A post from a person I follow on twitter got me thinking about tech product development…

This was on a Mac in the UK. With a UK keyboard. With the system locale set to UK. With the system language set to British English.
Yet the software offered an autocomplete using the American styling of “Mom”, seemingly ignoring the locale settings on the machine!
Okay, it’s not escaped me that Word for Mac is a MSFT product. So maybe this is about cultural insensitivity in tech (or maybe all) companies in general, but as this was on a Mac, I’m going to use Apple as an example of what could be done better.
Everyone remembers the Apple Maps launch debacle, right?
So many of the faux-pas could have been avoided if there was a bit of cultural sensitivity and local knowledge applied when sanity checking the mapping data, especially the place-mark data.
Firstly, there’s a GIGO problem at work here. Apple took in some seriously old source data.
For instance, the data was so out-of-date it contained companies long since closed down, gone bust, or merged with competitors. Yet, if there had been a bit of local clue applied, these could have been caught in the sanity checking of the data.
Here’s a few examples still there, which could have been eliminated this way, all in the locality in which I live:



I understand that Apple did conduct a beta of Maps, but if they did, they either didn’t have many beta testers in the UK, or the ability to let them correct bad data wasn’t great, or the feedback simply didn’t make it to the released version.
But, that’s okay, now it’s released, it can be corrected by crowd-sourcing – i.e. getting our paying customers to do our jobs for us – right?
Well, there is a “report a problem” option, but that doesn’t seem to be working well, either it’s too hard to report an inaccurate place-mark, there’s a colossal backlog of reports, or they are going straight to the bitbucket.
If only they had bothered to actually get some local knowledge, obvious clangers like these could have been sifted out early in the process.