So we want to speak about privacy/security?
I'm still waiting to see an user hurt by an Opera's security flaw
Attacks on the browser are becoming rare. Opera had security through obscurity before that (too small of a target to be worth targeting), and generally better coding than previous versions of IE.
The big attack vector nowadays is plugin/OS exploits that can be executed in browser. Like the attack here the other day. A Windows Help center vulnerability (malformed link) and a Java exploit. Opera wouldn't protect against either of those - the applet ran in Opera (I was on the latest Java version, so the exploit wasn't effective - however, few users keep java up to date & old versions are kept when new versions are installed. The Help Center exploit didn't work because I wasn't on XP, much less an unpatched Windows install - my antivirus flipped out though). By exploiting Flash, you can hit 99% of targets instead of some percentage under 50% (common plugins vs. targeting a browser). Browser manufacturers have gotten better with automatic updating, which only serves to make plugins a more attractive target.
NoScript + BetterPrivacy can save you from XSS attacks that might compromise your privacy or have unwanted effects (redirect to an attack page, or a shock site), and they can save you from malware that doesn't just exploit the browser.
Chrome prompted to run the Java plugin on GH, which is odd, because a lot of sites run Java unprompted for me. I'm not sure why. Chrome doesn't sandbox plugins (with the exception of Flash, which has recently been partially sandboxed), so any exploit that works on Java for NPAPI browsers (anything but IE, essentially) would have worked had you hit run in Chrome.
Anyhow, forgetting security, privacy is an issue. I can choose to allow Facebook.net requests from Facebook.com. If iMav were to embed a Facebook widget on every Geekhack page, I could deny those requests. Why might I do so? It gives Facebook tremendous insight as to how I spend my time on the web when I'm not on Facebook - where I visit, how long I spend there, et cetera. It's not just cookies that are the issue (there's
browser fingerprinting, Flash LSOs, HTML5 localStorage, etc.). It goes beyond simple Javascript whitelisting/blacklisting (NoScript has heurustics, HTTPS Enforcer engine, and a slew of other features; RequestPolicy controls Cross Site Requests). You might be giving 5 or more parties access to a relatively large amount of information; based on your activities, and partners who share data with Qualtrics, they can know how old you are, your education level, your gender, your location, etc. and build a profile around you. Qualtrics is free for publishers (premium features are extra), so a TON of sites use it. Same with Google Analytics.
You can use Opera, and be fine with it. I spoke with the devs on Reddit, gave them some praise and some flak. I have issues with some design issues and priorities taken, and I strongly believe that Firefox was lagging, but is now close enough to Opera that it doesn't matter beyond synthetic benchmarks (I will acknowledge that I don't have any really old machines in use right now, so I can't speak for memory usage on lower power machines). I didn't really care when FF4 finally took the Sunspider crown because, even by Mozilla's admission, it's a synthetic, unrealistic test that all browsers score closely in.
My point is:
There are plenty of reasons to use Opera. They did implement many features first (tabs were not among them, and Opera copied from the 3.7 mockups). Regardless of who is "stealing" from whom, it's absurd to complain about it. Software developers respond to their users, and if a feature is useful, the others will want to implement it. I could go on and on when/if Opera implements X-Do-Not-Track, and how they copied from Mozilla, but I don't care - it's a step forward for the web and Opera users alike. There are usage scenarios under which each browser is useful. I find the "kitchen sink" approach a tough sell, and appreciate Gecko's extensibility and vast variety of extensions. For that reason, Opera is installed on my PC, but I rarely use it.
Just my two cents.
muchadoaboutnothing raises one major point: Opera decides it adheres to standards and does it in a very hardcore way. Its solution to this is the User JavaScript thingie, where a script looks for nonstandard pages and tries to make them look as other browsers do. I find this strict adherence admirable actually, it's just a bit silly because Opera is not so big a player that it would have the impact they are hoping to get, that being "You fix your HTML".
Every browser does this. Firefox 4 has a standards mode too, and it triggers in pretty much the same ways (Valid doctype + character encoding = Strict standards mode).
Opera's issue, from what I've seen, is its performance in quirks mode (when it has to fall back).
Compare browsers on an old Pentium 4. You'll find Opera has them ALL beat in speed and responsiveness. This advantage diminishes the more modern your hardware gets, but that doesn't change the fact that it is the most economical with system resources. 11.10 has made tab creation just a tick slower thanks to a supposedly heavier (but fancier) speed dial, but it still has all other browsers beat here as well. I found Chromium to be notoriously slow when doing UI stuff like making tabs.
Again, I don't use any really old machines. I remember Opera ran well on my P4 machine when FF 3.5 suffered (caps blew, it's gone now).
I haven't found a browser that was entirely bugless. You're not just looking at the software, you're also looking at every webpage on the internet. Concerning security holes, are you kidding me?
True, but the big target is the plugins now.
ALL of the browsers get those fixed quickly. IE doesn't count. ActiveX is a single fat security flaw, so I'm not taking it seriously as a modern browser.
NPAPI (used by Gecko, Presto, WebKit, etc.) isn't inherently more secure than ActiveX; things essentially run with the same permission level. If
pepper takes off it could mean a lot, but in the meantime most exploits that work on ActiveX versions of Flash/Java/Shockwave/Adobe Reader/etc. work on the NPAPI versions too.
