Opera has 3% market share… on botnets?

According to a report by Finjan, Opera is used on 3% of a recently discovered botnet consisting of 1.9 million malware-infected PCs.

While a "fun" fact in itself, this number actually confirms the global Opera usage figure reported by StatCounter. However, the botnet figures have a much higher usage share for IE, and lower for Firefox, than StatCounter reports.

Again, we should remain skeptical of browser statistics because of all the error sources that exist. The botnet figures are very interesting, though, because the methodology is obviously very different to that of StatCounter and other browser statistics companies.

It does indicate that Net Applications's <1% for Opera must be way off, if nothing else. And yet, journalists and bloggers repeat the numbers from Net Applications without batting an eye.

Advertisements

29 thoughts on “Opera has 3% market share… on botnets?

  1. hmmm interestingMy brother and mom get hit by malware and Opera is uneffected by it (I did set the router to OpenDNS) When I get back home I have some more to clean up on my brothers computer

  2. Asa Dotzler writes:

    Another way to read that statistic is to assume the botnet is the deviation from the norm rather than Net Applications and that Opera users are more likely than average to be hosting malware and Firefox users are less likely than average to be hosting malware. An even less charitable way to read that statistic is that Opera is more susceptible than average to malware attacks and Firefox is less susceptible. – A

  3. Nice spin, but it doesn't match reality. Furthermore, while several cases of malware targeting Firefox have been reported in the wild, there are few or none for Opera.Net Applications is definitely a deviation. I have documented that very clearly in the past:http://my.opera.com/haavard/blog/index.dml/tag/net%20applicationsWhere StatCounter at least seems to approach numbers that might make sense, Net Applications is simply completely unreliable.Just because a flawed source is referred to by many who accept its numbers without an iota of critical thinking doesn't make that source reliable.

  4. Asa Dotzler writes:

    haavard, your blog posts on Net Applications are just a series of assertions. I don't see anything compelling there at all. But no matter which global number you decide is most accurate, they all agree that Opera has been basically flat for several years while Firefox, Safari, and Chrome are all growing. Chrome has outpaced Opera by a factor of 2 or 3 and Firefox is outpacing chrome by a factor of 4 or 5. As I've said before, being flat in usage share doesn't mean a browser isn't gaining new users. But to grow share, one needs to gain new users faster than the overall growth of users on the Web and Opera doesn't seem to be doing that. There's nothing wrong with that either. Opera is a profitable company and that's what matters, right? I mean, Opera ASA exists to make money and they're doing that so they're satisfying their mandate. Firefox exists not to make money, but to open the Web and to ensure that there's healthy competition and innovation. It takes more than single digit market share to do that so Mozilla needs share in a way that Opera doesn't because Opera's motivation (like all publicly traded companies) is simply to generate profit.- A

  5. Asa Dotzler who own Chrome, Google does and what kind of name does Google have with people who surf the net? A person might associate Chrome witht the name Google

    Firefox exists not to make money, but to open the Web and to ensure that there's healthy competition and innovation.

    Has the ff team contacted sites and let them know not to write in web standards code and not IE only coe?Just today I saw a site with this on it:

    Best viewed withInternet Explorer 5.0 or higherand Netscape 6.0 or higher.

    Yes the site is still active, I did sent them an email about it."Net Applications admits to skewed statistics"Safari market share tops 7%, but tracking firm admits data is skewedhttp://www.thestandard.com/news/2008/12/02/safari-tops-7-global-browser-market-share-sort

  6. Asa Dotzler writes:

    Chas4, ALL Web usage data is skewed. The best we can hope for is to understand the sample and how it impacts skew. That doesn't mean we stop considering the data. It doesn't mean we ignore useful information. UNLESS we have a marketing reason to bury our head in the sand and pretend that things are too unknowable to bother.- A

  7. Asa, my blog posts about Net Applications contain actual facts, not mere assertions. On the other hand, you have chosen to "bury your head in the sand" and make assertions based on unreliable statistics from Net Applications, but ignoring statistics to the contrary from StatCounter. Why are your assertions more correct than mine?StatCounter shows that Opera is up to 3% from less than 2% 9 months ago worldwide. In Europe, Opera is up from 3.69% 9 months ago to 7.72%, more than a doubling of the market share. According to StatCounter, Opera is indeed gaining market share.When I mention StatCounter, it is not because I think it's very reliable as such. It's because it contradicts data from a provider I know for a fact is even less reliable. My point is that people are spreading the market share myth based on false marketing claims rather than reliable data.You have chosen to use data which has been shown to be faulty in order to spread myths about Opera's market share. The message from Mozilla is that "Opera is insignificant, don't bother with it", and this is being repeated throughout the Web.Word of mouth is a powerful way to do marketing, and Mozilla certainly knows how to leverage its user base in order to push the Mozilla marketing message.So coming from an organization like Mozilla, which has a dominant position in the "alternative" browser market, this could be highly damaging to Opera. The constant repetition of the market share myth from the Mozilla marketing machinery is very similar to the FUD campaigns Microsoft has been know to use to deal with competition.When you know that all the available data is unreliable, you don't pretend that it isn't and go with whatever you feel like. Especially not when the one you choose has been revealed as being even more unreliable than some other sources out there.How did you like it when the numbers from Secunia were taken out of context and used to "prove" that Firefox is the least secure browser on the market? I'm sure you don't like having that claim plastered all over the Web. It's damaging to Mozilla's reputation.And I'm sure you, too, realize, that the constant repetition of the market share myth could be damaging to Opera.

  8. In South Africa, all computers are sold with Microsoft software pre-loaded. The average consumer knows nothing about computers and simply uses what he's provided with. A required 'computer literacy' course in this country teaches Word, Excell, and Windows. Basically, it teacher you how to switch on a computer that is pre-loaded with Microsoft software. Knowledge of Linux or Mac based systems is not a marketable skill here and neither is knowledge of Open Office. .I saw a netbook style computer the other day in a computer shop that had a linux operating system. It was less than half the price of it's nearest competitor. I was told that it was cheaper only because it had less power. (smaller ram, small hard drive). The real reason it was running Linux though, is that it couldn't run Vista. It simply did not have the power to run a resource hogging hefalump like Vista.If all cars came from the dealer with Ford engines, would that make Ford peoples choice in motor vehicle engines? give people true choices and well see a dramatic difference in the market share.-just my opinion.

  9. @haavard these statistics are – of course – only windows based: that means ie will loose in market share in the total number of all internet users (and ff would win).@glue I installed again linux on my hp mini 1000. the problem: drivers! speakers sound don't work; logitech mouse driver doesn't exists for linux (only win and mac –> not enough easy options!)… there is to many problems with drivers to get an easy operating system for the everage joe. on windows: download/take the cd and install everything that is on and then restart: it works! o.O

  10. Asa Dotzler writes:

    Haavard, if I was picking the stats based on partisanship, I'd go with StatCounter which gives Firefox a much higher share than Net Applications. You are obviously happier with StatCounter because it gives Opera a couple more points than Net Applications does. If I was just looking for positive spin, I'd do what you did and adopt StatCounter as my favorite measure since it gives Firefox about 10 more points and almost a full third of the market. But Net Applications has a much longer track record, a larger sample size, fewer unexplainable spikes, and more geographic diversity, so I'm much more confident in their numbers.As for your claims that I'm using Net Applications' numbers trying to advance a myth to hurt Opera, that's just silly. StatCounter puts Opera at almost meaningless market share too. I could just as easily use their number to insult Opera's usage. Both Net Applications and StatCounter agree that Firefox has more than one order of magnitude more share than Opera. My point is that one does not have to be selective in choosing a metric provider to point out that Opera has a pretty minuscule share of the market, a best case that's less than 1/10 the share of Firefox and less than 1/20th the share of IE and at worst case 1/30th the share of Firefox and 1/60th the share of IE.Now, I'm a lot more interested in how Firefox compares to IE since that's what matters to me and probably 99% of people who consider these things regularly so I'm generally not focusing on the so called "myth" of Opera's insiginificance. Opera's seldom more than a quick mention when I post analysis of monthly usage trends. If you'd like, though, as I go forward talking about the browser landscape, I can just stop mentioning Opera at all. After all, both Net Applications and StatCounter agree that Opera market share is pretty insignificant. I'm guessing, though, that you'd rather I continue to talk about all of the alternatives to IE and all of the standards-pushing browsers because it reminds people that Opera exists and has something to offer. It also makes the "other" category (not IE, Firefox, or Safari) look a bit larger and that's probably good for something.- A

  11. Originally posted by Asa Dotzler:

    You are obviously happier with StatCounter because it gives Opera a couple more points than Net Applications does.

    Now now, Asa, you know that this is a straw man. I did explain that I didn't refer to StatCounter because I think it's a reliable source as such, but because it paints a different picture than Net Applications, which I have shown to be extremely unreliable.My point being that browser stats are inherently unreliable.The numbers reported by Net Applications are definitely completely off. Remember, they used to list Opera with a much higher market share on more than one occasion, but then they suddenly changed it overnight and placed Opera way below the numbers from the day before.

    But Net Applications has a much longer track record, a larger sample size, fewer unexplainable spikes, and more geographic diversity, so I'm much more confident in their numbers.

    Net Applications is almost exclusively focusing on the United States, has a well known track record of manipulating their own statistics, has reported numbers that are provably false, and so on. You yet to address any of the points I have brought up in my post about Net Applications.StatCounter actually reports on individual countries, unlike Net Applications.

    StatCounter puts Opera at almost meaningless market share too. I could just as easily use their number to insult Opera's usage.

    3% worldwide and nearly 8% in Europe is far from meaningless, but accepting StatCounter's data as gospel would be almost as wrong as the way you are accepting Net Applications as gospel.

    My point is that one does not have to be selective in choosing a metric provider to point out that Opera has a pretty minuscule share of the market, a best case that's less than 1/10 the share of Firefox and less than 1/20th the share of IE and at worst case 1/30th the share of Firefox and 1/60th the share of IE.

    What is "the market"? Is Eastern Europe not part of "the market"?Yes, your point is that "Opera is insignificant", and you keep pushing that message. Thanks for admitting that you consciously push exactly that marketing message. I guess it's a good way to keep smaller independent competitors without the same marketing reach down. If you repeat something often enough…

    I'm guessing, though, that you'd rather I continue to talk about all of the alternatives to IE and all of the standards-pushing browsers because it reminds people that Opera exists and has something to offer.

    I know from experience that Mozilla's marketing chooses to ignore Opera when you make use of our inventions, and only mentions Opera in an attempt to play it down as a competitor. I'd rather see some honesty and replies that are actually relevant rather than attempts to change the subject, which was the reliability of browser statistics.

  12. Asa Dotzler writes:

    Haavard, you don't like any of the numbers because none of them paint Opera in a very good light. You're so unhappy with the numbers that you're willing to cite a higher rate of malware infections as something positive!I don't need to play Opera down as a competitor because Opera isn't much of a competitor. At best, Opera is an ally. At worst, it's insignificant. But it's really not competitive with Firefox. In all honesty, it's not even playing on the same field.That doesn't mean Opera isn't a viable and profitable company. It does mean that Opera either doesn't want to or is unable to compete effectively with Firefox and IE though. I've always assumed the more charitable explanation, that Opera was focused on different things like mobile handsets and server-side solutions. If that's not the case, and you (Opera) really are giving it your best to compete with Firefox and IE, that's an entirely different discussion.

  13. Asa Dotzler writes:Chas4, ALL Web usage data is skewed. The best we can hope for is to understand the sample and how it impacts skew. That doesn't mean we stop considering the data. It doesn't mean we ignore useful information. UNLESS we have a marketing reason to bury our head in the sand and pretend that things are too unknowable to bother.- A

    You said it yourself "ALL Web usage data is skewed" then why use the data?

  14. Asa: I'm really having a hard time taking you seriously now. Not only have you ignored my points (once again you have failed to address the points I have made and tried to change the subjet), but you have started twisting my words.I am not reporting a higher rate of malware infections as something positive, I am reporting it as an observation.There are clearly good numbers out there for Opera, which you chose to ignore in order to continue the PR spin of "Opera is irrelevant" (you certainly know how to stay on message, a good PR stragegy, even when the message is flawed). You also chose to ignore the facts that have been presented, which show Net Applications to be highly unreliable (more so than, e.g., StatCounter).I realize that you are frustrated with Mozilla's failure to gain a foothold on mobile phones, but your PR strategy here leaves a foul taste in my mouth. If Mozilla did not consider Opera to be competition, it wouldn't be spending its time spreading FUD.The last paragraph in your latest comment contains an interesting contradiction compared to what you have been saying in the past. I actually quoted you in my blog when Mozilla decided that Opera had taken all the heat, and that it was now safe to join the complaint because the flames had subsided. But now you are being "charitable", based on browser statistics that you admit are skewed, and yet you keep blindly relying on them?

  15. Asa Dotzler writes:

    Haavard, you and I disagree about the relevance of Net Application's browser usage report. You're willing to throw away or discount any data that isn't perfect and I'm not. I worked in market research for years and was quite good at it. I know that there is no such thing as perfect data but that even imperfect data can provide a lot of valuable insight and understanding. That's because I know how this all works quite well. And you're just dead wrong about "blindly relying on them." I've met with Net Applications. I've learned a lot about their methodologies and I have confidence that their numbers closely reflect the growth of Firefox, more closely and consistently than any other global source — including the fledgling reports from StatCounter. (I also like the numbers from XiTi for Western Europe and Gemius for other parts of Europe.) If I was "blindly" picking stats, I'd certainly go with StatCounter which shows Firefox doing 10 points better globally. But I'm not after the biggest number. I'm after the number that most closely reflects reality over the longest period of time. Net Applications fits very well with the data I have on Mozilla users and usage and it trends closely without the wild spikes and dips that StatCounter has already shown in the short few months its been reporting publicly.Or, to put it another way, your arguments and assertions don't change my opinion of Net Applications and aren't likely to do so no matter how many times you repeat them.And for mobile, I personally could not care any less about mobile. Others in my organization obviously disagree with me but I think that Opera's record speaks perfectly to why I don't care about success there. Opera's got fantastic share on mobile devices but it isn't showing up in Web usage because most people don't spend a lot of time browsing the web with phones. Even if Opera managed to get 100% of all phone browsers, that wouldn't account for a whole lot because most people don't browse much on phones. Today less than 2% of all surfing happens on phones. That's up from about 1% several years ago. It's a tiny pie and it's growing very slowly. Opera or Apple or Microsoft can claim all the success there that they want to but it amounts to an itsy bitsy tiny little fraction of overall Web usage. So I'm personally not the least bit frustrated with where Mozilla is on Mobile because Mobile just isn't interesting to me. I care about the Desktop where the real action is — where people actually spend real time online. I care about where 98% of the Web browsing happens and I'm pretty happy with where Mozilla and Firefox are with that group of people.Perhaps Opera is happy with where it is too. Good for you all. You've been shipping a desktop browser for more than 10 years and you're sitting somewhere between zero and three percent of Web usage — including your amazingly successful mobile effort. If that's good times for you all, great. It wasn't good enough for me when Mozilla was at one or two percent and that's why I and a handful of other people put so much time an energy into making Firefox and making it a success.

  16. Asa Dotzler writes:

    Chas4, sure it's growing, but mobile usage is still absolutely tiny compared to desktop browsing which is also growing and at a pretty decent clip.Just compare the Opera Mini pageview stats that Opera releases regularly with the rest of the Web and you'll see how tiny it really is.

  17. Originally posted by Asa Dotzler:

    Or, to put it another way, your arguments and assertions don't change my opinion of Net Applications and aren't likely to do so no matter how many times you repeat them.

    In other words, facts don't change your opinion. Yes, I have come to realize that. Instead of actually addressing any arguments and facts you have been presented with, you simply stay on message.

    You've been shipping a desktop browser for more than 10 years and you're sitting somewhere between zero and three percent of Web usage — including your amazingly successful mobile effort.

    Opera has been around for about 15 years, but the desktop version didn't become free of charge until late 2005, three and a half years ago. As you know by now, all of your numbers are speculation, but you sure know how to stay on message.Sadly, you actually end up contradicting yourself. Which is what I guess happens when the PR message is more important than reality. But when your PR message is damaging to the Web, you should realize that something is wrong, and when you are forced to twist people's words in order to win an argument, you know that you have already lost.I have no doubt that you will continue to rely on Net Applications despite all the facts you have been presented with.

    I've met with Net Applications. I've learned a lot about their methodologies and I have confidence that their numbers closely reflect the growth of Firefox, more closely and consistently than any other global source — including the fledgling reports from StatCounter.

    Ah, so that's the reason why you will contine to refer to Net Applications despite their shady history. Apparently you have some sort of business relationship with them? I can see why you couldn't switch to StatCouter even if you wanted to. Contracts are contracts.

  18. Originally posted by Asa Dotzler:

    Firefox exists not to make money, but to open the Web and to ensure that there's healthy competition and innovation.

    Originally posted by Asa Dotzler:

    I don't need to play Opera down as a competitor because Opera isn't much of a competitor. At best, Opera is an ally. At worst, it's insignificant. But it's really not competitive with Firefox. In all honesty, it's not even playing on the same field.

    :confused:emmm…Originally posted by haavard:

    Apparently you have some sort of business relationship with them? I can see why you couldn't switch to StatCouter even if you wanted to. Contracts are contracts.

    i see discussion have gone to some very low level, can we try being more civilized as you used to be gentlemen ?objectivity doesn't harm too

  19. Asa Dotzler you say that mobile web usage is tiny compared to the desktop, then why is Mozilla working on a mobile web browser?

  20. Someone needs to visit Africa and India. Consider this, in India, 800Million are registered voters. The most common way of surfing the net there is via mobile phone. The mobile phones of today are merely computers in a handheld case. :whistle:.It's also interesting to note that it's virtually impossible to discuss website development without mentioning the Opera browser. Every book I pick up on the subject mentions Opera. And often it must mention those things that only work in Opera.

  21. Asa Dotzler writes:

    Haavard said "Ah, so that's the reason why you will contine to refer to Net Applications despite their shady history. Apparently you have some sort of business relationship with them? I can see why you couldn't switch to StatCouter even if you wanted to. Contracts are contracts."Actually, we don't. We just like to meet and talk with people. You're the ones that put all of your pages under Net Applications trying to boost your numbers and when that didn't work you put StatCounter tracking on all of my.opera to boost your numbers there. We don't use either Net Applications or StatCounter for measuring Mozilla website traffic. We don't have a business relationship with either of them. You're the one that's had the relationship with both.

  22. What makes you think any Opera sites are using StatCounter?Net Applications is only used to provide statistics to blog owners for their own blog. The other of this site and other Opera sites aren't using it, as far as I can tell.Keep expanding your business relationship with Net Applications (which you call "talking to people"), though. You seem to be a good fit, considering how Mozilla is doing business and marketing.The way you have constantly avoided answering people's points through this discussion speaks for itself.

  23. Asa Dotzler writes:

    Oops, my mistake. I thought I read something over at OperaWatch comments saying that Opera was moving to a different provider and with all your regular bashing of Net Applications, that seemed reasonable to me. So why do you use Net Applications on my.opera.com? You could be using StatCounter, Google Analytics, or some for-pay service? Ominiture seems to be a pretty good service. If NA is such shit, why do you feed it to your community?

  24. I don't know why we are using Net Applications for user blogs. Maybe they were the only ones who offered something like that. If you look at the code, our sites seem to be using Google Analytics for general traffic analysis.

  25. I keep wondering why everyone in IT industry, especially browser vendors fall victim of a marketing game by net applications.It is a company who sells specialized hardware for web servers as far as I understand and their statistics have no meaning at all. It is more like IBM coming up with ''Here are the stats of people who goes to AIX powered web servers''Stat counter etc. are same deal too. Untrustable data which is only provided by companies who chooses to opt-in and only the people who visits/uses the sites.I don't say Opera has a huge market share or Firefox is less used. I just ask why freely advertise a company over their questionable stats? IMHO less technological but very scientific essays are way to go in stats business. Also thing to support shouldn't be Opera, Safari (Webkit) or Firefox. It should be only the web standards and nothing else. Brand shouldn't matter. A site saying ''Install Firefox to view this site'' is equally disturbing as ''Install IE'', at least to me on this very specific machine which Opera is indisputable browser.

  26. Firefox exists not to make money, …

    this words were true in the beginning, but The Times They Are A-Changin'

Comments are closed.