Coping with The Bullshit Web

I recently came across this post from Nick Heer castigating “the bullshit web.” A term he uses to describe the fairly despicable state of modern web applications. While I overwhelmingly agree with the points he lays out, especially in his disparaging remarks towards AMP, I think there’s more to be said about alternative approaches for web users to once again experience the web without the bullshit.

My personal computing setup is a bit towards the extreme, so I understand that most people will not utilize the approaches I favor, but I will list them here nonetheless.

  • Navigating to mobile sites by default: one easy trick which I’ve utilized on a number of international data connections is to default to using a web site’s mobile version, even though I’m on laptop (heresy!). A number of sites, excluding news sites, become much more pleasant for mobile users due to the constraints on CPU and network most mobile devices have.
  • NoScript: I have used NoScript, and it’s Chrome analogue ScriptSafe, for quite some time to block enormous amounts of JavaScript being loaded by various pages. This is a very big hammer to wield against the bullshit web, which makes it extremely effective but it requires a bit of understanding about what domains contain adware or surveillance scripts versus those which are simply part of the web application, despite being served from another domain. Special dishonorable mention here to all the CloudFront users whose CDN domains are completely opaque and impossible to determine whether they’re serving legitimate application code or adware.
  • Tor Browser: one aspect of the bullshit web which I find particularly obnoxious is the tracking, especially done on an IP basis. I’ve seen numerous people complain in forums, and even within my own family, that “something creepy” was going on when they saw an ad on an unrelated website for something they referenced in a Facebook post, or had Googled. While NoScript by itself does an admirable job preventing surveillance by tracker/beacon scripts, many larger web companies will track you based on the browsr’s fingerprint or your IP address. The Tor Browser utilizes the Tor network to provide an anonymization layer upon which I rely heavily. A non-trivial amount of my every day web traffic is routed over the Tor network, further helping me opt out of the bullshit web.
  • w3m: as I mentioned in my blog post last year In Defense of Being a Console Luddite, you would be surprised how much faster everything gets when you cut the web down to it’s most basic, content-centric, components. While only a small percentage of my traffic ends up going through the w3m text-based browser, I find it quite useful for quickly getting answers without ever leaving my terminal (where I’m typically working anyways).

Most of these approaches are good for the web-as-a-document-centric-universe, but perform much more poorly with the web-as-an-application-platform. For many web applications there is simply no practical way around a heavy browser-based experience (e.g. Confluence, Jira, Slack). For these applications I don’t have much advice except to “vote with your wallet”, or as is the case for many people, advocate internally for better tools which are lighter weight or support alternative nad more native interfaces.

Regardless of how you accomplish it, I hope you’ll echo Nick’s closing rally cry:

Death to the bullshit web.