LISTSERV at the University of Georgia
Menubar Imagemap
Home Browse Manage Request Manuals Register
Previous messageNext messagePrevious in topicNext in topicPrevious by same authorNext by same authorPrevious page (July 2008, week 1)Back to main SAS-L pageJoin or leave SAS-L (or change settings)ReplyPost a new messageSearchProportional fontNon-proportional font
Date:         Mon, 7 Jul 2008 11:50:52 -0500
Reply-To:     Mary <mlhoward@avalon.net>
Sender:       "SAS(r) Discussion" <SAS-L@LISTSERV.UGA.EDU>
From:         Mary <mlhoward@AVALON.NET>
Subject:      Re: A powerpoint slide show comparing R with SAS and SPSS
Comments: To: Alan Churchill <savian001@gmail.com>
Content-Type: text/plain; charset="UTF-8"

I googled it, and Google referred me to Michael A. Raithel's book, which gave me an answer!

Now I understand a little better; MXG and MICS are applications written in SAS, and do a lot of performance statistics, so mainframe shops want these applications to run rather than writing new ones, so WPS, being a SAS clone, could run these SAS applications so mainframe shops would not have to license SAS.

Even if the shop does have a database package with a programming language, such as Oracle, SQL Server or DB2 (with C/PL/1/Cobol), it would not automatically have anything written in Oracle or SQL Server's programming languages that would do the same thing as what is already written in SAS by purchasing MXG/MICS. So dropping SAS is like dropping a PL/1 license; any code written in that would have to be converted to something else, is that right?

This seems to me a more serious obstacle to dropping SAS (or at least not replacing it with WPS) than the statistical work like that done in pharmacy analysis, since the amount of code already written in SAS for statistical users may not be as large as something like MXG/MICS, and they could more easily convert over to SQL Server or Oracle to do data manipulation, perhaps keeping 1 SAS license around on 1 machine in order to rerun previous analysis written in SAS.

Comments from pharmacy types? Is there a lot of "prepackaged" SAS programs that you just rerun each time, or do you pretty much create new programs each time you work with new applications?

-Mary

----- Original Message ----- From: Alan Churchill To: 'Mary' ; SAS-L@LISTSERV.UGA.EDU Sent: Monday, July 07, 2008 11:15 AM Subject: RE: A powerpoint slide show comparing R with SAS and SPSS

A guess? Every mainframe site in the world if you toss MICS in as well (which runs on SAS, the same as MXG). How many sites? Thousands.

Yes, they all have RDBMSs, I am sure, but that doesn’t mean you can flip a switch and have MXG/MICS run on it. MXG is in excess of a million lines of SAS code, MICS is probably about the same. I don’t think you can use an RDBMS under the covers but one of the other MF’ers on here can probably enlighten us more. Last I knew, MXG still runs old style macros and has not been converted to the newer macro style. That could be old info by now.

Any MXG people out there that can weigh in on the above?

Alan

Alan Churchill Savian www.savian.net

From: Mary [mailto:mlhoward@avalon.net] Sent: Monday, July 07, 2008 10:04 AM To: Alan Churchill; SAS-L@LISTSERV.UGA.EDU Subject: Re: A powerpoint slide show comparing R with SAS and SPSS

How big a market is MXG? Why don't these shops yet have a database management system like Oracle or SQL Server, particularly if they are pushed away from SAS by its high pricing in some sectors?

-Mary

----- Original Message -----

From: Alan Churchill

To: 'Mary' ; SAS-L@LISTSERV.UGA.EDU

Sent: Monday, July 07, 2008 10:49 AM

Subject: RE: A powerpoint slide show comparing R with SAS and SPSS

A couple of things.

As I stated before, WPS was a play for MXG. If they pick up new business in the stat and SAS world, it would be icing but it isn't the primary move. In the MXG world, they can cause damage. SAS is in a lot of lines of business, though, and will shift as needed. ITRM is also a nice wrapper on MXG and WPS is not competing there. MXG, BTW, uses SAS datasets as storage mechanisms and not a RDBMS.

A lot of open source is done by people who just like to program (like me). Paul Oldenkamp has started a nice project for open source SAS and I plan on contributing a lot of code to that project. Others have joined as well.

Finally, a lot of companies do not like open source since there is no liability behind it. Sure, it's cheap, but companies like to be able to go after pockets if a calamity happens. They also like tech support and SLAs behind deployed infrastructure.

Alan

Alan Churchill Savian www.savian.net


Back to: Top of message | Previous page | Main SAS-L page