A look back and forward at Opera
my.opera.com
my.opera.com
Worse the areas they didn't pioneer they still fall short. For example I go really excited about some of the parts of their extension API... until I read Chrome's API which is significantly more fleshed out.
I still can't tear myself away from Opera though. I just can't get used to the extension-based replacements for much of the native functionality, be it mouse gestures, the RSS reader, notes (oh gosh. notes.). It would be nice if, as David points out in his article, they focused more on polish than on new features. At the moment Opera feels a lot like "old" Google with a whole bunch of creative, yet half-baked, ideas.
Do most users actually comparison shop browsers? Do they try out Opera, Chrome, and Safari and then decide they'd prefer one over the other because of some eye candy the others lack?
I'm sure some power users do, but I would bet that the average user just uses whatever's already installed at work, or whatever browser sounds like what they've heard other people talk about.
They don't have the time or interest to comparison shop (and probably wouldn't even know what to look for, or even that other browsers exist).
It's definitely true that the main reason why Opera has struggled on the desktop is because of a lack of polish. The UI has always felt clunky, non-native and awkward.
That is a major reason why I switched from Opera to Firefox (née Phoenix) many years ago and haven't looked back since. UI has always been their weak link, while their performance and rendering engine has always been top-notch.
Having tried Opera 11.5 for the first time, it's certainly a huge leap forward. It's not perfect, but it doesn't stick out like a sore thumb like previous versions. The downside is that it's a bit too close to Firefox/Chrome to provide a compelling reason to switch, and it still has too much irrelevent crap front-and-centre coupled with its own UI quirks (e.g. why is "Show menu" in the "File" menu and not in 'View'?). Even so, a vast improvement.
I think it's far too easy to blame their low marketshare on their miniscule marketing budget. Mozilla faced similar problems. Mozilla failed when they had a crap product in SeaMonkey (née Mozilla Suite) and succeded when they had a compelling product in Firefox. If Opera can create a similarly compelling product, I have no doubt that many people would happily switch.
I think Opera's best approach would be to focus on Opera Mobile. Mobile is the browser growth market, I suspect it's easier to convince users to try a mobile browser than a desktop one, and they should be able to translate mobile marketshare into desktop marketshare.
This is an important point, but I'm not sure that it's clear exactly why Firefox was so "compelling".
I remember when Firefox first came out, it was billed as a "faster Mozilla". I gave it a try, because I'd found that the original Mozilla browser was bloated and slow, and I was hoping Firefox might live up to the hype. But, no. Firefox was just as slow and bloated as Mozilla. In fact, from an end-user's perspective, they seemed pretty identical, except for the name. Yet people jumped all over Firefox like it was the Second Coming.
Now, I understand that Firefox does have one great advantage over Opera, which is that it's open source, while Opera is closed source. But the original Mozilla browser was also open source, and it didn't get a fraction of the interest Firefox got. Another advantage was that Firefox had extensions, but (if I recall correctly) the original Mozilla browser had them too.
So, can Firefox's success over the original Mozilla browser be chalked up completely to the name change? Would Opera have been more successful had they simply rebranded it with a sexier name, or claimed that the "new" browser they'd released was a "faster Opera"?
Firefox also shamelessly copied and polished up the best features from Opera (i.e. tabs and integrated search), attracting many early adopter Opera users, plus the fact it was free rather than ad-supported gave the browser a wider audience than Opera ever could.
Changing the name, the theme and the unused Communicator bloat (Mail, Composer, Chatzilla) was more symbolic in distancing Mozilla from the negative associations of the mostly terrible Communicator than anything else.
It's interesting to read the comments from Phoenix 0.2 on Slashdot (http://developers.slashdot.org/story/02/10/07/1739241/Phoeni...). Their comments system is terrible these days, though.
Another issue for Opera is that 'stability' fluctuates significantly with each release.* I'm talking about how some websites will stop functioning properly. I'll try upgrading to the beta. Sometimes the site I wanted to work will work, and then it's another site that stops working. I always have to use Opera as a complementary browser. Opera will find some new way to freeze or crash as a result of a certain webpage trying to load.
*(This can be attributed to poor web design, but it's just part of the reality of being the underdog that they should pay more attention to this.)
I'm becoming more happy with Dragonfly. I do wish that a color picker option existed outside of it though. I don't need or want widgets. Extensions for me are nice. Just a ruler and tweet button give me all that I need though - color picker would be nice.
I do believe that polishing the UI will help overall. I like the more minimalistic approach of 11.5 and the ability to fine tune so much (always an Opera feature). I think I will hang around Opera for awhile since I've added mini to iOS too and see how it can make life easier for me with the links syncing feature. However, there are many areas that would be certainly benefit from some UI and UE attention.
You're not alone thinking Opera has focused too much on creating new things ... internally people have also been thinking and saying this. I remember people complained regularly about Opera crashing when I worked there, although I personally have never had much problems with stability.
The lack of polish on the desktop version could also be because they have been putting a lot more resources on the mobile version ... after all, they bring in a lot more money. Opera is also a lot more successful in the mobile market, as the original article mentions. Opera Mobile is a great product, and Opera Mini is just amazing.
Opera probably still wants to push the browser as a platform, especially on devices (http://www.opera.com/business/devices/), and widgets and Unite are important for this.
FWIW I don't use Op on the desktop, primarily because I've become entrenched with FF since moving to it for web dev tools some time ago, but on phone and tablet I find Opera far better than other offerings.
-- Having the panel toggle set to the left edge of the screen is absolutely perfect. If I were to transplant one feature from Opera to Eclipse, for instance, that would be it.
-- Also, being able to set the tab cycle order (for alt-tab and when closing) is indispensable. Having to cycle in tab-bar order feels downright archaic.
-- The little features, like having notes sync between machines via opera link and integrated bit torrent, are similarly difficult to imagine going without.
There's plenty more...
After watching some non-technical people browse the web, I think all modern browsers are pretty much equivalent (as far as the user is concerned). Normal users don't install extensions, they don't care about easy bookmark access or browser speed (as long as the difference isn't in the order of magnitudes) or really any of the things that browsers might do differently. It's just slightly different window dressing around websites. So I think cleaning up the Opera user interface as the article mentions, while nice, won't make a difference to user numbers.
When products are effectively the same, marketing is the ONLY thing that matters (bundling aside).
Now having said all that, there is of course a big difference amongst web browser when it comes to power users and as a long time Opera user, I find that it enables me to browse the web in ways that are not possible in Firefox and Chrome.
1. Good defaults (in terms of functionality and settings). I can install Opera on a new computer, connect to Opera Link to import bookmarks / searches, change three settings and I have my complete web browsing experience set up on another computer. I don't have to install ANY extensions, widgets or what have you.
2. The reason that I don't need extension is partially because some functionality is already included in the browser, and partially because the address bar in conjunction with bookmark aliases replaces a ton of extensions I might need to have and acts in many ways like a command line to the web. Bookmark aliases let you specify some "command" that you type in in the address bar that invokes a bookmark. So I can type "hn" into the address bar, hit enter and it will take me to hacker news. In addition to that, the address bar consumes javascript. That means that you can set aliases for bookmarklets and for multi argument quick searches. So if I type "bm" (for bookmark) into the address bar, it will open the window to bookmark something on delicious. The same thing that would happen if you click on a delicious bookmarklet button... only you don't have to click on a button. And it will get synced when you set up Opera Link on a new install. And you can hook up any old snippet of javascript you can think of to such a command.
3. Find on page acts as an element selector. If you see a link on a page that you want to click on, you can do "Ctrl + F -> type in part of link name -> enter" to "click" on the link. I know that FF has extensions (vimperator?) to do something similar, but again... that's one less extension you need.
In short, with a very small combination of extra features, Opera will save you a million mouse clicks with zero setup / configuration hassle.
(I once wrote a blog post about multi argument address bar searches: http://www.pushingbits.net/posts/google-date-range-address-b...)
Anyway, the only extension I've ever really wanted for Opera, is Firebug. The lack of a good debugging tool has long been a good drawback, and even though Opera now has Dragonfly, it still doesn't feel as good as Firebug.
Are there some particular features I'm missing out on by only using Dragonfly?
2: Don't really need it. When I type "h" into the address bar, Hacker News is the first guess. Similarly for other sites I frequent.
3: Firefox has done that (with no extensions) for as long as I can remember; it even goes a step further: you can search only links. Chrome also does this.
1. Last time I tried that, it didn't sync extensions, extension settings, or bookmarklet buttons (or other settings). Unless they started doing that now (and judging by https://services.mozilla.com/ it seems like they don't), it'd still be much more of a hassle to set up a new FF install to be equivalent.
2. This only addresses a quarter of the point I was making and even for that... I don't like semantic search because you have to scan the suggested result to check that the first search result is in fact whatever you expected it to be. When I type "hn" and press enter, I don't have to wait for semantic search to do its thing, I don't have to make sure that some other site that starts with "h" or has an "h" in the name took the top result spot (and what if you have multiple pages you visit often that start with "h"?), I KNOW it will take me directly where I want to go. I can do Ctrl + T -> hn -> Enter in a fraction of a fraction of a second, while reading something else on another monitor.
3. Hah, I did actually try it out in FF before writing that part of the comment. Of course, you first have to enable it in preferences (and it was disabled in my preferences for some reason, though judging by the documentation that's not the default). Anyway, thanks for pointing this out. :)
Anyway, it's the combination and integration of little things that makes a big difference, not the little things themselves that might be replicated in some way in the other browsers minus the synergy. I know this is probably not obvious until you have really tried it yourself and the reflex reaction is to squint a lot and go: "My browser can do something very much like that, too!" or "It don't need THAT! It wouldn't make a difference."
Give it a try. It does.
I have tried to setup the same shortcuts in Firefox without success (actually, trying to personalize key bindings in Fx is a mess).