Re: What does maxquerylen in v$undostat really mean?

HansF wrote:
> Howard J. Rogers wrote:
>

>> HansF wrote:
>>
>>> Howard J. Rogers wrote:
>>>
>>
>> [snip]
>>
>>>
>>>
>>> Based on my experience,  I find Oracle tends to retain definitions 
>>> for a term - if they want to change the definition, they tend to 
>>> change the name.
>>
>>
>>
>> "Experience". "Tend". These are generalities. Documentation should 
>> provide specific answers to specific questions. Your comment misses 
>> the point that, without specific, accurate documentation, we cannot 
>> know *for certain* whether your experience is playing us false this 
>> time; or whether the perceived past "tendency" not to change 
>> definitions is in force this time round. We cannot know. Therefore we 
>> are guessing and relying on our subjective experiences and perception 
>> of tendencies... and that is precisely what documentation is there to 
>> avoid.

Hans: that's a silly statement. There is a context to my comments. Please try not to rip them from it.

The context is: can one *rely* on the documentation? Your post states, essentially, "in my experience, the documentation works like this...".

*In that context*, yes: I am discounting that specific bit of your experience. Because when it comes to the documentation, I shouldn't have to weigh up such things. The documentation should be right, or a bug-report should be raised, and it should be fixed. In *this* version. And why? Because documentation is specifically intended to be an explicit statement of objective fact. It shouldn't *need* to be filtered, or have its gaps plugged, with subjective experience as a guide to the plugging.

Do I dismiss or discount your experience *generally*? Of course not, and my words are quite clear about it: "we cannot know whether your experience is playing us false THIS TIME" means that your experience is a fine guide on many occasions. Just, perhaps, not THIS ONE.

>Perhaps you
> just discount experience with the exception of your own?

Look, you can get personally abusive if you want. But I'm sticking to the facts. The OP read the documentation. The documentation misses out the key piece of information he needs. He's advised to read the wrong version of the documentation to fill in the gaps. I point out that's risky, and doesn't actually provide assurance and certainty (as certain as it needs to be, anyway).

That's all. You're making waaaaaay too much out of it, otherwise.

> Use of the term 'tend' simply stems from my general dislike of
> absolutism - a concept foreign to some.

There's nothing absolutist in pointing out that going cross-version to fill a gap in the documentation is risky.

> But I defer to your superior wisdom and hereby refrain from further
> discussion on this, or any other topic, in this newsgroup.

I don't ask you to defer to anything. Just don't take things personally when it's actually matters of fact which are being discussed.

I repeat: the original advice to go cross-version in the documentation is risky. Not 100% wrong. Not 100% right. Not absolute at all. Just *risky*.

I also repeat that there is in this specific case no need to go cross-version, since other V$SQL... views *in the 9i documentation* show a column of the same name measured in milliseconds. But that, too, is a risky approach... though it's one I felt compelled to employ myself for this specific request.

HJR
> TTFN/Hans
Received on Sun Dec 12 2004 - 10:20:46 CST