Have you ever noticed that sometimes Firefox saves the image files as WEBP, even when that image has the extension JPG? Well, that’s stupid, but you can change that.

Using Firefox for the shitty browser of choice, go for instance to this web page. Open one of the images in a different tab, say shiny-brite-pink-chirstmas-tree-9.jpg.

Now try saving it as a file. Chances are, instead of shiny-brite-pink-chirstmas-tree-9.jpg, you’d get shiny-brite-pink-chirstmas-tree-9.webp!

There’s this reddit thread, Why Does Firefox Save Every Image as WebP?

Just as the title says. The only way I can get Firefox to NOT download JPEGs and PNGs as webp is to right-click, view image, and make sure there is nothing after the file extension in the URI. What’s going on here? Why would Firefox change the extension like this?

Searching for this problem I see people are blaming the web developers, but if a web developer sends a JPEG as a WebP, but then the extension says JPG, why would Firefox not just do the right thing and save the file as the extension I actually want it as?

The first answer (sorted by “Best first”) gives you the solution:

Firefox doesn’t have a converter: when you save, you are saving what Firefox received.

This problem is caused by server farms and content distribution networks optimizing for bandwidth. If a browser signals support for WebP, then the server will apply WebP compression on the fly to JPEG and PNG files without changing the file name or extension. This is transparent to you when viewing the page because Firefox can decode WebP compression. However, it does mean that the image cannot be saved in the original format because Firefox doesn’t have it.

Could I suggest trying my extension? It strips off the signal that Firefox likes WebP-format images, which should discourage servers from using the extra compression on JPEG and PNG images.

More info: https://addons.mozilla.org/firefox/addon/dont-accept-webp/

Indeed, while Firefox is “smart” to support WebP, it’s actually stupid to do so.


The reason?

In Windows, if you have Firefox installed as the default browser, any WebP file will open in Firefox!

In Linux, I noticed that WebP files are not supported by:

  • the default image viewer in XFCE, namely Ristretto
  • the default image viewer in MATE, namely Eye of MATE (eom)

The stupid thing is that the MATE team has forked Eye of GNOME (eog) from GNOME 2, but they didn’t fork gThumb, which has the following advantages:

  • gThumb is the lightest GTK-based image viewer to offer decent image editing
  • gThumb is the lightest GTK-based image viewer to support WebP

Oh, well. I can fix the WebP annoyance by installing gThumb (which is one of my must-have apps in Linux!), BUT I STILL DON’T WANT THIS FUCKING WEBP SHIT!

If they want to save bandwidth, instead of saving 15% by converting JPG to WEBP, they should use web pages that have most of the content useful content, not huge JS crap!

Here’s the breakdown of an ArsTechnica page, chosen to have images in the article, and with comments enabled (to have more HTML):

Yessir! 66% of the load is JavaScript! The JPEGs, even if large in pixels, are highly compressed; the JS, whatever they did, isn’t. And 99% of the JS is for advertising and tracking, not for navigation and animation!

Now, let’s consider The Guardian. Take a random page, and analyze it:

80% is JS, and less than 3% is images! The images are only taking 0.15 MB, while the JS is taking 4.2 MB! How come? Because the retards at The Guardian are using JPEGs with Q=45!

Open any image from their website and examine the end of the URL:


They made savings in the wrong department. Either way, fuck WebP!

To get JPG instead of WebP, Firefox must claim it doesn’t support WebP, and the simplest way to accomplish that is to install the extension Don’t “Accept” image/webp, by jscher2000 (who also offers a converter):

Why are they constantly screwing the Internet?