Infrequently Noted

Alex Russell on browsers, standards, and the process of progress.

Comments for When Free Isn't Cheap Enough

I'm not sure that this is really describable as a negative externality in the classic sense. No one is being actively harmed and being forced to take on these support costs. If you'd have argued that those running older browsers are more likely to have malware on their machine, and then causing direct harm to others, you'd possibly have a case.

Additionally, there isn't moral hazard here unless you want to make the argument that IE6 users are more likely to have a security problem, but aren't being forced to cover the full costs of that. In the existing case if users are running IE6, a developer is perfectly at rights to charge them more to use their application if they'd like to, or not service them.

Hey Andy,

Web developers and users of better browsers take on the costs that organizations who run older browsers don't bear (but should). Developers bear direct economic losses in the form of higher costs for the production of "lumpy" goods (based on the lowest-common-denominator effect noted) and pass those costs on to all of their clients, not just the ones using bad browsers. Obviously, I support differential pricing based on browser quality, but for many web apps that's not really workable...or at least it hasn't been historically. The costs are suffused throughout the development process: libraries are bigger and slower than they otherwise would be since they assume lowest-common-denominator. CSS techniques are stilted because of assumptions about browser support. Performance and features are capped at every turn, driving costs up for everyone.


by alex at
Have you guys field tested CF? I'm not sure what kind of budget google gives you, but I'd be interested in knowing what happens when someone tries to use CF at a big company.
by Justin Meyer at
And web developers also have costs to support a multitude of browsers. Desktop software vendors incur costs to sell software for both Windows and the Mac, much less Linux. I don't think you could describe the Mac user as causing a negative externality to Windows software developers because they aren't running Windows. Does this reduce the market for that developer, yes.

How are older browsers any different?

Andy: browsers are assumed to be substitute goods. OSes aren't (usually). Given that web sites today are "lumpy" in that they are often built to assume multiple client browsers and not just a single client, the analogy just doesn't hold. The dynamics and expectations of market participants are totally different.
by alex at
Developing for down-level browsers isn't as hard as it is made out to be in my honest opinion. Monolithic, overly complex pages are usually to blame and a little carefully crafted server-side code can make delivering appropriate HTML, CSS and Javascript for the browser in use quite simple.