Re: Fed Up with being a DBA

Answers interspersed. And so far no complaints. Perhaps we should collect this thread and publish it. ;-)

Noons wrote:

>>I'd be thrilled if before I saw 10g released I felt I know 1/20th of 9i.
>>
>>
>
>Exactly. I don't even think there has been enough experience in the
>market about new features and which are of real value. Apart from ones
>that started earlier and were bug-fixed/enhanced in 9i. Like LMTs and
>the plan stability stuff that started in 8i. And the XML stuff.
>

I've learned more about Oracle since I started teaching that I ever learned when all I did was use it. When using it I only needed to know the minimum functionality set required to do my job plus a bit more out of intellectual
curiosity ... I'm most certainly no Tom Kyte or Jonathan Lewis driven to delve into the inner workings. When the
day is over I'm on my sailboat.

But get in front of a room with 30-40 industry professionals, some with Masters degrees and PhDs, all with at
least 5 years experience in databases, some with 15-20 and you find a strong incentive to learn things you never though of before.

>10G is coming out too early. It better have the advantages and stability
>that were promised for 9i, or people will just stay away.
>

I still think 10G is largely defensive against the ability of the hardware vendors to dynamically swap CPUs. I don't think they had any choice.

>Besides, it's not an odd number release... <d&r>
>

Can't help you on that one except to say that if you consider the decimals after the 10 it is.

>>Oracle. I think it says more about their sense of insecurity than
>>anything else. The biggest problem I see in
>>the colonies with respect to Oracle are pricing which is too high for
>>
>>
> ^^^^^^^^
>the WHAT????
>:D
>

It is a very big secret ... actually I can't remember. That was last night and this is this morning and I've not yet grabbed a cup of coffee.

>>Access. What the industry needs is a certification that has real
>>meaning.
>>
>>
>
>
>and the Masters ain't it, I'm afraid. As long as certifications are
>keyed to a version of Oracle they won't offer perceived value. As soon as
>Oracle removes the version number from OCP, I'll be in. Until then, I and
>quite a few others won't bother. Been there done that with the MSCE stuff:
>it's just a $$$ grab exercise.
>

One of my private clients certifies surgeons in the States. I wrote the software they use to track the various steps and hurdles. There is nothing this industry can even conceive of that comes close to the level of expertise they require to become Board Certified. And I've seen the numbers of the percentage they reject ... and it is staggering. Just to give you an idea here is a very short list of the more significant hurdles

1. After graduating medical school and puttingin a minimum of 7 years of residency you can apply
2. Submit all transcripts from all education
3. Submit information on all license information.
4. Submit information on all complaints lodged against you of any type whatsoever.
5. Submit a 6 months log of every diagnosis and medical procedure you performed by date and explaining the outcome
6. Take a written exam
7. Pass two oral exams given by expert examiners that ... to give you the flavour in Oracle parlance would be something like:

It is 3am in the morning and your pager just went off. You phone into the office and someone tells you the database has crashed. What do you do todiagnose and treat the instance? Then the candidate would need to go through the entire process, orally, of examining log files, etc. with the examiner saying things like ... what if the log file had ORA-00600 or what if the .... The average exam is 2-3 hours.

8. Resubmit yourself every 5 years and if you fail you lose your license.

And they have a staff that researches every item submitted and verifies it.

Now that is certification.

>Of course, refresher courses for new versions are needed: that is what
>lawyers, doctors, etc do: they get basic training then stay actual with
>refreshers. The re-certification as Oracle has it now is not the same.
>The message they are sending out is that once 10g comes out, one's 9i
>certification isn't worth squat. Regardless of my opinion about OCP,
>that is grossly unfair on those that have taken the certification.
>

Let me add one essential here. Lawyers and physicians do it because it is required. Definitely here in the States. If they don't have continuing education they lose their ability to practice. Heck even tax preparers are required to engage in continuing ed. Some developers and DBAs (not all of them of course) seem to think that once they have a job the only requirement is that they show up.

>>Larry can bundle an O/S and an office
>>suite and he's got what is arguably the best RDBMS on the market.
>>
>>
>
>Mind you, with a bit of tweaking by a near 50 year old cracker,
>Lunix can do as much damage as Windows. Or a LOT more...
>(bless all that C I coded back when...)
>

And you can write a virus in PL/SQL too. But the new O/S always has one advantage ove the old one ... no legacy. You aren't going to look into the equivalent of winnt\system32 and find antiques like append.exe and chkdsk.exe. Microsoft is still selling DOS with a pretty shell.

>>challenge Oracle. Also, one must remember that those that shop based on
>>price will be the ones that buy SQL Server, not Oracle. So I'd expect
>>the biggest hit, as with Linux, will be on Microsoft.
>>
>>
>
>I see your point. But I'll have to say I still have to see
>a Windows shop running MySQL. The ones I know running MySQL
>are all Lunix...
>

I think that relates to the mentality and the technical ability. Those that can handle Linux are going to be more willing to try MySQL. But if it starts to have a market presence those interested will be the CFOs that worship the list price rather than TCO.

>Oracle, like so many other USA software makers, is NOTORIOUS for NEVER
>allowing external anything. If it's not invented in-house, it's wrong
>or it has never existed. The famous "not-invented here" syndrome.
>

I don't see that in Oracle as much as the fact that they have viewed Oracle Ed as a profit center. If they viewed education for education's sake as a benefit it would be different. It is the question of short-term profits vs. long term. Oracle Ed will bring in a few million dollars for awhile and then managers will become dissatified with paying $2K+ to send their people to a class and get back people that still can't develop an app. And it doesn't help the customer and it doesn't help Oracle's shareholders long term. If they supported real education programs they would forgo the short-term cash flow and receive a long-term benefit.

I view my job at the university as teaching people to think and problem solve. The tool we use is Oracle. But the people that come out of the class pass their final not by answering written questions ... there aren't any ... but by building an application from scratch. And to give you an idea of what they learn ... fully 1/3 of my students used CAST(MULTISET on their final. And I never covered the syntax in the class. They learned to think. Oracle will benefit. And I am immensely proud of them.

>Hang in there. What you guys are doing is sorely needed. Stuff
>what Oracle Education says. (No offence, HJR)
>

Actually I suspect HJR would be among the first to say so. One of the frustrations with being an instructor for Oracle is having to teach can curriculum developed by people that know, in Howard's case, not 5% of what he knows.

>>You'll have a hard time finding a book or course with a title like
>>'Oracle 7.x Warrior" published in the current
>>millenium.
>>
>>
>
>That's where we disagree: book titles!
>Mine would be "The Complete Oracle Warrior". :)
>

And all in say 500-600 pages? ;-)

-- 
Daniel Morgan
http://www.outreach.washington.edu/ext/certificates/oad/oad_crs.asp
http://www.outreach.washington.edu/ext/certificates/aoa/aoa_crs.asp
damorgan_at_x.washington.edu
(replace 'x' with a 'u' to reply)