Features Creep, Part 2 - Sun, 8 Sep 2013

I’m pleased I was able with my last post to goad James Tanner into responding: Respond? Who Me? Of Course.

I do respect his writings. He writes how he thinks and always makes great points and does not demand that everybody agree with him. I thought he did a fabulous job as a last-minute fill-in keynote speaker at RootsTech 2013. He makes it hard to stay up to date with his genealogical posts as he writes several comprehensive articles a day on all sorts of interesting topics. I’m happy if I can get out one post a week.

So the ping pong ball is back on my side of the table. I’m adding a little spin to the ball:

1. There are people who want a fingerprint identification system on their iPhone. Apple wouldn’t bother adding it unless they thought some people wanted it. If you don’t want it, don’t deride Apple for giving other people what they want. As long as Apple adds new features in a way that basically keeps the features you don’t want hidden from you, then it shouldn’t matter to you. But if the iPhone becomes harder for you to use because of a feature they added that you don’t use, then the problem as I stated earlier is bad design by Apple to implement this new feature.

2. You bring up another point and it is a good one. Features are added in order to sell new versions of the software. Yes, that is what they do. Developers try to add enough features so that every current user will find at least one “must have” feature that prompts them to upgrade. They make more money every year or two off their existing customers that way, in addition to the new people attracted to their program. And yes, they also do things to make older versions obsolete. It’s the usual pattern for most software companies. And as you say, it is frustrating when they do this.

   This is a marketing method to increase revenue. But this type of feature creep “problem” does not imply that the program is made worse. More often, the program is improved each version. With good design, the new version can still be as easy (or even easier) to use than the previous. But the developers have to stick to putting good design first before the willy nilly adding of features just for the sake of getting another release out.

3. Microsoft’s Ribbon does not fix poorly designed features. It does not help you figure out the use of features. What it does do is present the features to you in logical groups as you need them. This is a radical improvement over menus and toolbars that are static and inadequate for access to the hundreds of features programs now have.
4. You want a genealogy program with a workable merge function. Currently they are all hit-and-miss. Well, you’ve hit upon one of my biases. I am totally against the whole concept of merging other data with your own. I feel you should keep other people’s work separate. I believe the concept of virtual merging (keeping the data separate and letting a program display them together) is the correct way to do it. Promoting the merging of other people’s data with yours is one of the 6 Bad Things About Today’s Genealogy Software.

   However, if in fact you happen to have another dataset that you need to merge, I do not believe you should let any automated technique do it for you. I feel the only correct way is to inspect the people/places/things in the files to be merged and manually deciding yourself on which items are the correct or incorrect ones. Maybe both are needed. Maybe one will give you more ideas and clues. At each step, you must verify all the information yourself anyway. So you might as well do it the correct way the first time. Merging mechanically is allowing a computer to do a person’s work. There is no call for it. And if you say that you’ve got hundreds or thousands of people to merge, then see the previous paragraph.

5. You say: “Let’s hope someone makes a really good, simple, elegant, functional genealogy program that can become the new standard.”

   Behold: I’m working on it.