Why are USB cable standards so tangled?
The harder the industry tries to standardise, the more muddled it gets
In 2015, Benson Leung made a spreadsheet. The Google engineer had recently fried his brand new Pixelbook, one of the first laptops to be powered by the (then) new USB-C standard of cable. Unfortunately, in the time between the announcement of USB-C and the first products to use it, hundreds of Chinese companies had started manufacturing cheap cables - many of which were manufactured without the capacitor required to ensure that they were able to correctly supply the 3 amps of current required to power devices. Result: fried motherboards. Leung began compiling an open spreadsheet of ‘safe’ and ‘unsafe’ USB-C cables - a move that forced Amazon to purge the dodgy ones from its store.
Eventually, USB-IF, the implementers forum for the standard, introduced a marque to show compliant products, but even today, after two further attempts to create an easy-to-follow standard, USB-C is a hot mess, requiring an understanding of the USB 3.x protocol, the wattages required to power different products, and an understanding of other ‘piggyback’ protocols such as Thunderbolt and DisplayPort. The cable that was supposed to do everything has ended up confusing things even more.
It’s an extreme example of a growing problem in the tech sector - namely that so-called standards are rarely standardised. Back in the noughties, most messenger apps were based on a standard called XMPP (or to you and me, Jabber). Now, everyone has fragmented and like the Tower of Babel, we’re all left talking to people in our own silos.
The most recent entry to the canon is Wi-Fi 6, so named because it's the sixth generation of Wi-Fi standard. However, if Wi-Fi 5 supports the 5GHz band, then logically, Wi-Fi 6 supports the 6GHz band, right? Wrong. Only Wi-Fi 6e products use the new band - splitting the standard in two from the off.
HDMI is even more of a minefield. As we move toward 4K as the norm for video content, with 8K waiting in the wings, you need to have an HDMI cable fast enough to support them. The problem is, it’s almost impossible to identify the difference between the latest HDMI 2.2 cable from a ten year old HDMI 1.2 cable - they’re rarely labelled in the same way that Ethernet cables are, for example.
Yet it’s not as hard as these examples make it sound. Look at Bluetooth, currently on version 5.2. Bluetooth SIG has made sure that not only is every new version completely backwards compatible, but often, older Bluetooth chips can take advantage of newer features when connected to a newer generation device.
But to return to where we started, the USB-C debacle has been perhaps the biggest example of how standards have slipped. USB-IF has made two attempts to clarify the different types of USB-C and what they’re capable of, but it remains a complete minefield - a seemingly reliable cable may still not be capable of carrying Thunderbolt, or it may not support your phone brand’s fast charging solution.
Last year, the European Union launched an enquiry into how USB-C got so tangled up, and a working group is currently looking at ways to make it clearer to consumers what they’re buying. The fact that it has taken a government regulator to take this in hand speaks volumes.
The moral of the story is that, unlike software, hardware can’t work on a DevOps basis, debugging problems on the fly as they appear. New standards need to be finalised from the outset, with the past and the future in mind. The alternative is ever decreasing circles into a world where nothing is truly compatible with anything else - and that would be a sad state of affairs.
2022 State of the multi-cloud report
What are the biggest multi-cloud motivations for decision-makers, and what are the leading challengesFree Download
The Total Economic Impact™ of IBM robotic process automation
Cost savings and business benefits enabled by robotic process automationFree Download
Multi-cloud data integration for data leaders
A holistic data-fabric approach to multi-cloud integrationFree Download
MLOps and trustworthy AI for data leaders
A data fabric approach to MLOps and trustworthy AIFree Download