If anyone wants a real-world high profile example of the advantages of releasing often (in an Agile-like way) look no further than the release schedules of Internet Explorer (IE) and the rest; aka: “The Browser Wars”.

As anyone who’s even vaguely associated with web development will know, IE’s current perception isn’t enviable – particularly on the subject of HTML5.  In a nutshell, the next release of IE (IE9, which will include some support for HTML5 related features) is tentatively scheduled for 2011, where-as IE’s competitors are releasing HTML5 related features now. 

It’s not hard to find blog posts and various articles on this general subject – they seem to keep popping up regularly; the latest nail in this somewhat over-nail coffin appeared the other day on WebMonkey – which in turn references a really excellent piece by Peter Bright over at arstechnica.com.  The thing about these posts is that they gave that metaphorical penny in my mind the last wee push it needed so as to drop: it’s surely has to be one of the great classic examples of what happens when you don’t release early and often in a competitive market.

Exhibit “A”: Release History

Peter Brights piece compared IE9’s expected release date with the vastly more regular updates of FireFox in very recent history; as interesting (or depressing) as this is, the complete picture is even more remarkable.

Sources: All dates were taken from Wikipedia or the relevant browsers website.  The versions and dates given are for stable public releases – often just for the Windows platform.

• 1995, Sutherland and Schwaber jointly presented a paper describing Scrum at OOPSLA ’95 (Wikipedia: SCRUM).
• October 1999, Kent Beck publishes "Extreme Programming Explained" (Wikipedia: Extreme Programming).
• February 2001, The Agile Manifesto is drafted (Wikipedia: Agile Manifesto).

The dates and versions shown here are approximate only but still give us a good comparative overview of the relative release histories.

For reference, I’ve also included some Agile milestones.  Agile adoption has been on the increase for a while but really started to gain momentum in the late 1990’s and early 2000’s in particular – roughly coinciding with the release of the Agile Manifesto.  This isn't as easy to measure as software releases, but clearly we live in a mor agile world.

Let’s compare the browsers release history for a recent 10 year period, specifically since February 2001 (signing of the Agile Manifesto) to February 2010 (108 months).

As we can see, for this period Microsoft is releasing a new version (roughly) every 36 months on average – not particularly agile; neither is Apple – presumably busy with other things.  At the other end of the scale Opera have been the most prolific 8 releases over 108 months (an average just behind FireFox), with Chrome by far the most regular releasers of new versions (on average).

Perception and Rationale

As mentioned or alluded to earlier, lament at Microsoft’s poor release schedule is fairly common – but it’d be wrong to assume Microsoft didn’t know what it was doing.  At the end of the day the people at Microsoft are just like those at Google, Apple and elsewhere: they’re not stupid.  Yes everyone makes mistakes now and then but people also naturally have differences of opinion, different strategies and so on. 

The bind that Microsoft seem to be in, clearly explained by Peter Bright, is balancing the popular cry of users and the demands of the market against the desire to build a stable platform.  As anyone in tech will know: the continuously changing IT landscape is a major challenge to keep up with and in that sense a degree of stability is welcome.

Of course, the majority of consumers and developers don’t have this at the fore-front of their minds; they have their own issues to face – issues which are confronting them right now.  To quote Peter Bright:

“The current stable releases of Safari, Chrome, Firefox, and Opera are all streets ahead of the current stable version of IE. Make no mistake: these other browsers do not provide complete, systematic, exhaustive implementations of these specifications (though Opera's SVG support is not far off). But they are already providing extensive capabilities, not to mention impressive performance, to Web developers. And they're doing so today.”

Peter Bright: http://arstechnica.com/microsoft/news/2010/03/can-microsoft-really-build-a-better-browser.ars/

Which is where Agile become very relevant: get something out early.  It doesn’t have to be perfect – just good enough to be usable and to start true engagement with users.

So, while there might sometimes be a good rationale for not being agile, we can see that isn’t always without cost.