5 ms·

Because I’d get wildly incorrect analytics that way from all the bots with bogus user agents.

• swiley 5y ago
  You don't think nojs and librejs are skewing your client side analytics?
  • jaywalk 5y ago
    No, not at all. You can't even compare the tiny amount of people using plugins like that to the vast amount of info you can lose out on by doing no client-side analytics.
  • KronisLV 5y ago
    I doubt they would in any significant way, since the majority of users have JavaScript enabled in their browsers. [1][2] Furthermore, if you look at the LibreJS project page [3] and then click on the link of the Firefox extension [4], the user count will explain that the vast majority of people on the web won't have it installed and perhaps won't even have heard of it. If anything, extensions like uBlock Origin [5] would be of more concern in regards to the client side analytics, since they usually have more downloads. In summary: unless you're running a site that's tailored particularly for free software enthusiasts or the crowd that are overwhelmingly likely to disable JS, then no, it won't matter. Sources: [1] https://stackoverflow.com/questions/9478737/browser-statistics-on-javascript-disabled [2] https://deliberatedigital.com/blockmetry/javascript-disabled [3] https://www.gnu.org/software/librejs/ [4] https://addons.mozilla.org/en-US/firefox/addon/librejs/ [5] https://addons.mozilla.org/en-US/firefox/addon/ublock-origin
    • throwaway2048 5y ago
      plenty of adblocking filters block even self hosted analytics JS packages.
      • KronisLV 5y ago
        Indeed, that's why i'm saying that they're a bigger problem if you want your analytics to be truthful than LibreJS or people with JS disabled outright, since overall it's more likely that they have an ad blocker of some sort, rather than no support for launching JS at all.
        • swiley 5y ago
          The point is that users will mess with javascript on your page (especially analytics.) And as someone pointed out bellow, bots that run javascript are becoming increasingly common. Either way you're going to have to do some work with the data (and probably include statistics from the server side anyway.)
          • KronisLV 5y ago
            But if 1-5% of your data is affected, is it even worth doing anything about it? Server side logs could be useful for detecting abnormalities indeed, but for getting overall usage statistics, i doubt the occasional bot or user with disabled/blocked JS execution is at all relevant.
            • swiley 5y ago
              How do you know it's 1-5% if you're not even looking?
              • KronisLV 5y ago
                Something along the lines of https://en.wikipedia.org/wiki/Law_of_large_numbers https://en.wikipedia.org/wiki/Law_of_large_numbers Essentially, if 1-5% (example number, the actual one probably differs) of the general population have ad blockers, or will not execute the JS that you want them to, then it stands to reason that the demographics of your site would also be representative of the overall picture.* As for how those numbers would be figured out in the first place - i'm sure that plenty of larger sites and parties with more resources have figured out ways to aggregate this information and have published it (like the statistics in the posts above). So essentially you don't have to look, because others already have. *unless the audience of your site is really niche and therefore differ in this regard from the average. For example, Facebook would be a decent representation, whereas Hacker News would not.
  • eli 5y ago
    Of course. I also increasingly have bots that run JS, pushing GA numbers up. Nothings perfect. But I get a LOT of dumb bogus bot traffic. On some pages it would dwarf real traffic. I don’t even bother to log the requests that hit the cache.