I've used Mozilla's flagship browser since before Phoenix, the first version of Firefox, existed. Been using Chrome on Linux for a few months now, and I've mostly moved my day to day habitual browsing to it. Although there are some "Linux just seems to suck at doing this still" issues, like having audio always just work, they aren't unique to Chrome. Still, I've found a few things that I miss from Firefox, or I wish chrome would do given its additional features.

For general purpose browsing:

  • "/" quick search a-la VIM or Firefox
  • URL autocompletion more like Firefox's "awesome" bar
  • middle-click on page body goes-to URL in X clipboard
  • ability to save textarea resize as a preference on a per domain basis
  • some kind of UI to ad-hoc add sites to "ignore this bad SSL cert" white list
  • when I click a dropdown, I want to be able to type the selection right away, not have to click away and leave the dropdown selected

Developmenty things:

  • view selection source
  • slightly less buggy DOM inspector
  • selection based DOM inspector w/ highlighting, like firebug's
  • don't view XML documents by simply stripping all tags and showing a space separated list of text nodes

The top item from each category are the main barriers to me adopting Chrome for most of the duties that I still use Firefox for. In general though, there are a lot of things I like about Chrome. V8 has been an improvement on most websites, anecdotally. WebKit does a few things far better than the versions of Gecko that I have been using, chief among them is scaling the text size up and down, which is something I tend to do quite often for some reason. I like that resources are hyperlinked when I am viewing source.

I'm not sure the actual "weight" of actions, but everything that is there feels snappier. This could all do with Firefox's record keeping; some of my sqlite files for URL history, etc, exceed 100MB in Firefox, and I find myself often seeing IO wait spike for seemingly no reason when using Firefox, whereas I don't normally experience this in Chrome.

The decision to use processes for each tab masks to some extent the true memory footprint of Chrome (whereas Firefox is always easy to see), but it also means that I tend to reclaim memory when I close tabs, rather than growing indefinitely to whatever the maximum capacity is. I can generally run Chrome alongside Firefox, whereas if I was running two distinct Firefoxes I'd probably get myself into some trouble.

In the long run, subsequent revisions of Firefox might surpass Chrome in being slim, but that's not the trajectory that software usually takes. I'm not sure what Chrome's security record is like on Windows, but I'd recommend checking it out for most users. For developers, you'll probably still need a copy of Firefox nearby, but for Linux users who never liked Konqueror, it's great to regularly experience sites in WebKit and Gecko.

Oct 24 2009