Welcome.
My name is Len Sumnler. I have worked in IT since I graduated college from Loyola University in Chicago in 1974. I have worked in large Fortune 500 companies, large and small consulting firms, and small privately owned companies. I have coded in IBM Cobol, Wang Cobol, IBM assembler, FoxPro, Dbase, Paradox, MS Access, C, Clarion Topspeed, Perl, Ruby, and Python. Database experience includes ISAM, VSAM, IMS DB/DC, MS Sql, MySQL, PostgresSQL, Sybase, and Sqlite. Business domains include, accounting & manufacturing, insuranse (life, property/casualty, auto), and healthcare(physician back office, medical billing, claims).
Generally speaking I have always been a business application programmer. I code tons of reports and do a lot of CRUD type database stuff. I know, I know to many how boring, but I have always loved this work. Why, because as mundane as many programmers may find this most non techies (that is the business comsumers of what I produce) love the work I do. In addition, I generally work for smaller companies where I have complete control over the design of the application and the database. Here is the other little secret although 80% of what I produce is the same thing over and over again it only account for 20% of my work load. Most of the really interesting coding is on the business logic that makes each individual system I create unique and gives my customer, users, and clients that competitive edge, whether thats cutting expenses, new way of looking at their data, simplifying the UI for the way they do business, or creating ad hoc user reporting systems. Oh by the way this 20% of the application in generally where I spend 80% of my time.
The time I have spent with users in system design and analysis has been some of the most rewarding. I have always had a different take on users. I assume they don’t know shit about what I do or how computers work, or why I need to have records with unique keys nor should they; that is not their job, its MINE. I’m there to solve their problems and I do that extremely well.
As an example of what I do one of the systems I am most proud of was a system written entirely in Excel, using VBA, and macros. Why you ask, because corporate only allowed development in ASP/Net and even then it had to be signed off on and this project was done under the radar and in truth didn’t need all of the complexity. The system was a cage cashier system for a local casino that facilitated a cage cashier to keep their money draw stocked with the appropriate amount of cash during their shift and then close out their drawer at the end of their shift. Prior the the Excel system this was all done manually through dozens of three part forms manually posted to various intermediary and end of shift balance forms and all calculated by hand. I reproduced all of the forms in Excel down to the form numbers, automated all of the calculations and balance sheet postings and when the system was complete trained the entire cage cashiering staff in a one hour training session. I was a hero to the cage cashiers, cage supervisors, and head of the department when OT dropped by 25%.
Just as a side note. I don’t really believe in long project, you know the 3 to 5 year projects. I have no problem with 1, 3, and 5 year goals but an IT project should be in the 3 to 6 month range. That is, I deliver usable code in 3 to 6 month and working prototype code I can show and let them hammer on ASAP. To this end I have done and used many tricks but usually they fall into one basic category of template or framework type code. In my early days I would create my own templates of fragments of code I would reuse. Larger chunks of code may even include tokens that I could in-mass change. In the late 90′s I found a development product that ran under MS DOS and then Windows called called Clarion Topspeed which automated this process for CRUD and reporting to a level I have yet to encounter anywhere else either Windows or Linux. Sadly, the company never produced a Linux product and has taken a new direction into targeting the .Net environment. Now I have nothing against .Net but I due question the companies rational for going after this market when you can buy the MS .Net products for around the same price.
So this has lead me on my quest to find another programming environment to do the work I love and expand out into the web environment. This has lead me to Python. The shortcoming of python are all pretty much the same as most of the alternatives I’ve look at GUI development is a pain in the ass as most of the languages require hand coding the screen coding and doing general business reports have been a pain again requiring hand coding. The up side is python is a wonderful language to code in due to its concise nature, readability, available libraries, cross platform capability, and quick turnaround. The question of slow execution speed generally is not a problem in general for 98% of the work I do and in those few cases I have now problem dropping down to C. Remember for most of my work my applications are generally waiting on the user to type or make some type of decision. I don’t really expect it to be a problem in the web environment either for about the same reason. It has an object orientated nature but at the same time does not force you to develop your code using OOP, it is an option. I am neither a fan or a critique OOP I think it can be useful if used in the proper way, I am just not sold on the idea that everything must be in OOP. This could get to be a long discussion and maybe I save that for a post at some later date.
