Phil Rack is the creator of a Bridge to R and SAS Bridge to R which enables both WPS and SAS softwares to connect to R. He is also a WPS Reseller. WPS is a base SAS equivalent that can take in SAS code , SAS datasets, write SAS code, and create SAS datasets ( and also create its own format)- at the cost of 660 $ a license ( and almost one tenth of a SAS Institute installation on network servers). Having worked in SAS language and analytics consulting for almost 26 years ,Phil runs www.minequest.com besides running the SAS Consultants network that mentors analytics consultant globally ( I am an ex- member :))
Ajay- What has been your career journey. What advice would you give to someone entering a science career after high school?
Phil- I started out consulting full-time in 1983. I left an analytics job with McMillan-McGraw-Hill Publishing because I didn’t believe the company was investing in BI tools and training as it should. That was pretty early in terms of when BI was becoming important.
Many companies at that point saw BI as only two things.
(a) Ability to forecast sales and
(b) ad-hoc reporting with sums/totals and percentages.
It was obvious to me that I had to make a change …
Phil Rack is the creator of a Bridge to R and SAS Bridge to R which enables both WPS and SAS softwares to connect to R. He is also a WPS Reseller. WPS is a base SAS equivalent that can take in SAS code , SAS datasets, write SAS code, and create SAS datasets ( and also create its own format)- at the cost of 660 $ a license ( and almost one tenth of a SAS Institute installation on network servers). Having worked in SAS language and analytics consulting for almost 26 years ,Phil runs www.minequest.com besides running the SAS Consultants network that mentors analytics consultant globally ( I am an ex- member :))
Ajay- What has been your career journey. What advice would you give to someone entering a science career after high school?
Phil- I started out consulting full-time in 1983. I left an analytics job with McMillan-McGraw-Hill Publishing because I didn’t believe the company was investing in BI tools and training as it should. That was pretty early in terms of when BI was becoming important.
Many companies at that point saw BI as only two things.
(a) Ability to forecast sales and
(b) ad-hoc reporting with sums/totals and percentages.
It was obvious to me that I had to make a change to do the kind of work I wanted to do. In terms of training, I was formally trained as a demographer and did my graduate studies at Ohio State so I received a pretty good dose of quantitative subject matter as well as a unique perspective on the social implications of markets and geography. If I had to do it over again, I would probably take more course work in the subject area of the “Family.” I’m always amazed how many times the work I do in banking and finance revolves around the family lifecycle.
Ajay- What has been the biggest project success you have seen in your consulting practice?
Phil- This goes back three years to a project where I was working on Basel II compliance with a commercial banking client that I just loved working with.
A few months into the engagement, they pulled me aside and asked me to put together an automotive portfolio stress test for them. This bank had very large loan exposures to the auto market for second and third tier suppliers to the Big Three as well as international auto manufacturers.
The Risk Management group and I sat down for a couple of days and pulled together a project plan and an outline of what we needed to be able implement a dynamic Auto Risk Stress Test Model for this portfolio. The software used was SAS/Base and Excel and the program allowed us to modify 50 to 60 parameters to model different scenarios. All together, it took perhaps three weeks to implement and it was amazingly indicative of the fall out of the auto industry as well as foreshadowing some of the financial carnage in South Eastern Michigan such as lower property values and unemployment.
Ajay- “It is not what you know, it is whom you know.” Comment please as an SAS consultant.
Phil- In terms of my business, 80-90% of the work I do is either based on prior work that I’ve done for that company or through referrals. If you want to have a successful consulting career, you really have to pay attention to developing your network. I’ve taken advantage of social gatherings such as charity events and other social mixers to try to extend my network. I hand out a lot of business cards every year. Formal organizations exist here in Columbus, Ohio such as TechColumbus.org that is a dynamite organization that helps small tech businesses in the area of networking, financing, access to different hardware platforms for testing, etc… I have mixed emotions about the value of some social networks however.
I see so many individuals on LinkedIn that have 5,000 connections that I have to wonder what it is these folks really do. Who has the time to read all the updates and postings for 5,000 people and still be able to get work out the door? ( Note from Ajay- I have 6300 connections on LinkedIn . Ouch !!)
Ajay- What motivated you to write the SAS to R and WPS to R bridges? (Which IS your favorite analytical tool, since you are active in all three?)
Phil- It started out as a “proof-of-concept” exercise and it’s just keeps growing. The WPS to R Bridge is a piece of software that I wrote originally for WPS users to access R from within the WPS Workbench. For those who are unfamiliar with WPS, it’s a SAS/Base alternative that is extremely compatible with your existing SAS/Base software and your code is just plug-and-play. WPS doesn’t have the statistical capabilities of SAS such as SAS/STAT, ETS, OR, etc… so the idea was to write a bridge so that WPS users wouldn’t have to learn a new GUI/IDE to use R. The Bridge gives WPS users access to R graphics as well as any of the R statistical libraries but it has the advantage of the superior data handling of the SAS language. One of the new features is the ability of the WPS to R Bridge to run R programs in parallel. Depending on your hardware, you can easily run six to a dozen R programs simultaneously and collect the R listing and log files back into the WPS listing and log in the order you submitted the programs.
I did write a Bridge to R for SAS users but very few SAS users have expressed interest in it. I suppose that SAS users are happy enough paying the fat licensing fees to SAS that it just doesn’t matter to them. I have to say, my favorite tool at the moment is WPS. I find the interface/workbench to be so superior to what SAS has to offer that I now find myself writing code in WPS and then taking it over to SAS if that’s what the client requires.
Ajay- What do you think about internet based delivery and social networking including communities and lists changing the software product cycle?
Phil- This somewhat goes back to question #3 in terms of communities. I think it has its value as a place to share your concerns and find answers to difficult programming issues. Now, Internet delivery and cloud computing I find very interesting. I think there’s some strong advantages to using the cloud to provide services to your clients. If you look at the SAS pricing model, they really take it to you financially if you want to use your license to be a DSP (data service provider) or put your code on an intranet/internet. For some reason, SAS is just hostile when it comes to small and medium sized businesses. Companies like World Programming who license WPS have a much more realistic idea of licensing in that you can expose your WPS license to your intranet/internet and not have to pay 10x the fees that SAS charges. WPS doesn’t charge additional fees for those who are DSP’s either and there are quite a few of them in the Pharma domain.
Beyond security challenges associated with cloud computing, I think SaaS that provides analytical services such as high performance forecasting and name and address cleanup and verification are ripe for the picking. One other issue I see with cloud computing is when you have tens of gigs of data that you have to move from your desktop or server to the cloud. The infrastructure just isn’t fast enough, or let’s say reasonably priced, to allow for moving this amount of data to really scale well.
Ajay- How does MineQuest intend to influence the analytical software paradigm?
Phil- I think the role for MineQuest in the next few years is twofold.
We’ll keep offering services to banks and other financial service firms in the area of Operational Risk and SAS programming.
The other area is to help these large financial service companies realize that they can save millions of dollars by moving their SAS Server licenses to WPS. This also allows the smaller businesses who have steered away from SAS software because of cost to begin using WPS and not take such a big financial hit. I find it exciting to think how this will also open the job market for the thousands of SAS programmers out there already.
The BI battles are taking place on the desktop and Windows Servers and MineQuest has invested a lot of time and effort in creating macro libraries to help these organizations migrate their code to WPS and access R for advanced statistical capabilities.
We believe that the bread and butter software for almost any financial organization in the BI realm ultimately revolves around the SAS language for reporting, summarization and disbursement of data and we plan to continue to serve that market.
( Ajay –
SAS language uses mainly Procs and Data step for output and input.Base SAS is a product copyrighted by the SAS Institute (www.sas.com) .SAS Institute has been leading the analytics world since the 70’s.WPS is copyright of World Programming Company (WPC) (www.teamwpc.co.uk/products/wps ) )