"Perl demands more programming expertise"

There's the rub. Web based multi-threaded high volume transactional applications are megabucks to develop in SAS but can be rolled out in no time by a perl programmer adding a few modules to an Apache server. But then such a programmer might also charge megabucks in the long term...

HTH.

On Thu, 11 May 2006 22:09:19 -0700, David L Cassell <davidlcassell@MSN.COM> wrote:

>geoffrobinson@GMAIL.COM wrote: >>Hi, >> >>I was trying to study the varying strengths and weaknesses of perl and >>SAS. I did some searching, and it seemed people mentioned there were >>strengths and weaknesses. However, I could find anyone who actually >>spelled those out (they were probably warry of flame wars). >> >>Does anyone have any opinions? I'm especially interested in jobs which >>involve going through large amounts of data. What would give perl an >>edge? What would give SAS? >> >>I'm also thinking about standards based compliance, support from a >>given community, and price. In other words, SAS may work reasonably >>better but I avoid vendor lock-in with Perl and I don't have to pay for >>it. When is that outweighed by other considerations? > >For me, the question is always "what do you want to do?" > >If you want to do systems admin work, then Perl has a clear edge, >regardless of the platform. > >If you want to do serious statistical analysis, SAS has an insurmountable >lead over Perl. But I assume you would do serious stat work in R or >XLispStat if you like open-source tools. > >So, what do you want to do with your language? > >Perl will try to do everything in RAM. This gives it a big speed advantage, >but it means that you have to specifically handle large-dataset issues >yourself. At least you have that option with Perl. But SAS is better >designed for truly large data sets. > >Do you want to connect to external data sources? Both do that. >Well. But SAS adds more costs, while Perl demands more programming >expertise. > >Do you want to do web work? Do you want to do OLTP? Do you >want to do data warehousing? Data cleaning? Database management? >The two packages have differing strengths and weaknesses. > >I find that they complement each other nicely, so for me the >question is not either/or. I choose both. > >HTH, >David >-- >David L. Cassell >mathematical statistician >Design Pathways >3115 NW Norwood Pl. >Corvallis OR 97330 > >_________________________________________________________________ >On the road to retirement? Check out MSN Life Events for advice on how to >get there! http://lifeevents.msn.com/category.aspx?cid=Retirement