Chad Perrin: SOB

6 June 2009

Opera Sucks

Filed under: Geek,Mezilla,Review,Security — apotheon @ 04:21

I don’t like Opera as a browser. I pointed this out, very briefly, in Web browsers suck. I was recently asked in another venue why I don’t like Opera, though, and felt like it would be a waste if I let the list of critiques I provided vanish into the anonymous pile of discussion there. I’ll share them here, for the sake of posterity:

  • Default keyboard shortcut support isn’t as complete or “friendly” as that of Chrome, Firefox, or Internet Explorer.

  • Tab placement isn’t contextually linked to displayed content very well.

  • Toolbar layout isn’t very customizable.

  • For the same display elements with default skins, Opera toolbars take up more space.

  • The text-search interface uses the same ugly-ass, inconvenient format as in most MS Windows applications.

  • It seems to focus more on ACID compliance than on implementation of (X)HTML and CSS standards that people actually use.

  • It’s closed source software, and thus less trustworthy.

I may come up with more complaints about Opera later. If so, I’ll add them to the list.

There are some positives to Opera, of course. I just don’t find them to be good enough to overcome the negatives. The examples that occur to me are:

  • It seems to be fairly quick — maybe a little quicker than Firefox, about on par with Google Chrome. I imagine I’ll see some speed improvement from Chrome in the near future, though, especially once extension support gets added (it’s already in a development version) so I can install extensions that block unwanted content.

  • It has reasonably configurable security-related preferences, maybe on par with those of Firefox. That would make it more configurable in this respect than Google Chrome, which appears to be the slow kid on the block when it comes to individual security and privacy preferences at the moment. Chrome’s Incognito mode is great, but I’d like the configurability to be able to hit a sweet spot between the normal mode and Incognito mode for most of my browsing.

I don’t know if I’ll come up with any more positives worth mentioning.

7 Comments

  1. Postives…

    Sessions: I tend to have 30-40 separate browser windows which I keep around for days or weeks at a time. Sometimes, it’s necessary to reboot Windows (patches, etc). I save the session, reboot, restart Opera and all my windows come back. Also, when Opera crashes (obviously a bad thing, but so does Firefox), it’s very good about restoring the previous session from where I left off.

    Bittorrent: the client just works and I don’t have to install anything, or put up with that irritating unstable and “patch chatty” Azureus.

    Negatives…

    Like I said, I tend to leave my browser running for days or weeks at a time. Every so often Opera gets into this mode where it will spike the CPU load every 30 seconds or so for several seconds at a time, making the browser go catatonic for the duration. Only fix seems to be to quit the browser and restart. I’ve seen this behavior through all versions since I started using Opera last fall.

    In general…

    I split my time between Firefox and Opera. FF seems to work better with some sites like gmail or facebook.

    Comment by LJ — 7 June 2009 @ 11:28

  2. Sessions: It’s much the same on Firefox. My current tab session comes back every time.

    BitTorrent: I actually do anything involving BitTorrent on a separate computer, using a command line client I access via SSH, so I don’t have to worry about interrupting downloads when I put my laptop to sleep or when the browser crashes or anything like that. I don’t really use BitTorrent enough to much care, anyway.

    I split my time between Firefox and Opera. FF seems to work better with some sites like gmail or facebook.

    The fact that some sites don’t work as well in Opera (see my complaint above about standards support and compliance) is part of the reason I just use Firefox and don’t bother with Opera.

    Comment by apotheon — 7 June 2009 @ 11:12

  3. Standards support is my #1. Doing anything with styling dynamically with JavaScript requires either (a) standing on my head to get it to work tolerably well with Opera, or (b) “screw Opera”.

    Comment by Chip Camden — 8 June 2009 @ 11:19

  4. It’s not just javascript with Opera, either. Anything CSS 2.1 and HTML4.1+ is a major pain to get working in Opera.

    Comment by Joseph A. Nagy Jr — 9 June 2009 @ 06:40

  5. I’d like to see you use nspluginwrapper and HTMLUnit for a while, and then hear your comments.

    You might also experiment with xvfb and various screenshot takers.

    Image browsers are faster than web browsers. A folder full of images (e.g. of web pages) can be scanned very fast.

    Ultimately web “browsing” is mostly looking for a location (e.g. a resource identifier) in order to receive or to send data (e.g., by issuing HTTP commands using the identifier).

    The “looking” part can be accomplished at high speed (as fast as the human eye can process input: probably nanoseconds). The “interactive” part, e.g. issuing the commands, is the slow step. They need not necessarily be coupled.

    If we were interested in maximizing speed and efficiency, we could gather locations first, and issue commands second.

    If one wants to browse (“crawl”) the web, by reading a page at a time, and clicking from one hyperlink to another, there are automated ways to do that, and these save enormous amounts of time. When one uses a search engine, he’s leveraging the power of (someone else’s) automated crawling. How many people do you know who browse the web without using a search engine?

    Or, at least, a compiled list of hyperlinks (e.g. like Yahoo!)?

    With the size of today’s web, automation is not optional. One does not approach the web without utilizing this (even if only by proxy, through a search engine or a directory).

    Comment by argv — 17 July 2009 @ 06:04

  6. I’d like to see you use nspluginwrapper and HTMLUnit for a while, and then hear your comments.

    Why? I actually try to avoid Java.

    You might also experiment with xvfb and various screenshot takers.

    Why? I’m not sure what you’re getting at.

    Image browsers are faster than web browsers. A folder full of images (e.g. of web pages) can be scanned very fast.

    Please explain how this is relevant to any of the above.

    With the size of today’s web, automation is not optional. One does not approach the web without utilizing this (even if only by proxy, through a search engine or a directory).

    There are times when things need to be done one step at a time, with human interaction, though.

    Comment by apotheon — 18 July 2009 @ 04:13

  7. I’d like to see you use nspluginwrapper and HTMLUnit for a while, and then hear your comments.

    What? What does java have to do with CSS2.1+ and HTML4.01+?

    You might also experiment with xvfb and various screenshot takers.

    I use GIMP or the print screen key.

    Image browsers are faster than web browsers. A folder full of images (e.g. of web pages) can be scanned very fast.

    Image browsers? Never heard of such a beast. Images take longer to load then plain text. The bigger the image the longer the lag.

    How many people do you know who browse the web without using a search engine?

    I do on occasion. I wind up in fun places that way.

    Or, at least, a compiled list of hyperlinks (e.g. like Yahoo!)?

    I don’t use Yahoo.

    Comment by Joseph A Nagy Jr — 18 July 2009 @ 05:01

RSS feed for comments on this post.

Sorry, the comment form is closed at this time.

All original content Copyright Chad Perrin: Distributed under the terms of the Open Works License