Rounding the corners of a box

I very much doubt that number - NOT that you can rely on percentage given that percentage is NOT the number of users.

Let's ASSUME for a moment that wikipedia is quoting real numbers:
http://en.wikipedia.org/wiki/Usage_share_of_web_browsers

First off you have to question the accuracy of the numbers - is that 50% jump in Q3 2006 you can see on most stats from more people using firefox, or is it from FF's pre-fetch artificially inflating the numbers? Is Opera's low numbers from not a lot of people using Opera, or that Opera users often have to 'mask as FF' or 'mask as IE' to get around faulty browser sniffers? (that are often so *** stupid they identified Opera 10 beta as Opera 1.0)

But that's only the tip of the iceberg on how irrelevant percentage can be - Sure, the percent of the market using IE has dropped from 90% in 2005 to around 70% for general purpose websites (as per TheCounter.com) - but over that time frame the number of internet users has gone from 16% of the world population to 23%... Think about that....

2005 - 90% of 1,032,640,000 internet users is 929,376,000 people using IE.

2009 - 70% of 1,427,093,220 internet users is 1,094,104,802 people using IE.

Net result? Roughly 165 million more people are using IE today than were three years ago... The number of people using IE hasn't dropped one damned bit - thus are born the lie of percentage statistics. It's one of those things you learn in business school: It is entirely possible to lose market share while still having a net and even healthy growth in client base! The moment you ask "Yes, but a percentage of what numbers" when the total numbers involved changes every year, it all falls apart.

Looking at the growth of internet use worldwid, and the growth rate of IE, a realistic guesstimate for the number of people using IE in 2012 would be around 1.25 billion, which if the growth rate of population continues at the 2.55% yearly it has been and the population estimates are right, would be out of 2.23 billion internet users... in other words 56%.

Assuming the trends hold - it's possible they won't given that this is the first year industrialized nations saw a leveling out in internet user base growth... and the US actually for the first time saw an actual decline!

So 33% or less? Bullcookies. Think about your numbers before making wild claims - I'd be shocked to see it go below 50%, and even if it does there will be more people using it in the future than there are today.

I'm gonna turn this post into a blog entry on my Opera blog once I'm done dicking with it's skin - So many people are easily decieved by percentages and don't understand trend prediction, it's a good topic to cover.

 

But the number of people using the internet has increased and more of those people are using non-IE browsers. You can't only look at one range and say that's the domain

People said that 5 years ago when I said IE would lose its 95% market share lead. I also said Microsoft would no longer matter anymore. Soon after, Paul Graham wrote an article "Microsoft is dead" where he said the same thing and IE has lost a third of its position in the market. I'm pretty good at this. Or at least I'm better than everyone who said "bullcookies" 5 years ago.

 

I wouldn't have said bullshit to them losing share - share is always in flux and nobody can sit there at 95% forever... Besides, the whole "Does it work in IE, who gives a *** about the rest" bugged me just as much as the "Down with evil corporations" whackjobs you hear in regards to M$.

I would instead have pointed out that share is irrelevant if you ask the question "A share of what"

You go back to that 95% lead from 2004, how many internet users were there then? That's just under 14% of the world population was using the internet at that time - given the world population figures for 2004 was 6.38 billion that works out to 893 million IE users at that time. Since then IE may have dropped 25% share, but at the same time increased it's user base 22% to that 1.09 billion.

Which is why market share usually means jack **** - the size of the market is always in flux. FF and alternative browsers in general have done really well in niche markets like the 'big geek on campus' or developing nations - but in developed nations outside the geek community, not so much.

... and really to average Carlos Mencia Grey (#DDD) user, all these alternative browsers don't offer a damned thing they care about they aren't getting from IE7 or 8...

 

That will be something that I doubt will come true. What you are predicting is IE losing half its market share within 3 years. Firefox and Safari have both more or less stagnant at their current market share. Unless something major happen, I don't think your prediction will occur. What many end-users are concerned with is whether the website works or not, not whether the website conform to web standards. In short, the end results matter the most.

 

Two major things:
1) The mobile web
2) ChromeOS

Correct. See above.

What you are forgetting is this. Web developers hate IE. Web developers promote other browsers to everyone. IE only works well with Microsoft software. IE does not work well with the web. Now, the web is the OS. IE doesn't fit. Every other browser does.

Then there's momentum. IE has been trending downwards for 5 years while other browsers have trended upwards for 5 years. Friends don't let friends use IE. (Example, I started using Firefox 5 years ago because my son kept pushing me to use it. His gaming friends pushed him to use it.) Would you let your wife use IE? No, you wouldn't. You're smarter than that.

 

A claim made for over a decade - the mobile web is actually here, and it's running XP Home on 8-10" displays at 800x480 or 1024x600... and in general IE runs better than FF in those narrow environments. (just as XP Home is proving a much better OS on them than and of those custom linsux distro's ever did)

Oh yes, running applications in an interpreted language manipulating the DOM using a bloated and inefficient format for storing and decrypting the UI (HTML) inside a browser engine atop a X11 implementation atop linux is going to be SO efficient... No wait... efficient isn't the word - what's the opposite of efficient again?

People aren't running out to buy 3ghz quad core processors just so they can run crapplications that are bandwidth locked and have all the performance of native applications on a 486/66 running windows 3.1

You tried editing a REAL spreadsheet of more than 100 cells at spreadsheets.google.com? Ever tried doing more than ten pages of document in their word processor? You can't even scroll up and down without it being choppy! They're **** toys and compared to native applications will always remain toys since javascript is an interpreted language - and even the scripting engine in chrome does little to nothing to alleviate that. One might as well have written them all in line numbered non-compiled 1980's basic!

It's a bit like thin clients - they NEVER catch on until they become big enough to run the big boy OS instead of the goofy toy OS. How many thin clients can we count that went the way of the dodo the past 20 years? Hell, it wasn't until "nettops" that we had thin clients anybody showed the least interest in (not calling them thin clients helped too), and that's because most of them with the 1.6 atom and GMA950 (or even better nV Ion) have as much horsepower as a mid-range desktop from four years ago. (and aren't based around some obscure cpu or upscaled PDA OS)

NO IT HASN'T!!! It's been CLIMBING IN USE. EVERY browser has been climbing in use - not one of them has seen a downward trend unless you count Netscape. The ONLY way you can consider it to be declining is if you only count it by A MEANINGLESS PERCENTAGE!!! Hell, you sit down and do the math, and IE's growth rate has been pretty much flat since FF passed the 10% share mark. That's not a downward trend when you start out over 90% market.

There has not been a single drop in the number of people using IE the past DECADE. - even when FF and others show market gains. Until we see an actual decline in use, your statements are... well... I'll be nice - based on false assumptions and incomplete facts - and boils down to little more than rampant paranoid evil corporation/Microsoft bashing...

I mean, can't you at least bash the REAL evil corporations - you know, Apple and Adobe. Oh wait, that's my Shtick.

 

I am a web developer. Because my country uses IE more than any other browser, and by a ginormous percentage, I build ALL my pages for IE. All of them. Even my personal page, even though I'm the only one who looks at it (and never in IE). So do most of the rest of us. I told a friend of mine to install Firefox, then I went ahead and installed FF and Ie7 on her computer (she had ie6 but on win xp) but she doesn't notice any difference. If I hadn't installed it, she would not have upgraded. How many people do we do this to? It's a bit like walking into someone's house and telling them they need new shoes, then taking their old shoes and throwing them away and giving them glitzy new shoes. It's a good thing, but kinda rude and pretentious.

The trend in web design in my country is... View Source. There is no Jeffery Zeldman here. Even those who Do know how to do it Right are doing it for large corporations via big design firms. This will not touch the small developer such as the one who developed the site linked above (no, none of us did that framed mess). Frontpage was supposed to be dead, wasn't it? It's alive and well in my country, no matter what Microsoft says. IE6 is alive and well, though luckily there was a drop as people upgraded OSes to Vista and 7. However, those are People Who Upgrade.

People Who Don't Upgrade are like my uncle. He is still using Windows 98 with IE5.5 and has no qualms with any of it except he would like to change his ISP. Thing is, IE5.5 can't even render (at all) his ISP's home page : ) Otherwise, he ddoes not notice.

He's not a power user, not a mobile user, not a facebook user (I dunno but I suspect IE5.5 doesn't work so well on fb being that it's such a bloated piece of JS garbage). He's the Joe Sixpack mentioned above. He can see all the sites he wants to, because the sites he visits are not fancy. He doesn't know what mouse gestures are (heck neither do I), and he doesn't care. What can Firefox do for him? (Luckily my cousin, his son, installed Mozilla (not Firefox) a long time ago and that browser CAN render the ISP's home page : ) However my uncle isn't completely sure what Mozilla is and still clicks on the big blue e.

You've got a long way before reaching these people and their numbers aren't all that small.

*edit it seems the site I linked above is in the middle of their changeover we were informed of... or they forgot to pay for their domain name : ) If it comes up a red site, then the frames are gone and instead it'll be tables with spacer gifs in the menu navigation : ) Unless they've redone the beta I saw.

 

I presume you mean netbooks. I'm talking about mobile phones/tablets/slates as well as netbooks. IE is puny on phones. ChromeOS will run circles around Windows on netbooks.

Microsoft could only hope to do as much. IE can't even properly use the DOM at all much less anything past level 1.

Now you're starting to sound like Microsoft explaining away their shortcomings. You're not going to run AutoCad on a local file with your netbook. Of course, you might run AutoCad on your office computer and view the files on your phone while you're on the road.

Nope.

All the time.

Don't have that problem.

You presume the total app and all apps will be run with javascript but it's only the client end and the user UI that will do so. For example, I occasionally run Gimp or OO from my FreeBSD notebook while on the road. I don't have those installed on that notebook but run them using X forwarding. I don't recommend doing that all the time but it works. However, we're not talking about games or high end processing. We're talking about what most people do: surf the web, read their mail and text. You don't need a 3Ghz proc for that and you definitely don't need a $200 operating system from Microsoft.

Times they be achangin'.

Let's look at it this way: IE has a declining share of a growing market. Other browsers have an increasing share of a growing market. As more people are joining the market, they are increasingly using other browsers and NOT IE. So, apparently, out of every 10 new users, six or more of them are NOT choosing IE. That's a bad sign when more new users are choosing your competitor.

But any product will always get users. Even the worst in the world will gain some following. If you sold crap in a box, someone would buy it, but would you claim success by that increasing number of buyers if one month you sold 10 and the next you sold 12 while your competitor sold more perfume in a box than you did?

Business is judged by its share of the market compared to others. Even in this thread people will say "Oh, IE will never die because it's got more share than other browsers" so you can't say you it won't die (become insignificant) with a decreasing market share/percentage.

 

Google's Javascript isn't considered terribly efficient in the first place. More than one person has suggested that they write Javascript like they write their native language, Java.

 

Thanks for the link mate, been seeking for it for quite a while now, very useful indeed.

 

Why do you think they put so much effort into optimizing their own engine to run it.

Instead of just writing better script that will work cross-browser reliably and not using script when it's unnecessary, let's release a 'new' browser that has a user interface that's like a trip in the wayback machine to IE3 with the only innovative or original feature being an uber-optimized javascript engine that can blaze through code no matter how badly we write it...

Great idea there guys.