Codd 1970 'domain' does not mean Date 2016 'type' [was: burble about Date's IM]
Quote from johnwcowan on November 6, 2019, 5:15 pm... expressions have static types, values have dynamic types. (Even Dave called your interpretation "unusual".)
Even Dave...?
I'm not sure whether that's good or bad... . ..
I find both of your interpretations to be a bit... Nuanced. As I wrote before, I think we'd all uncontroversially agree if we were asked to identify the dynamically-typed and statically-typed parts of various languages. It's only when trying to identify them in general that the discussion corrodes.
Are you sure you don't mean that expressions (may) have dynamic types, but values have static types?
Dynamically-typed languages tend to be characterised by dynamically-typed variables and parameters. Statically-typed languages tend to be characterised by statically-typed variables and parameters, though some otherwise statically-typed languages permit explicit creation of dynamically-typed variables and parameters (e.g., C#'s dynamic keyword.)
Quote from Dave Voorhis on November 6, 2019, 5:27 pmEven Dave...?
I'm not sure whether that's good or bad... . ..
I meant it positively, as in "Dave, the master of tact on this forum".
Are you sure you don't mean that expressions (may) have dynamic types, but values have static types?
Values have fixed types in the vast majority of languages (there are a few, like Self, in which the type of a (mutable) object can be changed at run time), and in a statically typed language without inheritance the type of an expression determines the type of the expression's value perfectly, and there is no point in talking about two kinds of types. But in statically typed languages with inheritance, an expression evaluates to a value whose type is a supertype of the expression's type, and in dynamically typed languages, an expression can evaluate to any type at all.
Note that compile-time types need not be declared types in languages that do compile-time type inference, making "declared type" an extremely misleading term. In Python-plus-mypy, you get compile-time (lint-time) errors if an expression can be shown to be ill-typed even in the total absence of type declarations, although mypy is capable of taking advantage of them when they exist.
Dynamically-typed languages tend to be characterised by dynamically-typed variables and parameters. Statically-typed languages tend to be characterised by statically-typed variables and parameters, though some otherwise statically-typed languages permit explicit creation of dynamically-typed variables and parameters (e.g., C#'s dynamic keyword.)
I would say that's the definition of "dynamically/statically typed language", and insofar as C# has dynamically typed variables and parameters, it is not a fully statically typed language, just mostly.
Quote from johnwcowan on November 6, 2019, 6:03 pmQuote from Dave Voorhis on November 6, 2019, 5:27 pmEven Dave...?
I'm not sure whether that's good or bad... . ..
I meant it positively, as in "Dave, the master of tact on this forum".
Ah. Well. That's ok then. :-)
Are you sure you don't mean that expressions (may) have dynamic types, but values have static types?
Values have fixed types in the vast majority of languages (there are a few, like Self, in which the type of a (mutable) object can be changed at run time)
If an object's type changes as a result of mutation, then the object is a variable. Indeed, a mutable object is a variable (or set of variables) whether its type changes or not.
and in a statically typed language without inheritance the type of an expression determines the type of the expression's value perfectly, and there is no point in talking about two kinds of types. But in statically typed languages with inheritance, an expression evaluates to a value whose type is a supertype of the expression's type,
In statically typed languages with inheritance, an expression evaluates to a value whose type is the expression's type, but is assignable to a variable of the supertype because the type is the supertype. Because inheritance...
and in dynamically typed languages, an expression can evaluate to any type at all.
In dynamically-typed languages, an expression evaluates to a value of the type it denotes, and that value can be assigned to any variable.
My point in intentionally quibbling here is to highlight the fact that as long as we speak informally, loosely and vaguely, there will be innumerable informal, loose, and vague interpretations. Informal, loose, and vague interpretations are an open door to interminable quarrelling and quibbling, and that won't get us anywhere. Therefore, I suggest that either these discussions be around a specific language or languages -- where, as I've said before, we'll probably all agree (or at least have clear points of disagreement) -- or it should be formalised so that clear rigour can replace vague looseness.
Either way, it should take place in its own thread, too.
Quote from AntC on November 6, 2019, 9:02 amDavid, pull your head in. "Fighting endlessly" is something you do to an unacceptable extent.
Hmmm. I don't think this is warranted. Really.
Since I started this thread, and set the title, and made the first post, I'm the one to decide what this thread is "supposed to be about" from "right back at the beginning". And it certainly wasn't about "dynamically typed" -- on which you and John have contributed opinions "endlessly", but I'm none the wiser. (And thank you but I don't want any more.) Go start another thread, and stop hijacking the ones I start.
I've now read all the way back through this thread. The original topic was about domains, but over the 12 pages many deviations and detours have been taken, by yourself no less than others.
The first 'fighting' was Dave challenging my use of the term 'typeless'. We disagreed, but having stated respective positions I have no intention of 'fighting endlessly'.
A further 'fighting' was JWC use of 'dynamic types' which you disputed and I had no part in. Brian, Erwin, others all had an oar in. I did no 'fighting'.
My problem was JWC saying this:
It's a mistake to think that dynamic types occur only in dynamically typed languages. Java classes are dynamic types; so are C types (and they correspond exactly to the static types of the expressions that return them). The term dynamically typed language is a bit of a misnomer, for it refers to static typing: it is (as I said above) a language with just one static type to which all expressions belong.
I think this is nonsense, or at least deserves a clear explanation with careful definitions before it can make sense. He refuses to provide it, and along the way has 'fought' each of my attempts to gain clarity. Yes, I have persisted, but I am not attacking or defending a position, just trying to push for clarity and understanding.
My sincere apologies if you think it was I who hijacked your thread, or that it was I who was doing the 'fighting endlessly'. It was not my intention.
Anyway, no matter whatever you had in mind the TTM (with our without the IM) is a system of static types and values that belong to those types, so I guess it really doesn't matter.
Quite. Off topic.
Perhaps that last statement is closer to where the thread started than much that precedes it. But this forum software is not well suited to branching and forking threads of discussion, so perhaps this one has passed its use-by date.
Quote from dandl on November 3, 2019, 2:48 amI think what you're saying has a reasonable basis, that the principle is worthy but the naming is at issue.
It's clearly an aspect of philosophy, which means that it must surely have been discussed endlessly since the time of the Greeks (without the need for any conclusion, of course). Do you have a reference?
A reference? That the physical universe came into being at the beginning of time is what I was taught in grade school back in the '70s. It's what's called "common knowledge." But we're not here to contemplate infinity. What's relevant here is that that which is being discussed is by definition part of the Universe of Discourse, and that the physical universe is not independent of time. Which means that the physical universe, and everything that can be perceived therein is by definition concrete.
Brian
Quote from Brian S on November 16, 2019, 3:47 amQuote from dandl on November 3, 2019, 2:48 amI think what you're saying has a reasonable basis, that the principle is worthy but the naming is at issue.
It's clearly an aspect of philosophy, which means that it must surely have been discussed endlessly since the time of the Greeks (without the need for any conclusion, of course). Do you have a reference?
A reference? That the physical universe came into being at the beginning of time is what I was taught in grade school back in the '70s. It's what's called "common knowledge." But we're not here to contemplate infinity. What's relevant here is that that which is being discussed is by definition part of the Universe of Discourse, and that the physical universe is not independent of time. Which means that the physical universe, and everything that can be perceived therein is by definition concrete.
Brian
[This post seems out of context -- I'm not sure what preceded it.]
The curious thing about common knowledge is that it too is almost impossible to pin down. If we belong to a community there may be knowledge commonly held by that community, and much of it it may be dead wrong. What keeps a bicycle upright?
If we belong to different communities we may each quote common knowledge and find we disagree at a basic level. Is it polite to slurp your soup?
No, I wasn't taught that in grade school (whatever period that may correspond to). I think most people would think the universe started at a point in time (or was created), and would find the idea of a 'beginning of time' hard to accept. I learned this particular concept as one accompanying a particular cosmogenic view (Big Bang), so I do know of it, but I view it as conjecture, not fact.
I accept the Universe of Discourse as a convenient convention: the complete range of objects, events, attributes, relations, ideas, etc, that are expressed, assumed, or implied in a discussion. I accept the idea of the Physical Universe, and the idea of Time. I do not accept the particular relationship you construct between them, or the logical conclusion you draw. It just ain't that simple!
I was easily able to find philosophical discussion on the distinction between abstract and concrete eg wiki.
There are parallels with what you say, but also differences. Is a concrete example itself concrete or abstract? Is the Theory of Relativity concrete or abstract? It is certainly fixed in time, if not space. Is a subatomic particle concrete? There is lots more like this here.
So I accept your proposal as a convenient way of dealing with certain kinds of data analysis, but not more.
Quote from Brian S on November 16, 2019, 3:47 amThat the physical universe came into being at the beginning of time is what I was taught in grade school back in the '70s.
Quote from dandl on November 16, 2019, 7:42 amI think most people would think the universe started at a point in time (or was created), and would find the idea of a 'beginning of time' hard to accept. I learned this particular concept as one accompanying a particular cosmogenic view (Big Bang), so I do know of it, but I view it as conjecture, not fact.
In the scientific community, it's generally -- but perhaps not universally -- accepted that time started when the universe came into being, not the other way around. Whether you believe in a "big bang" or not, it's evident that time requires (at least) one universe in order to exist.
I accept the Universe of Discourse as a convenient convention: the complete range of objects, events, attributes, relations, ideas, etc, that are expressed, assumed, or implied in a discussion. I accept the idea of the Physical Universe, and the idea of Time. I do not accept the particular relationship you construct between them, or the logical conclusion you draw. It just ain't that simple!
Indeed. Conflation of philosophy with database theory is unhelpful. Relational database theory is a self-standing theory based on set theory and first-order predicate logic. Philosophy may provide useful applications of relational theory, but doesn't need to be included within relational theory.
In other words, if you wish to record data about things that have an existence in time vs concepts that do not, then a relational database will almost certainly be appropriate and you don't need to alter relational database theory to do so.
Quote from Dave Voorhis on November 16, 2019, 8:21 amQuote from Brian S on November 16, 2019, 3:47 amThat the physical universe came into being at the beginning of time is what I was taught in grade school back in the '70s.
Quote from dandl on November 16, 2019, 7:42 amI think most people would think the universe started at a point in time (or was created), and would find the idea of a 'beginning of time' hard to accept. I learned this particular concept as one accompanying a particular cosmogenic view (Big Bang), so I do know of it, but I view it as conjecture, not fact.
In the scientific community, it's generally -- but perhaps not universally -- accepted that time started when the universe came into being, not the other way around. Whether you believe in a "big bang" or not, it's evident that time requires (at least) one universe in order to exist.
I don't 'believe' in anything much. I leave that to the religions.
Big Bang is a good theory, but it relies on some rather extreme assumptions and has a number of awkward loose ends. It also violates a principle of symmetry, in that it requires that the universe look different now as against in the past or future. So I remain sceptical, and hopeful.
I accept the Universe of Discourse as a convenient convention: the complete range of objects, events, attributes, relations, ideas, etc, that are expressed, assumed, or implied in a discussion. I accept the idea of the Physical Universe, and the idea of Time. I do not accept the particular relationship you construct between them, or the logical conclusion you draw. It just ain't that simple!
Indeed. Conflation of philosophy with database theory is unhelpful. Relational database theory is a self-standing theory based on set theory and first-order predicate logic. Philosophy may provide useful applications of relational theory, but doesn't need to be included within relational theory.
Wrong way around: relational theory is a useful branch of logic, one of many topics addressed by philosophy. Philosophy does not concern itself much with the practical applications of a theory.
In other words, if you wish to record data about things that have an existence in time vs concepts that do not, then a relational database will almost certainly be appropriate and you don't need to alter relational database theory to do so.
Not sure what to make of that. A relational database seems to be a good fit for data that fits the basic model of fixed attributes, known quantities and regular recurrence. It seems a less good fit for data that is qualitative, irregular and uncertain. I can't see that time is an important distinguishing feature.
Quote from dandl on November 16, 2019, 12:20 pmQuote from Dave Voorhis on November 16, 2019, 8:21 amQuote from Brian S on November 16, 2019, 3:47 amThat the physical universe came into being at the beginning of time is what I was taught in grade school back in the '70s.
Quote from dandl on November 16, 2019, 7:42 amI think most people would think the universe started at a point in time (or was created), and would find the idea of a 'beginning of time' hard to accept. I learned this particular concept as one accompanying a particular cosmogenic view (Big Bang), so I do know of it, but I view it as conjecture, not fact.
In the scientific community, it's generally -- but perhaps not universally -- accepted that time started when the universe came into being, not the other way around. Whether you believe in a "big bang" or not, it's evident that time requires (at least) one universe in order to exist.
I don't 'believe' in anything much. I leave that to the religions.
Anything you hold to be true is a belief. That's what "belief" means.
A "fact" is a belief with (typically corroborated and agreed) evidence. Are there things you hold to be facts?
If so, they are the things you believe.
Big Bang is a good theory, but it relies on some rather extreme assumptions and has a number of awkward loose ends. It also violates a principle of symmetry, in that it requires that the universe look different now as against in the past or future. So I remain sceptical, and hopeful.
Hoyle et al.'s steady-state model has largely been discredited. It appears the universe did look different in the past and will probably look different in the future.
I accept the Universe of Discourse as a convenient convention: the complete range of objects, events, attributes, relations, ideas, etc, that are expressed, assumed, or implied in a discussion. I accept the idea of the Physical Universe, and the idea of Time. I do not accept the particular relationship you construct between them, or the logical conclusion you draw. It just ain't that simple!
Indeed. Conflation of philosophy with database theory is unhelpful. Relational database theory is a self-standing theory based on set theory and first-order predicate logic. Philosophy may provide useful applications of relational theory, but doesn't need to be included within relational theory.
Wrong way around: relational theory is a useful branch of logic, one of many topics addressed by philosophy. Philosophy does not concern itself much with the practical applications of a theory.
Logic isn't philosophical, though formal logic is used by certain branches of philosophy. What Brian appeared to be suggesting is that the relational model needs to take into account a philosophical interpretation of logic.
It doesn't.
In other words, if you wish to record data about things that have an existence in time vs concepts that do not, then a relational database will almost certainly be appropriate and you don't need to alter relational database theory to do so.
Not sure what to make of that. A relational database seems to be a good fit for data that fits the basic model of fixed attributes, known quantities and regular recurrence. It seems a less good fit for data that is qualitative, irregular and uncertain. I can't see that time is an important distinguishing feature.
The temporal database folks seem to think time is an important distinguishing feature, but they achieve that by building on the relational model rather than changing it.
My point is that the relational model -- as is -- is a good fit for record keeping. It is a model for data. Strange debates1 seem to result when (what appear to be) attempts are made to use it to model some aspect of reality (or fictionality resembling reality), rather than data about (perceived or intentionally fictional) reality.
--
1 Is this one of them? Not sure...
Quote from Dave Voorhis on November 16, 2019, 12:34 pmQuote from dandl on November 16, 2019, 12:20 pmIn the scientific community, it's generally -- but perhaps not universally -- accepted that time started when the universe came into being, not the other way around. Whether you believe in a "big bang" or not, it's evident that time requires (at least) one universe in order to exist.
I don't 'believe' in anything much. I leave that to the religions.
Anything you hold to be true is a belief. That's what "belief" means.
No, mostly it doesn't. Unfortunately this simple word has a lot of shades of meaning, but mostly it's taken to refer to an attitude or opinion to something said or written by someone. I found lots of definitions, but this is as good as any:
(Merriam Webster) Belief: a state or habit of mind in which trust or confidence is placed in some person or thing.
So I am careful to say that I don't 'believe' that 2+2=4. I hold it to be a true and correct consequence of the application of the laws of mathematics, but it is not a 'belief'. You are welcome to use the word in other senses of meaning, but that's how I use it.
So very little of what I know about the real physical world is a belief. Each is either a fact (an observation) or the consequence of applying logic (based on theory or model) to facts, and is innately true or false or uncertain. I may place trust and confidence in various parts of that process (people gathering facts, drawing conclusions), but often these will be in conflict. I don't 'believe' unless I'm forced to make a choice in order to take some action.
A "fact" is a belief with (typically corroborated and agreed) evidence. Are there things you hold to be facts?
It's not that simple. A scientific fact is an observation, a datum. A legal fact is one established by a court. Every document or artefact is a fact, even if what it says is false. A historic fact is just a story made to match some other fact. In ordinary usage a fact is something held to be true based on some kind of connection with reality, as against a theory or speculation. But if someone says "That's a fact" they just mean that it's true, not that it's a fact.
A good general use of 'fact' is to identify consensus knowledge, as the basis for debate or action. Such facts need not be true, but they must be agreed to have any value.
Facts are even messier than beliefs, but a proper use of fact should never confuse it with belief.
If so, they are the things you believe.
Not at all. I have here a business card for a person I know. The card is a fact, and it contains various pieces of information such as his phone number, which is another fact. I accept this as provisionally true because this my best available evidence, and I will use this phone number if I need to call him. But the card is quite old and the number may be wrong. I have no basis for a belief, I simply have a practical basis for dealing with world.
Big Bang is a good theory, but it relies on some rather extreme assumptions and has a number of awkward loose ends. It also violates a principle of symmetry, in that it requires that the universe look different now as against in the past or future. So I remain sceptical, and hopeful.
Hoyle et al.'s steady-state model has largely been discredited. It appears the universe did look different in the past and will probably look different in the future.
This is a reasonable conjecture based on available evidence (facts) and prevailing theory (models and logic). On balance for now my expectation is that this theory will be supplanted by one that is more 'universal', but almost certainly not Hoyle. I am deeply troubled by a number of aspects of the model, and with physics itself, particular as it deals with time. Fortunately there is nothing on which I need to take action, so it remains idle speculation. I don't have to 'believe' anything.
Just as an aside, if you accept BB, try using it to calculate the current size of the Universe. Hint: bigger than 14K LY. Does that make any kind of sense?
I accept the Universe of Discourse as a convenient convention: the complete range of objects, events, attributes, relations, ideas, etc, that are expressed, assumed, or implied in a discussion. I accept the idea of the Physical Universe, and the idea of Time. I do not accept the particular relationship you construct between them, or the logical conclusion you draw. It just ain't that simple!
Indeed. Conflation of philosophy with database theory is unhelpful. Relational database theory is a self-standing theory based on set theory and first-order predicate logic. Philosophy may provide useful applications of relational theory, but doesn't need to be included within relational theory.
Wrong way around: relational theory is a useful branch of logic, one of many topics addressed by philosophy. Philosophy does not concern itself much with the practical applications of a theory.
Logic isn't philosophical, though formal logic is used by certain branches of philosophy.
Really? wiki.
What Brian appeared to be suggesting is that the relational model needs to take into account a philosophical interpretation of logic.
It doesn't.
Agreed.
In other words, if you wish to record data about things that have an existence in time vs concepts that do not, then a relational database will almost certainly be appropriate and you don't need to alter relational database theory to do so.
Not sure what to make of that. A relational database seems to be a good fit for data that fits the basic model of fixed attributes, known quantities and regular recurrence. It seems a less good fit for data that is qualitative, irregular and uncertain. I can't see that time is an important distinguishing feature.
The temporal database folks seem to think time is an important distinguishing feature, but they achieve that by building on the relational model rather than changing it.
My point is that the relational model -- as is -- is a good fit for record keeping. It is a model for data. Strange debates1 seem to result when (what appear to be) attempts are made to use it to model some aspect of reality (or fictionality resembling reality), rather than data about (perceived or intentionally fictional) reality.
Agreed. Although with that proviso: there are kinds of record keeping for which the relational model is a less good fit than others.