August 6th, 2009

It started with this:

(A)lot of folks are lobbying their companies, clients, and partners to drop support for IE 6.

My issue is how they’re doing it. The more dickish folks are throwing up an insulting roadblock, patronizing IE 6 users. Those folks should be immediately fired, and smacked upside the head.

Maybe you can think of IE6 as a perfectly viable user agent for consuming content, but cost prohibitive for rendering top-tier experience design. Serve your print styles to IE6 if you don’t want to offer design support. Or serve a basic white-on-black stylesheet. Or no stylesheet.

Essentially, Toby’s point is that cute, crude or otherwise offensive messages relating to the fact the user should upgrade their browser reflect poorly on a website, and possibly the brand behind them. He believes that a much better solution is to either serve up special stylesheets that relegate the visitor’s experience to a less than pleasurable one or serve up no styles at all. Either way, he says, let them in and don’t be mean about it.

And I say, “horse poop.”

While in any sane world Toby would be right, I believe that, as horrible as it may be to pester, poke and prod visitor’s to your website about their choice of browser, the sad truth is that it may be the only way to get the monkey of IE6 off our collective backs.

For context, we have to understand that IE6 is rarely a browser of choice. If a user is continuing to use it as their primary browser, they will fall into one of three delicious categories:

  1. Can’t-Do-No-Better: The user is on an older machine that is incapable of supporting a newer browser, or at least, they think this is case.

  2. Don’t-Know-No-Better: As the vast majority of major websites continue to support IE6, the user has no reason to upgrade their web browser, and no real understanding of why they would even need to.

  3. Not-Allowed-To-Do-No-Better: Corporate users whose IT departments have standardized on IE6 for any number of reasons. It could be a simple matter of stability concerns (stay with what you know), a conflict with internal software that was written specifically for IE6, security concerns (which is absolutely ridiculous) or a cost concern, as even free browsers require costly manpower to implement across even a small organization.

Of these three groups, the only one that is beyond our grasp as developers is Group 1. Older machines can’t run newer software. The world is a cruel and dark place and woe be its inhabitants.

However, Group 2 is very clearly someone we can affect. If our site prompts them to upgrade their browser, maybe they will, and slowly the percentage of folks who fall in this camp will dwindle.

Which brings us to Group 3. The group that everyone says we should cower in fear over, or treat with kid gloves. “It’s not their fault!” they say, “Why be mean to them?”

It’s at this point in the lecture where we talk about basic incentives.

Let’s assume, more than likely correctly, that there is no tangible reason this employee couldn’t use another browser. That the decision is arbitrary, based on either a cost/benefit analysis, a lazy IT department, or that basic fundamental resistance any business has to change. For the business to upgrade their web browsers, some outside force will have to either make the cost equation balance, spirit the IT department to life, or push the company in some other way to consider change.

Let’s also be clear about two things: Everything is about tradeoffs.

What I advocate below I don’t advocate out of some dysfunctional hatred of IE6, nor out of some utopian ideal of HTML5 magically becoming viable by Q1 of 2010. I advocate it because, when looking at where my time as a developer is generally spent, the proportion that inevitably falls to IE6 is almost always out of whack with the revenue associated with supporting those users.

Generally, about 10-25% of my hours spent working on markup are spent either testing, debugging or hacking around numerous IE6 deficiencies. For the cost to work out for the client, the revenue they’d receive from IE6 users has to at least cover my cost of development, and hopefully exceed it. Otherwise, even if the cost were covered, the time spent working on IE6 as opposed to adding additional features or perfecting an existing interaction to better compete is simply not worth it.

There are sites, generally large and established ones, who have no option but to support IE6 for the time being. Even though numerous online businesses specifically don’t support IE6, and do quite well despite, for many sites, the loss of potentially one quarter of their page views or paid accounts is a loss they aren’t willing to bear. Generally, these sites are slower to evolve, and more stable in features and interactions. The sunk cost of any development time supporting IE6 is spread out over several years, and the additional users more than make up for the headaches associated with supporting them.

However, most sites do not have this problem. Just as most sites do not need to spend thousands per month on a server farm, the vast majority of sites do not need to worry about the one-quarter of internet users who continue to use IE6. They need to be more worried about attracting, impressing and retaining new users above all else.

To do that, sites must be able to quickly adapt, add new features, repair old ones, respond to feedback and improve at the fastest pace possible. Sites must consider that newer browsers generally offer better support of newer Web standards that allow their developers to create increasingly gorgeous, detailed and impressive experiences at a lower cost.

“Supporting IE6 costs you money” is true for every site. “Supporting IE6 will make you money” is true for only a handful.

Getting to Yes.

Returning to previous generalized groupings, the question we have to ask is, if the very existence of Group 3 hurts both our community and the ability of our clients or our businesses to grow, how do we convert them?

The most basic law of nature, human or otherwise, is that, “objects at rest tend to stay sitting right there using their lame old browser and making my life hell”.

Companies need an impetus to change their policy. One of four things will do this.

  1. The company upgrades browsers as they upgrade their user’s machines. This is inevitable, and yes, simply waiting around is a perfectly valid strategy if you’re either in that handful of sites or don’t care. But you’re talking another 2-4 years of supporting IE6.

  2. The company upgrades their browsers out of necessity. Should any major application the company uses ever drop support for a given browser, you can be assured that the company will immediately plan to upgrade their user’s machines.

  3. The company upgrades their browsers out of user demand. As much as employees feel like their IT department is a cold, unforgiving wasteland of condescending, unsupportive social-rejects who like nothing better than to get in their way, those same social-rejects ultimately are there to support their users. And given enough employee demand, especially from the right employees, the department will begin to upgrade user’s browsers.

  4. The CEO’s favorite web site stops working in IE6. Upgrades will happen the next week.

While item 2 can probably be enacted without any snarky messages, I submit that items 3 and 4 will work infinitely faster with snarky messages. Basic human nature dictates that.

If a user sees a single message, from a single site, that their browser is not new enough to visit the site, they might dismiss that site and move on. If they see a site which is merely “broken”, or appears quite flat and lacking in interaction as Toby advocates, they may dismiss that site and move on.

If the user is constantly bombarded with actual dialogs explaining that their browser is out of date, and it’s interfering with their ability to use their favorite sites, you can be assured the IT department will soon be barraged with requests to upgrade user’s browsers.

And the squeaky wheel gets the grease.

Lubricate and Isolate

What’s more hilarious than anything else about this discussion is that we’ve already gone through this as a community before. Remember the browser wars?

As much as we’d love to repaint history as Microsoft somehow forcing us to upgrade, the truth is that Navigator continued to run on our machines, but developers began to increasingly use IE only technologies and develop only for IE. As time dragged on, the number of users who used IE shrunk, while the speed at which we developed new and better experiences grew, and it happened over the course of maybe two years. There was a collective push towards more modern browsers, a very loud, badged and advocated push, and that push from the development community helped push the user community to come along.

It’s a play we’ve already staged once, and yet numerous people are attempting to rewrite the script. Or pretend it’s a different play entirely.

All of the so-called “business concerns” are concerns I heard raised six years ago when I began to recommend clients ignore Navigator and its ilk. They’re the same arguments everyone heard. And thankfully, back then, a good majority of us saw the benefits of standards and pushed through that concern. We sold the possibility of faster development and newer technologies to our clients, insisted this was the way forward, did so loudly, and sure enough, most people came along with us.

Why this time is so magically different is beyond me.