OK, you had the grace to admit your minor mistake, so let me try once more to show you your major one.
The relational calculus of which you spoke is one way to manipulate it!
Exactly! And it's not expressive enough. Pure and simple. That's my point.
Nope, read Codd's paper if you don't believe me.
Basically the RM is (taken from DBDebunk.com):
Theory: predicate logic and set mathematics
Structure: R-tables (precise definition!)
Integrity: domain, column, table and database integrity
Manipulation: R-operations (restrict, project, join, etc.)
I'm sorry, I don't know what point you think you're proving (to someone who has read Codd's paper thank you very much, as well as having spent years developing commercial database applications) but that just restates that data is stored in relations (yes, which are a part of set theory) and manipulated by the relational calculus (the operators of which are essentially those listed under 'manipulation').
Furthermore, as I've said, that the relational calculus exists as a predicate (understand: first order) logic over these.
That is NOT expressive enough.
What example of business data would require you to store different entities in the same relation?
You sound like an IBM salesman in the 1970s!
What do you mean "business data"? (this is a thread about the semantic web).
What do you mean "want to store [...] in the same relation"? I don't want to store them in a relation, they're nodes of a tree!
The RM is a logical model -- if you were a DBMS maker and you really wanted to implement (e.g. on the physical level) a hierarchy as a series of nodes and pointers you can. However, from an application standpoint you don't need to store the hierarchy in any other fashion.
Yes, it's a logical model that provides an interface to the user.
Not just an interface for inserting data, but retrieving and manipuating it as well! The relational calculus is a (surprise, surprise) a system of calculation over RELATIONS!. If I can't express my manipulations within that model, I'm stuffed! (Or rather stuck doing manipulations by hand in a computationally complete programming language extension... say PL/SQL)
The RM provides for declarative constraints AND declarative manipulation tools which are far simpler than the procedural counterparts (and as we all know, the simpler the better, less buggy, etc.).
The constraints RM provides for my data are the type of relations and the referential integrity between them; only these are preserved by the operations of the relational calculus. That nearly all commercial database implementations provide extra constraints that the engine is checking alongside sound relational calculus manipulations is an admission that the model is too weak!
As for your manipulations, they're only declarative if they're expressible in the logical model... And mine are NOT!
The constraints that I'm talking about are precisely the tree structure itself, the form of nodes and their linkage (just as the constraints you can express it the relational model are the forms of relations and the links between them - I'm not trying to claim that all data fits the model I'm defending, btw - that's your job!)
You do have a point, though: what if further constraints are needed than those that can be expressed structurally? Well, given hierarchical data (again, I'm not trying to claim that all data fits this), if my logical model is that of inductive types, I can express further constraints over this structure (cf. XSD and constraints expressed in XPATH, XPATH directly being over the XML structure). If, otoh, I've lost the structure (by flattening into relational form) then, if my constraints have a structural component, how the hell are you going to allow me to express them? Again, you can't (by anything that builds conservatively over the relational model).
The relational model is not set theory, it's the relational calculus.
How can you possibly say I've provided no substance when I've given you an example (both structure and a trivial manipulation), and all you've done is talk about representation?
What more I want is to manipulate my data, not just represent it!
In my example it's precisely my point that the different nodes cannot be in the same relation (and cannot be queried over if a first-order manner)... but the point is that they are constructors of the same inductive data type! This is the problem in your inexpressive representation. I'm sorry, but putting this down to my confused design is pure garbage - as the anonymous poster (iirc) said, your argument is basically along the lines: "if it doesn't fit my model it doesn't belong"!
Once again I know that you can flatten trees into relations. What I'm saying (over and over), is that if you flatten them, and if you query and insert relations, you lose the ability to manipulate your data in the form is actually takes.
(And, yes, a tree can be viewed as a specialised graph - an acyclic one. You said that there are no graphs that have cycles, which would mean that graphs and trees are identified. I'm sorry, but I think you're a moronic little troll with no understanding of what you're talking about and I'm not going to discuss this further...)
As it happens DC and Virginia are the only places I've yet visited, but that my prejudice was enforced is precisely because when I was down in Langley: people's eyes widened when I asked about the bus (clearly, the cleaners use them, not the white elite); people did a double take when I said I didn't need a lift to go get some groceries, I'd walk; I couldn't even get onto the NASA compound unless I was in a car!
In all of these cases not using a car would have been more than practical.
Now, I did visit the stretched-out leafy suburbs in DC, so I do see what you mean. All the same, I chose to live in a smaller house than I might afford (in a country where people often commute 3 hours a day and I have before) partly in order that I can walk to work. How often do you hear an American say that? (Not singling you out exclusively, South Africans are the same for instance, but the story already provided that leaning.)
American population 290,342,554
Chinese population 1,286,975,468
(CIA Factbook, current)
American oil consumption: 19.7
Chinese oil consumption: 5.3
(Million barrels per day - Non-OPEC Fact Sheet, 2002)
By my calculations that means that Americans use more than 16 times as much per person.
Yeah, nice one, mate!
Am I missing something, though, or have you not said on your blog whether the fuel's actually effective?
I wanted to mention the UK City Councils that are experimenting with running the buses on biodiesel - do you have any links?
Come on, you're splitting hairs. You know as well as the rest of us that the term 'emissions' is used in a particular established (albeit inconsistent with the real meaning) sense... Do you troll around every time someone says 'zero emissions'? If so you must have an interesting life...
As for the net effect on CO2, even if you are so ignorant you might as least have read the thread above you; a direct response to you shouldn't even have been necessary.
Doesn't matter what your 'need' for foreign oil is when it's all gone!
It's sad that that's what it might take for your consumption to slow. It's heartening that it might destroy your economy if you do leave it until the last minute to change...
I think the point is that SUVs are unneccessarily large and fuel-greedy - a general criticism of your cars that Europeans have made for decades (and don't try to kid me that Americans even think of using public transport in most states).
What's undeniable is that there are only finite reserves of fossil fuels and it's about damned time that Americans, as one of the biggers consumers, did something about implementing these already-proven alteratives.
"The question would be:
What would you do with such a database?"
This kind of data is everywhere. Why do you think XML exists? Why do you think functional programming languages have such a following?
But more practically, from what it seems, those would not be in the same relation because they do not have the same schema. So the question is sort of moot.
Huh? You say I can store a tree, but that I can't query for a (sub)tree is 'moot'?!?
provided there are no cycles in the graph (which would make it not a graph by definition)
I think you're confusing graphs and trees.
Why do you think that hierarchies cannot be modeled in a relation or that you cannot manipulate them in a relational system?
Maybe I should talk to the wall instead... no, I'll tell you what, I'll get back to my banana farming!
"Bizarro" language? For manipulating inductive (and algebraic) data types? Perhaps if you've been living in a cave...
OK, imagine a tree (if that's how you like to see this), where the node type has three different constructors:
A - Has a string attribute and one child node
B - Has an integer attribute and two child nodes
C - Has a real number attribute and no child nodes
Now I pick a particular node, b, of type B.
First I want to return the subtree rooted at b.
How can I do this when your language for manipulations only returns for me relations? (And these nodes won't go into the same relation even if you could recurse through the children to arbitrary depth to find them.)
Second, by extension, I want to remove one of the subtrees under b and replace it with a copy of the other.
These are trivial and neccessary manipulations, but are they trivial once you've thrown away your structure?...
Back to the same old misunderstanding in a circle.
Well, I'm sold - let's reduce all structure down to the relational model (just because we can) and then require a computationally complete language to compute, by hand, all the constraints and manipulations. (No, I'm not talking about a manipulation which is first-order over relational calculus... that was the point!)
So what RDBMS has an alternative to SQL where I can copy the subtree beneath a particular node in an instance of arbitrary size within an inductive data type?
I mean that we can do more given the power that we have. Like complex constraints on, and relationships between, hierarchical (algebraic) data, not just trivial contraints (identification of parts with keys and un-standardised extensions) on flattened representations.
Man, I've just seen the light - inductive data structures in programming, too...
Throw it all out and stick with the relational model!
It's such a happy coincidence that what was easy to index and manipulate with the computational power and storage requirements of the 1970s is the perfect model for all data forever. Hooray, no need to go into work tomorrow: I'll just grow bananas... out of my ass!
I think that was the point being made (and in a useful way, i.e. with the very argument I also use to explain this to people who neither understand the construction intuitively nor technically).
The Umax/Yamada player did not die, this is precisely my point! After the 6100, which I bought, a 6600 with further capabilities has been released. Both are actively supported with firmware upgrades and there's a large community of users, represented for instance at my-yamada. It just doesn't have the build-quality of big name, large development budget, hardware.
Don't even talk to me about MP3 players - Apple made the iPod so they could sell music, not because of the hardware demand! In fact the major players stayed away from MP3 players, to start with, and all we got was garbage... except for Creative - for once (tell me if I'm wrong) here was someone who did have lots of manufacturing capability, but didn't have a commercial interest in keeping the status quo (again, correct me if I'm wrong), and furthermore did have an interest in developing digital music. Again, it's about the bigger picture, not about the isolated, blinkered application of the dogma that capitalism is a positive and unstoppable force.
Again, we agree about something - there is no covert conspiracy (that's what I was saying, you're the only one trying to put those words in my mouth). All I'm saying is that capitalism does not apply only in little vacuums is not a positive force for the individual. Some of the only people who believe otherwise are those that identify too much with their country: "what's good for America is good for me!" (I'm sorry if this offends you, or if I'm wrong about you, but as far as my education, I have a flawless academic record at an English public school, a first class honours degree, followed by three research contracts and one teaching contract at a good university where I'm working on my PhD thesis... let's not take the shit-flinging any further!)
Right, because Sony, Pioneer, these kind of companies have no connection to DVD Video publishing, huh?
And if they don't make the devices, capitalism dictates that someone else will step in and do as good a job, right? (No offense, but you're American, right?)
Take a similar existing example: there's a huge demand for standalone video players with alterative codecs (DivX, XviD, Vorbis etc.) and packages (OGG, MKV). What's the capitalist response? The big players are not interested and the demand is (not) met by the Umax/Yamada player (and not a lot else), which sucks!
It's not about 'them and us', not even about 'them, us and a third party' (RIAA), it's just about 'them' and their many and intertwined commercial interests...
Number one distinction between PayPal and a bank - there's no independent regulator to whom I can (could) go when PayPal starts demanding that I give up my own privacy in order to get access to my money!
Yes. And aside from the pure technical issues, Hollywood simply does not want us to have a compatible double-layer video medium for obvious reasons...
I'll believe it when I see it!
You have a cheap DVD-player with a tiny buffer apparently.
Whereas you clearly have a massive one... buffer, that is!
(No, seriously, whip it out and let's see, you've really impressed me and, I'm sure, everyone else...)
Slashdot Mirror
OK, you had the grace to admit your minor mistake, so let me try once more to show you your major one.
Exactly! And it's not expressive enough. Pure and simple. That's my point.
I'm sorry, I don't know what point you think you're proving (to someone who has read Codd's paper thank you very much, as well as having spent years developing commercial database applications) but that just restates that data is stored in relations (yes, which are a part of set theory) and manipulated by the relational calculus (the operators of which are essentially those listed under 'manipulation').
Furthermore, as I've said, that the relational calculus exists as a predicate (understand: first order) logic over these.
That is NOT expressive enough.
You sound like an IBM salesman in the 1970s!
What do you mean "business data"? (this is a thread about the semantic web).
What do you mean "want to store [...] in the same relation"? I don't want to store them in a relation, they're nodes of a tree!
Yes, it's a logical model that provides an interface to the user.
Not just an interface for inserting data, but retrieving and manipuating it as well! The relational calculus is a (surprise, surprise) a system of calculation over RELATIONS!. If I can't express my manipulations within that model, I'm stuffed! (Or rather stuck doing manipulations by hand in a computationally complete programming language extension... say PL/SQL)
The constraints RM provides for my data are the type of relations and the referential integrity between them; only these are preserved by the operations of the relational calculus. That nearly all commercial database implementations provide extra constraints that the engine is checking alongside sound relational calculus manipulations is an admission that the model is too weak!
As for your manipulations, they're only declarative if they're expressible in the logical model... And mine are NOT!
The constraints that I'm talking about are precisely the tree structure itself, the form of nodes and their linkage (just as the constraints you can express it the relational model are the forms of relations and the links between them - I'm not trying to claim that all data fits the model I'm defending, btw - that's your job!)
You do have a point, though: what if further constraints are needed than those that can be expressed structurally? Well, given hierarchical data (again, I'm not trying to claim that all data fits this), if my logical model is that of inductive types, I can express further constraints over this structure (cf. XSD and constraints expressed in XPATH, XPATH directly being over the XML structure). If, otoh, I've lost the structure (by flattening into relational form) then, if my constraints have a structural component, how the hell are you going to allow me to express them? Again, you can't (by anything that builds conservatively over the relational model).
The relational model is not set theory, it's the relational calculus.
How can you possibly say I've provided no substance when I've given you an example (both structure and a trivial manipulation), and all you've done is talk about representation?
What more I want is to manipulate my data, not just represent it!
In my example it's precisely my point that the different nodes cannot be in the same relation (and cannot be queried over if a first-order manner)... but the point is that they are constructors of the same inductive data type! This is the problem in your inexpressive representation. I'm sorry, but putting this down to my confused design is pure garbage - as the anonymous poster (iirc) said, your argument is basically along the lines: "if it doesn't fit my model it doesn't belong"!
Once again I know that you can flatten trees into relations. What I'm saying (over and over), is that if you flatten them, and if you query and insert relations, you lose the ability to manipulate your data in the form is actually takes.
(And, yes, a tree can be viewed as a specialised graph - an acyclic one. You said that there are no graphs that have cycles, which would mean that graphs and trees are identified. I'm sorry, but I think you're a moronic little troll with no understanding of what you're talking about and I'm not going to discuss this further...)
As it happens DC and Virginia are the only places I've yet visited, but that my prejudice was enforced is precisely because when I was down in Langley: people's eyes widened when I asked about the bus (clearly, the cleaners use them, not the white elite); people did a double take when I said I didn't need a lift to go get some groceries, I'd walk; I couldn't even get onto the NASA compound unless I was in a car!
In all of these cases not using a car would have been more than practical.
Now, I did visit the stretched-out leafy suburbs in DC, so I do see what you mean. All the same, I chose to live in a smaller house than I might afford (in a country where people often commute 3 hours a day and I have before) partly in order that I can walk to work. How often do you hear an American say that? (Not singling you out exclusively, South Africans are the same for instance, but the story already provided that leaning.)
American population 290,342,554 Chinese population 1,286,975,468 (CIA Factbook, current) American oil consumption: 19.7 Chinese oil consumption: 5.3 (Million barrels per day - Non-OPEC Fact Sheet, 2002) By my calculations that means that Americans use more than 16 times as much per person.
He was talking to the DSL provider who, more than likely, were using PPPoE...
Yeah, nice one, mate! Am I missing something, though, or have you not said on your blog whether the fuel's actually effective? I wanted to mention the UK City Councils that are experimenting with running the buses on biodiesel - do you have any links?
Come on, you're splitting hairs. You know as well as the rest of us that the term 'emissions' is used in a particular established (albeit inconsistent with the real meaning) sense... Do you troll around every time someone says 'zero emissions'? If so you must have an interesting life...
As for the net effect on CO2, even if you are so ignorant you might as least have read the thread above you; a direct response to you shouldn't even have been necessary.
I've a feeling you're right, but neither you or the poster you're responding to have tried to back up your opinions (disguised as facts).
And none of those could not be produced by other (renewable) means...
You're taking the high ground by suggesting that your country concentrate on using up the planet's limited resources in other ways?!?
You deserve a medal...
Doesn't matter what your 'need' for foreign oil is when it's all gone! It's sad that that's what it might take for your consumption to slow. It's heartening that it might destroy your economy if you do leave it until the last minute to change...
I think the point is that SUVs are unneccessarily large and fuel-greedy - a general criticism of your cars that Europeans have made for decades (and don't try to kid me that Americans even think of using public transport in most states).
What's undeniable is that there are only finite reserves of fossil fuels and it's about damned time that Americans, as one of the biggers consumers, did something about implementing these already-proven alteratives.
This kind of data is everywhere. Why do you think XML exists? Why do you think functional programming languages have such a following?
Huh? You say I can store a tree, but that I can't query for a (sub)tree is 'moot'?!?
I think you're confusing graphs and trees.
Maybe I should talk to the wall instead... no, I'll tell you what, I'll get back to my banana farming!
"Bizarro" language? For manipulating inductive (and algebraic) data types? Perhaps if you've been living in a cave... OK, imagine a tree (if that's how you like to see this), where the node type has three different constructors: A - Has a string attribute and one child node B - Has an integer attribute and two child nodes C - Has a real number attribute and no child nodes Now I pick a particular node, b, of type B. First I want to return the subtree rooted at b. How can I do this when your language for manipulations only returns for me relations? (And these nodes won't go into the same relation even if you could recurse through the children to arbitrary depth to find them.) Second, by extension, I want to remove one of the subtrees under b and replace it with a copy of the other. These are trivial and neccessary manipulations, but are they trivial once you've thrown away your structure?...
Back to the same old misunderstanding in a circle. Well, I'm sold - let's reduce all structure down to the relational model (just because we can) and then require a computationally complete language to compute, by hand, all the constraints and manipulations. (No, I'm not talking about a manipulation which is first-order over relational calculus... that was the point!)
So what RDBMS has an alternative to SQL where I can copy the subtree beneath a particular node in an instance of arbitrary size within an inductive data type?
I mean that we can do more given the power that we have. Like complex constraints on, and relationships between, hierarchical (algebraic) data, not just trivial contraints (identification of parts with keys and un-standardised extensions) on flattened representations.
Man, I've just seen the light - inductive data structures in programming, too... Throw it all out and stick with the relational model! It's such a happy coincidence that what was easy to index and manipulate with the computational power and storage requirements of the 1970s is the perfect model for all data forever. Hooray, no need to go into work tomorrow: I'll just grow bananas... out of my ass!
I think that was the point being made (and in a useful way, i.e. with the very argument I also use to explain this to people who neither understand the construction intuitively nor technically).
The Umax/Yamada player did not die, this is precisely my point! After the 6100, which I bought, a 6600 with further capabilities has been released. Both are actively supported with firmware upgrades and there's a large community of users, represented for instance at my-yamada. It just doesn't have the build-quality of big name, large development budget, hardware.
Don't even talk to me about MP3 players - Apple made the iPod so they could sell music, not because of the hardware demand! In fact the major players stayed away from MP3 players, to start with, and all we got was garbage... except for Creative - for once (tell me if I'm wrong) here was someone who did have lots of manufacturing capability, but didn't have a commercial interest in keeping the status quo (again, correct me if I'm wrong), and furthermore did have an interest in developing digital music. Again, it's about the bigger picture, not about the isolated, blinkered application of the dogma that capitalism is a positive and unstoppable force.
Again, we agree about something - there is no covert conspiracy (that's what I was saying, you're the only one trying to put those words in my mouth). All I'm saying is that capitalism does not apply only in little vacuums is not a positive force for the individual. Some of the only people who believe otherwise are those that identify too much with their country: "what's good for America is good for me!" (I'm sorry if this offends you, or if I'm wrong about you, but as far as my education, I have a flawless academic record at an English public school, a first class honours degree, followed by three research contracts and one teaching contract at a good university where I'm working on my PhD thesis... let's not take the shit-flinging any further!)
Right, because Sony, Pioneer, these kind of companies have no connection to DVD Video publishing, huh? And if they don't make the devices, capitalism dictates that someone else will step in and do as good a job, right? (No offense, but you're American, right?) Take a similar existing example: there's a huge demand for standalone video players with alterative codecs (DivX, XviD, Vorbis etc.) and packages (OGG, MKV). What's the capitalist response? The big players are not interested and the demand is (not) met by the Umax/Yamada player (and not a lot else), which sucks! It's not about 'them and us', not even about 'them, us and a third party' (RIAA), it's just about 'them' and their many and intertwined commercial interests...
The name Quark is trademarked. Do they acknowledge either Murray Gell-Mann or James Joyce?... (Maybe another law suit is looming ;)
Number one distinction between PayPal and a bank - there's no independent regulator to whom I can (could) go when PayPal starts demanding that I give up my own privacy in order to get access to my money!
Yes. And aside from the pure technical issues, Hollywood simply does not want us to have a compatible double-layer video medium for obvious reasons... I'll believe it when I see it!