It has been almost a decade since I first learnt the SAS language. I have learnt statistical programming languages before that (namely SPSS) and have been proficient in training in other languages (like R). In addition I have been exposed (either due to necessity or sheer project pressure) to use and incorporate elements of SQL, Python, Octave as additional languages and packages like KXEN and JMP for supporting analytics.
If you ask me what has been the most profitable decision I ever made in my decade long career as an analytics professional- it would be I chose to learn SAS and chose to train myself rigorously in it.
The answer for that is simple- demand for SAS language professionals continues to exceed supply. While some thought that newer languages like R, Python or even WPS would diminish the demand or need to train on SAS products, the answer is just like Microsoft Windows is an essential skill while Linux is a desirable skill for a corporate career; similarly SAS is an essential language in analytics.
Some people complain that SAS language products cost more. If you are complaining about the cost, it means that you are stuck in a low value commercial product. Take this simple example. If you improving the conversion rate of a direct mail campaign by 1 percent, it would simply provide a profit of a few thousand dollars to your client or company. That covers your annual SAS language license fees! Is there any other product that charges you for one year and you get your money back in one month? Of course, if you are not dealing with high performance analytics you should choose other options.
Yes, if you are in academics and research there is enough evidence to state that R is a better language to be proficient in. But as of 2012, SAS language users continue to exceed analytics professionals in all other languages put together. I know both SAS and R- R is a boon from heaven for my research. When I am advising corporate clients, they pay for the software and I end up recommending SAS language over other products simply for the ease of usage and training.
Let us end this debate forever analytically. Take the noted article at http://r4stats.com/articles/popularity/ on comparing SAS and other languages. SAS language is ahead in one parameter by a scale of 4:1 to SPSS and 12: 1 over R. Yes that parameter is Number of Jobs.
So I you are going for advanced analytics learn R. But if you just want a job, a steady analytics career and are planning to enter this sun-rise industry of business analytics, you might just want to read little SAS books as an initial step.
(The Author is a trainer and writer on both SAS and R language. Choosing between languages is a matter of context and practical usage for him rather than ideology. These are his personal views only!)