Web dev at the end of the world, from Hveragerði, Iceland

You can’t solve people problems with software

You can apply various technologies as a part of the solution, but unless the people part of the problem is addressed specifically, at best what you’ve done is punt the problem down the road.

New software or a new standard will not solve a problem that could have been sold already by legacy software and standards. New tech is most effective at solving problems caused by the limitations of older tech. It’s less helpful when the problem is caused by misaligned incentives, indifferent or hostile culture, or plain old animosity.

There are exceptions to this, of course. Without new software, hardware innovations would be useless, but even then, it’s the hardware that’s solving the problem. But pure software innovation? A new standard? It’d have to be a major paradigm-shifting breakthrough to do anything more than solve the investor’s too-much-money problem.

The reason why software pioneers so often fail while fast followers succeed isn’t because followers iterate on the technology. It’s because the pioneer was solving a tech problem while the follower was using the now-solved tech problem to solve a people problem—often attacking a specific cultural or systemic issue.

Facebook’s innovation over MySpace wasn’t in its software (although that was competently done) but cultural and in it’s social signalling.

Slack could have been built back in the 90s. Its innovation was in how it addresses ongoing organisational changes in the modern workplace. It wasn’t that it achieve some sort of technical breakthrough.

An example from web development history: most web developers throughout the history of the web have displayed a complete indifference towards maintaining a semantic structure in markup. They had no interest in marking headings as headings. They couldn’t care less about the semantic meaning of tables. Blockquotes were just styled DIV elements, if you were lucky. This continued to be true even as the technology improved and CSS made tables less necessary. It took two non-technical changes to shift overall developer consensus.

  1. Massive cultural propaganda. The idols of web design relentlessly campaigned on the importance of semantic code.
  2. Google’s search engine gave semantic code substantial financial/economical value.

Culture and incentives. Those were what change the web developer industry’s standard practices. It wasn’t a better technology. If the leading lights of the community had crawled into a W3C hole and tried to solve the problem by making CSS less dumb (and, let’s face it, it is incredibly dumb), we’d still be laying pages out with tables and Google wouldn’t have even tried to leverage semantic markup, citing developer disinterest.

We are surrounded by people problems that web and e-book developers are trying to solve with software and new standards.

One of those problems is money. The incentives of ad-supported web media do not align with the interests of their readers nor do they gel that well with the needs of advertisers. Ad blockers are an attempt to solve this problem with software. It won’t work. Even if ad blockers couldn’t be circumvented easily (unlikely) most of the traffic will bypass them simply by virtue of Facebook’s and Twitter’s hesitance to to use an in-app browser that supports them. Even if that weren’t true, even if they were 100% effective on all websites anywhere on any device, the problem would resurface because we’d still have the same system of needs and incentives that created the current system.

We still have media that needs an easy, scalable revenue stream. We still have advertisers who, for whatever reason, want a venue for brand advertising. We still have ad networks that connect the two. We still have readers who will click on any old inane thing no matter how dumb it sounds. The only thing the hypothetical perfect ad blocker would net us is that we’d replace websites with intrusive ads with apps with intrusive ads. (Just go play a free-to-play iPhone game for examples of how awful and annoying in-app advertising can be.) We haven’t actually solved a thing. To effect true change we need to tackle one of the people variables.

Which means:

  1. Try to figure out new revenue streams for media. The problem there is that most web media is less intellectually engaging than a gossip column that has been cycled several times between Chinese and English in Google Translate. Except it isn’t as much fun to read. The biggest problem preventing new revenue streams for web media is that most web media isn’t worth paying for, even as a micropayment. Change that and the rest will follow.

  2. Try and convince brand advertisers that they don’t actually need media websites or ad networks. Which would probably bankrupt web media and ad networks. Which is fine by me but might bother some.

  3. Bring in legislation that reins in the worst misbehaviour of ad networks. And make sure the penalties are harsh to the point of threatening the viability of the company being punished. Which is not going to happen because our current political system is deathly allergic to actually punishing corporate wrongdoing unless it really, really has to (see Volkswagen, Deepwater Horizon).

If we only apply technology to the problem, we just end up pushing it down under the surface for a while.

All of the above applies to ebooks as well. Copy and paste is missing or broken on many ebook platforms, not because of a limitation in the software but because somebody decided it should be broken. Many of these platforms have poor accessibility, not because HTML suddenly becomes less semantic when wrapped in a zip file but because the platform vendor simply does not care. There is no interoperability because important players don’t want interoperability. None of the major problems holding back ebooks can be solved with new software or new standards. You cannot solve indifference with new standards or better software. You can’t solve a deliberate crippling of user-facing features by holding meetings at the W3C. You can’t make publishers spend more on ebooks—improve their semantic structure—by introducing new tech to solve a problem they never gave a toss about in the first place and could have solved years ago with legacy tech.

If you think that the problem of web media and ebooks can be solved by yet another W3C recommendation or an Apple-endorsed browser extension then all we have to look forward to is another 15 years of intrusive ads and broken ebook platforms.

You can also find me on Mastodon and Bluesky