An email today triggered a thought I’ve had before, but never bothered to put into writing.
Once you learn to use a piece of software, are you tied to it forever, for better or for worse?
This is an multi-faceted problem, so I’m not even going to pretend like I have an answer. However, it’s worth investigating, especially in the context of fanboiism.
Learning anything requires effort, and presumably, after investing effort, one expects a return. Software is no different, regardless of the task the software performs. Learning software creates an automatic bias because once a level of mastery is achieved, it becomes a marketable skill.
People don’t usually devalue their skills.
Functionally, though how can anyone fairly judge software A against software B? If you’re accustomed to using A, you’ll focus on how B does things differently in a negative way because you have to reinvest to learn the new way.
B’s process could be more efficient, but not until you learn how to do it. So, it’s virtually impossible to know which is better unless you give B a fair shot and invest equally. How often does that happen?
This isn’t always true, but think about your own experience. I’m sure there are levels of attachment that you’ll never break.
The cost of retraining is one very important concern with enterprise software, whether it’s retraining users to use something new or training IT to support something new, change costs money.
The email from earlier today compared Android (and the Android SDK) unfavorably to the iPhone and its SDK. I’ve not used either, but it was a bit surprising to me, considering the black box that is the App store approval process.
I got another dose today as I documented how to use the WebCenter bookmarklet Rich (@rmanalan) built. I didn’t realize it, since I don’t use Windows/IE regularly anymore, but IE makes it much tougher to add a bookmarklet. First, you can’t drag and drop code, and second, you have to confirm that you want to use it.
I understand why. It was just a reminder of how different a browser IE is.
What do you think? Find the comments.
One answer to the problem is to provide a “transition mode” in the new product to help people converting from the old product. Back in the 1990s when Lotus was king and Microsoft Excel was back in the pack, Excel offered a “Lotus compatibility mode” that allowed Lotus users to use the same commands that they used in the old program. As time went on, these users would presumably learn how wonderful Excel was and do things in native mode, but this offering helped users of the old product adopt the new one.
Needless to say, there are plenty of examples of this. It's a pity that Word 2007 didn't offer a “Word 2003 compatibility mode,” though.
P.S. In my industry, the automated fingerprint identification systems industry, government agencies usually contract to use a particular vendor's AFIS for 5 to 10 years, give or take, before they go out to bid again. In the majority of the cases, agencies elect to stay with their current vendor – not only because of the retraining involved, but because of the data itself. While there are standards for interchange of data between different AFIS vendors, we all use proprietary data storage features to some extent, meaning that even in the best of circumstances, it costs more money and time to move from one vendor's system to another vendor's system, rather than sticking with a next-generation system from the same vendor. (FCC disclosure: these views are my own and are not the views of MorphoTrak bla bla bla. I don't think Louis Gray has a badge for this one.)
Heh, the proprietary data issue is common to many types of software, and even when migration is possible, it's not ever easy or pain free, e.g. have you ever exported a Keynote deck to PPT?
I haven't seen much of the compatibility mode hat-tipping lately, but that's part of the Cupertino philosophy. They're at the heart of a lot of software warring lately.