Last time i saw a stats page its getting upwards of 19% .. i am one of these 19% . I think its far superior to IE (speed and security) . Since moving over i havent had any spyware to remove . IE is crappy :D hehe
maybe someone else has some more hard facts.
i noticed mozilla is gettin real popular too :)
My site is non-nerd and typically reports:
From my Analog browser summary report:
no.: reqs: %reqs: pages: %pages: Gbytes: %bytes: browser
---: ------: ------: -----: ------: ------: ------: -------
1: 223790: 79.40%: 67043: 67.55%: 2.43: 82.40%: MSIE
: 128611: 45.63%: 37394: 37.68%: 1.39: 47.15%: MSIE/6
: 89608: 31.79%: 27010: 27.21%: 0.99: 33.45%: MSIE/5
: 5515: 1.96%: 2588: 2.61%: 0.05: 1.78%: MSIE/4
: 16: 0.01%: 11: 0.01%: 0.00: 0.01%: MSIE/3
: 6: : 6: 0.01%: 0.00: : MSIE/7
2: 22113: 7.85%: 7890: 7.95%: 0.25: 8.52%: Netscape
: 11948: 4.24%: 4142: 4.17%: 0.13: 4.56%: Netscape/4
: 7520: 2.67%: 2757: 2.78%: 0.08: 2.88%: Netscape/7
: 2312: 0.82%: 844: 0.85%: 0.03: 1.01%: Netscape/6
: 299: 0.11%: 125: 0.13%: 0.00: 0.06%: Netscape/3
5: 6765: 2.40%: 3231: 3.26%: 0.05: 1.56%: Mozilla
: 850: 0.30%: 225: 0.23%: 0.01: 0.38%: Mozilla/1
: 448: 0.16%: 164: 0.17%: 0.01: 0.17%: Mozilla/0
9: 1289: 0.46%: 345: 0.35%: 0.02: 0.56%: Opera
: 887: 0.31%: 234: 0.24%: 0.01: 0.45%: Opera/6
: 336: 0.12%: 99: 0.10%: 0.00: 0.10%: Opera/7
: 66: 0.02%: 12: 0.01%: 0.00: 0.02%: Opera/5
I use Opera 7.03 spoofing MSIE 6.0 but my accesses are excluded from my Analog reports.
[edited by: WarmGlow at 7:52 am (utc) on May 27, 2003]
A good number of loggers do not track Opera properly. Including WebTrends, SawMill, Counter.com, and Fast/Mach5 as far as I know. Opera 6 and Opera 7 can produce 5 different Agent strings.
Now if my memory is not completely deficient (hope it's not), the default for Opera is to identify as IE.
The question then would be how many Opera users leave the default values. As Opera users are normally more technical than the average, I would say it is not too high a percentage, but does anyone have hard facts (guess statistics will also be acceptable, although I wouldn't call them hard facts)?
I think these stats are as good as any.
< 0.5% is in tune with my own experiences unless it is a site with an obvious bias - like this one :)
George, your stats don't seem to take spoofing into account (don't say anything about it).
georgeek, w3schools gets their stats report directly from theCounter.com which is one of the stats sites that does not report the correct Opera statistics as it uses the spoofed "string." However, even with that said, theCounter.com's current stats (w3school's list is outdated) report Opera at approximately 1.5%. Keep in mind, many sites that actually use theCounter.com's tracker are not very advanced websites and typically appeal to a "less-than-technical" crowd.
Yes, many Opera users ARE more tech savvy than the norm, but this is exactly the reason many do in fact leave their Opera ID'd as IE6 (default). This feature is not merely for spoofing, but in fact serves several purposes including triggering certain built in features that will allow Opera to emulate IE's (ahem, unique) code interpretations. Another obvious use is to avoid site lock-outs for those still-too-numerous sites that sniff either for NN variants or IE variants, but ignore all others.
M$N.com itself was sniffing for Opera and serving errant style sheets that caused content to display improperly. This documented, intentional, misleading practice lead many Opera users (those with Opera set to ID as Opera) to believe the browser was at fault, and could not properly render M$ pages. Changing Opera's ID to MSIE6, miracously enabled the improperly rendered pages to render properly.
This, along with other malicious, frightened-by-a-real-browser response tactics was the moving force behind Opera's release of Opera Bork. The Opera Bork edition took those same M$ pages and translated them into the language of the famous Swedish Chef Bork character from Sesame Street. It was a hilarious bit of revenge!
M$'s true fear of Opera is the stiff competition Opera is handing M$ in the (Internet enabled)cellular phone market. Opera's small screen rendering is a groundbreaking technology (truly!) and eliminates the need for coding WAP, instead taking standard HTML pages and reformatting them to display on the growing number of small screen devices. Nokia, Erickson and others are all in the Opera camp.
If you have Opera 7, simply key "Shift + F11" to see small screen rendering in action. This is a "must have" feature for serious webmasters.
Biased? Perhaps... informed? Definitely!
Wow, never had tried the small screen. Nice feature, although it seems to me you would still have to have a huge cellular phone to display that.
|famous Swedish Chef Bork character from Sesame Street. |
Oh, I thought it was something from Star Trek...
Spoofing: There are indeed still some pages that do not render correctly with Opera ID, but I for instance usually surf as Opera and only change when getting to one of those pages. It is so easy to change, why should I not contribute to raise Opera usage stats?
Regarding Smartphones and Small Screen Rendering: [opera.com ]
More, with screen shots of popular phone models using Opera: [opera.com ]
Add Kyocera to the Opera camp as well... the list grows!
By the way, I keep my settings to ID as Opera. I find extremely fewer and fewer sites where I have any problems using this setting.
I log browser usage for several non-techie sites. My script is not spoofed by Opera spoofing as IE. Opera accounts for 0.4% of visits, Mozilla and Netscape 6+ for 1.3%.
Personally, Opera is my favorite. It's fast, safe and has lots of nice UI features. The main thing I miss is the Googlebar (for checking PR). Also, for some weird reason all postings on this board keep being 'new'...
I say it looks very, very promising.
What is interesting, is that I'm finding a lot of Opera users who are not technical in background, nor particularly Internet savvy.
Buzz factor... cool skins, outrageous levels of customization, and all the "stuff" ignites little sparks on the consumer end.
So to cut a long story short - it's not that easy to determine these stats then. It's kind of a shame but I understand why.
In most cases it seems it could well be 5% plus of the market. If only 5% of Opera users change the browser string to identify as Opera and that makes up say (roughly) 1.5% of all browsers then the percentage of Opera browsers could be up around?25%?
If my out of practise maths are anything to go by...
No wonder stats are so skewed towards IE. Not only does Opera send an IE browser string, but also OmniWeb and Safari by default. I'm not so familiar with Mozilla and what it does but it certainly makes it difficult to determine how much of a hold IE has on the market...
I really wish I could say ALL the problem was related to legacy sniffer scripts, but unfortunately this is not the case. There are more than a few newer sites that are still coded as if there were only two main browser flavors.
Here is the real issue however.. IF you code for Web Standards, and use IE as your baseline, but check your code in a Standards compiant browser (Opera/Moz), then browser stats lose some of their importance. It will not be such a "do or die" issue... just one of simple user prefernce.
|Here is the real issue however.. |
Not really an issue for me/us as our site doesn't push the boundarys of any particular browser, more of a curiousity since I love Opera (Win), am inclined to be anti-MS and I wanted to see how well Opera was tapping into the market. I appreciate your feedback though and do see your point papabaer.
My site, a quite non-nerd church site, receives visits from two and only two Opera browsers. I'm one of them and my helper is the other. I don't think I've ever seen another. Frankly, I wish Opera defaulted to identifying itself as Opera. It's too bad because I rather prefer the way Opera 7 handles some of my CSS, especially the styling of the title and author of articles. IE does it acceptably, but they display just differently enough that I wish everyone saw it the way *it's supposed to be*. grrr..... As far as I can tell, Opera 7 is interpreting the CSS properly, and IE isn't. No big deal, but....
|I wish Opera defaulted to identifying itself as Opera. |
Opera always identifies itself as Opera. Identification is done in the useragent-string the browser sends to the server. For Opera acting as Opera, it looks like this:
Opera/7.10 (Windows ME; U) [en]
For Opera acting as IE:
Mozilla/4.0 (compatible; MSIE 6.0; Windows ME) Opera 7.10 [en]
As you can see, both strings contain the word Opera. It is bad sniffing if the second string is labelled as an IE string.
|A good number of loggers do not track Opera properly. Including WebTrends, SawMill, Counter.com, and Fast/Mach5 as far as I know. |
Brett_Tabke This is true but Sawmill (which I use) is believed to handle Opera correctly in virtually all cases from Sawmill v6.5 onwards.