HPR4516: Browser User Agent

Hacker Public Radio

This show has been flagged as Clean by the host.

A while ago I visited a web site that is an archive for old historical documents. It is a data base from where documents can be downloaded as pdf-files. As a visitor I can login to the archive as a guest.

When I find a document of interest after search I can right click on the pdf icon and download the document.

But I can not. No download when I click on the pdf icon. This is the first time I visit this archive so I do not know exactly how it works.

It was time to investigate the root cause.

I use Firefox on a Linux machine. I tested several methods to see if any would solve the issue. I tried by changing settings for pop-ups. I changed Firefox security settings. I disabled VPN. I disabled Firefox extensions for blocking trackers.

I tested also the Chromium browser as well as the Epiphany browser.

All those methods resulted in no difference; nothing happened.

I was in contact to a friend on Windows and also staff at the Archive, both telling me that from Windows it worked, including with Firefox. So now I knew the archive works and that it works on Firefox.

So I thought, can this somehow be related to Linux?

I asked in a Linux forum if someone on Linux could test to download.

One person in the forum tested and solved the problem. I would not define it as a Linux problem, but a site design that affected Linux users.

The trouble was something I had not thought of: The User Agent.

When my browser contact a web site, my browser can tell the site what kind of browser I use, which operating system I use and more. This information can be used by the site to optimize the presentation of the content for me.

For to me unknown reason, maybe a mistake, maybe related to some old design compatibility issue, this site did not accept the information my user agent provided.

The solution is to tell the web site I am something else.

In the browser has Developer tools. When opening the developer tools from meny, typically a developer window with tools and analytics is opened at the bottom part of the web page. One tool is network. Within network I can simulate how the site looks on different devices, with different network connections and also with different User Agents.

The tools are similar in both Firefox and Chromium. The exact list of standard options differ and Chromium has more options preloaded.

In this specific case, when I in Developer tools on Firefox change User Agent to Chrome Desktop, which relates to Chrome on Windows as I understand, and then reload the web page I can now download the pdf file from the archive.

When this web site interpret me to be this other type of user, they can understand each other without any issue.

I have reported back to the site owner and hopefully this specific site will not need this work around for the browser User Agent in the future.

Provide feedback on this episode.