Slashdot Mirror
Firefox Breaks 50,000,000 Barrier
MrDrBob writes "Today at 16:59 GMT (8:58 AM PST) Mozilla Firefox received its 50,000,000th download. To celebrate, SpreadFirefox.com has created a special page, where you can watch the downloads continue to climb in real time. Three cheers for Firefox! May it go on swiftly to 100,000,000!"
AFAIK FF updates itself now... At least on Windowz.... FC gets its updates from YUM / up2date, so that shouldn't register either.
In fact, many distros include FF so that doesn't even count on the download numbers.
09 F9 11 02 9D 74 E3 5B D8 41 56 C5 63 56 88 C0
Probably no where near the number of users that have it from their FC2 and FC3 disks or by apt-get install firefox or emerge firefox. I'd guess it is close to even.
Paying taxes to buy civilization is like paying a hooker to buy love.
I just saw the counter go down. Cool, they're even counting "returns".
Actually, the javascript only updates once a minute, and when it does, it computes a new rate, so the increments you see are only estimates until the next update.
bp
We're not counting downloads served by the Firefox update system. Neither are we counting the millions of downloads from download.com (they actually host Firefox as well) or from other download repositories. We're also not counting people who go directly to FTP without going through our "bouncer" tool (the app that directs you to the most appropriate mirror). There's lots we're not counting here. It's not meant to be a count of users. It's just a measure of how many people use our system to download Firefox.
- A
Shouldn't we send the person who did that download some flowers or balloons or something?
Actually, we're rewarding the members of the Firefox community who helped that guy find Firefox. The big prize will go to the Spread Firefox affiliate who delivered the click that coincided with the 50 millionth download. http://www.spreadfirefox.com/fifty.html
- A
Seeing as I am downloading it directly from portage, this will not be counted by their download counter. I guess that many of the other posters have forgotten this and have probably underestimated the number of unique users by excluding most Linux users.
I'll probably be modded down for this...
And it can't be a coincidence that the page doesn't display properly in Internet Explorer!!
Look here. This Page Is Valid XHTML 1.0 Strict! No conspiracy theories around here. Valid HTML is difficult to get looking right in IE.
If your download went through the URLs listed in these links, and you're not talking about using Firefox's built-in update service, then your download was probably counted. http://www.mozilla.org/products/firefox/all.html
If your download did not go through one of those links, then you aren't counted.
- A
If you look at the source, you'll see that it is actually pulling a real number from an RSS feed.
It then defaults to 2 downloads a second, incrementing the timer by that rate every second. When it grabs the feed again a minute later, it then uses the "real" rate of downloads from the first grab to the second grab and starts incrementing by that amount.
It then continues to do that for as long as the page is up.
So the number is real-ish.
You are in a maze of twisty little relative jumps, all alike.
I should add a minor correction, that page lists the current version, so if you got 1.0, 1.0.1, or 1.0.2 from similar links, you'll also be counted. All those links go through our mirror tool that distributes load across our dozens of mirrors. We're not counting downloads that don't go through that tool.
- A
the counter isn't exactly "real".
:
// The initial rate, in downloads/second.
Take a look at the code and you will see that it gets a seed from the server and calculates a rate:
if (last_time && time - last_time != 0) {
download_rate = calculate_rate(time, count);
tick_time = Math.round(1000.0 / download_rate);
} else {
download_rate = initial_rate;
tick_time = Math.round(1000.0 / download_rate);
}
It uses the rate so that the browser is somewhat accurate, but not truly.
Also of note, the default rate is 2.0/second
var initial_rate = 2.0;
I only came here to do two things; kick some ass, and drink some beer...looks like we're almost out of beer.
We're not celebrating the person who made the download. We're celebrating the community and the specific community member who helped deliver that new user to Firefox. See the affiliate program at SpreadFirefox.com http://www.spreadfirefox.com/
- A
Does anyone actually know how many visits slashdot gives a site that is on posted on the front page? any guesses?
;)
The guy who did the Christmas and Haloween lights prank did a rather nice analysis on the incoming bandwidth from Slashdot and other media sites.
Pasted below is Slashdot's statistics:
5 min: 781
10 min: 1,604
1 hour: 11,699
2 hours: 21,651
4 hours: 35,895
8 hours: 53,720
24 hours: 90,607
2 days: 94,830
week: 98,054
month: 117,210
Take it with a grain of salt though...the analysis might be another hoax
I'm a virgo and on Slashdot. Coincidence? Yes.
Why do so many Gentoo users get this confused.
Gentoo does not host the packages that you download. They host the ebuilds and the ebuilds tell portage where to download the packages from.
So, you downloaded it directly from mozilla. Check the ebuild yourself if you don't believe me.
nil
This is what OneStat has to say.
OK, I checked and you are right. From the ebuild:
/ fi refox/releases/${MY_PV}/source/firefox-${MY_PV}-so urce.tar.bz2"
SRC_URI="http://ftp.mozilla.org/pub/mozilla.org
Sorry! But I bet some distros host their own copies?
I'll probably be modded down for this...
According to the website, Jon had to turn back. He was swimming with his PR manager who was in a raft, but after the raft sprung a leak Jon had to rescue him, sadly ending his Atlantic crossing in the meantime.
Quite a funny writeup, my favourite:
"As much as I don't want to talk behind a colleague's back, there is no doubt that we would never have let Eskil assist Jon in the raft had we known he can neither swim nor read maps," says an embarrassed Tor Odland, Opera's Communications Director.
I've read that they don't count updates. So if you click the little red arrow in firefox to update, it doesn't increase that number.
1. Microsoft IE 86.63 %
2. Mozilla Firefox 8.69 %
3. Apple Safari 1.26 %
4. Netscape 1.08 %
5. Opera 1.03 %
Am I wrong to be surprised that straight Mozilla (not the FireFox edition) doesn't even register, while Opera does?
--
make install -not war
It also says right there on the page that updates are not included.
First, they have stressed several times on the spreadfirefox website that they are counting only downloads. They aren't trying to pretend that these are usage statistics, they are simply counting the number of times the application has been downloaded. From what I remember of calculus, if you downloaded it 20 times in a row on one computer, that's 20 downloads.
Second, they have also stated that downloads initiated through the update widget at the top right corner don't count towards the stats. Your upgrades to 1.0.2 and 1.0.3 only counted if you downloaded them from the main page.
Computers need to explode more often.
Indeed, they do. Most of the Debian-based distros use a Debian build hosted by either the Debian Foundation or the distro's supporting company (for example, Ubuntu's version of the .deb and the Ubuntu Backports project's .deb for Firefox are both hosted on their own respective servers).
Also, if you want an RPM build, you've got to get that yourself, unless things have changed since the last time I ran an RPM-based distro. I'm not sure about Autopackage's hosting, but it's probably on Mozilla's servers.
Gentoo is the only distribution that I can think of off the top of my head with its own independent (not tar.gz/tar.bz/tar.bz2) package management system that takes from the project's servers.
Haec merda tauri est. Ceterum censeo Carthaginem esse delendam.
The counter syncs with the "actual number" every 60 seconds, which means it will never be more than a couple hundred downloads off at any given time, and in fact it's usually very accurate once it's been running for a minute. Surely you don't expect us to poll our servers every tenth of a second, right?
Given the inherent margin of error with the "ACTUAL actual number," and the fact that we're dealing with a number as high as 50 million downloads and a delta as low as about 200, our consciences are quite clean labelling this the "actual number." We average around 240 downloads/minute.
Somehow I don't think we'd have linked to the page that describes exactly how the counter is implemented in the footer of the letter if we were trying to hide that information.