Mozilla Browsers Gains on IE
Saturday July 10th, 2004
Anonymous writes in with a link to a story on PC World.com talking about how the recent IE security vulnerabilities have started to impact current browser marketshare. It goes on to state that Mozilla/Netscape browsers share of the market according to WebSideStory has gone up 26% in the last month, gaining almost an entire percentage point overall, with IE losing 1 percent of its share. It attributes this to people trying out Mozilla based browsers because of concerns for security, and points out, "Once people start examining the features of Mozilla versus Internet Explorer instead of looking at a brand name... I think they'll see there's a lot more value."
The w3schools stats page shows a lot of promise for Mozilla browsers and Opera both showing decent growth: <http://www.w3schools.com/…owsers/browsers_stats.asp>
Where do w3schools get their statistics from, anyway?
#13 Re: Re: Compulsory w3schools link
Sunday July 11th, 2004 3:20 AM
(The statistics above are extracted from W3Schools' log-files, but we are also monitoring other sources around the Internet to assure the quality of these figures)
And they do have a bit saying that all statistics are unreliable
If their web browser use statistics are derived from the people who visit their site very likely do not represent "John/Mary Doe" web surfer. The w3shools is a site for persons wanting to learn or improve their web site development skills. This is likely a group who would be more likely to be aware of alternatives to IE, and most important, skillful and willing enough to install an laternate to IE.
Of the alternates Mozilla leads the pack--likely because it is free (not ad supported, or pay) rather than that better. If Opera was totally free without the ads, would Mozilla still ldead the pack with such a large lead?
I agree with the first paragraph and disagree with the second.
The first paragraph is both true, and a very good thing -- it means that web developers (well, at least those that care somewhat about standards) are using Mozilla (and Opera)....which is not that surprising....
About the second -- I don't know if that's entirely true, though it's definitely a factor in the moving-away-from-IE-because-CERT-said-so group. I use Mozilla over Opera not so much because of the features it has that are better than Opera, or because the Opera UI is horrible, but because I was weaned on Netscape, and from Netscape 6.whatever to the Suite was a natural step (and later led to Fx, or Fb at the time). Had I used Opera before, say, Netscape 6, I would probably still be there...
#8 Who knows...
by wvh <email@example.com>
Saturday July 10th, 2004 9:25 PM
That's not a relevant question, really. At least, any answer would be pretty much a random guess. It isn't free. If it would have been free, if Opera would have been Mozilla, this might have been the operazine forum. Maybe there wouldn't have been any post-Netscape releases anymore, in that case. But Opera isn't Mozilla; it isn't opensource, and it isn't free.
I prefer Mozilla (through Galeon) because it intergrates better with my -mostly- Gtk desktop, and because it is opensource (in which I believe, philosophically). If both would have been opensource, my decision would be based on an undoubtedly totally irrational fringe feature. They are both very good browsers, and purely on technical merits (such as rendering and security), I wouldn't dare to pick one over the other. I would actually advise people (on Windows, non-opensource non-technical regular computer users) to try out both. Whatever, as long as they dump IE. :)
Mozilla has a much better (a *real*) mailclient though. Somehow that improves Mozilla-The-Browser by association - only one store to shop.
looks like mozilla's friend Apache is also doing well <http://news.netcraft.com/…es/web_server_survey.html> And that's important because if Microsoft won the server war, alternative browsers such as Mozilla and opera would be shut out that way
#7 This repost is merely a snapshot
Saturday July 10th, 2004 9:14 PM
.... of what's benn going on for at least 6 months. Read the Press Box section of onestat.com ( <http://www.onestat.com/html/aboutus_pressbox.html> ) and you'll see that IE has been losing market shareto Mozilla at least since the beginning of the year. With the recent security announcements and FireFox 1.0 on the horizon, I don't doubt that IE usage will drop below 90% by the end of the year.
#17 Re: This repost is merely a snapshot
Sunday July 11th, 2004 11:26 AM
TheCounter.com shows the same trend:
Mozilla: May 2.45% June 2.61% July 2.91% (up to July 11)
IE: May 94.98% June 94.54% July 94.13% (up to July 11)
Just when it started looking like Mozilla usage stopped doubling every year, Mozilla usage has nearly made up for several months of slow growth. Here's to another year of Mozilla usage doubling!
(please cancel this comment on edit)
Not surprised there. I've been using FireFox/Mozilla and Thunderbird for a long time now and they are both better than IE. I personally wouldn't use IE unless I had to because of the security bugs in IE.
#11 Does CERT list "many" Security Holes in Mozilla
Sunday July 11th, 2004 1:26 AM
A coleague of mine said that there are as many security holes in Mozilla as in IE, just that they are not publicized. He stated as proof that CERT lists many Mozilla secirity issues.
I *briefly* looked at a search for "mozilla" in CERT, and the two issues i looked at were resolved. It might be useful to analyze all CERT issues and formulate some effective "standard response" to such (valid?) criticism.
He also claimed that mozilla holes are little known because mozilla.org keeps them "secret" (security by obscurity isn't good, is it?). An effective one-liner response would be helpful.
#14 Re: Does CERT list "many" Security Holes in Mozill
Sunday July 11th, 2004 5:23 AM
"As many" isn't right (not sure about CERT, but on other security sites, IE has at least double the number), but Mozilla security issues do exist.
Mozilla keeps issues secret until they are publicised or exploited (i.e. people know about them anyway), or until they are fixed in Mozilla releases and releases based on Mozilla by other distributors. But Microsoft doesn't publicise issues until they are exploited and/or fixed either.
So, it's true that Microsoft flaws are more likely to be found/exploited, because IE is more widely used. But I don't think there's any proof that the number and severity of known flaws would the same, even if the level of attention was the same.
I'm not sure a "standard reponse" is needed - people can formulate short responses based on the point that's made...
#16 Re: Does CERT list "many" Security Holes in Mozill
Sunday July 11th, 2004 9:30 AM
Security by obscurity is a perfectly reasonable second line of defense against attacks. The problem with obscurity is when you use that as the first line of defense. An example would be a cryptographic alogrithm that is "secure" because the algorithm is kept secret. Such cryptographic systems have been easily defeated. Think DeCSS.
But if you have a security hole in your program due to a bug, you by all means want to keep it as secret as possible until nearly all systems are patched with a fix. In this case, fixing the problem is the first line of defense, and the obscurity is a second line of defense. When the security flaw does exist and is installed in millions of computers worldwide, what else can you do until a fix is available except keep it a secret?
By the way, if Mozilla holes are secret, how does *he* know about them? ;-)
#20 Re: Re: Does CERT list "many" Security Holes in Mo
Sunday July 11th, 2004 6:28 PM
one of the problems with security by obscurity is that it creates the forbidden fruit temptation. There is more of a challenge to find something in closed software. Then there's the whole anarchist mentality against the big 'evil' companies (my personal view is that Microsoft is not actually 'evil' which why I pretty much yawned at the whole DOJ thing. I view it as a stumbling block to innovation. while granting that they are capable of innovating when they are pushed hard enough, they end up stagnating after winning. It's a big challenge for us and the only way to counter that is to produce a unambiguosly technically superior product)
#21 Re: Re: Re: Does CERT list "many" Security Holes i
Sunday July 11th, 2004 9:51 PM
> one of the problems with security by obscurity is that it creates the forbidden fruit temptation.
But certainly it's harder to exploit a security hole if that hole is kept secret, right? Of course, in the meantime you need to be actually plugging up that hole as fast as possible.
#12 Lets fix some other bugs first
by dgtlmoon <firstname.lastname@example.org>
Sunday July 11th, 2004 2:32 AM
Why dont we focus on fixing some serious bugs (that have been open for two years) thats stopping mozilla from being taken seriously in the office? <http://bugzilla.mozilla.org/show_bug.cgi?id=154892> .
Ive just removed mozilla from my office after instructing a collegue to roll it out (40 desktop installations) because of some severe printing issues that appear to be getting little if no attention.
Where do we go from here?
I can't agree more. There are a few of us who actually print out stuff from the web, and I've been hit several times by that bug.
Brendan Eich said that mozilla.org is considering bringing back the Netscape Bug Bounty program which pays people for finding security bugs
I'm waiting for the June Google Zeitgeist to come out to see if Mozilla has finally risen above MSIE 5.5. <http://www.google.com/press/zeitgeist.html>
The counter/stat-tracker sites - OneStat, thecounter, etc. - generally bear little relation to real browser usage. I don't know why their numbers are so biased (presumably because counters like that are only used on amateur sites) but they consistently appear to be wrong.
Of course, that doesn't mean it's not valid to look at trends in those numbers - i.e. 'Mozilla usage is increasing slightly' - which seems to be this story. But if you want some absolute percentage numbers then Google is a better bet: you know what you're getting. (I'm not saying it is a perfect represenation of the entire Web, but you know what it is: it's people who use Google. Which is a lot of people across a lot of groups.)
By the way it would indeed be interesting if Gecko does rise above IE5.5, but that'd be more due to the meteoric decline of the old IE versions than the success of Gecko :>
The really interesting thing is that the WebSideStory and OneStat statistics have always shown IE with marketshare at or above Windows. IE would need nearly 100% marketshare on both Windows and the Mac in order to make the WSS/OS stats valid. In reality, it appears that IE had just under 90% share at its peak, with current stats down around 85% or below.