Date: Fri, 11 Feb 2011 11:52:40 -0800
Reply-To: Sterling Paramore <gnilrets@GMAIL.COM>
Sender: "SAS(r) Discussion" <SAS-L@LISTSERV.UGA.EDU>
From: Sterling Paramore <gnilrets@GMAIL.COM>
Subject: Re: Thoughts on GUI Development
In-Reply-To: <D318FFABEC06504E9EB926116EF6CC0606B60E09@VAPWVBE022.us.ad.wellpoint.com>
Content-Type: text/plain; charset=ISO-8859-1
I should be getting 4.3 later this month so I'll be excited to check it out.
The webinar looked a little underwhelming though.
On Fri, Feb 11, 2011 at 10:54 AM, Chaitovsky, Avi-Gil <
Avi-Gil.Chaitovsky@anthem.com> wrote:
> While I haven't had a chance to use it hands-on (since my company is
> still using EG 4.2), EG 4.3 apparently is able to analyze code and
> "translate" it into the GUI.
>
> Chris Hemdinger, who was one of the people behind EG, recently gave a
> webinar on the new features in EG 4.3, including this one. You can
> watch it at http://www.sas.com/reg/web/corp/1259797.
>
> Avi
>
> Avi-Gil Chaitovsky
> WellPoint, Inc.
> Empire BlueCross BlueShield
> HealthCare Analytics
> Sr. Business Information Analyst
> 165 Broadway, 14th Floor
> New York, NY 10006
> Avi-Gil.Chaitovsky@anthem.com
> (212) 476-1483
>
> -----Original Message-----
> From: Sterling Paramore [mailto:gnilrets@GMAIL.COM]
> Sent: Friday, February 11, 2011 12:58 PM
> Subject: Thoughts on GUI Development
>
> SAS-L,
>
> I started learning SAS just a little over two years ago to organize and
> study health insurance claims. I came from a computational chemistry
> background, and am very comfortable with command lines and ancient
> computer
> languages like FORTRAN77 (yes, as a grad student we were doing
> cutting-edge
> molecular simulations using a language version developed 3 years before
> I
> was born! - there were no good/free FORTRAN90 compilers until around
> 2006).
> When I showed up at my new job, my supervisor had just purchased SAS BI
> and
> started handing me projects solve with Enterprise Guide. Knowing
> nothing
> about SAS at the time and only a little about data structure, I made
> heavy
> use of the GUI for the first 3-4 months. As the tasks I needed to
> accomplish became more and more complex, the GUI became more and more
> limiting and exponentially more frustrating. So I started learning
> datastep
> programming and have been much more satisfied with SAS programming ever
> since.
>
> However, there are some really nice aspects of GUI's that are absent in
> code. The ability to glance at a screen and immediately see the
> relationships between datasets is exceptionally powerful. Lately I've
> had
> to reverse-engineer some inherited code and I spent hours mapping out
> dataset relationships in Visio so I could trace out how changing one
> component would affect the final output.
>
> It got me to thinking that most of these code-writing GUIs are developed
> backwards. It seems that GUI developers spend a lot of time trying to
> force
> code to conform to their ideas of a GUI when they should be spending
> more
> time writing GUIs that interpret code written by humans and then feed
> any
> manipulation back down into code. We almost need two different types of
> compilers: one that interprets code and speaks to the computer and
> another
> that interprets code and speaks to the visual cortex.
>
> I doubt these ideas are new. Does anyone know of projects (SAS or
> otherwise) that are attempting to do this kind of thing?
>
> -Sterling
>
>
> CONFIDENTIALITY NOTICE: This e-mail message, including any attachments, is
> for the sole use of the intended recipient(s) and may contain confidential
> and privileged information or otherwise protected by law. Any unauthorized
> review, use, disclosure or distribution is prohibited. If you are not the
> intended recipient, please contact the sender by reply e-mail and destroy
> all copies of the original message.
>
|