Web standards
Jamie is the head of product development at MirrorWeb, focusing
on new archiving technologies. You can follow all of his tweets
PROFILE@mightyshakerjnr.From HTTP/1 to CSS, the early internet was built
on open standards. It is thanks to the wider
technology community’s desire to work together to
build a shared fabric of modern communication that
we are here today. Unfortunately nowadays, with a
technological culture where an increasing amount of
online activity takes place via integrated products and
apps rather than through websites, things simply
aren’t so straightforward.
Truth be told, we have absolutely no idea what the
internet will look like 20 years from now. We don’t
know what browsers we will be using, what features
they will support or even if we will be using browsers
at all. The further we stray from open standards, the
more likely it is that future generations will not be
able to see the content of the world wide web as it
stands today.
One example of this is Google’s AMP project. While
it is open source, it’s not an open standard: Google
is the arbiter of what the AMP speciĽcation looks
like and that speciĽcation is designed solely forGoogle’s beneĽt. It replaces standards-backed HTML
with a subset of tags that Google deems helpful to
itself. While I’m a fan of how Google’s investment in
Chrome has transformed modern browser technology,
its approach to AMP should worry proponents of the
open web.
When the technologies of today’s web are obsolete,
there’s an increasing fear that our children won’t be
able to retrace our steps and Ľnd out more about the
world we lived in. It’s the great irony of the digital
age: in a time where more information is recorded
than ever before, we risk not being able to replay
anything at all.
Archiving companies and institutions need
to sit down and deĽne a standard for archiving
API responses. So much of how we interact with
the web – whether that’s through platforms like
Twitter, WhatsApp or Instagram – is carried out via
API-backed native applications and yet we have no
standardised way to store and replay this content. At
the moment, the Web ARChive (WARC) format is the
ISO standard for storing and preserving web data but
no suitable equivalent exists for APIs. At MirrorWeb,
we already have an internal standard for how we
archive APIs and personally I would welcome a wider
industry discussion on the most eļective way to do
this going forward.
Developers, on the other hand, need to think
heavily about making platforms more accessible and
making sure they better conform to web standards. A
lot of modern websites are not designed for the kind
of tooling that web crawlers and screen readers use.
Building accessible websites isn’t just a nice thing
to do: it’s the law and it’s time we started holding
companies more accountable to that. Making websites
accessible may take time but it should still be an
essential part of your workľow. In the short term,
you’re helping out your customers and that can only
be a good thing. In the long term, you’re making sure
that future generations can learn about our fondness
for cat GIFs.
As a developer working in digital preservation, I
know only too well the challenges that the web has
ahead of it. It is essential that we take action soon to
avoid an internet dark age and it’s only through a
concerted community eļort that we’ll avoid loss of
public information on a scale not seen since medieval
times. If you need any more convincing, just think of
the fallout when GeoCities shut down: can we really
aļord to have a similar thing happen to the entirety
of the web?PREVENTING THE
INTERNET DARK AGE
Jamie Hoyle on how open standards can
stop data loss on a massive scaleSTANDARDS