Modern Browsers Ship

http://bethesignal.org/blog/2011/02/19/modern-browsers-ship/

by jdub • 15 years ago

127 99 15 years ago

btipling• 15 years ago

Is the height supposed to mean anything? Why did they start with Safari? It's the first modern web browser? It points to a page that says IE9 is not a modern browser yet shows IE7 and IE8.

jdub9• 15 years ago

No, the icons just indicate releases. They're at different vertical levels so they don't obscure each other. :-)

I decided to start in 2006 to catch Firefox 2 and Internet Explorer 7, otherwise you don't fully see the gaping hole which represents the Dark Ages of the modern web. (Safari was added later, and 2.0 happened to slot in at the start of 2006.)

phoboslab• 15 years ago

I think it would be better to have every browser on it's own line. Especially with IE and Safari as their icons look quite the same from a distance.

Edit: My attempt at it. Not as pretty, but a bit more readable:

http://www.phoboslab.org/crap/modern-browsers-ship.png

simonw2• 15 years ago

It would be pretty cool if you extended it back to 2001, when IE6 was released - then you could also show the rise and fall of Mozilla and the birth of Firefox.

Even better, roll it back to 1991!

Edit: and another pony request - how about putting the release date in the tooltips (yyyy-mm-dd) to clarify any confusion about when exactly it was?

jinushaun• 15 years ago

IE7/8 are there to show scale. The premise is that modern browsers release frequently. You can see that in the time it took IE to release 2 versions, the other browsers released at least double that same amount. Thus IE is not a modern browser.

jdub9• 15 years ago

The contrast I most hoped to illustrate was Chrome vs. everything else. :-)

Hovertruck• 15 years ago

I don't think the height means anything. As far as starting with Safari, they probably just picked 2006 as the start point and it was the first one with a release that year.

The reason it shows IE7 and IE8 is to illustrate that all of the modern browsers are shipping pretty regularly, while IE only has two releases on there.

chaosfox• 15 years ago

where is opera ?

jarin• 15 years ago

Hanging out on 2.2% of computers

rimantas• 15 years ago

Opera's popularity is highly geographically dependent, unlike any other browser, I think. It is very popular in Russia and Ukraine for example (almost on par with Firefox). See http://my.opera.com/dstorey/blog/2009/03/16/a-look-at-deskto... and http://www.opera.com/press/releases/2010/04/14/

jdub9• 15 years ago

Adding it now. :-)

thomasz• 15 years ago

Last time I checked it was crawling around 2.5 % market share.

davej3• 15 years ago

This chart doesn't show the 5 year sabbatical that Microsoft decided to take between August 2001 and October 2006 (The amount of time between releases of IE6 and IE7).

I reckon they shipped IE7 at exactly the most damaging time for the web because If they waited another year or so then it might have been long enough to kill off IE altogether.

rapind3• 15 years ago

That sabbatical is precisely why I will never ever promote Internet Explorer. It was ridiculous. I don't even care if suddenly MS produces the best technical browser available (highly doubtful, but even so). I don't care if it's a new team because it's the same company, and yes much of the same leadership. They blew it and they don't get another chance.

They had the best browser back when IE4 launched. Given the amount of pain IE has caused over the years, it surprises me that any self respecting web developer could ever forgive them.

"IE you are dead to me."

masklinn4• 15 years ago

> That sabbatical is precisely why I will never ever promote Internet Explorer. It was ridiculous.

Some (including standardistas and other web luminaries) argue it was a good thing overall (though it got long in the tooth around 2005-2006) because it allowed the web to settle and developers to better understand what they were working with (because the target got static, it became possible to explore in depth, not just in breadth).

> They had the best browser back when IE4 launched.

They also had the best browser back when IE5 launched and back when IE6 launched.

InclinedPlane2• 15 years ago

It was horrible. But let's keep some perspective on why that sabbatical existed. It wasn't merely because MS had decided they dominated the browser market and didn't need to innovate anymore.

More than anything it was because MS was afraid of continuing IE development. IE is what got MS into all the regulatory anti-trust trouble in the first place, MS was too afraid to make a new mis-step with IE which could land them in even more trouble.

Andys• 15 years ago

This same argument is why I still have it in for Apple, despite their current success in consumer electronics.

The Apple I know is the one from the 1990s who had things locked down so tightly, hardware and software, bit loyal developers who put their best efforts into the platform, as if to destroy any ecosystem that built up around their successful products.

And I still feel that things haven't changed - if you can make money from iOS apps today, that is wonderful - but be prepared to get bitten one day.

rbanffy3• 15 years ago

> I reckon they shipped IE7 at exactly the most damaging time for the web

Mission accomplished, then.

WalterGR2• 15 years ago

This chart doesn't show the 5 year sabbatical that Microsoft decided to take between August 2001 and October 2006.

Major versions of IE tend to be (are always?) synced to OS releases. What effect did the Longhorn Reset have on the sabbatical Microsoft "decided to" take?

Does anyone know? I can't imagine I'm the only one to ask this in the intervening 4.5 years, but my Google searches aren't finding anything.

cryptoz• 15 years ago

My understanding is that once MS won the '90s browser wars, they disbanded the IE team entirely. Only when new competition arrived did they restart development. I could be wrong, though...

abhijitr• 15 years ago

After IE6, a good chunk of the team was repurposed to work on the next-gen presentation technology for Longhorn, which eventually became WPF/XAML.

Another part of the team went on to build some ill-fated re-skinnings of IE, such as MSN explorer.

So yes, the Longhorn train wreck was a big reason for the sabbatical, but not the only one.

Semiapies• 15 years ago

Maybe because five years ago was five years ago. I see nothing wrong with focusing on the most immediate five years.

yread• 15 years ago

Hmm Opera is missing.

2006

9.1

2008

9.5

2009

10.10

2010

10.5

zootm• 15 years ago

It's in there now, although it only shows major version numbers (generally not unfair, although a major version of Chrome doesn't necessarily mean the same as a major version of, for example, Firefox).

Edit: Actually, I just noticed that Opera 9.1 and 9.5 are there, along with Firefox 3.5, so scratch that!

Tyrannosaurs2• 15 years ago

This is my main issue with it. It shows releases which increment major/minor version number but not all releases (by this definition) are equal.

Ultimately all it says is Chrome do lots of smaller releases, IE few bigger ones, the others in the middle.

Qz• 15 years ago

This 'Modern Browser' stuff is getting kind of old. The question is relatively meaningless, but if it were meaningful, the answer would be more complex than something you can diagram in ~5 minutes.

jdub9• 15 years ago

Think of this as an answer, not the answer. :-)

jokermatt9995• 15 years ago

"Modern browser", "open/not open", "killer", "dead", and recently, "Quora" articles can probably be ignored if they're the focus of the title. They're pretty much always petty arguments. I think I'd respect these sort of articles a lot better if they skipped the rhetoric and just focused on their point. "IE is not a modern browser" is a great soundbite, but it also means that's what rebuttal articles will focus on. "IE is holding back the web by not implementing features the majority of other browsers are" is less catchy, but it conveys the actual focus of the article. Sadly, I think most tech journalism would distill an article like that into another meaningless soundbite.

podperson2• 15 years ago

One of the things that killed Macromedia Director Schockwave (remember that?) which was in many respects superior to Flash (SWF stands for "shockwave flash" -- Flash originally rode on Director's coat-tails) was too-frequent releases. As a Director coder you never knew what platform you needed to target and the end-user experience usually comprised being pestered to update their plugin before seeing your content.

Flash forward to 2010-11 and Google is doing the same thing with Android and Chrome (and Firefox is following suit). This is actually a BAD thing, even though we all have broadband, etc. now.

There's some kind of happy medium between significant, reasonably well-paced releases and shoving stuff out the door constantly.

"Real artists ship," but they don't spam you with incremental releases.

jdub9• 15 years ago

I think that might be true in cases where you're not building on such a widely distributed and incredibly backwards-compatible / gracefully-degrading platform such as the web.

nlogn• 15 years ago

And Android for that matter. I really don't see podperson's point; what Android builds have broken backwards compatibility?

davej3• 15 years ago

The paradigm changes when browser vendors can auto-update with releases that are a few hundred KBs in sizes (i.e. Google Chrome).

tremendo• 15 years ago

My impression was always that Macromedia acquired FutureSplash (shortened to Flash) when it became clear it could kill Shockwave, which was much slower (14Kbps connections then). Flash also had a rapid release cycle, which did not have any negative impact in its adoption and success, releases were backwards compatible. Chrome's adoption is certainly more significant now, also not affected by many releases. For Android, I'll wait and see.

rbanffy3• 15 years ago

> This is actually a BAD thing

I am not aware newer browsers break backwards compatibility. Newer browsers usually work better at the things they were already good at. If you want to target a larger audience, pull your analytics and draw your line at the set you want to serve. Usually any reasonable line will cover mostly everybody.

This is the kind of excuse Microsoft uses - when you can't improve your software, you call it "stable" and "mature".

Elepsis• 15 years ago

It's really not that hard to find examples of this. Chrome recently broke rounded corners, and they're about to break h.264 video. But it's certainly more fun to attack Microsoft in every post.

palish• 15 years ago

I've never once been pestered by Chrome. Updates are automatic and invisible.

Groxx2• 15 years ago

I was just about to say this.

Shockwave / Flash / nearly every other application in the world tells you your software is old, and asks if you want to update it, and it will keep telling you this until you check the "don't tell me again" box. Which few people do. So instead they get irritated that it always pops up, and get used to clicking "no" because they didn't want to deal with it then, and don't want to deal with it now. A year passes, they get a virus through an old exploit, and they blame the software instead of their habits / their IT department's dogma (which is not always incorrect, but frequently problematic).

Chrome, on the other hand, keeps itself up-to-date in the background, at all times, and doesn't ask you to deal with it. It doesn't even inform you it's doing so, aside from a minuscule little up-arrow on the wrench that disappears on the next launch - no progress bars, no delay, no "restart now", no alerts, nothing. There's no question if someone's using version X or Y; they have the latest of their branch. Period. Or they'll get it in a day or two, and problem solved.

Transparent updates, especially when they're nigh-totally transparent, are in an entirely different world than "do you want to update yes/no/cancel". It's apples and granite, or oranges and wolverines - there's no comparison ("apples are sweeter than wolverines").

EdiX• 15 years ago

The problem with automatic and invisible update is this: say you are using chrome in your company as the interface to an internal tool of some sort. Since the tool is used internally it's messy and one day chrome changes and now, automatically and invisibly, no-one can use it.

I'm not saying chrome's way doesn't have its upsides but there is also a good argument for mantaining older releases.

simonw2• 15 years ago

Wow, 2007 was a bad year for browser releases.

defroost• 15 years ago

The "Grade A" browser ratings are only slightly more meaningful than the "Modern Browser" label. And IE fans (apologists?) can rationalize from now until Sunday, but web developers who project after project are forced to write CSS Hacks, conditional comments, etc. know that improvements to IE have been slow and minimal. The "modern", standards compliant browsers are the ones that continue to improve and push the envelope re: HTML5/CSS3 features. All this talk about IE doing well by not adding HTML5 features because the W3 spec may change is just an excuse for continued suckiness. You'd think the fact that so many people (me not included) use IE give MS a greater pressure to keep pace with the Chromes and Safaris of the world, but I guess they don't agree.

Joeri• 15 years ago

"web developers who project after project are forced to write CSS Hacks, conditional comments, etc. know that improvements to IE have been slow and minimal"

Not my experience. IE8 seems to have its act together, at least on CSS 2.1 compliance. Granted, conditional comments are still necessary for IE6 and IE7, but MS has made big improvements.

bkhl• 15 years ago

Same here. I agree that IE6&7 were bad, but I didn't have to "hack around" for IE8 and IE9. Of course, if you do need to hack around, then you are probably not doing what is considered as "standard". A lot of things that we, early adapters, talk about (e.g. HTML5) aren't standard, yet.

This modern browser crap is getting old. I have no problem browsing internet with both IE9, Chrome, and Firefox.

ozten• 15 years ago

Using this logic, lynx is fairly modern having released within the year:

June 2010 lynx 2.8.7 Oct 2006 lynx 2.8.6 Feb 2004 lynx 2.8.5 ...

lynx, now with WebGL, WebSockets (disabled) and IndexedDb!

Detrus• 15 years ago

It's a bit strange comparing Chrome's releases with the others. Firefox had many more pestering minor updates in that time, which are comparable to Chrome's, only Firefox labeled them differently. Safari also had major changes between version numbers.

To get an accurate picture the diffs between updates should be compared, Chrome made version numbers meaningless.

blocke2• 15 years ago

Firefox's minor updates were not really feature updates.

On the chart Chrome whole number releases are new feature releases. Firefox 3.5 and 3.6 are seperate because they are similar feature releases. 3.6.0 to 3.6.1 are not.

geekfactor2• 15 years ago

There's a certain arbitrariness to this analysis, since all that is being considered is an increment in the major version number. If the author were looking at operating systems this way, Apple hasn't released a new operating system in nearly 10 years! (OS 10.0 Cheetah was released on 3/24/01)

Not to offend anyone, but Chrome major version releases tend to feel relatively incremental compared to what I had come to expect from Firefox.

code_duck2• 15 years ago

I use Chrome every day, and would be unable to list one major difference between Chrome 6 and 7. Or 5 and 6. Or 7 and 8.

robryan• 15 years ago

The developer tools have been steadily improving, apart from that I can't recall to much exactly.

Will be interesting to see if Firefox can keep to their new release schedule after 4. Otherwise it seems that Chrome just iterates to fast compared to Firefox. Chrome 9 is a long way ahead of Firefox 3.6, sure Firefox 4 might even it up but if every release has this kind of beta time Chrome will just keep skipping ahead.

InclinedPlane2• 15 years ago

Why would that be offensive? Chrome major version number changes are a: typically invisible to the user, and b: chosen for merely engineering reasons. If tech blogs didn't exist to tell people about major version changes in chrome most user's wouldn't have the slightest clue what version they were running.

tl;dr Chrome's major version releases are supposed to be incremental.

geekfactor2• 15 years ago

Ask the person who down-voted me.

jdub9• 15 years ago

These really do reflect the most significant releases of each browser, regardless of arbitrary major/minor/micro version number changes... Firefox 3.6 was as major a release in that universe as Chrome 8 was in theirs.

Check the Wikipedia pages for each major browser to see just how much changed in the versions represented. (Perhaps I should link them!)

blocke2• 15 years ago

All the nitpicking people are doing on this and ignoring the point behind that graphic...

The Point: "Modern Browsers" are the browsers that actually get around to actually @#$@ing SHIPPING new features and NOT TAKING FOREVER TO DO IT. Iterate faster and get new crap to users NOW.

user24• 15 years ago

> get new crap to users NOW

thing is, 'new crap' is also unfamiliar crap. Unfamiliar to developers. 5 browsers, 5 teams of developers, results is 5 different ways of doing things. -moz-border-radius and the like are BAD for the web, because everyone shipped new crap without any kind of agreement on how it was supposed to work. And now we've got vendor specific code in our websites. That's not how the web is supposed to work.

Impatient browser vendors shipping unfinished specs like css3 are as bad as proprietary spec extensions like 'filter' etc.

Modern Browsers Ship

Related Stories