Ben,

I spend almost all of my SQL Server time in SQL Server 2005 which is much, much improved over 2000. The GUI in 2000 was horrible. The installation of SQL Server 2005 is also much improved.

I love EG. It's just a wonderful product and a great way for people to get started. I use it to generate ODS and GRAPH code. With it, I think you can get by with SAS in a day.

Alan Churchill Savian "Bridging SAS and Microsoft Technologies" www.savian.net

-----Original Message----- From: SAS(r) Discussion [mailto:SAS-L@LISTSERV.UGA.EDU] On Behalf Of ben.powell@CLA.CO.UK Sent: Monday, October 10, 2005 9:33 AM To: SAS-L@LISTSERV.UGA.EDU Subject: Re: SAS vs. SQL

My problem with oracle was always curiosity: install is fairly straightforward, on windows at least (I wouldn't recommend it on linux if you don't have RHEL, although I should try again with CentOS given that is supposed to be a clone) but the moment you are there, "simple oracle jobs" like network configuration become a tarpit very quickly. Unsurprisingly this is not something you could accuse SQL Server of ;-) I think I had a fairly fully featured database up and running there from the 90 day free trial within 10 minutes.

Ease of install and use is something SAS has in common with SQL Server, but not in a GUI way (leaving aside EG). The menus achieve relatively little. The bulk of what SAS does is found through code. The code itself is quite logical so very quickly it becomes a question of syntax and vocab. You can learn to "get by" in SAS in three months of a couple of hours a day.

So while Oracle has a steep learning curve I've not yet had the time to climb; SQL Server has an "inverse learning curve" - it teaches you rather than you learning it; SAS has (imo, but maybe that's according to likes and dislikes) a shallow learning curve, but I suspect you reach the plateau only after a much longer time than the other two, if you take the whole SAS suite into account.

Mixing my metaphors? J'accuse moi!

Rgds.

On Mon, 10 Oct 2005 07:37:39 -0600, Alan Churchill <SASL001@SAVIAN.NET> wrote:

>No axe to grind on Oracle, it's just not my preference. It could be a lot >easier to use and install. My opinion. > > >Alan Churchill >Savian "Bridging SAS and Microsoft Technologies" >www.savian.net > > >-----Original Message----- >From: SAS(r) Discussion [mailto:SAS-L@LISTSERV.UGA.EDU] On Behalf Of >ben.powell@CLA.CO.UK >Sent: Monday, October 10, 2005 4:25 AM >To: SAS-L@LISTSERV.UGA.EDU >Subject: Re: SAS vs. SQL > >Why the axe to grind against Oracle? > >I think this is an esoteric question. A case could be made for each of >SAS/SQL Server/Oracle or vanilla SQL in whatever package for a given set of >circumstances. Currently there's room in the market for all the above >mentioned. > >On another level: Oracle is the most database centric; SAS is the most >analysis centric; SQL Server is the most Microsoft centric (buy up >variations of whatever is going on in the market and role it out beneath the >banner (anyone on here know the name of technical lead behind SQL Server? >You can't because you're thinking of a cabal of lawyers - which is not to >say its not still a great product). > > >Rgds. > > >On Sun, 9 Oct 2005 18:45:32 -0600, Alan Churchill <SASL001@SAVIAN.NET> >wrote: > >>tpc.org is no longer just OLTP: they cover a number of categories. >> >>So many people (at one point, myself included) view the SAS vs any RDBMS >>vendor as a 'winner-take-all' chain match: I no longer hold that view. The >2 >>sides work well together if done properly. >> >>Btw, I'm not an Oracle fan ;-] You could probably have guessed that. >> >>Alan Churchill >>Savian "Bridging SAS and Microsoft Technologies" >>www.savian.net >> >> >>-----Original Message----- >>From: SAS(r) Discussion [mailto:SAS-L@LISTSERV.UGA.EDU] On Behalf Of David >L >>Cassell >>Sent: Sunday, October 09, 2005 6:24 PM >>To: SAS-L@LISTSERV.UGA.EDU >>Subject: Re: SAS vs. SQL >> >>Alan quickly responded to me: >>dc>>"If you are looking at MS SQL Server, then SAS will query its native >>files >>dc>>significantly faster then MS SQL will query *its* native files." >>> >>>I don't believe this is true. >> >>It used to be. But then, I'm pretty old, so my data may be long past the >>expiration >>date on the carton. (Just like me!) >> >>> I've tested it and conferred with SAS on >>>methodology. Indexed SQL Server tables are very, very fast. Even >>>non-indexed >>>tables are lightning quick. >> >>Good. I wish some other databases would follow suit. >> >>>www.tpc.org >>> >>>highlights the databases and who is on top in various categories. >> >>But I specifically said that I was talking about non-OLTP, didn't I? >> >>TPC's benchmarks are very definitely OLTP. (At least, those are the only >>ones >>I'm familiar with.) And lots of databases gear up for fast OLTP, and as a >>result >>fall down at the other end. Can you say 'Oracle'? There. I knew you >>could. >> >>>I would say though that I don't think that someone should say use this or >>>that exclusively. SQL Server and SAS can work well together if done right >>>and they each compliment the other IMO. Both bring a tremendous amount of >>>power in their own right. >> >>Agreed. SAS is not a complete RDBMS, and doesn't claim to be. Lots of >>people really need OLTP, commit/rollback, and other features. >> >>David >>-- >>David L. Cassell >>mathematical statistician >>Design Pathways >>3115 NW Norwood Pl. >>Corvallis OR 97330 >> >>_________________________________________________________________ >>Don't just search. Find. Check out the new MSN Search! >>http://search.msn.click-url.com/go/onm00200636ave/direct/01/