Re: How to perform a health check?

wizofoz2k_at_yahoo.com.au (Noons) wrote in message news:<73e20c6c.0410071407.2f29a9f9_at_posting.google.com>...
> joel-garry_at_home.com (Joel Garry) wrote in message news:<91884734.0410051033.1147b01e_at_posting.google.com>...
>
> >
> > What you say is true and correct. However, standards committees are
> > just groups of vendors, aren't they?
>
> Not at all. But even it were so, that's better than a single vendor.
>
> > I think the real problem is you
> > can't have standards (of the sort I think you propose) before the db
> > products because the products are evolving too fast.
>
> It's got nothing to do with the products. That is the whole point!
> While we continue to associate standards with products, it won't work.
> It's got to do with the technology. Administering a database has not
> changed that much in general principles in the last 30 years. Because
> it is an activity that has nothing to do with the specific technology.

Hmmm, something is bothering me about this (probably something like not dealing with SP's that long ago), but I can't put my finger on it, so I'll have to say, as usual, you are right.

>
> I'll offer a metaphore (!):
> You drive a car nowadays the same way you drove a car 30 years ago:
> here is a steering wheel, here is an accelerator, brake, etcetc.
> And when you take a driver's training course and test, no one is
> asking you to only drive a Ford Fiesta 2004 model. It's for ALL
> makes and models of cars!
>
> Of course, IF you want to get the best out of a given model and make
> of car,it pays to know a little more about it. That is why BMW for
> example has advanced driver training courses for their buyers.
> Perfectly kosher. But the BASIC driver training is now more than
> standardised and has NOTHING to do with the vendors.
>
> > Remember all the
> > hoopla about unix style operating systems? All that got blown away by
> > linux. Because linux was written by a bunch of people saying "get it
> > done now and make it work with what is out there." Then
> > sorta-standardized after the fact.
>
> And yet, remarkably, it is to Unix that Linux owes it's elegance
> and command-line interface. The kernel itself may be different
> or use different technology. But the basics of operation of
> Unix are all there: file metaphore, programmable shell.
> Anyone with a basic knowledge of Unix can pick up Linux in no time
> at all and with minimal training. Of course, if that anyone then
> wants to write a device driver, a highly specialised activity,
> then it pays to get training on the specifics.

Elegance, yes, academia... go back and look at the Tannenbaum/microkernel arguments! Linux was successful precisely because Linus hacked out the basic stuff, and allowed distributed development.

And try to get umpteen years of unix experience past a HRhunter who must have linux :-(  

>
>
>
> > If you are advocating academic-style standards setting, that will just
> > be way too slow and will never become useful,
>
> Totally disagree. It is PRECISELY because we are ignoring academic
> principles that training in IT is a shambles.
> There is a reason why universities and academia have been so useful
> in every field of human knowledge. We are ignoring it at our
> detriment.

I said "never," and now feel like those guys who said people will never go faster than 25MPH. *:o)

But really, I've ranted before on how stupid it is to train C programmers to write application code. IT as an industry is (well, should be) very different in purpose from academia. Now, if there were an academia purposed towards industry (in what we are talking about here), we might have something. Maybe Daniel's way (specific training plus associations) can get there. Unfortunately, what I see happening when universities try to get too close to industry, money gets in the way of education. I see this happening local to me both in the biotech and the computer fields - you know all those rants you make about java developers v db types? Guess where it is coming from:  university/business cooperation means money for buzzwords. (US) Universities make big bucks off of intellectual property these days. Good for some people, but I don't think it is good for real academia in the long term, it tends to create conservative cabals. But of course, IP is it's own ball of worms.

jg

--
@home.com is bogus.
http://www.signonsandiego.com/news/business/20041007-9999-1b7connect.html