JL Computer Consultancy

In a recent thread on AskTom,   http://asktom.oracle.com/pls/ask/f?p=4950:8:::::F4950_P8_DISPLAYID:35336203098853 Tom was asked what he thought about an article by Don Burleson that offered the readers the opportunity to enter a competition to write a program that could reliably predict tables and indexes that could measurably benefit from reorganization. The metrics that Don proposed as methods for identifying such tables and indexes were listed under a heading of “The Precepts”. 

The original article is at http://dba-oracle.com/oracle_news/2005_2_17_table_index_rebuild_reorganization.htm rel=nofollow but it has been modified a little since the start of Tom’s thread, so comments in the thread may no longer appear to be totally consistent with the current version of Don’s article.

Tom’s response was a review of the four ‘precepts’, starting “But I have a problem with them.” In summary :

#1      I say "no, that is not a good governing rule". It is a fact, but one fact of dozens of facts.

#2      While this precept is "true" on the face of it, it is not really relevant in the long term.

#3a     Well, if the table has truly chained rows - a reorg will...... do nothing.

#3b     I might do it once, and fix the obvious mistake we made with PCTFREE on this table.

#3c     For the "low block density", this precept would be "true".  But once again, I would be asking 

        "why" -- why is the table 50% white space?

#4      They (freelists) cannot really get "out of balance"

Don’s initial response to the thread addressed one of Tom’s critiques. Don claims to “see unbalanced freelist lengths all the time”.

The next major follow-up from Don was that two articles on Metalink said that you could get excess space in objects if you had freelists – so how come Tom knew better. (This, despite the fact that Tom had said you could use extra space, but you didn’t get the performance impact that Don had claimed). Don also explained that the purpose of the thread was to decide whether or not it was possible to build a predictive tool.  (This, despite the fact that Tom’s comments had been all about the relevance of precepts, not the general principle about whether or not a predictive tool could be built).

The next major follow-up from Don is titled “My precepts are perfectly valid”. This summarizes to:

“I know I’m right, because I’ve seen it; it must be a good idea because Oracle Data Mining applies statistical methods to data; why do you proclaim yourself a scientist when your qualifications are mathematics; here’s an advert for a book from my publishing company.”

This is where it gets interesting – because Don has also written an article on one of his sites called “Are all Oracle Scientists created equal?” This has prompted me to wonder what the expression “Oracle Scientists” might mean, and how you decide whether a particular person falls into that category.

What is a scientist?

If I had to name two people who represent the ideal of the scientist, one would be Richard Feynman, and the other would be Richard Dawkins. So naturally, I feel inclined to turn to them to see how they might treat the question.

This is a comment from Richard Feynman in a piece entitled “The Value of Science”

“… Except, possibly, that some children catch on. And when a child catches on to an idea like that, we have a scientist.”

So it seems that Richard Feynman, one of the greatest scientists of the 20^th century, thinks that being a scientist is about a state of mind, not a piece of paper. And once you’re into being a scientist here’s a thought from Richard Feynman about how you do science:

“The first principle is that you must not fool yourself - and you are the easiest person to fool.”

Turning then to Richard Dawkins, what does he tell us about scientific endeavours?

“What really makes scientists special is less their knowledge than their method of acquiring it – a method that anyone could adopt with advantage.”

Again we see a man, at the top of his field, who points out that being a scientist is about how you behave … moreover, anyone can do it! And does Dawkins have a comment that echoes the moral guideline of Feynman? How about this for a scientific standard of behaviour:

Science has no method for deciding what is ethical. That is a matter for individuals and society. But science can clarify the questions being asked. … This usually amounts to the useful ‘You cannot have it both ways’ style of arguing.

I don’t think I can give you a definitive statement of what a scientist is – but I think I understand why the expression ‘Oracle scientist’ means something to many of the people who work with Oracle databases.

I’ve met many ‘Oracle scientists’ – and there are far more Oracle scientists than there are members of the Oak Table – and they seem to have two things in common: their enthusiasm for “the idea” espoused by Richard Feynmann, and the approach (that Richard Dawkins reminds us anyone could adopt with advantage) they use to enhance their knowledge.

Are all Oracle Scientists created equal?

The title is, of course, the title of Don’s article – and I’d like to apply the Feynman/Dawkins thoughts to it.

The article has been changing steadily over the last few days, so don’t be too surprised if several of the items below have been edited out by the time you read it. Any text in [square brackets] are my editorial additions to clarify the context.

You can’t have it both ways - 1

Quote:  “The web has evolved into a dangerous place today, and today you have to find out who you are talking to.”

On the other hand

Quote:     “Whenever I engage an alleged “scientist” or “expert” in a discussion, the first thing I do is Google their academic and research background.”

So the web is dangerous and you have to find out who you are talking to – but the place to go to find out who you are talking to is the web? The expression “circular reference” springs to mind.

It’s remarkably easy to set up all sorts of misleading information on the Internet. All you need is a few dollars to buy a domain name, a story about who you are, and a few pretty graphic designs for the web pages. For example, a little while ago a foundation for the accreditation of serious Oracle professionals (U.S. only) appeared on the Web. The name was Aesop, or something like that. The trouble was, everything was pretty anonymous “to ensure fair play”. The only item I ever saw with an Aesop accreditation was one of Don Burleson’s books, and co-incidentally the ISP for Aesop just happened to be the same as the ISP for several of Don’s websites.

You can’t have it both ways - 2

Quote:  “Cary [Millsap] is the "real-deal" Oracle scientist.”

On the other hand

Quote:     “[question from reader] Are you an Oracle scientist? [Answer from Don]I'm not exactly sure what the term means, actually”

So Don doesn’t know what an Oracle scientist is – but he does know that Cary is the “real-deal” Oracle scientist

You can’t have it both ways - 3

Quote:  “You can instantly spot a fake Oracle scientist when they start insulting their fellow scientists”

 On the other hand

Quote:     “[question from reader] Are you an Oracle scientist? [Answer from Don]I'm not exactly sure what the term means, actually”

So Don doesn’t know what an Oracle scientist is – but he can instantly spot a fake one. Strangely, though, it seems that his method for spotting fakes is concerned with bad manners rather than quality of work.

Sir Isaac Newton is often described as a scientist, as is Robert Hooke – but you should see how rude they were about each other.  Newton’s most famous comment (about standing on the shoulders of giants) isn’t modesty,  it’s a snide comment on the fact that Robert Hooke had a crooked back and severe stoop. (See Big Bang by Simon Singh ISBN 0 00 715251 5).

It is even stranger that the reference Don supplies simply points to a newsgroup item from an individual who spends his time posting advertisements for a product (contrary to that newsgroup’s charter) and being offensive. How does this posting allow Don to determine that the poster is an “Oracle scientist” in the first place? The “logic” is simple:

               This person is clearly an offensive individual.

               This person is on an Oracle newsgroup

               This person is trying to sell an Oracle related product

               We will state that this person is an Oracle scientist.

               Hence:     rude Oracle scientists are fake Oracle scientists.

In fact (see example 1 above) a search on the web suggests that the author of the abusive language is a highly qualified academic with inside experience in the kernel group at Oracle corporation, a paper published at SIGMOD, and some interesting sounding positions with big companies under his belt. In fact this person seems to meet all the requirements of being acceptable according to Don Burleson’s critieria of “engaging in discussion”.

You can’t have it both ways - 4

Quote:     “one Oracle scientist felt compelled to interject my company's dress code”

On the other hand

Quote:     “[question from reader] Are you an Oracle scientist? [Answer from Don]I'm not exactly sure what the term means, actually”

Same argument – Don knows what an Oracle scientist is when it suits him. (Although, in this case, it looks as if the qualifier may simply be that anyone who posts a follow-up on AskTom is deemed to be an Oracle scientist if Don thinks he can extract some emotional mileage from it).

What really makes scientists special is less their knowledge than their method of acquiring it - 1

Quote:  “Whenever I engage an alleged “scientist” or “expert” in a discussion, the first thing I do is Google their academic and research background.”

Quote:     “In this world of fakers and posers, all Oracle professionals need to have strong “BS” radar and a quick Google search can tell you how much weight to give to the assertions of any Oracle “scientist”.”

Quote: (Since deleted): “For example, when I first encountered Jonathan Lewis I could find out nothing about his background or qualifications, and just like

Cary Millsap notes, he was an “unknown” commodity to me.  Later, when I discovered that Lewis has a master’s degree from Oxford, it confirmed my suspicion that Lewis was a brilliant fellow.”

Addendum:  My error – I thought the third comment was actually in the original article; in fact it was in the AskTom thread in a review section from Don that echoes the general sentiment of the article.

It is of course difficult to assess the quality of what someone is saying on a technical topic if you are not proficient in that topic. I can understand that Don might feel the need to check someone’s qualifications before believing what they say. It is interesting to note that Cary managed to work out that I was “brilliant” by judging what I had to say, but Don had to see an old qualification.

What really makes scientists special is less their knowledge than their method of acquiring it - 2

Quote:  “For example, I was recently approached by a stranger named Scott Martin (www.tlingua.com), who made some bold assertions about creating an Artificial Intelligence (AI) engine for Oracle. Suspicious at first, I Googled him.”

Personally, when Scott first sent me an email, I just clicked on the link to his website and read the papers he had written about Oracle. That gave me a pretty clear idea about his level of competence, and his ability to explain and demonstrate the things he had discovered.

You can’t have it both ways - 5

Quote:     “When I do a Google search for some self-proclaimed Oracle scientists, [my emphasis] I can’t find out much about their science background”

On the other hand

Quote:     “Jonathan Lewis - Respected Oracle author with a Masters Degree in Mathematics from Oxford University.”

At present, the site seems to give me the Burleson seal of approval. However, if you search for my academic background you may have trouble finding it. I don’t even bother to mention my M.A. on my website because the material I have published about Oracle says much more about what I can do than the degree would. In passing, I don’t have any significant science qualifications – bizarre though it may sound, my Maths degree counts as an Arts degree at Oxford.

What really makes scientists special is less their knowledge than their method of acquiring it - 3

Quote:  “Is a publication by an Oracle Scientist the pontification of a Harvard PhD or the ruminations of a high-school dropout?  Frankly, I’d like to know.”

Why?  If the material is sound, the presentation lucid, and the conclusion correct, reproducible, and potentially useful, why should it matter who wrote it?  (This is a topic I will be visiting with a worked example some time in the next couple of weeks).

What really makes scientists special is less their knowledge than their method of acquiring it - 4

Quote:  “[From Kent Crotty - Oracle consultant] I would have to agree with Don.   To lend any legitimacy to arguments made by any ‘scientist’, that individual must have the backing of research and qualifications. The very definition of the word demands this:”

Personally, when I think a line of argument is worth following, I try to follow it. This “reader’s comment” included a URL to the word scientist in the Merriam-Webster Online Dictionary. http://www.webster.com/cgi-bin/dictionary?va=scientist  The “very definition” reads:

1 : a person learned in science and especially natural science : a scientific investigator

The word scientific is then defined as:

               of, relating to, or exhibiting the methods or principles of science.

And the word science is defined as:

1 : the state of knowing : knowledge as distinguished from ignorance or misunderstanding
2 a : a department of systematized knowledge as an object of study <the science of theology> b : something (as a sport or technique) that may be studied or learned like systematized knowledge <have it down to a science>
3 a : knowledge or a system of knowledge covering general truths or the operation of general laws especially as obtained and tested through scientific method b : such knowledge or such a system of knowledge concerned with the physical world and its phenomena : NATURAL SCIENCE
4 : a system or method reconciling practical ends with scientific laws <culinary science>

I don’t see the bit about “the backing of research” or “qualifications”, mostly it seems to talk about knowledge and the acquisition of knowledge. It looks as if Kent Crotty has acquired Don Burleson’s habit of quoting references without reading them first. So I would like to thank Kent for bringing to my attention that there is an actual definition of a scientist as: “an investigator exhibiting the methods or principles of science” – this seems to be a pretty good statement of how the members of the Oak Table (and many other Oracle practitioners that I have met) tend to behave.

So what is an Oracle Scientist?

With thanks to Don Burleson and Kent Crotty for their guidance, I would like to offer the following description:

               An Oracle Scientist is an individual who chooses to apply the methods or principles of science to enhance his or her knowledge of the Oracle RDBMS.

Footnote

At a later stage in the thread that prompted me to write this note, Don Burleson provided a script to “show the unbalanced freelists” that he sees all the time. I will be reviewing this script in a future article – and will also make a few comments about how freelists work.

Footnote 2

I have notified Don Burleson by email that this article is on my website, and offered him the opportunity to reply on this website, or by giving me a URL to a response on his own website.

Back to Index of Topics