You sit down a user and teach him some apps, what a file system is and how to navigate. Fine. Now he creates a bunch of documents, contacts and such. All his documents are organized nicely, and contacts and such. Now lets say he wants to organize a conference.
He has to assemble the data. He has to go through one tree, get his documents together go through another tree, find all the people he knows, filter through them, and again in an address book. Throughout the maintenance he has to go through a process of grouping it together. Now lets say he meets people and adds them to his contact list. He has to go through a process of creating a new list of people related to the party, but not the initial invitation. Eventually, he decides to go work for one of these people, now he has to put a relation in that he is working with them.
Eventually, you create a lot of shortcuts in windows. Shortcuts to shortcuts to shortcuts. Now lets say you want to search all the people you work with you met at this party. You have to start navigating.
Now lets take the other extreme. Everything in a database. You have to design a schema for everything, but a query is really quick.
What I'm thinking of, is a generic interface, that may use a database underneath, or a filesystem, that allows you to easily put and pull relations of data quickly and show them to you easily. Not necessarily in a 3d interface, but a nice interface.
Someone pointed out Tinderbox Now take something like this, except instead of notes, put documents and application files there (no, not dll's, spreadsheets, contacts, your c code, recipies). Then give the ability to relate them, group them and name the relations. Finally be able to search the relations, groups and the content themselves. Now you have a knowledge system, much like the human mind. I can easily pull up data on say, love, or the perl projects i've done with certain criteria.
I don't care about the underlying technology, I'm thinking of it from a requirements level downward.
You forgot to mention domain name servers and the ip's to find out what their IP's are. All the root servers know about slashdot are what its ip's are and what the ip's of the dns servers are along with the fact that the name is slashdot.org.
All the root servers are gigantic signs with posts pointing general directions to find out more specific information.
I don't want it from a file system, i'd want it from my interface that may work with a file system. I don't like having to make copies in real life or post it notes that tell me where things are (akin to symlinks).
I'd like to be able to have all my phone numbers, names, text.. all my knowledge and be able to search by relation.
Easy one on showing how your mind works. Happy. I associate volleyball, my girlfriend, highschool, some parts of college... etc..
Now lemme think about a party i'm having. Friends (a set of people), party food (cake, snacks), cleaning the apt.. Now if I could query on computer, everything related to the party and navigate the info cleanly, great.
Now let's say I randomyl want to find my friend's phone number, i can query that too. Point being, it's not about the foundation on what the system is built, it's the human interaction that needs work.
The problem with having such a hirearchy is representing it cleanly. Creating a tree, then drawing a bunch of lines/links changes it from a tree to a blob. You can't even search by the links that exist.
Yes, but requiring training alienates the people who can't accept the training. Anyone who knows english to even a 2nd grade level should be able to make a request.
Not really. I have a document spec for when I implemented something in folder "project x".. Shit, I need it again for "project y" for a substancial time. i could waste paper or move the document. Grr, i haveo to make a copy.
It doesn't completely work. And sometimes, folders get too big and are hard to move around.
Yes! for the love of God yes! Exactly what i mean. I'm glad you didn't relate it to sql and what not. If this has a search mechanism that is english like (think star trek), kill off the visual representation of links, i think it'd be less intimidating. You could link documents etc...
Yes, but not word like. Something where you can have drop downs or navigate somewhat visually. And don't convert what i'm saying into sql.. sql is flawed fort he common man since it is not conversational english and is so syntax driven.
History on me, i was an admin for a while, a unix programmer as well.
See, if you could actually name the type of link it was, not the filename link, but what type of link. Project X, project Y, Group Z, and have an interface to provide that reverse lookup, that would own.
I would guess that you would create a link and say, these common objects use this link to a pool of "things" which would be optimally stored in their own way. When you wanna find something, find it by its object name, or link type or something.
Ah, but that's the point. It should be a fascade to the user. Make the user think things are really organized like he expects them to be so that the person is more comfortable.
When people with no experience try to work with folder son a file system, it doesn't always make sense even though to us, it's perfect.
I wouldn't use attributes as something to assign to, but quite reverse. Assign things to attributes. Such as, my phone number belongs to "me". My good friend might be in my "friends" relation, a project relation, etc etc...
*sigh* No, NOT GREP. You'd have sets of relations on things and you'd say, i want all these types of items that involve a certain relation. And I don't mean a filesystem relation but, all things related to project X, or all things related to my entertainment centre.
Think beyond files and cli. Think about having a bunch of things that have relations, ideas and what not that can be search on. Yes, grep might come into play on the lowlevel, but i'm sorry, my mom won't ever know grep.
See, the idea of keeping it in a file system which is visible to the user is what bugs me. My desk isn't a stack of recursively contained containers. It's a bunch of things that relate in terms of paperwork.
Yes, all my project stuff go in similar places, but there are times when i need to cross file things so i can easily access them, such as business cards with numbers, etc.. Problem with the physical paradyme is that it lends itself to duplication of the physical, 2 business cards in two places.
I saw that show. I don't think he means that in terms of organization for the "OS" that it didn't use some sorta tree form.
The computer he seem to describe would be able to pull up the information based on what you wanted based on a request, not on some method of searching for a file.
Slashdot Mirror
Granted, an 8 year old can be taught to behave, but what is going to stop a kindergartener from doing something distructive to a $1000 machine?
Have you tried the Xalan type stuff? http://xml.apache.org
Sorry mom. :P or is an emoticon too harsh for you either.
You sit down a user and teach him some apps, what a file system is and how to navigate. Fine. Now he creates a bunch of documents, contacts and such. All his documents are organized nicely, and contacts and such. Now lets say he wants to organize a conference.
He has to assemble the data. He has to go through one tree, get his documents together go through another tree, find all the people he knows, filter through them, and again in an address book. Throughout the maintenance he has to go through a process of grouping it together. Now lets say he meets people and adds them to his contact list. He has to go through a process of creating a new list of people related to the party, but not the initial invitation. Eventually, he decides to go work for one of these people, now he has to put a relation in that he is working with them.
Eventually, you create a lot of shortcuts in windows. Shortcuts to shortcuts to shortcuts. Now lets say you want to search all the people you work with you met at this party. You have to start navigating.
Now lets take the other extreme. Everything in a database. You have to design a schema for everything, but a query is really quick.
What I'm thinking of, is a generic interface, that may use a database underneath, or a filesystem, that allows you to easily put and pull relations of data quickly and show them to you easily. Not necessarily in a 3d interface, but a nice interface.
Someone pointed out Tinderbox Now take something like this, except instead of notes, put documents and application files there (no, not dll's, spreadsheets, contacts, your c code, recipies). Then give the ability to relate them, group them and name the relations. Finally be able to search the relations, groups and the content themselves. Now you have a knowledge system, much like the human mind. I can easily pull up data on say, love, or the perl projects i've done with certain criteria.
I don't care about the underlying technology, I'm thinking of it from a requirements level downward.
You forgot to mention domain name servers and the ip's to find out what their IP's are. All the root servers know about slashdot are what its ip's are and what the ip's of the dns servers are along with the fact that the name is slashdot.org.
All the root servers are gigantic signs with posts pointing general directions to find out more specific information.
I don't want it from a file system, i'd want it from my interface that may work with a file system. I don't like having to make copies in real life or post it notes that tell me where things are (akin to symlinks).
I'd like to be able to have all my phone numbers, names, text.. all my knowledge and be able to search by relation.
Easy one on showing how your mind works. Happy. I associate volleyball, my girlfriend, highschool, some parts of college... etc..
Now lemme think about a party i'm having. Friends (a set of people), party food (cake, snacks), cleaning the apt.. Now if I could query on computer, everything related to the party and navigate the info cleanly, great.
Now let's say I randomyl want to find my friend's phone number, i can query that too. Point being, it's not about the foundation on what the system is built, it's the human interaction that needs work.
The problem with having such a hirearchy is representing it cleanly. Creating a tree, then drawing a bunch of lines/links changes it from a tree to a blob. You can't even search by the links that exist.
Yes, but requiring training alienates the people who can't accept the training. Anyone who knows english to even a 2nd grade level should be able to make a request.
Heh, I didn't say any of it is easy. ;)
:)
"Yeah, I'll have this done by tomorrow."
Yes, but then you could limit it to "me" "4-th grade" "friends" and get a lot of info quickly.
Mind you what if you have schoolwork that has projects, or projects for clients, or you are doing a project for school that is sponsored by a client?
Not really. I have a document spec for when I implemented something in folder "project x".. Shit, I need it again for "project y" for a substancial time. i could waste paper or move the document. Grr, i haveo to make a copy.
It doesn't completely work. And sometimes, folders get too big and are hard to move around.
Doesn't the S in SAP stand for spanish? It'd be funny if they changed it to a J in this case ;D
That's ok, mine is turning into a firehazzard ;)
Yes! for the love of God yes! Exactly what i mean. I'm glad you didn't relate it to sql and what not.
If this has a search mechanism that is english like (think star trek), kill off the visual representation of links, i think it'd be less intimidating. You could link documents etc...
Yes, but not word like. Something where you can have drop downs or navigate somewhat visually. And don't convert what i'm saying into sql.. sql is flawed fort he common man since it is not conversational english and is so syntax driven.
History on me, i was an admin for a while, a unix programmer as well.
See, if you could actually name the type of link it was, not the filename link, but what type of link. Project X, project Y, Group Z, and have an interface to provide that reverse lookup, that would own.
I would guess that you would create a link and say, these common objects use this link to a pool of "things" which would be optimally stored in their own way. When you wanna find something, find it by its object name, or link type or something.
*ponders and broods*
Ah, but that's the point. It should be a fascade to the user. Make the user think things are really organized like he expects them to be so that the person is more comfortable.
When people with no experience try to work with folder son a file system, it doesn't always make sense even though to us, it's perfect.
No, not like a beos file system.
Nonononoohn.. no slashes, no "home".
I wouldn't use attributes as something to assign to, but quite reverse. Assign things to attributes. Such as, my phone number belongs to "me". My good friend might be in my "friends" relation, a project relation, etc etc...
and my friend would have its basic attributes.
Bingo, except the underlying OS doesn't have to be a database, though it might use one for relating data :P
*sigh* No, NOT GREP. You'd have sets of relations on things and you'd say, i want all these types of items that involve a certain relation. And I don't mean a filesystem relation but, all things related to project X, or all things related to my entertainment centre.
Think beyond files and cli. Think about having a bunch of things that have relations, ideas and what not that can be search on. Yes, grep might come into play on the lowlevel, but i'm sorry, my mom won't ever know grep.
See, the idea of keeping it in a file system which is visible to the user is what bugs me. My desk isn't a stack of recursively contained containers. It's a bunch of things that relate in terms of paperwork.
Yes, all my project stuff go in similar places, but there are times when i need to cross file things so i can easily access them, such as business cards with numbers, etc.. Problem with the physical paradyme is that it lends itself to duplication of the physical, 2 business cards in two places.
I saw that show. I don't think he means that in terms of organization for the "OS" that it didn't use some sorta tree form.
The computer he seem to describe would be able to pull up the information based on what you wanted based on a request, not on some method of searching for a file.
You are right. 100%. I was making a joke about the gameboy part, but I just wanted to point out why it was bit and not byte.
:)
Guess the humour went over some people's heads
Thus the lowercase letter b. if it were gigabyte, it'd be GB, like gameboy.