Web Performance Guidelines: New Zealand

August 4, 2024

This blog is a revised version of an earlier post which used a different set of terminology. What I called standards are now considered to be guidelines in this blog and that meaning is adjusted accordingly.

This post seeks to examine some underlying justification and practical application of public web performance guidelines. Currently no such publicly adopted guideline exists in New Zealand public policy.

What do I mean by web performance? Broadly I mean objective and subjective measurements of a websites performance. These are detailed very well in the MDN Resource.

Speed supports other rights, but not a right itself?

There have been movements to recognize a right to internet access but there is not much precedent for a right to a particularly good internet experience. However there is a broad and growing field of evidence that web performance matters to how users interact with services at the margins.

Accessibility: A rights based framework, based on capability, to ensure access to services. Foundational to user experience.

There are notable advances in the area of web accessibility in legal and social policy which has some relevance to web performance. New Zealand has the Web Accessibility Standard 1.1 which all public service and non-public service agencies must meet.

These standards are intended to support people:

with low vision
with reading, learning or intellectual disabilities
who use mobile and touch-based devices, voice assistant and speech recognition software.

While accessibility is a right, web performance is maybe a nicer to have idea than considered as a right. However, guidelines could still be very beneficial.

Web performance includes a variety of objective concerns. These include page weight i.e the total size of the page being downloaded, which web performance experts will tell you, matters at the margins.

Some write web performance is an unalloyed good and no one has ever complained that a website is too fast.

Some arguments and links can be shared in favor of viewing web peformance as a public interest issue:

Public interest

Poor performance can, and does, lead to exclusion. This point is extremely well documented by now, but warrants repeating. Sites that use an excess of resources, whether on the network or on the device, don’t just cause slow experiences, but can leave entire groups of people out. - Tim Kadlec

Efficient Government

If you’re paying for the bandwidth every time a hefty file is downloaded, your monthly bill could get pretty big.

So apart from the indirect business benefits of happy users converting to happy customers, there can be a real nuts’n’bolts bottom-line saving to be made by having a snappy website. - Jeremy Keith

Making Guidelines

Considering web accessibility standards to inform the content of a New Zealand specific set of guidelines

The New Zealand web accessibility standard contains a flexible definition of a website that could be copied if necessary for a starting point. I think it demonstrates that these definitions are not as hard as we imagine them to be sometimes.

It also contains excellent language for describing “High-stakes information or services” that determine whether things get a preferential focus. I think any web performance guidelines would have something like this definition:

Online information or services whose inaccessibility at the time of publication could reasonably be expected to have a negative impact on an individual’s emergency preparedness and response, health and safety, or critical civil and political rights, entitlements, services, or obligations.

An example of a limited exception from normal rules that makes sense is for “3.2 Complex visual maps”:

3.2.1 Complex visual maps that associate information with one or more points or shapes that cannot reasonably be represented by common identifiers such as postal addresses or the names of specific places or regions are exempt from the requirement… An example of a complex map is one that shows the distribution of different underground mineral deposits in New Zealand.

As I read this I just imagine, what if there were public interest focused web performance guidelines with a similar kind of developed policy feel?

Core web vitals and other common metrics, or not? Imagining the basis of guidelines

There are many potential futures. In one future, there is a set of web vitals, or core metrics, which agencies must meet or work to meet. It is hard to say what it should look like but there is significant room to imagine.

Examples:

Exceptions: “Outage events, or unplanned service degradation”

It would be possible to write into the policy that no agency should be considered to be breaching the guidelines it if there was evidence that the performance problem was caused by an outage or something out of the agencies control

Preferences: “Largest Contentful Paint on High Stakes Information web pages”

It would be possible to write into the policy a preference for a few selective metrics under certain conditions, such as delivering high stakes information.

Compliance: Periods of time to deployment

It would be possible to write into the policy a compliance schedule, where agencies are to make changes in result to non compliant performance on some kind of appropriate timeline.

Who could it apply to?

Public service agencies with a mandate to improve performance and organizations wanting to do business with government

Gentle enforcement guidelines

There would be enormous easy wins in this area, as many public agency websites are already very effective and lean and basic with their usage of resources already. Generous or significant amounts of lead up time could be given to comply with the guidelines.

Conclusion

Web performance is in the public interest. We can learn from prior documents in the area of web accessibility and general arguments for the public good of web performance to consider making guidelines.