April 8, 2014. That’s the date that Microsoft has set to end all support of Windows XP. Yes, XP will almost be old enough to be Bar-Mitzvah’d at that point. Oy vey.
That date is important to mobile developers because XP is the boat anchor that is holding back full-throated supported for HTML5-based apps.
Managing support of older browser platforms – especially when you have to balance desktop and mobile browsers – is a pain for everyone and the costs and resource requirements of doing so are significant. Last month Google declared that it is ending support for Firefox 3.5, IE7 and Safari 3 as of August 1.
While it’s hard to find reliable statistics on OS market share, XP sadly still plays a prominent role in many enterprises. And the limitations of XP are what’s holding many developers hostage to this day. A Forrester report from as recent as Q3 2010 indicated that a whopping 75% of PCs are still running Windows XP.
That is in addition to Microsoft’s move to add support for HTML5 in Internet Explorer 9 and 10. But alas, neither of those browsers are available for Windows XP users. Factoring in the limitations of XP and the resulting significant share of non-HTML5 compliant browsers IE 6, 7 and 8, application developers can’t fully commit to HTML5.
Here are some graphics I found on HTML5 browser compliance that I thought were worth sharing. They paint a pretty clear picture (Source: : http://www.findmebyip.com/litmus).
Therefore they stuck with the challenge of having to develop multiple presentation layers, rather than being able to consolidate under the HTML5/CSS3/JS umbrella. This naturally multiplies development, testing and support costs which can rob the ability of development organizations to invest those resources in other activities that move the product/service forward. Not good, right?
Yet this is exactly what Adobe should be thanking Microsoft for. Many of the older desktop and web applications were built using Flash and Flex. So developers are handcuffed to these platforms and it makes investing in new versions and support for mobile devices (remember that iOS doesn’t support Flash) more challenging. And there are 3 more years until XP goes away and therefore 3 more years that Flash hovers over the industry like a spectre.
Now my colleague Avinash Kaza writes that even if Flash goes away it’s not a big deal as Flash drives a relatively small portion of Adobe’s revenues and that tools are where the money is. That is true to an extent. But Flash has had a halo effect on Adobe’s business far greater than the revenues indicate. By making Flash player free, it transformed Adobe into a web publishing platform, that drove the sales of its tools. Even though Adobe probably makes the very best design tools for HTML5, if Flash were to go away Adobe would lose that halo and be positioned as a great tools company.
I was just talking with a client who is trying to figure out how to bring their Flash-based app to mobile, with the iPad as a key target platform. Unfortunately there are no easy paths forward. In the same boat as many other companies, they don’t have a lot of budget and many of their customers are accessing the app via IE 6, 7 & 8. They did ask me for projections of when XP will fade in the enterprise so that we could build a product roadmap to migrate the product towards HTML5. I thought I’d share some of the information I found so far:
- XP still has a significant role in the enterprise. Forrester’s data shows a 75% share for XP, but forecasted that Windows 7 would exchange share with XP. This clearly hasn’t happened to the extent forecasted, but the chart that follows shows a much smaller number, but still significant progress. I believe that the Pingdom number includes consumer PCs too because of the relatively high share for Vista which most enterprises avoided like the plague (with good reason). We might have been in a very different place if Vista wasn’t such a disaster of a product. In my opinion, Vista cast a pall over Windows 7 that stunted enterprise adoption.
- Windows 8 is not due until (early) 2012. The demo and D9 looked great. I have not been able to find any info on projected enterprise adoption. While I love the Metro UI and tile interface, I wonder if the stark differences in UI will delay enterprise adoption because of anticipated higher training costs and impacts on user productivity during the initial transition.
- Here are current browser share that seem to show that gains by IE9 are correlated with Windows 7 adoption. Also to be noted is that Firefox users (based on the Gecko rendering engine) are pretty quick to update to the latest version (Firefox 3.6 has pretty good HTML5 support and Firefox 4.0 is even better) — Source: http://www.w3schools.com/browsers/browsers_stats.asp.
Did you find this information useful? Are you struggling with the same issues?
If you have chosen a particular technical direction, please share with the group and make us all smarter for it!
 Forrester Research. November 2010 “Updated 2010: Windows 7 Commercial Adoption Outlook”