Why all the complexity with version numbers?

I run Piwik on a few of my sites to collect statistics. They released a couple of new versions this week, with the latest being 2.16.5.

I don't understand the need for a version number to include three components. Why wasn't this 2.17? What made this an x.x.x release and not an x.x. release? For that matter, why wasn't this simply version 3?

No one seems to share a strategy. Internet Explorer, for example, was first released in 1995 and took 21 years to get to version 11 (on Windows), with a few sub-versions. When Microsoft released Edge, the replacement for IE, they dropped any reference to version numbers.

That was very much in keeping with the other web browsers that dominate the market: Chrome and Firefox. Chrome was only released eight years ago, yet it up to version 53. They use some sub-versions for specific builds, but otherwise they just keep raising it. Firefox is up to version 49.

The first version of Windows was released in 1985 and is now up to version 10 (though likely won't go any higher). They skipped 9 in favour of 8.1 and Microsoft only started using version numbers again with the release of Windows 7.

MacOS (as it's now known) is famously stuck on version 10 (X), and has been since its release in 2001. We're up to 10.12 currently. Why just keep updating the sub-version? Who knows? Presumably because Steve Jobs liked the roman numeral for ten and didn't see a reason to switch (he wasn't selling software).

I can see some benefit to using version numbers from a sales and referencing perspective, but most people simply don't care if they're using 1.10.99 or 34. Most of them (and I say this as an experienced support person) simply won't know.

So how about we agree that a major and a single sub-version is enough. Unless it's a patch, just keep incrementing the major version number. Who cares if we end up with version 394?

0 Comments

Post a Comment