Net Applications again: Opera vs. Chrome - actual numbers vs. claimed market share

Readers' comments 21

Left under the original article, as archived.

• 

  sms985Friday, December 12, 2008 1:19:57 PM

  Good point, let me just ask:
  How do you actually count you active users? Is it by number of downloads, traffic on sites...?
• 

  Petter NilsenFriday, December 12, 2008 1:41:56 PM

  I guess Netapplications don't measure Russia:
  
  http://www.liveinternet.ru/stat/ru/browsers.html?period=month&slice=rus&id=18&id=20&id=17&id=checked&show=%D0%BF%D0%B5%D1%80%D0%B5%D1%81%D1%82%D1%80%D0%BE%D0%B8%D1%82%D1%8C+%D0%B3%D1%80%D0%B0%D1%84%D0%B8%D0%BA&per_page=10&report=browsers.html%3Fslice%3Drus%3Bperiod%3Dmonth
  
  or Europe:
  
  http://www.pubcon.com/redirect.cgi?f=118&d=3800000&url=http://www.xitimonitor.com/en-us/browsers-barometer/browsers-barometer-september-2008/index-1-2-3-145.html
  
  
• 

  HaavardFriday, December 12, 2008 1:43:48 PM

  tomas.sprlak: I believe it's the update system. Each Opera installation checks automatically for updates. The numbers can be calculated based on that.
• 

  sms985Friday, December 12, 2008 2:01:38 PM

  tahnks
• 

  Unregistered userFriday, December 12, 2008 3:47:23 PM

  Anonymous writes: I was gonna say, gathering numbers from downloads would be inaccurate, since many people may try out Opera and then never use it. Counting from the update system makes more sense. As a Linux user, I turned off auto-update, and rely on my package manager to keep Opera up to date. Does that skew the results? As mitchman2 pointed out, it's probably because NetApps only monitors a small subset of the web, that Chrome has a higher market share.
• 

  PaiTraktFriday, December 12, 2008 4:39:50 PM

  Anonymous: I'm guessing Opera still polls some Opera-server for browser.js-updates.
• 

  Charles SchlossFriday, December 12, 2008 5:26:26 PM

  Congratulations to Google for Chrome launch on Windows
  
  Does Net Applications check the full browser string?
  
  
  here is a differentset of Browser Statistics
  
  http://www.w3schools.com/browsers/browsers_stats.asp
• 

  wupperbayerFriday, December 12, 2008 6:03:34 PM

  @mitchman: It seems that Net Applications is mostly basing its usage statistics on US-based websites. Yes, they say they’re monitoring browser usage worldwide, but I doubt it. Maybe they have some European websites they’re monitoring, but most of them are US-american ones.
  
  If not so, Opera’s usage share had to be near zero in the US (because of the 4% in Europe an 20(!)% in Russia.).
• 

  MorghusFriday, December 12, 2008 7:10:57 PM

  Doesn't China and India have a pretty sweet amount of Opera users?
• 

  AlexeyFriday, December 12, 2008 7:31:55 PM

  @Anonymous
  >As a Linux user, I turned off auto-update, and rely on my package manager to keep Opera up to date
  
  Your update is from your repo. It does not count at all.
  
  @wupperbayer
  Don't forget that russian ≈21% (+ ≈9% Opera Mini = 30%) are on small russian web, USA usage couldn't be close to zero.
• 

  InvestorFriday, December 12, 2008 7:34:19 PM

  If Desktop market grows, Opera market share doesn't have to grow to add new users.
  
  Users and usage are two different things. Opera is first, second or third browser of choice on Desktops. Used all the time, most of the time, or some times.
  
  The number of average fulltime users may be calculated from quarterly revenue data...
  
  We have read CEO statement suggesting Opera gets $ 1 per user per year from search deals...
  
  Desktop revenue Q 3 was Nok 28 mil.
  Average $ 1 / Nok rate in Q3 was 5,4
  
  Number of average fulltime users is then 20,7 mil.
  
  (((28 mil x 4 Q) / rate 5,4) / $ 1= 20,7 mil)
  
  Let's see if that number grows, quarter on quarter
  
• 

  Unregistered userFriday, December 12, 2008 8:07:18 PM

  Anonymous writes: Haavard, you're either mis-understanding what NA is measuring or you're trying to mislead your readers. NA measurs _usage_ and not _users_. Opera could have 10 times the _users_ (presumably defined as a computer that connects to an opera update server at least once a day - though I don't see that made explicit anywhere by Opera) and if those _users_ spent fewer hours online (visiting fewer sites) than Chrome _users_, Opera could have less _usage_. Imagine that I'm an Opera user and I'm also a Firefox user. If I spend half of my time on each, I'm counted as a _user_ of both but my _usage_ would be split across the two. Now imagine that I'm a full-time Chrome or Safari user. I'm now counted as one user but with twice as much usage as I was as either an Opera user or a Firefox user. None of this is terribly difficult unless you want to make it that way. NA has always shown IE crushing Firefox in usage, but that doesn't cause everyone at Mozilla to try to discredit their data at every opportunity. It's a valuable datapoint that browser makers can add to the various other pieces of information about users and usage that helps them to understand the Web better. NA has always shown IE crushing Safari in usage, but that doesn't cause everyone at apple to try to discredit their data at every opportunity. And I'm not saying that they are avoiding sample bias, no one can, but I am saying data that's somewhat close, and probably more comprehensive than any other available data, is better than no data at all. Finally, even if you don't believe that the ratios or the specific numbers are accurate, there's no doubt a set of trends that are meaningful. IE is falling, at a pretty good rate. Firefox and Safari are growing at a pretty good rate. Opera is mostly static. Those trends are meaningful and to discount them as if they were invented out of thin air is foolish.
• 

  Unregistered userSaturday, December 13, 2008 2:48:17 AM

  Anonymous writes: yeah, trend is rather clear: while ff (and lately safari) grows, opera is at the same place. that means, that for former IE users, FF is more appealing alternative than opera. rather than discrediting this source of information, opera needs to do something about the later - make it more attractive to IE users. mostly by making it compatible with real world wide web (gmails/yahoo mails etc kind of stuff [deviantart does not work since second world war..]) and adding extensions. opera' lack of support for some VERY useful and popular extensions is keeping users away as for the chrome. well, their ads are EVERYWHERE, so their user base grows. many MANY users have never heard of Opera. you expect anyone else than you to 'spread the world'? I'm absolutely sure, that chrome is able to reach 5% market share in half a year.
• 

  InvestorSaturday, December 13, 2008 7:27:18 AM

  > Anonymous writes: as for the chrome. well, their ads are EVERYWHERE, so their user base grows. many MANY users have never heard of Opera. you expect anyone else than you to 'spread the world'? I'm absolutely sure, that chrome is able to reach 5% market share in half a year. <
  
  Opera being unknown in that user segment, Chrome will have to steal market share from IE or FF, right ?
  
  So Opera market share may still grow in other segments.
  
  Yes, we agree !
  
  
• 

  Unregistered userSaturday, December 13, 2008 10:13:56 AM

  Anonymous writes: Yeah, Chrome's ads everywhere will make more people switch to another browser and that's good for Opera. Now for some moronic stuff: "NA measurs _usage_ and not _users_." That's idiotic, considering the fact that Opera's caching means that it fetches stuff more rarely than other browsers, making it severely under-counted. Combine that with the fact that it needs to spoof as other browsers all over the place and you get the point. "while ff (and lately safari) grows, opera is at the same place" Bullshit, Opera has grown 55% in a year. "mostly by making it compatible with real world wide web" This is bullshit. Educate yourself, moron: http://my.opera.com/community/forums/topic.dml?id=259677
• 

  Unregistered userSaturday, December 13, 2008 2:02:17 PM

  Anonymous writes: "That's idiotic, considering the fact that Opera's caching means that it fetches stuff more rarely than other browsers, making it severely under-counted. Combine that with the fact that it needs to spoof as other browsers all over the place and you get the point." blah blah - stat codes use standard enforced methods of avoiding caching. if opera caches these scripts anyway it is Opera breaking standards (it can be easily checked anyway). btw, 'opera caching' is a MYTH. it was once, long ago, better than average, now it is standard behaviour of all browsers "Bullshit, Opera has grown 55% in a year." blah blah - it is 55% USER growth, not marketshare growth. as long as marketshare stays sill around 1% mark nobody is going to support opera if it costs more that 5$ to do so "This is bullshit. Educate yourself, moron: http://my.opera.com/community/forums/topic.dml?id=259677" blah blah - sorry, but it fault is on both sides - opera is subborn and does not support some stuff (like oncontextmenu) on purpose, or is very slow to implement other (get/set in javascript.. callee issues etc. or overflow-x/y. these are implemented now, but it took opera few years too long for webdevelopers to take notice. and btw. your wording suggest extremely low self-confidence, arrogance, lack of arguments and foremost - very low level of culture. please, do not bother replying me.
• 

  Unregistered userSaturday, December 13, 2008 5:56:33 PM

  Anonymous writes: Who's the one with low self-confidence? I'd say it's the Firefox fanboy with his irrational hatred of Opera, spreading to blogs including Opera Watch, Favbrowser, and so on. Your pathetic excuses for Net Applications, who have admitted that their stats are skewed, and have been busted editing their own stat multiple times, just goes to show how much of a fanboy you are. Net Applications is useless because of Opera's caching, because of Opera being forced to spoof, and because they actively edit their own stats and lie about what they represent. "stat codes use standard enforced methods of avoiding caching" Which is obviously wild speculation on your part. "it was once, long ago, better than average, now it is standard behaviour of all browsers" Nonsense. Stop being an irrational moron, and educate yourself.
• 

  Khaled KhalilSunday, December 14, 2008 10:53:52 AM

  if google have its own stats i would like to read it.
  despite it is now a competitor on the browser side, also it have a history of neglecting opera, as far as i know, google is the the default search engine of many browsers (including opera, firefox, safari, among many others) and therefore it is their first financier (excluding safari which is owned to a relatively big corporation), hence google should know 'exactly' how many searches were performed by one of those browsers (also by their toolbar for internet explorer).
  of course it would not be a real state for (at least) two reasons:
  technically:
  1-not all users use the default search engine the way it is supposed, many users likes to open the search engine web page, and few like to change the defaults.
  2-even if google is default on many, it is not for all, i mean by that exclusion mainly internet explorer, which is what it is.
  politically:
  sorry, i don't mistrust someone until i have a good reason to doubt.
  still, i find it interesting to know.
• 

  Unregistered userSunday, December 14, 2008 11:56:17 AM

  Anonymous writes: This is simple case of not liking the message so trying to discredit the messenger. If Opera fans, including the Opera itself which uses NetApplications HitsLink (if all Opera blogs are using HitsLink, it will help to increase Opera market share numbers at NetApplications) need to look at what is actually happening and not what they wish was happening. If Opera ASA is happy with the product's reach (users and usage) then they should be happy and tell their fans to be happy. If they are not happy with the product's reach, then they should make changes to improve its reach. Trying to confuse everyone about what the reach is doesn't help. Either accept it is what it is and be happy or work to make it better than it is and stop trying to lie to fans about Opera's popularity. Opera is a great web browser and people should stop worrying about numbers anyway. If you like it then you should use it and if you don't like it then you should use something else. Stop being silly.
• 

  Unregistered userSunday, December 14, 2008 12:28:38 PM

  Anonymous writes: "This is simple case of not liking the message so trying to discredit the messenger." This is a simple case of a Net Applications epmployee using a red herring to distract people from the fact that NA's numbers are obviously completely bogus, and worse yet, edited manually by NA staff to make the numbers look the way they want them to. "if all Opera blogs are using HitsLink, it will help to increase Opera market share numbers at NetApplications" Except Net Applications clearly decided to block *.opera.com since that's what you saw just after Opera started using the service. "Trying to confuse everyone about what the reach is doesn't help. Either accept it is what it is and be happy or work to make it better than it is and stop trying to lie to fans about Opera's popularity." This is just a pathetic attempt to change the topic because this troll can't answer the actual facts that are presented.
• 

  Khaled KhalilSunday, December 14, 2008 6:51:11 PM

  Stop being silly.

  Stop commenting on posts that you are not interested in.
  

  If Opera ASA is happy with the product's reach (users and usage) then they should be happy and tell their fans to be happy. If they are not happy with the product's reach, then they should make changes to improve its reach.

  good point, i appreciate your opinion
  anyone, if you are happy of your social life/position/wealth/etc. go kill yourself, there is nothing remaining to do in this life.
  if not, you are not very happy for what you are, go kill yourself for this worthless llife.
  very simple isn't it
  
  i am very sorry for this off-topic comment, but i had to tell someone something about why we are here, i am not celebrating nor working hard for opera, just discussing something about transparency and fidelity, while it seems to be an ethical or moral debat, it is actually a technical qualifier for this business.
  
  now according to someone, i may be have a better thing to do now