Oracle FAQ Your Portal to the Oracle Knowledge Grid
HOME | ASK QUESTION | ADD INFO | SEARCH | E-MAIL US
 

Home -> Community -> Usenet -> c.d.o.server -> Re: SQL server Vs Oracle

Re: SQL server Vs Oracle

From: David W. Fenton <dXXXfenton_at_bway.net>
Date: Wed, 19 May 1999 01:26:53 GMT
Message-ID: <xro03.235$ql.23124@typ12.nn.bcandid.com>


Stephen Harris (sweh_at_mpn.com) wrote:
: Arvin Meyer (a_at_m.com) wrote:
: : Stephen Harris wrote in message <7hs76p$1bd$1_at_nebula.mpn.com>...
:
: : Microsoft is now dogfooding all production code for NT on it. It is not only
: : very stable, but were it not for the lack of some third party drivers, it
:
: Damn, I seem to remember someone saying that about NT4 which _still_ isn't
: sufficiently stable for my liking (I'm not being inflamatory here, just
: stating a personal opinion; an OS that can't manage memory and "leaks" needs
: fixing). Any NT4 system under SP3 is vulnerable every possible which way.
: I will consider W2K SP3, which I estimate to be mid 2002.
:
: You may find it interesting to read what large analyst companies (eg Gartner)
: are saying as well. You may find it's along the same lines.
:
: This one is most telling:
: http://webserv.vnunet.com/www_user/plsql/pkg_vnu_news.right_frame?p_story=82737
:
: These also relate:
: http://webserv.vnunet.com/www_user/plsql/pkg_vnu_news.right_frame?p_story=82799
: http://webserv.vnunet.com/www_user/plsql/pkg_vnu_news.right_frame?p_story=82801
: http://webserv.vnunet.com/www_user/plsql/pkg_vnu_news.right_frame?p_story=82917
:
: An interesting addendum:
: http://webserv.vnunet.com/www_user/plsql/pkg_vnu_news.right_frame?p_story=83051
:
: As a balance, they also say Linux won't beat NT :-)
: http://webserv.vnunet.com/www_user/plsql/pkg_vnu_news.right_frame?p_story=82800
:
: : There is NO user handholding. A NetworkPC uses a SmartNIC. The machine is
:
: Errm? Migration of user data? Training in the new app? Educating them in
: the difference between an NT domain login and a Win95 local login? NO user
: handholding? Sorry, no.

  1. If there is significant user data on the workstations, tell them they have 2 months to move it to the server. Don't migrate *any* user data (assuming, of course, that you've properly configured those workstations in the first place, setting the default folder as the user's folder on the server, and using policy manager to prevent too much mucking around with their setup).
  2. Teaching someone to logon to NT takes 2 minutes. I've trained the boss in how to use the NT laptop. The difference between the two logons should never come up. They should never *see* a Windows logon again *if* you properly configure your workstations to *require* domain login (you can even prevent them from cancelling out, but that's a real PITA).

[]

: : but my back is still in an old body. Better that you start learning the
: : newer technology, than saying "Ha! Totally impossible", or some young kid
: : will have your job.
:
: Please don't start attacking me. My job is far more secure than you will ever
: realise because I am _not_ stuck in any technology rut. My comments are
: based on actually knowledge of operations _taking_place_now_. They are not
: someone's ideal scenario, they are opinions based on having to support these
: 500 users (we have two users that just had dumb terminals until last year
: because they were too stupid to turn a PC on and login to the server).

Sounds to me as if you're not managing your company's resources sufficiently well. You shouldn't have allowed them to store local data in the first place (or are you migrating from Win3.x? I've forgotten). So, you get all the department heads to spearhead a data migration over the course of two months. Send an IT person around to each department to run a 1-hour training class in migrating data onto the server (you *do* have the storage space, no?).

If you were upgrading to P400s with twice the memory and twice the HDD space you've described, you would not need to replace these machines in two years.

Am I talking through my hat? No. I've been in this business for the last 10 years, more or less, and I used to think like you do. I learned about five years ago that it's simply counterproductive. Get the good equipment out there now, and the productivity improvement will come *now*.

: Technology is a wonderful thing, but users (like your back after shifting all
: those boxes) doesn't match up. It is the physical reality of the user
: handholding that slows the rollout down.

If you're changing OS and Office suite, then you need a training budget, too. With 500 workstations, you should be able to get a decent deal on 2-3 hours of training for all employees, as a package deal (I'd say a lot less than $100/person at NYC prices). And 2-3 hours is all that's needed. Office97 is just not that different. Win95 *is* quite different, but I've found that, despite my initial reservations, training people to use Win9x is *much* easier than training them to use Win3.x. Since you're migrating to NT4, it's basically the same interface.

For people who need more than the "getting started" lessons, you should come up with more advanced training.

And there *will* be lots of benefits, especially in terms of connectivity (Internet, etc.). This will show real returns almost immediately, and the users will see it.

: If you manage an IT department, change 500 boxes in a week and your users
: don't complain then either you have very competent users, or a large machine
: gun nest outside your door. In my company you wouldn't even have a job if you
: tried it.

Well, one would assume that users have been preparing for such a rollout for about 3 months before it actually happens.

If not, well, the users aren't the problem.

: This is my last post on this matter because the discussion has delved even
: further away from DBMS's and has started to bore me. I did say this should
: be in an advocacy group...

No one is advocating anything other than good business practices. Use your company's money and resources how you like, but experience tells me that spending more up front (though carefully planned and properly allocated) will give more benefit over the long haul.

--

David W. Fenton                        http://www.bway.net/~dfenton

dfenton at bway dot net                http://www.bway.net/~dfassoc

Received on Tue May 18 1999 - 20:26:53 CDT

Original text of this message

HOME | ASK QUESTION | ADD INFO | SEARCH | E-MAIL US