Yes, this is a nice feature added to a basically-reasonable implementation of a PDF viewer. I think the objection is that that PDF viewer should be an actual independent application, not baked into a browser that already is too many things to too many people. It's like Chrome including a basic antivirus function (https://support.google.com/chrome/answer/2765944?co=GENIE.Pl...) - yes it's useful, yes I trust it more than a lot of AV products, but no I don't think it's reasonable to bundle it into the program that's supposed to be here to render web pages for me. (Similar arguments, to varying degrees, are made against WebRTC and Pocket)
I really don't see why it should be an independent application. I mean it's not like we expect a PNG viewer or HTML5 video viewer to be a separate application in a browser. Being able to view (and in this case fill/interact with) PDFs is pretty much a basic necessity on the web. Beyond the core HN crowd, almost nobody cares to have a 3rd party application that they have to install to view PDFs in their browser. Having a lightweight and secure PDF viewer that is also not made by some 3rd party company that could be collecting any amount of data on you is a good thing in general.
> I mean it's not like we expect a PNG viewer or HTML5 video viewer to be a separate application in a browser.
PDFs are generally an actual document, separate from the site they're on. If images and videos weren't a part of the web pages being viewed, I would be quite skeptical of including them in the browser. I mean, there are JS viewers for STL files (https://www.viewstl.com/) - should browsers include a 2D modeling environment?
> Beyond the core HN crowd, almost nobody cares to have a 3rd party application that they have to install to view PDFs in their browser.
See, I have the exact opposite experience; I've had less-technical family complain to me they were annoyed at Firefox because it stopped just opening PDFs in Adobe and forced them into a crippled slow viewer inside itself. Unfortunately, I can't tell which of us is in a bubble.
> Having a lightweight and secure PDF viewer that is also not made by some 3rd party company that could be collecting any amount of data on you is a good thing in general.
That many PDF viewers are awful is an argument for making a better PDF viewer, but not for baking it into a browser.
> PDFs are generally an actual document, separate from the site they're on. If images and videos weren't a part of the web pages being viewed, I would be quite skeptical of including them in the browser.
See, I don't really agree with that because to me, PDFs are a pretty core part of content on the internet that users browse to via their browser. Pretty much every restaurant makes their menu available on their website as a PDF document. Almost all users will interact with PDF documents while browsing the web at some point or the other. Otoh, a tiny fraction will even know what an STL file is, let alone care about opening/viewing one. So that comparison really isn't a fair one.
> That many PDF viewers are awful is an argument for making a better PDF viewer, but not for baking it into a browser.
That's a bit of an odd statement. If anything, it proves exactly why this is a good move from Mozilla. The PDF standard has been around forever, and yet there is a dearth of free, high-quality PDF viewers that aren't bloated or filled with ads or spyware or trying to get you to upgrade to a paid version of their software. So Mozilla has finally taken matters into their own hands and provided a pretty good, light-weight and integrated solution that will do the job for most users. Power users who care can still enable other software via the plugin system as their default PDF viewer. I'm not sure how you can blame Mozilla for addressing a very real deficiency in the state of available software for PDF viewing.
