Oh, dear — Chris Date is displeased with me
Chris Date is quite annoyed with me, and has taken issue with various things I’ve written. Some of his reasoning is hard to follow. For example, he said something to the effect that it would be silly for him to ever say anything misleading, because he’d immediately be caught out. Uh, Chris – you’re the guy who’s berating the terrible level of education and understanding in a field for which YOU WROTE THE DEFINITIVE TEXTBOOK (which has sold “over 700,000 copies”). If your readers can’t even understand the correct things you say in your book, why should they be able to instantly spot the errors?
Even odder is Date’s multi-paragraph diatribe about a specific two-word phrase I supposedly used. Maybe if he’d looked and read what I’d really said, it would have made more sense to him. Or maybe not. He also bragged about not knowing who I am, thus revealing a couple of things. First, he can’t be bothered to use Google any more than he can be to actually read a blog post he’s criticizing. Second, he’s pretty out of touch with the actual DBMS industry, something which can be independently inferred from his frequently bizarre statements about DBMS vendors and their technology. (Chris Date may still be one of the world’s great experts on the use of DBMS, but when it comes to actually building them, he seems to be pathetically ignorant.)
Other things he says in that piece, however, are sufficiently coherent to warrant an attempt at response. So here goes.
He starts by asking, in effect, “Why are you picking on me?” Well, I didn’t start out by intending to pick on Chris Date. But Fabian Pascal engaged me in flaming discussion, and when he found himself out of his depth, quickly resorted to what he seemingly thought was a discussion-ender, namely dropping the sanctified name of “Chris Date.” So I did a little poking around, and discovered that Date indeed has explicitly endorsed Pascal and Pascal’s website, posts on Pascal’s website “on a fairly regular basis,” lets Pascal speak for him, and generally seems in my opinion to be responsible for and indeed to apparently agree with the views Pascal ascribes to him.
Beyond that, I’m picking on Date because he is misleading his paying audience(s) on the subject of TransRelational technology. In that regard he’s acting like so many other DBMS vendor marketing spokespeople, who may not have been precisely lying, in that they probably deluded themselves before peddling nonsense to others. But at least John Cullinane and Dave Peterschmidt were shipping actual, useful products. Nor did they charge for seminar admittances or book sales just so that people could hear their pitches.
Moving on, Date claims that his views are scientific in general nature. However, he ignores overwhelming evidence against them, namely the last 20 years of development and use of relational DBMS, and presents almost no empirical evidence for them. So yes, I stand by my claim that the pure-relational fanatics are “reasoning” in a quasi-religious manner much more than they are reasoning scientifically (in any sense of “science”).
This ties into the claims that to be against relational theory is to be against logic, that relational theory is based on science that was established over 2,000 years ago, and so on. The most outrageous versions I’ve seen of such hokum indeed came from Pascal’s keyboard rather than Date’s. But as noted above, Chris Date supports Fabian Pascal, and anyhow what Date himself says is bad enough.
About the most expansive valid claim that can be made along those lines is “Sound mathematical predictions can be made about the behavior of systems built in conformance to the Relational Model, and the same is not at this time equally true of systems that do not conform to the RM. In particular, it is possible to make mathematically sound assurances about data integrity.” That’s a strong argument for using the RM in certain contexts – quite a few contexts, actually. But it hardly refutes the central claims of the DBMS2 agenda, nor is it a strong argument against the use of today’s SQL-oriented DBMS. And it certainly doesn’t excuse the bitter insults Date and Pascal regularly hurl at both the DBMS user and DBMS developer communities. Nor does it justify their vitriolic attacks on me. Nor does it support many other things that Date and/or Pascal have said.
I think this covers most of what Date said in his screed. Oh, he had a lot more numbered points than that, but they seemed to repeatedly touch on the issues mentioned above. As for the less personal, and much more important, aspects of our disagreement – well, this whole blog repeatedly covers those subjects.
Finally, a bit of self-examination here – is my disagreement with Date really that important, or am I like Captain Queeg with the missing strawberries, trying to reenact the successes of my younger years (e.g., Cullinet and Sybase)? Well, Required Technologies surely is not an important target – but nor have I focused on it. My real target is any claim that a large enterprise should obsessively try to run itself on One Grand Centralized relational database (especially one that actually ships commercially; I’ve been assuming that no enterprise would try to bet the farm on an idealized “true relational” product that lacks the key buying criterion of “it actually exists.”). It just so happens that the shrillest initial opposition to this idea came from Chris Date’s spokesman Fabian Pascal, and it’s Pascal who brought Date into the conversation.
Comments
26 Responses to “Oh, dear — Chris Date is displeased with me”
Leave a Reply
This post raises several questions, over and over again.
It doesn’t sound that way – he seemed bewildered that you dragged his name into your criticisms of Pascal. Lumping them together is pointless, at least in this context.
Googling might reveal your resume and some writing – what useful information might he derive from that about your “criticisms” of him?
“Pretty much out of touch” is vague – can you elaborate? The “inference” refers to “frequently bizarre statements” – any examples? Is theoretical a synonym of “bizarre”? Your use of these terms isn’t grounded in anything concrete that I can find. I agree that the claim to “science and mathematics” requires further qualification, but I don’t see what principle or statement or purpose your criticisms of Date hang on. Those with Pascal got quickly personal on both sides, and who “fired first” is irrelevant to a discussion about Date.
This is strange – an expert on Theory X (in any field) must also be an expert on building products based on Theory X (not that much is actually based on relational)? And what do you base your “pathetically ignorant” comment on? From reading his books, he discusses in adequate (not exhaustive) detail implications of relational theory and algebra on optimization, data migration, and several other “real-life” problems.
Endorsing someone’s site, and some of their writings, doesn’t imply a blanket endorsement of everything they write or say. That sort of generality serves no debate well, nor is it either useful or kind to level attacks meant for Pascal at Date. And what does “lets Pascal speak for him” mean? I don’t think that’s what’s going on, whatever it is.
What views are you talking about that are so offensive? I’ve read a good number of your posts, and I still don’t know. Is it mostly about TransRelational? Aside from that, what’s the dispute?
Date referenced TransRelational only in his 8th edition of “Introduction to Database Systems,” not before. He’s writing a book on it now. There’s plenty of online articles, and books with no reference to TransRelational. He doesn’t appear to be selling it.
So now no one can write a book on any topic unless he or she has “shipped a product” based on that subject? This really would be a massive assault on… well, any sort of education, theory, abstract subject matter, etc. Is the “product” (a sold good, I assume) critical to the theory it’s based on? I agree that a product built and shipped is some validation of its underlying theory – but it’s neither a sufficient nor even a necessary condition for a valid theory. You’ve elsewhere disputed the value of software that doesn’t meet some (unspecified) market share – so selling a product isn’t enough, but rather the profit margin?
I’m just really confused on what makes an idea “valuable” to you.
What did he say that’s so bad? I hate to keep asking, but you keep referring to statements I can’t find. And I don’t think the “support” that Date gives to Pascal extends to everything on dbdebunk.com. For example, I have no clue at all about Date’s politics, and wouldn’t assume that Date endorses the political links on dbdebunk.com. Pascal links to writing he does at various sites; I don’t see Date explicitly endorsing many of those. You seem to take an association for the purpose of promoting relational theory as unequivocal support for all viewpoints.
Such as? Pascal points out the lack of education, and gives numerous online examples of ignorance of relational theory (specifically when people use the word “relation” incorrectly). Any search for the word “relational” on Google will reveal (at least) 90% incorrect usage. So lack of education, in relational matters, is a fact. I don’t think point that out is an insult – you could actually measure it, albeit somewhat crudely. (The 90% is my guess only.)
In any event… that’s Pascal. Date is much more diplomatic. What “insults” are you talking about?
Vitriolic? Hardly. Again, perhaps Pascal – not Date. I just re-read his response to you, and if there’s bile there, please point it out.
I will venture that Date and Pascal have both supported their argument far more strongly than you’ve supported, for example, your discussion of DBMS2. It depends on your definition of “support,” of course.
In any event, I’m still disappointed at the lack of coherent enumeration of your beefs with Date. I can see a heated set of emails – Date is obviously somewhat upset in his post – but nothing I’d call vitriol.
Eric,
I’m not holding Date responsible for all of Pascal’s views. I am, however, holding Date responsible for Date’s views as attributed to him by Pascal. I don’t see how you can argue with that. I’m further holding him responsible for Pascal’s formulation of views that they both share; if you think that’s unfair, I suggest you suggest to Date that he spell out the differences between his and Pascal’s opinions on the subjects they both focus on, perhaps in the way he sometimes explains his differences with Codd.
The TransRelational(TM) hucksterism is addressed in other threads, and I won’t follow up on it in the comments here.
The reason I call Date’s shrill comments on current DBMS products “bizarre” is that they radically exaggerate the degree of intelligence it is practical to engineer into commercial sofware, at least in a performant way. Other than a lot of invective, and some very odd general comments like the one about how great commercial DBMS’ caching systems are (they aren’t — and if they really were, Oracle wouldn’t have just bought TimesTen), he and his followers offer few specifics EXCEPT for TransRelational(TM). Basically, Date is just insulting vendors for not yet having solved extremely difficult AI problems — and unless he’s a total sleazoid, which I greatly doubt, the reason he’s doing so probably has a lot to do with not understanding how tough the problems really are. I.e., when Pascal and Date accuse others of “ignorance” on that score, they themselves are actually the ignorant ones in the discussion.
All that said, I’m working on some posts in which I confess to myself knowing less than I would like to about certain areas of DBMS engineering, so you might want to take a look at those when they’re available.
I’m not. I just don’t know which views you’re talking about specifically – besides the verbal jousting.
Commercial DBMSs? Like operating systems and languages, I think it’s far better in general to engineer a great deal of “intelligence” into those products in order to ensure the users (developers) have a great deal of simple power at their disposal. I’ve read some (not many) design reviews of various aspects of Oracle, for example – they’re doing some very complicated things, some forced by the awkward inconsistencies of SQL. I think that relations per se give you a much better formal basis for building that complexity – they make the complex simpler. Date explains that fairly well in his books. They’re not AI – not even close. Cost-based optimization has already explored many of the same areas in gory detail.
I can’t comment on the caching comments Date has made – can’t recall seeing them.
Date’s writing at least points out that it’s not necessary to solve all of the problems – that the lack of proper logical-physical separation, and the bizarre corners of SQL, force them to solve problems that didn’t have to exist in the first place.
Date’s rather short “Database In Depth” has some discussion; his Introduction, much longer, has more about the potential (and easy to examine) impact of relations on DBMS capabilities. I’m not saying this is proof, but when I compare what he imagines with what vendors actually do, I don’t think relations pose problems any more complex than what vendors have already done.
From the original post:
I disagree. If you’re talking purely in terms of acceptance or popularity, then fine (though you could easily point out that the ascendance of SQL, sad as it is, demonstrates the overwhelming superiority of something even vaguely relational over its forebears). But to call any of it “overwhelming” evidence on relational’s technical merits isn’t accurate – unless I missed something big.
I think you’re treating the market value of software products as the phenomena to be observed; in that case you’re more in the realm of economic journalism. There is a value judgement here: that a model that adheres to useful disciplines, and helps to formalize the informal, is useful for building systems. Here’s where I think relational does it best:
1. it sticks close to set theory and predicate logic, while also making room for type theory (in its various forms). The relational aspects are much simpler than more complex areas like graph theory
2. it correlates well with business rules as expressed by users (predicates), as opposed to something like OO that forces a single tool for all jobs
By walking this tightrope between users and math, I think relational hits the sweet spot. Tools like Alloy, and The Third Manifesto’s type system, show the way relational can make proper use of OO and user extensions.
So there is a value judgement here: that a sound mathematical model, if it’s powerful enough, solves problems and enables solutions better than an incomplete model (or an implementation without reference to model, as in the case of SQL). I don’t think that’s an indefensible value judgement, based on science. Engineering will always address issues the scientists didn’t; that doesn’t obviate the science. It’s a question of level of abstraction.
Eric,
When I talk about the overwhelming evidence provided by the commercial marketplace, I’m talking of a couple of things. First of all, the vendors have lots of smart employees, many with deep understanding of language theory and other areas of computer science, and if they could figure out a way to deliver what Date is calling for I’m confident they would. Second, the enormous amount of use commercial RDBMS are put to suggests that maybe they aren’t really as bad as Date says they are. He says they’re too hard to program against. I say that programming is hard in general, and these particular products aren’t necessarily the Root of Most Evil.
More specifically, Date’s critique of commercial RDBMS engineering seems to be concentrated in two main areas. First, he wants what amounts to runtime type-checking, so as to prohibit or at least inoculate against certain kinds of programmer error. This is probably the one he stresses more in Database in Depth. Well, it’s been almost 20 years since I went down in flames being on the wrong side of the strong-typing debate, and I’m not going to address that in a lot more detail than I did above. The vendors are well aware of the merits of all sides of the type-checking debate, and it’s one I trust the vendors and market to sort out.
His other main area of critique mirrors a puzzlement I’ve long had — why aren’t cost-based optimizers more intelligent? He wants two things of them. First, he wants less simplistic cost estimates. Well, so do I. But I’ve been tracking the actual engineering decisions by the vendors in those areas for a long time, I’m sure they would accelerate the improvements if they could, and I even have a pretty good idea what makes the problem so hard: There’s a lot more complexity in the logical-to-physical mapping than one might think, due mainly to the slow rate of disk rotation. In 40 years of Moore’s Law doublings in other measures of compute capacity, disk rotation speed has doubled ONCE. Really. From about 7500 RPM to about 15000. (Or is that per second? Whoops. Anyhow, I’ll firm that up soon and blog about in connection with memory-centric data management technology.)
Second, he wants them to do at least as good a job as they do today, but on bigger problems, so that logical modeling truly doesn’t have to be physical-aware. Well, I’m sorry, but optimizations that grand simply have NEVER worked in commercial computing. This doesn’t mean researchers shouldn’t try really hard to finally make them work. But it does mean that one shouldn’t sit on the sidelines and shout insults at the people who haven’t solved the problem yet. And I think that’s exactly what Date and Pascal are doing.
The caching story is pretty much the same as the one for cost-based optimizers. I just happen to be more sensitive to it because I’ve been focusing some of my research on the area. At a guess, I think Date would say it’s secondary to his other points.
They would if it were a priority. I don’t think the theory is a priority to most. Managers and marketing direct the programmers’ efforts, and once SQL took off, it pushed other options aside. This happens in every industry. Also, I don’t think language design was high on the DBMS vendors’ lists; even if it is now, the traction SQL has, and XML has, is high.
Type checking of column values? SQL does this, and XML Schema (a serious improvement over DTDs) does as well – in fact, I’d saying XML Schema is entirely based on typing. Where would you find “loose typing” most valuable?
I agree that “raw sources” must remain either where they are (“natural” form), or be saved off as CLOB or BLOB or whatever. I also think there’s a large amount of value in extracting real data from those – defining real relations/tables. Far more useful than trying to generalize SQL and XPath and substring across all possible “raw” domains.
Wrong side? I didn’t know there was one. Java’s type system is weak (which is why it’s so heavily armored) – there are far better languages (e.g. Haskell, ML family).
I’m not so sure. I don’t think their customers demand enough. Objects are awkward at best in Oracle, and think The Third Manifesto is a call for some decent type-definition capabilities. I don’t see the market sorting out much here, as the arguments are esoteric; the effects are only felt very far downstream.
(As a background, FWIW, I’m currently a Java/J2EE developer working in XML integration, with a background in Oracle SQL and MS/VB. Although XML is what I do, it’s ugly and confused just as a message format. If it ever becomes a primary focus of data storage – I can’t call it a database – I’ll throw myself into the drink. SQL is much better. Relational is better still. If I had a programming language with relations built in – like Alloy hints at – it would eliminate much code. Relational would be useful for general-purpose programming, not just data storage, because queries and constraints are things we have to do with a lot of effort. Encapsulation is useful across components, not within them, and in any event constraints serve much of the same purpose.)
I agree with you here – I think the success of SQL says something (not proof, just evidence) of its perceived value over prior approaches like networks and hierarchies. My opinion is (I think) similar to Date’s – that while we went partway toward the goal, we gave up early on a prototype (SQL) that IBM pushed for reasons I don’t know. I think Date’s comparison of SQL against relational yields some genuine problems, and are more than just carping.
Programming is hard in general, but you can see many languages and platforms pushing (clumsily) toward declarative techniques, even if it means generating procedural from there. Relational constraints and query mechanisms, to me, are a strong way of doing just that, even if you’re using them in memory as opposed to against a data store.
Relative to SQL, you’re probably right. Date points out where those differences make optimization more difficult. I understand there’s a lot of complexity, and that engineers work hard to accelerate things. But vendors also have a great deal of legacy, both in SQL and in their own products. No one wants to break backward-compatibility.
Out of curiosity (and not advocating it), have you examined the details of TransRelational, relative to the engineering decisions you have been tracking? I’ve seen only one detailed critique of TransRelational, but it was limited and flawed. I’ve not seen much else.
Sure they have – witness speedups in the Java Virtual Machine. Not the same issue, but an example of order-of-magnitude improvements. I think what the engineers are trying to do is harder than what Date proposes, which means the wrong problems are being tackled by smart people.
Pascal certainly criticizes a lot, and is clearly tired of repeating his arguments. As far as I can recall, Date has always been more patient with explaining his reasons, and criticizing rather than insulting. I could be wrong, but Date doesn’t usually comment on things he hasn’t thought about.
I guess it comes down to this:
1. I don’t trust the market to “sort out” anything, especially when the issues are complex
2. Smart people don’t always make smart decisions, because technical problems become compelling even when solving them is irrelevant (I’m prone to this myself, as a developer)
3. Marketing and management push engineers in faddish directions, to follow the hype
4. Re-engineering an existing product is rarely viewed as cost-effective
I do believe that relational has to be proved in products, and that it can (being a technically superior solution). Whether it will or not is a different question. Although I like Java, for example, given the languages available decades ago (Lisp, Smalltalk), and other options today (OCaml, Haskell, etc.) the present state of our languages is pathetic. I think the same is true in DBMSs.
Well, LISP is where I went down in flames. And I could go on at length about why LISP really isn’t nearly as well-suited for commercial use as we all once thought, but I’ll skip it.
As for my opinions about what the vendors do or don’t think about, this is a lot more than just speculation. I’ve been talking with and consulting to the heads of engineering, CTOs, and bosses of same at these companies for over 20 years. And by the way, a DBMS is ultimately a big fat language interpreter, so there is a LOT of languages talent in DBMS development organizations. They’re not oblivious to these issues at all. Yes, the SQL juggernaut was for marketing rather than technical reasons, but every major DBMS vendor has come out with multiple languages and app dev tools all along. They’re not shy about doing what they think gives genuine technical advantage.
Regarding TransRelational — the patent and other public descriptions generally refer to a memory-centric, query-heavy design. Actual development adapted these ideas for a more disk-based, update-friendly approach, in a columnar architecture. (Pascal can fulminate against my use of that word all he likes, but it’s well-founded, given the caveat that “columnar” can mean many things by the time the system is finally implemented.) The memory-centric version is pointer-crazy — one of the big virtues of putting things in memory is that you can use pointers freely — so it can’t be translated to disk without heavy adaption. I don’t know exactly what changes they made to the publicized ideas.
It’s a pity that the CEO and the investors at Required couldn’t get along. I don’t think the problem was the technical bait-and-switch (building something different from what they said they would when they raised the money); that happens all the time. Rather, it’s the classic issue of a founder refusing, rightly or wrongly, to be replaced by “professional” management. I do not have a strong opinion as to who was “more” at fault. I do have the strong opinion that the technology is/was NOT so radical as to easily survive a three-year shutdown and ongoing corporate dysfunction and yet wind up with an “orders of magnitude” breakthrough coming to market.
And I’ll stop at that, because while I don’t use the full journalists’ playbook, I do have an issue here with protection of sources.
Meanwhile, Fabian Pascal has just publicly confessed to his ignorance, and I quote:
“Can’t pronounce on products, I do not follow them.”
Of course, he immediately went on to comment anyway, stating his “guess” about DBMS technology. But that’s typical. In Pascal’s world, logic and consistency are things he demands of other people.
Curt,
Regarding logic and consistency, not to mention knowledge, I think
that Eric has demonstrated very clearly what they are, and that
you don’t possess neither. I DK where he takes the patience from,
I deem it a waste on you. But I am glad that there are smart
knowledgeable people that prove your emptiness.
value in education and it
shows. You are simply a limited knowledge mercenary who, taken out
of products, markets and jargon, have nothing to offer except to
those who are as limited like you.
You can complain without all you want about you
being insulted, but I do nothing but pointing out
your ignorance, baseless accusations, and doing
it bluntly and explicitly.
—————————————–
EDITOR’S NOTE: The above comment is NOT edited in any way by me. The reader — if there is indeed any who cares in the first place — may speculate as s/he wills about Pascal’s state of mind at the time of posting it.
I will, however, make note of Chris Date’s hypocrisy in lampooning me for very minor real and supposed grammatical errors, yet enthusiastically endorsing the work of the person who writes such gibberish. (The stuff Pascal posts on his own site, while generally more coherent than the above, is still pretty fractured. While one can usually cut through the clutter and see what he means, sometimes the grammatical chaos serves to conceal substantive illogic.)
Why can’t you stand behind your own words and
must insert them in my own comment, to confuse
the reader.
As to grammatical errors, I am showing you the
respect you deserve. I don’t bother to go back
and correct text that is addressed to you.
I dare you to point out ANY comment of yours
in which you addressed any of the substantive
criticisms that were made about you, and one
argument you won with evidence and logic. In fact
Eric has demonstrated point by point that all you
do is dump empty celkoisms and complaints without
ever addressing your proven limitations.
I’ll let others decide how typos and grammatical
errors due to unwillingness to waste time typing
anything to the likes of you says anything about
one’s state of mind. But at least we have a mind;
you’re a vociferous ignoramus.
As to Chris’s logic and consistency, it does
not need to be defended from the likes of you,
but just to show you again how your gotchas are
zilch: when he says he’ll be found out, he does
not mean by the ignorant masses, to which you
belong. But rather those few thinking
practitioners to whom he and I address
ourselves e.g. Eric and others.
But then, how can you possibly see that
distinction?
You don’t edit, but you insert your comments
into mine, and you just deleted or failed to
post a reply from me. Let’s see if this one will
survive.
Complaining about grammatical errors instead of
addressing criticism sof your stupid,
unsubstantiated claims is your SOP. You’re not
capable of much else. Substance and logic — how
would know what those are?
You know absolutely nothing of the substance of TRM, except a few
details about the business, and even those are
shaky. You ask people to trust you and your hidden sources when
I have proven you a liar with evidence at TDAN.COM, while you
accuse me and Chris of lying without ANY evidence.
The difference between us is that I know what
I DK, and I don’t talk about it (e.g products); you, OTOH talk
non-stop about things of which you have no clue.
What is more, what products did you ship that you
consider yourself any better?
Anybody with half a brain who reads your exchange
w. Eric can see how he shredded almost everything
you said to pieces, yet you have not understood
any of that, and never addressed it.
You are a big mouth, empty head with a keyboard and an
uneducated audience, a liar and badmouther to boot.
[…] 5. Being on advisory boards almost always involves compensation or the expectation of compensation. Anybody who asserts otherwise is dishonest or naive. But then, the only folks I’ve ever seen assert otherwise are Fabian Pascal and (sort of) Chris Date. […]
Fabian,
If somebody has read this far into the comments thread, they’ve probably also read the posts on which it’s based.
So there’s really no point in me recapitulating them. While you’ve seen fit to pretend they never happened, I doubt anybody will be fooled.
It’s funny that Fabian and his ilk claim to have “science” on their side when all real scientists and engineers know that reality wins. Theory is great, and it can be nice, neat and seemingly incontrovertible, but if it doesn’t match reality, then there’s something missing from the theory. All they’ve got are some nice neat little equations that unfortunately don’t reflect the real world.
>> but if it doesn’t match reality, then there’s something missing from the theory.
well, of course not. all new theory negates reality to some degree. Special Relativity departs from reality, as
as experienced by humans. but it was real enough to incinerate 200,000 Japanese.
>> All they’ve got are some nice neat little equations that unfortunately don’t reflect the real world.
and what might that be?? XML as neato data store??
Robert,
Actually, the forces involved in nuclear fission have substantially nothing to do with Special Relativity or even, IIRC, General Relativity.
I’d say that the big substantive problem with the relational purists’ view is more like that shown by the old joke about the guy looking for his lost keys under the lamppost. “Where did you lose your keys,” he is asked. “Oh, about 30 feet down the sidewalk.” “Then why are you looking for them under the lamppost?” “The light is better here.”
Date, Pascal, et al. have convinced themselves that the problems that CAN be solved by a neat little mathematical theory — namely, those of achieving data integrity with a minimum of programming effort — are far more important than those that so far cannot — e.g., getting the rest of your programming challenge solved with a minimum of programming effort, running your systems with a minimum of operating cost, etc.
Indeed, I think even their core point is no longer true. For a long time, they were right that enterprises did too little careful relational database design. But now that most transactional applications are purchased instead of built inhouse, I have to ask — if an enterprise is getting into trouble via a bad database design, why are they coding a project inhouse large enough that this can even occur? And that’s even apart from the problem of having to work with preexisting legacy database designs, whether good, bad, or merely incompatible.
“Actually, the forces involved in nuclear fission have substantially nothing to do with Special Relativity or even, IIRC, General Relativity.”
That might be true with respect to force, but are you suggesting that the humungous energy released by fission has nothing to do with Relativity’s prediction that mass is energy and that the energy in a given mass is proportional to a very large number, namely the speed of light, squared?
“all real scientists and engineers know that reality wins”
Your statement might apply to physical scientists like physicists and chemists. However, many mathematicians would disagree with your broad generalization. Non-euclidean geometry was perfectly valid science even back when we seemed to live in a euclidean world. By your reasoning, Reimann was magically transformed from a non-scientist into a scientist when Einstein published his first theory.
Even then, one might argue by your reasoning that neither was a scientist until others developed technologies capable of testing Einstein’s theory.
One wonders what magic event made “A real scientist” a real scientist.
Bob,
You caught me forgetting some of my history of physics. Yeah, e = mc^2 was derived in connection with Special Relativity. And yes, it was used as a strong clue to determine which isotopes had energy stored that could be released via fission (or fusion). My bad.
As for Riemann being a scientist — I wouldn’t say that he was. He was as pure a mathematician as they come. In general, I agree with the modern consensus that mathematics is NOT a science, even though it is not uncommon to do science by doing mathematics, to prove theorems in the course of scientific exploration, etc.
Basically, science is about coming up with elegant, concise, often mathematical descriptions of natural phenomena, which then can be used to make correct predictions of other natural phenomena. (This includes engineered natural phenomena.) Further, mathematics can be used to show the equivalence between, incompatibility of, etc. various scientific claims and conjectures. (Those are the cases in which a mathematical result is likely to actually BE a scientific result too.)
Even so, mathematics isn’t science any more than making a profit is crime; the overlap in both cases is considerable, but overlap is not at all the same thing as equivalence.
I’m a little confused with your statements.
Mathematics is not itself a science, yet science depends upon it in order to validate assertions.
Thought I would do a quick search on some definitions of Mathematics.
According to modern consensus (and one you endorse), Mathematics is not a science.
This is by no means a comprehensive list (just a quick search), but mathemathics is defined as a science in the descriptions:
http://dictionary.reference.com/search?q=mathematics
http://www.britannica.com/ebc/article-9371530?query=mathematics&ct=
I’d like to know the make up of this modern consensus is that you refer to also.
Anyway, back to work.
JZ
edit: interestingly, the definition in wikipedia has a section on whether mathematics is really a science or not (and some of those for and against), but I wouldnt say based on this alone that modern consensus says that mathematics is not a science.
Yet another area where there is no black or white, but just various shades of grey.
http://en.wikipedia.org/wiki/Mathematics#Is_mathematics_a_science.3F
Well, I wouldn’t take Wordnet’s definitions too seriously, for a variety of reasons, the main one being that writing a careful definition isn’t the point of Wordnet, any more than it is in, say, a Scrabble dictionary. A definition is just a label somebody dashes off as a form of documentation for the real substance, namely the formal relationships. (Nor would I be impressed in the likely event that Wordnet contains the relationship “Mathematics is_a Science.” I attended a “College of Mathematics and Physical Sciences” as part of Ohio State University, and “Mathematics is_a Math_or_Physical_Science might be too verbose. I don’t know, however; while I know what Wordnet does, I haven’t actually taken a look at it.)
And the Wikipedia excerpt really supports my viewpoint.
The Britannica quote is a sterner challenge to my claim of consensus.
I guess I have to retreat and say that there’s a modern consensus among mathematicians. What people in other disciplines — even related ones — think evidently varies more widely.
Curt,
As it relates to mathemeatics, I feel I must speak in your defense. Math is a language, not a science. It is the universal language used to describe objects, events, relations, etc. It is a language which is intimately married to science, but it is not science.
Anyone who has studie Goedel is aware that no mathematical system is completely closed. There are always exceptions that the system cannot handle. That is the reality we are dealing with.
The relational model will not endure. No matter how many heretics are executed.
I for one find C.J. Date’s argument’s both valid and sound, and agree with almost everything he says from a theoretical perspective. In fact, the model he proposes (or elaborates upon, rather) is much cleaner than anything currently out there in the market. Why not accept that as a goal to work toward, but make more of an effort in doing so (as opposed to continuing to hack away at a language — SQL — that never did the job “right” in the first place).
Specifically for Curt, re: type constraints. My understanding of Date’s writings is that column constraints are quite different that type constraints. Type constraints apply to a type (which is orthogonal to a relation) — whereas column constraints (or “relation” constraints), per se, apply at the table level. Different ballpark altogether. Type constraints ensure consistency of type across tables (relations), rather than simply at the table (relation) level. This protects, among other things, integrity violations within joins.
I just finished Database in Depth, and as a working Oracle and Sybase professional with programming and UNIX expertise, I’d say that what he’s stated and proposed is very solid.
I would, however, be interested in hearing /concrete/ examples of where Date’s logic is flawed or simply unworkable. And by “concrete” I mean mathematical and supported by evidence, not merely “well, that’s just how it is in ‘real life’.”
All too often — and I cite this post explicitly — people resort to vague generalization or assumptions that have no factual or quantatative basis. And if they do have said basis, please show it. (This is not a specific criticism of anyone here, but I did notice that nothing in the above gives specific examples, other than statements like, “it’s difficult” or “today’s databases work well.” What is difficult? Which databases work well, and at what?)
My $.02.
Adam
Adam,
Thanks for commenting.
I’d encourage you to poke through the rest of the Database Theory & Practice topic, e.g. the post on the unsuitability of the relational model for handling text.
“Anyone who has studie Goedel is aware that no mathematical system is completely closed. There are always exceptions that the system cannot handle. That is the reality we are dealing with.
The relational model will not endure. No matter how many heretics are executed.”
He he, then by the same logic:
The arithmetics will not endure.
The number theory will not endure.
The probability theory will not endure.
The integral calculus will not endure.
etc… etc… etc…
No matter how many heretics are executed. LOL
[…] behind the hoohah around Transrelational(TM) technology from Required Technologies, Inc., and Chris Date’s highly regrettable promotion of same. Now I’ve been able to get more detail from another former executive of the company. […]