Hacker News Reader

Opera was actually strict pro-standard.. sometimes for their own bad.

Possibly. But they're also the only browser I'm aware of apart from IE that supports "document.all". And (at least as of 2007), they were still using IE's quirks mode box model if you didn't specify a doctype.

As I said, they did a good job of trying to please everybody. You could code for Firefox or IE and have a reasonable expectation that it would work on Opera too.

That's a good thing, by the way.

What do you mean, "possibly"?

Either you agree that Opera implemented standards as strictly as possible (really the only way when you your own rendering engine with such a small market share) while aiming for compatibility at the same time (which is implemented by copying certain quirks--and O isn't the only one that does that--as well as having a large database of popular sites that are fixed with bits of custom JS), or you disagree that Opera aimed and successfully did implement web standards (which is provably false).

Saying "possibly" makes it sound like you don't really know. Did you consider, while typing that remark about the downvotes (bad form, and doesn't add anything to the discussion, except for polarizing it even more), that it might actually be caused by your self-admitted lack of knowledge on the subject?

Sure thing, those people may be "Opera devotees", because those are the people that know what they're talking about maybe?

You claim that top post was "not actually an anti-Opera comment". What about your remark that "They'll continue to be the browser that nobody cares about"? That was you baiting all of those European and Russian Opera users, that you may or may not have noticed are also here on HN. If you had said "the browser nobody in the US actually cares about" then yes, I could not have disagreed. But this general claim is a tad ignorant, the global 1.3% market share is not at all geographically evenly spread, and goes up to 47% in certain countries. That's a long shot from saying nobody cares about it.

(BTW I may be an Opera devotee--a little bit less than I was a few years ago though--but I'm not downvoting you)

>What do you mean, "possibly"?

I'm guessing OP meant possibly for the "sometimes for their own bad" part of the comment.

>Either you agree that Opera implemented standards as strictly as possible (really the only way when you your own rendering engine with such a small market share) while aiming for compatibility at the same time

Noble aim but that doesn't work in the consumer world. I am pretty open-minded, so I am seriously trying to use Opera every 2 or 3 years. And each time I really start to enjoy using it but after one or two days I get horribly disappointed: a significant number of websites I like just doesn't work. Not just loss in function, they are completely unusuable, for instance I get a white page.

Last time I did this try is probably 2 years ago. Last time I observed this behaviour with another non-IE-browser: at least 5 years ago, probably more like 8.

There was a time when certain websites only worked with IE. Somehow Mozilla and friends adapted... but what happened to Opera? I cannot stop thinking that Opera somehow got stuck in the 90s. They are certainly innovative, they were the CSS pioneers, everybody copied tabs and other cool features.

But still... When your rendering is incompatible to the real world, nobody will use you. So I can only agree: smart move. Maybe this will make Opera a serious competitor in the Browser market again.

Anyway, I also agree, Opera still has a user base. I work on an in-house tool and we have users using Opera, so I test against it.

> Noble aim but that doesn't work in the consumer world. I am pretty open-minded, so I am seriously trying to use Opera every 2 or 3 years. And each time I really start to enjoy using it but after one or two days I get horribly disappointed: a significant number of websites I like just doesn't work. Not just loss in function, they are completely unusuable, for instance I get a white page.

You're saying that's the fault of Opera when it's possibly the fault of lazy devs.

You used to (still have?) devs testing for Opera and then preventing users going any further. Users would spoof the agent id string. So then devs started doing more tests, and rejecting Opera.

(https://groups.google.com/group/alt.sysadmin.recovery/tree/b...)

>You're saying that's the fault of Opera when it's possibly the fault of lazy devs.

Obviously it's "hobby devs" breaking the standards and Opera having low tolerance for that. As a matter of fact, a large part of the web's success is due to hobbyists. The hobbyists were only able to contribute to the success because it was and still is so easy. You don't have to be a HTML or JS guru to create a dynamic web page. On the other hand this allows for many devs that never even heard of HTML validators, JS linters etc.

I think it's good that most HTML/JS implementations are very fault tolerant. It's good and the decline of XHTML, XSLT and friends underlines that.

>You used to (still have?) devs testing for Opera and then preventing users going any further. Users would spoof the agent id string. So then devs started doing more tests, and rejecting Opera. >(https://groups.google.com/group/alt.sysadmin.recovery/tree/b...)

In 2003 Mozilla and friends had the same problem sometimes too.

> Opera having low tolerance for that

Except that is not true. Opera has a high tolerance, probably much higher than any other browser (out of necessity). But that doesn't help much if you're being blocked or sites constantly find new bugs in other browsers they decide to work around and thereby breaking it in any browser that is not one of the top 3.

All browsers support document.all; see [1].

It is true that Opera tried to reverse engineer more of the IE DOM than most others, for example we were the only ones to have attachEvent and friends. But in recent years that caused problems with sites that made false equivalences like "has attachEvent -> must be IE" and so non-standard features were being actively removed.

It is also true that Opera has been heavily involved in the standards process; both in codifying the legacy and in more forward-looking initiatives to add novel functionality to the platform. For example Hixie was working at Opera when he started work on what became HTML5. Opera's Simon Pieters wrote a spec for CSS quirks mode [2]. We have also been involved in numerous other specs, and in the beginnings of the effort to produce a comprehensive vendor-neutral testsuite for web platform technologies.

[1] http://software.hixie.ch/utilities/js/live-dom-viewer/saved/... [2] http://quirks.spec.whatwg.org/

Yea, it is unfortunate that Netscape "Mariner", which was also going to support DOM level 1, got cancelled.

> Webkit isn't "Chrome in a box" either.

The Chromium Embedded Framework (http://en.wikipedia.org/wiki/Chromium_Embedded_Framework) isn't too far from that. I'm not sure if Opera is using this project in particular, but they do mention using Chromium.

It seems you are right, other comments mentioned using V8 as well.

> There are a lot of markets where Opera has 10, 20, or 30% share.

As there are so many would you mind listing a few?

Russia. Was 20%+ last year, now 10%+ http://gs.statcounter.com/#browser-RU-monthly-201201-201301 Or Belarus 30%+: http://gs.statcounter.com/#browser-BY-monthly-201201-201301

And Britain at 20% (was surprized myself!) http://gs.statcounter.com/#browser-GB-monthly-201208-201301

Edit: my bad, the graph probably didn't refresh properly

20% in the UK would surprise me too, but that graph shows Opera at 0.56%, and seemingly declining.

StatCounter is a terrible source for anything outside of North America and Western Europe.

As highlighted by my comment, it's middle and eastern Europe :

Ukraine 20% (36%), Slovakia 6% (13%), Poland 7% (12%), Czech Republic 5% (10%), Russia 14% (32%), Belarus 32% (45%), Kazachstan 32% (16%)

Also, Mobile browsers are not insignificant market: (whole) Europe 24% (50%)

Numbers from statcounter.com as of January 2013 (January 2010). I picked 2010 as it was about the time Chrome started it's push around here (I'm in Czech Republic at the moment) - ads in radio, tv, outdoors, on homepages of google.com, youtube.com; Opera was often blocked from accessing gmail, apps,; was deprived of features (by user-agent sniffing) google-wide. And FF was well established at the time, so it didn't carve much to the Opera's share.

Those are all immature web markets. By the time the majority of web users in those countries spend a significant amount of money online, they probably wont be using Opera anymore. If you're looking for more file-sharing users, sure, target these markets. But if you want to make money today then don't waste your time. (I've lived in, started/run businesses, and produced web products in Eastern Europe).

thank god for people like you who spread the word about how you can't make money on the web in those markets :) all I can say that you did not try hard enough :)

The European mobile web market is 'immature'. #shithnsays

Not what I meant at all. I just meant that in my experience, the counties listed have relatively low median disposable incomes, and are generally not inclined to pay for web products.

Have you ever heard about ads?

Why wouldn't they be using Opera anymore? They will be used to Opera, and the brand will be well known to most people.

Why do you think Google is desperate to enter these markets right now?

>I picked 2010 as it was about the time Chrome started it's push around here (I'm in Czech Republic at the moment) - ads in radio, tv, outdoors, on homepages of google.com, youtube.com; Opera was often blocked from accessing gmail, apps,; was deprived of features (by user-agent sniffing) google-wide

I think by far the biggest impact was bundling Chrome and autoinstalling it with Java and Flash updates. Google must have paid top dollar for it which Firefox and Opera had no chance of matching.

> Opera was actually strict pro-standard.. sometimes for their own bad.

Not quite. They always made every effort to be as compatible as possible with real world sites.