Exit Opera

Opera — apparently tired of having relevance and influence over the evolution of the web — is quitting the “build a browser” business in order to join the “customize someone else’s browser” business.

This has triggered much discussion of monocultures and the health of the web.  In comparison, I haven’t seen much discussion about how it’s simply an awful decision for Opera itself.  Well, not until I read the comment thread of this Ars Technica story (about lay-offs), some of which nicely covered several aspects I’ve been thinking about.

First: the announcement said that “this is primarily an ‘under the hood’ change”.  I’m skeptical about that.  tigas commented:

I await the first beta of OperaKit. Rumored casualties are Carakan (confirmed), Dragonfly (confirmed), Fit-to-Width/Seamless Zoom and Single-Column/Small Screen mode (all but confirmed, those depend on Presto’s dynamic layout code), Instant “Back” and “Undo” (probably too dependent on Presto to be able to be retrofitted to WebKit), IRC and Torrent clients

This doesn’t surprise me.  Rendering and JS engines are not identical black boxes.  They’re complex things with very large surface areas.  Switching them must be invasive and have a plethora of knock-on effects. I don’t know if Opera has announced a release date for their first customized version of Chromium, but it wouldn’t surprise me if it took longer to get working than many people expect.  Especially if they just laid off a bunch of developers, including numerous old-timers.

Furthermore, I suspect that new version will feel and behave dramatically differently to the old version, and that’s going to annoy a lot of their users.  For example, fluxtatic commented:

Crap…if Dragonfly goes, I’ll shut automatic updates off and stay with version 12.x until it gets too dangerous. I’ll have to hope in the meantime all the little things that make Opera better stick around or come back (tab restore, right-click to open new tab on the tab bar, automatic restoration of last session, being able to save sessions, etc, etc.)

Think of those loyal Opera fans who would defend it to the death in every discussion about browsers on the web.  Do you think they’ll have the same passion for a customized Chromium?  lithiumfrost summed it up best:

Personally, I think Opera can now disappear. Their announcement made for a suicide note in my mind, and I can’t imagine what kind of value-add they can offer now that they threw in the towel on their own layout engine and Javascript compiler. A lot of the value that Opera offered was in it’s small download size and very fast layout engine along with very good standards compliance. They innovated, and they really did so with Presto.

Now that they are becoming just another Chrome-clone, I can’t see the point in continuing to use it. Why? I might as well download Chrome. They’re doing the actual browser work now, and have a lot more experience and expertise in the development.

I’d like to see them open source Presto now, but I my guess is that they probably would have announced that with the change if that’s the way they were leaning. What a waste. Farewell Opera; it’s been a good ride.

What a waste, indeed.

10 Responses to Exit Opera

  1. maybe mozilla can hire some of those developers ?

  2. perhaps the charitable side of the mozilla foundation could buy the old opera browser fork (seperate from Opera the company) and opensource the code and allow the community to back it, kinda like the deal from Microsoft to save apple from bankruptcy many many years ago….. sometimes you have to pay to keep a competitor on their feet to keep some competition in the market! if microsoft hadn’t paid up back then they probably would have been declared a monopoly and broken up.

  3. i daresay mozilla can use the UI fluidity of Presto under heavy CPU use.

  4. I’m afraid for Opera mobile on android, it had a reflow on zoom perfect and fast. No other browser I know was that good. And even on my netbook it was very fast.

  5. Presto was far from influential in real terms. Only industry insiders – in particular it seems, a plethora of Mozillians – who gave it undue credibility as an ‘influential’ product. No market share, no influence. At my workplace we develop to W3 standards then hack until we get an acceptable result on any browseer with 5% market share or more. Presto never got more than 3% so it was never given a second thought.

    I strongly suspect that a lot of those shedding tears for Presto kept dreaming that, after several decades, it would magically start gaining popularity and become another army in the allied fight against IE. If this was the case, how come Firefox, Chrome and even Safari have come along well after Presto was born and slammed IE but Presto remained irrelevant?

    As for Mozilla wasting any resources on a dead rendering engine, that money would be better spent employing as many of the 45 odd ex-Opera developers as possible. Their fresh approach to Gecko could add a great deal of objectivity. That is if Mozilla doesn’t assign them to duplicate another chunk of existing code, as they have done with the dev tools resources duplicating Firebug.

    • Nicholas Nethercote

      Everyone knows Opera has a small market share. But their influence was large when it came to web standards, particularly given that a new HTML feature generally won’t be considered for standardization unless it’s implemented in at least two different engines. Opera used to have one of those engines; no more.

  6. To bad making a rendering engine isn’t a cheap but profitable endeavor like creating your own brand of clothing. In which case no one would hesitate to make their own. Pretty soon we’d have hundreds of lines of rendering engines. No rendering engine would be the same except for the web standards they support.

    OMG! What are we waiting for Mozilla! Lets figure out a way to do this, pronto! Yay polyculture!

  7. Can Mozilla just hire some of those experience folks to work on Servo? Obviously hacking on Gecko is too much work to learn. Start Fresh with Servo and they have a big new challenge.

    If Money is a problem I am sure there will be donation made from the public.

  8. This post from an opera dev indicates that open sourcing Presto is still a possibility.
    https://brendaneich.com/2013/02/why-mozilla-matters/#comment-13430