to be consistent … buggers.
Poe of group da 2016-02-24 07:29:45 [item 487#46606]
Well in this case it seems to be both, that is why I wondered about convention. It is both a setting and a member of settings. Right now it is “settings” (plural) and I guess I’ll just leave it that way. Probably will not be real easy to change much later down the road though as it is getting used more and more in the code.
 
?

Seth says
seth of group seth 2016-02-24 07:40:48 [item 487#46608]
As to the state of the art of mentography … i have presumed that i would want a new computer tool to draw them … like i used the iHmap tool to draw these http://robustai.net/mentography/  .   But really it is all about how fast one can draw and correct a graph and publish a graph.  Lately i have fond that paper and pen and an iPhone is the the fastest.  Though i think i need a sharper pen.  Alternatively a larger graph becomes tedious to redraw when it needs to be enhanced.  Oh well … mentography is like soo 80s … now people just write quads and be done with it.  Same thing really. 
 
Poe of group da 2016-02-24 07:45:36 [item 487#46609]
Yes. There actually are good tools that keep the arrows connected and allow you to move them around. In fact, though you couldn’t label arrows, floodles are just like that … both in free form mode and in auto formatted mode.

I prefer quad syntax. It is easy to understand as a human and also easy to see from the programs point of view. It is the best merge of both. Graphs are very easy for humans, but take mental gymnastics to see from the codes point of view.
the  imap tool i was using did allow for labeling arrows.  to do mentography you must label the relationsips.

Seth says
seth of group seth 2016-02-24 07:50:21 [item 487#46610]
incidentally the best grammar to write quads is still … i maintain … http://robustai.net/sailor/grammar/Quads.html smug  … though i think that should be enhanced to allow it to be embedded in thoughts here … with curlies … {all snow is white} … or … {thinking.domains seth is great} … or … {thinking.domains nathan is god}.  
Poe of group da 2016-02-24 07:57:07 [item 487#46612]
With a quick look, the examples don’t seem very readable to me as quads.
I think a good quad format should be automatically readable, not require an adjoining document to understand.
seth of group seth 2016-02-24 08:10:29 [item 487#46614]
well yes, a quick review of the examples reveals that i was thiking in a totally different universe in 2002 than obtains for us now.   The first one is a triple … showing that the context can be omitted withing some process which could imply it.   some of the others are demonstrating that the things that were being said on the semantic web of the time could well be expressed in the grammar that i was proposing. 

there is a trick that i employed that removes the dreaded namespace part of a RDF url making a quad absolutely human readable and easily writeable … yet still translatable directly into the linked data of the semantic web and big data.   That is between the lines here.    Should anyone ever get excited about deploying quads in thoughts  here … and yes it is a distraction at the moment … i would want that “trick” totally deployed.  It is what makes this a human evolvable language … understandable …. and sharable between humans and computer domains. 
Poe of group da 2016-02-24 08:20:52 [item 487#46615]
Okay. Well if you are excited about redrafting to modern standards I could probably throw it in as a live reference.
seth of group seth 2016-02-24 08:22:16 [item 487#46617]
hmmm ….
seth of group seth 2016-02-24 08:42:22 [item 487#46621]
well i could get totally excited about it but i refuse to.   why?  well just because i am more excited about actually deploying some thinking domains. 

but were i to get excited i would want the quads to actually go into a quads sql table in a protected manner.   and i think that involves making them put a button on a thought which could only be pressed by somonone with permission to do so and then only once … and that push woould assert the quad to the database. 

then too i don’t think the quads should be combined with the references.   … why? … well because they don’t need to be parsed at all … that would slow the system down. they need to be more controlled.   also it would be nice to be able to embed references inside of quads … eg
{seth-bookmarks 20000 about food}◉
… where the author clicking the “◉” would cause the quad to be posted to the database … and then might turn into a ◎.
Poe of group da 2016-02-24 08:48:18 [item 487#46622]
Okay. That’s not just a live reference, that’s a whole feature. And it shouldn’t be the thinking.domain system quad table, it should be a general quad table for the quad feature. So yes, it’s a wish list feature, I agree.
?

Seth says
seth of group seth 2016-02-24 07:50:21 [item 487#46610]
incidentally the best grammar to write quads is still … i maintain … http://robustai.net/sailor/grammar/Quads.html smug  … though i think that should be enhanced to allow it to be embedded in thoughts here … with curlies … {all snow is white} … or … {thinking.domains seth is great} … or … {thinking.domains nathan is god}.  
Poe of group da 2016-02-24 07:57:07 [item 487#46612]
With a quick look, the examples don’t seem very readable to me as quads.
I think a good quad format should be automatically readable, not require an adjoining document to understand.
Poe of group da 2016-02-24 08:00:25 [item 487#46613]
Stock floodles are mind maps, not metographs. Labeling the arrows in mind maps rarely makes sense … but I often thought of it as an option. When I go back and rework floodles I will be considering that. The problem is that it does not have a direct dictionary mapping … there is no natural place to store a label in a JSON object.
seth of group seth 2016-02-24 08:21:01 [item 487#46616]
my complaint with floodles has always been that you can only draw trees with them … and cannot practically draw networks … and a mentograph is a network … and knowledge is a network.   i know you kept claiming there was a way to do it … but it was not convenient … it was tedious … required a mental challenge  … and would never catch on.  
Poe of group da 2016-02-24 08:24:34 [item 487#46618]
I don’t know about that. Sounds like just an editor plugin issue. I only had one editor plugin and seems like I often said that. My editor was a mind map editor. But seems like the structure supported networks and just a different editor that made the network thingy easier to access would be all it needs … same editor could probably add the optional arrow labels.
seth of group seth 2016-02-24 08:26:20 [item 487#46619]
?
Poe of group da 2016-02-24 08:34:43 [item 487#46620]
My idea for a long time now, well before I started work on thinking.domains, is simply to get rid of the floodleverse and have individual floodles that exist as a document in any other document controlling system. Here in thinking domains, they would plug in as an alternate thought type with their own editor instead of the RTE, that’s all. That’s what I would go for next with floodles. Then someday, rework the floodleverse to have a better rights model … like the one I am developing here for instance.
?

Holmes says
MR of group mark 2016-02-24 08:59:54 [item 487#46626]
I liked mentography & the genius that put together CyberMind™ – was a great way to think about 2nd generation atoms of computer software. Study UML if you want to see how complex relationships, arrows & visual modeling can be before you try to fit everything into 3 or 4 chunks (triads or quads ...). LISP & Prolog were the AI languages of day – pre triads & procedural software.  Of course UML would assume you want to deal with or bother with specifications anyway.  MOF goes from the specs directly to the software – think about it – meta it – generate it –> done. 
Now back to fbi½.34
fade to grey tuit
 
seth of group seth 2016-02-24 09:01:54 [item 487#46627]
sure … why not … wink
MR of group mark 2016-02-24 09:03:19 [item 487#46628]
Yep – whatever 
seth of group seth 2016-02-24 09:03:32 [item 487#46629]
?
Poe of group da 2016-02-24 09:14:15 [item 487#46630]
People been banging on this nail for a long time. Since it is not going in, it is probably pointed at a knot. Need to move it over an inch or two. ?
MR of group mark 2016-02-24 09:16:29 [item 487#46631]
.. or maybe notice others have already nailed it! .. like those who invented fire & the wheel.
Oh no. None of those you mention nail-it anyway. All of those are defiantly bent nails!

Seth says
MR of group mark 2016-02-24 08:59:54 [item 487#46626]
I liked mentography & the genius that put together CyberMind™ – was a great way to think about 2nd generation atoms of computer software. Study UML if you want to see how complex relationships, arrows & visual modeling can be before you try to fit everything into 3 or 4 chunks (triads or quads ...). LISP & Prolog were the AI languages of day – pre triads & procedural software.  Of course UML would assume you want to deal with or bother with specifications anyway.  MOF goes from the specs directly to the software – think about it – meta it – generate it –> done. 
Now back to fbi½.34
fade to grey tuit
 
seth of group seth 2016-02-24 09:01:54 [item 487#46627]
sure … why not … wink
MR of group mark 2016-02-24 09:03:19 [item 487#46628]
Yep – whatever 
seth of group seth 2016-02-24 09:03:32 [item 487#46629]
?
Poe of group da 2016-02-24 09:14:15 [item 487#46630]
People been banging on this nail for a long time. Since it is not going in, it is probably pointed at a knot. Need to move it over an inch or two. ?
MR of group mark 2016-02-24 09:16:29 [item 487#46631]
.. or maybe notice others have already nailed it! .. like those who invented fire & the wheel.
yep the sematic nail has not worked to go in yet.   surprise omg ...this morning i almost see how by moving it a bit it just might catch hold.   and no mark, this particular nail has never took hold in the web ...not in a way that really works … yet many have tried to hammer it in. 

See Also

  1. Thought Thought, Feeling, and Will with 382 viewings related by tag "ontology".
  2. Thought Wisdom - It's What's Missing from a simple NOW based Ontology with 100 viewings related by tag "ontology".
  3. Thought about: Unhacking Wars - comment 67183 with 70 viewings related by tag "ontology".
  4. Thought An Event is something that the news reports on with 42 viewings related by tag "quads".
  5. Thought On the matter of "as itself for itself" with 27 viewings related by tag "ontology".
  6. Thought Can we feel our humanity? with 25 viewings related by tag "ontology".
  7. Thought Definition of Responsibility - self as cause with 14 viewings related by tag "ontology".
  8. Thought Thinking Domain Quads with 13 viewings related by tag "quads".
  9. Thought CyberMind Roots with 13 viewings related by tag "mentography".
  10. Thought What is Cybermind ? with 11 viewings related by tag "mentography".
  11. Thought about: cmap cloud & cmaptools in the cloud | cmap with 7 viewings related by tag "mentography".
  12. Thought symloop with 6 viewings related by tag "mentography".
  13. Thought Quads with 3 viewings related by tag "quads".
  14. Thought mentography of items with 3 viewings related by tag "mentography".
  15. Thought A mentograph of a Sticky Cyber Molecule with 3 viewings related by tag "mentography".
  16. Thought What is the mentography of JSON ? with 3 viewings related by tag "mentography".
  17. Thought Why quads? Quads vs RDF with 2 viewings related by tag "quads".
  18. Thought Loui Jover: Interesting Art Style .... with 1 viewings related by tag "mentography".
  19. Thought about: A Periodic Table of Visualization Methods with 1 viewings related by tag "mentography".
  20. Thought Tai Shu Yi King Commentary Brain uploaded with 1 viewings related by tag "ontology".
  21. Thought DJUG meeting speaker Grady Booch with 1 viewings related by tag "mentography".
  22. Thought A new ontology with 1 viewings related by tag "ontology".
  23. Thought Concept Net with 1 viewings related by tag "ontology".
  24. Thought General Rights with 1 viewings related by tag "rights".
  25. Thought items vs quads gives us forms with 0 viewings related by tag "quads".
  26. Thought Dualities listed with 0 viewings related by tag "ontology".
  27. Thought Repeating Quads Structure on tagtalking dev blog with 0 viewings related by tag "quads".
  28. Thought about: U.S. Seeks Silence on CIA Prisons with 0 viewings related by tag "rights".
  29. Thought T-based vs E-based being systems with 0 viewings related by tag "ontology".
  30. Thought form vs data with 0 viewings related by tag "quads".
  31. Thought bug: i can think in this close group when i come from the news. with 0 viewings related by tag "rights".
  32. Thought testing please be patient with 0 viewings related by tag "rights".
  33. Thought mentography of context with 0 viewings related by tag "mentography".
  34. Thought about: The Right's of Photographers (when shooting in public places) | static photography with 0 viewings related by tag "rights".
  35. Thought Going Meta with 0 viewings related by tag "ontology".
  36. Thought quads table in the database with 0 viewings related by tag "quads".
  37. Thought Relationship with 0 viewings related by tag "ontology".
  38. Thought [title (19025)] with 0 viewings related by tag "ontology".
  39. Thought [title (19026)] with 0 viewings related by tag "ontology".
  40. Thought about: ConceptNet with 0 viewings related by tag "ontology".
  41. Thought Differing ontology contexts with 0 viewings related by tag "ontology".
  42. Thought Invitations To The Steering Committee with 0 viewings related by tag "rights".
  43. Thought Quads Factoring with 0 viewings related by tag "quads".
  44. Thought about: a dialogue ... with 0 viewings related by tag "ontology".
  45. Thought Rights table with 0 viewings related by tag "rights".
  46. Thought Seriously considering developing tagtalking.net based upon ... with 0 viewings related by tag "quads".
  47. Thought Ontology with 0 viewings related by tag "ontology".
  48. Thought Pentals with 0 viewings related by tag "quads".
  49. Thought group quads project with 0 viewings related by tag "quads".
  50. Thought For humans to read and edit quads, they should be displayed and edited with humanly understood words. with 0 viewings related by tag "quads".