Baby Steps and Little Bites when implementing NEW government software

Ellis was at it again, writing a great column. You can mull through it, but this really stuck out;

Unlike Sioux Falls, Clovis took a bite-by-bite approach to rolling out EnerGov. While Sioux Falls is using EnerGov for multiple tasks, Clovis started first with building permits. Next, it plans to add code enforcement, and then move on to another task in a step-by-step process that is less complex than in Sioux Falls.

I have found in my professional life that diving head first into a project with multiple elements all at once never ends well. Taking on projects in a step progression is always the best way to go. That was always my beef with ‘Shape Places’ it should have been broken up into sections and reviewed and voted on in those different sections.

Do I think the city should move forward on the new software? Yes. But their approach of releasing it all at once is not a good idea. Ask Obama about the ACA website. But typical of a ram rodding administration.

Even if this goes live next month, it will probably have tons of bugs. Take SIRE for instance, it has been live for several years and still does not work properly. And when you ask city staff to fix it, they are either in denial or point fingers. I see the same happening with this new software. Get ready for a big f’ing mess.

A local IT veteran put their 2-Cents in;

Having worked on these types of projects over many years you learn very few ‘managers’ actually know anything about what makes their departments work.

Very few managers are qualified to direct the daily efforts of their department much less design software. In the 40 years of designing complicated projects, I have decided to question everything anyone ever says to me. I heavily weigh the statements and direction of ‘managers’ so it is usually worth less than the regular line person. The only people who know how things work are the line people who sit and actually input the data.

Most systems are designed to get certain answers over actually making the data correct. You see, managers only want to show ‘improvements’ under their watch. They usually do not want old data added to the system because it might prove their old reports might be wrong if examined under the ‘new’ matrix. The line people want it done correct and very accurate. The managers want numbers shown for reports with little effort. There is very little room for a software developer to do the job right with these divergent needs.

The line person who has to enter the data will be required to reenter a lot of data not ‘acceptable’ to managers. If a report is generated using the data these line people put in not accepted by a manager, a low level head will roll before it is publicized.

Someone like our mayor will want the reports to be pretty and superficial.

A Kevin Smith or Tracy Turbak want only positive information to be produced for the public use. This data cannot be allowed to embarrass them. If the wrong type of data is allowed to be seen by public, their jobs could be on the line.

There is no way for an outside developer of software to win in these situations. The outside developer is be controlled by idiots of the first degree. Tyler likely has a salesman in charge of the software writers. So we have a convergence of two salesmen: a mayor and a Tyler project manager both wanting this to be over with.

So who is allowed to put a buffer between the two? Now the city managers need more consultants to interfere in the process. The new consultants are brought in to find different ways to tell the Tyler group how to make the reports pretty and hide the bad data.

The managers of large scale businesses and governments do not want real data being reported. They want a way to protect their budgets and personnel numbers. They do not want to protect their employees only their quantities. This software program will prove be a disaster if ever fully investigated. You do not hire more consultants to control consultants if the person in charge of the project actually knows what the real end result is.

The only thing we will get out of this software program is excuses, partial answers, hidden data and protected jobs. This writer is not involved in this project but having decades of working in many others, we developers despise working for paper pushers who do not understand their jobs. We developers learn to make too many bad compromises in the name of just getting away from the incompetence.

We know we will be watching someone years from now finding a way to make the junk work.

As in running any business, a city software plan should work if it is rolled out with a goal aimed toward perfection in data first and not what color the report font is when sized in 20 point. We citizens want correct, verifiable data useful to our lives. We do not report generating software to make the managers look better when snowing the City Council on Tuesdays.