11
mksana
5y

Please delete your browser cache.

Wtf is up with this shit?
Maybe I'm just having a streak of bad luck, but in recent days, I ran into this particular issue time and time again.
First with one of our own products - the user appearently not always was shown the newest version due to stuff being cached in the browser.
Fair enough, we had our web-dev find a solution to that, which he did. Until this is rolled out, the only resolution is to clear the browser cache.

I also ran into this same issue on multiple other fronts. For example, there's a remote connection to one of our clients I had to establish via browser. The backend was a bit unresponsive, and somehow I ended up in a situation where my login was rejected. The only solution? Clear your browser cache.

Then we have confluence and jira in the company. Same issue. All of a sudden, I could no longer log in. Worked fine in another browser.
Delete your browser cache.

Is it just that most frontend developers out there are incompetent at what they do or is this stuff broken by design? I don't recall having to clear my browser cache very frequently - in fact, I'm pretty sure I haven't done it for years on one of my PCs at home. What changed?

Ah well, maybe it was just a streak of bad luck. But still ...
/Rant

Comments
  • 3
    With more and more client side code and libraries there is a higher demand for local caching for performance and a lot of libraries to help.

    This exposes a new set of developers to the problem of to much caching.

    I have seen the same and part of it will probably lessen once libraries and developers identify which parts should not be cached :)
  • 3
    What about adding a hash to a resource? 🤔

    Or perhaps a serviceworker is running
  • 2
    I believe it's the Pandora's box we call "Cache all the things", I blame pagespeed testing a bit for this

    Unfortunately by doing this somethings end up unreliable and a lot of webdevs don't care as they'll have their cache disabled when doing dev/testing which leads to "but it worked on my machine".

    Hashing resources or using cache control does work well if used correctly and is added to resources that can not or should not be cached, sadly his takes time to identify elements or scripts that require to be uncached.
  • 2
    I just version the filenames of my static resources like images, scripts and stylesheets.
  • 1
    @Fast-Nop Automatically right?
  • 0
    @alexbrooklyn Half. I don't want to complicate the build process any more than absolutely necessary, and the project is small enough to get away with that.

    What's automated is e.g. the styles included in the document headers, that's one line in the template. Or scripts that I need in several places. After re-compiling stuff, the actual pages will be updated accordingly.

    Images aren't because they've never changed, and the version is just there to have a consistent naming scheme in case they would change.
  • 1
    @fast-nop: That's exactly the approach our web developer used.

    However, can't do much with the UIs I have to use developed by others.

    Appearently there are a number of approaches to tackling this, yet even professional companies sometimes seem to fail at properly implementing them.

    Reminds me of a discussion I had with the dev of a software that's around more or less in the same shape for 20 years. He was so positive about the web ...but not for the reasons you'd expect. His point was that users have gotten so much more accepting of having to wait or facing errors due to the web forcing them to see these things as the norm rather than exception or even problem. :D
Add Comment