Hi Arthur -

I think this sort of user-domain specific documentation is a niche generally filled by books-by-users. Also the "SAS Communities" idea on sas.com is in this vein.

I agree that SAS online could certainly improve. Clicking html lists that take minutes to load and having the documentation pane not link to the chapter or book title page are cumbersome.

Paul Choate DDS Data Extraction (916) 654-2160

-----Original Message----- From: Arthur Tabachneck [mailto:art297@NETSCAPE.NET] Sent: Thursday, August 12, 2004 4:27 PM To: SAS-L@LISTSERV.UGA.EDU; Paul Choate Cc: Arthur Tabachneck Subject: Re: Is SAS growing Linearly or Exponentially?

I'm sending this in the hope that some birdies happen to be listening or, better yet, if Dr. JG happens to be following this thread.

I agree with what has been said thus far in this thread (except for whether the trend has been linear or exponential, as I'd think linear might be more appropriate), but seriously question whether SAS is spending enough on documentation.

SAS has been steadily gaining an increased representation in at least both the analytical and general IT processing domains. And, given its ever increasing power and capability, I think it deserves its ever increasing user base.

Making its documentation readily available on-line, too, I believe was an extremely good move.

However, I have major problems with its documentation. Given the diversity of audiences that SAS has amassed, different groups need to have different information available. While some (on this board) have said (slightly paraphrased) that "we" are not paid to know, but rather where to find the needed answers, many of the list's posts are the result of many of us not knowing where to find what we need (other than here).

Don't get me wrong, I would hope that this list never stops or reduces its level of activity. And yes, many folks will simply never RTFMs, but many of the list's posts are from people who DO RTFMs. What I disagree with is the decisions SAS made regarding that we all know everything about what we are doing AND that we all need the same information.

Years ago, as a research-based graduate student, the original (maroon colored) SPSS manual was my bible as it provided a description of each statistical algorithm, what it was for, what assumptions needed to be met, what the consequences were of changing the various defaults, etc. The next generation SPSS manuals also lost their previous degree of (researcher) useability.

I don't even dare to think of how many different types of SAS users there currently exist. I know some who are mainly devoted to pure programming, actuarial science, statisticians, researchers, pharaceutical analysts, data miners and general IT professionals, but I am sure that there are far more groups whom I haven't met.

Personally, I need to read three currently not available manuals, namely ones devoted to actuarial science, statistics, and research. And, quite frankly, I'd not only read those, but the one devoted to IT as well.

My first ever publication looked like a stack of bound IBM cards, but was an attempt by the West Virginia Vocational Rehabilitation Research Center to get other researchers to know some of the available power of what was then known as the Statistical Analysis System. Many of the user created manuals, since then, appear to be trying to accomplish the same objective.

My appologies for the long post but, to end my diatrad, would it not be beneficial for all of us if SAS were to provide separate documentation for its various user groups, provide the information we need to look up, and let us use the current b-board to excel rather than mostly to explain?

Art