Java Penetration Statistics

[DzzD]

http://www.tekool.net/blog/2009/01/23/java-jre-version-penetration-stats-with-google-analytics/

1.4.2+: 700 (69.86%)
1.5+: 661 (65.97%)
1.6+: 568 (56.67%)

this would mean that 30% user use under 1.4.2 ?!

it seems when statistic come from non java website

http://forums.java.net/jive/thread.jspa?messageID=317425

EDIT : the 14th post say 67% under 1.4.2 , that's a lot

[zammbi]

1.6.0_00 to 1.6.0_09: 530 (52.89%)
none: 300 (29.94%)
1.5.0: 93 (9.28%)
1.6.0_10: 38 (3.79%)
1.4.2: 39 (3.89%)
1.8.0: 2 (0.2%)

Java 1.8 
What happened to 1.7 heh...

[bienator]

I was also wondering about Java 1.8. It either a bug in the deployment toolkit or something mac specific. I got 86 visitors with java 1.8 in the last 30 days - all mac users.

Games published by our own members! Check 'em out!

[SimonH]

none: 300 (29.94%)

That '30% have no java' seems a bit high to me - eg for 5359 visits I got 4683 have java (of some sort!) 676 don't have java. That's 14% - still way too high though!

[zammbi]

Well this is only one source with a sample size of 1002.

A actuate conclusion would need to be taken by multiple sites and larger sample size.

[Eli Delventhal]

I was also wondering about Java 1.8. It either a bug in the deployment toolkit or something mac specific. I got 86 visitors with java 1.8 in the last 30 days - all mac users.

Maybe Apple felt so bad about having us permanently behind with 1.6 that they made a time machine, went into the future, and then ported 1.8.

[DzzD]

That '30% have no java' seems a bit high to me - eg for 5359 visits I got 4683 have java (of some sort!) 676 don't have java. That's 14% - still way too high though!

no,  this does not mean they have not java, that means they have a java version under 1.4.2, also getting statistic from non-Java related website give better result, for example I got only 7% but my own website is very java specific wich make statistic false.

[brackeen]

More stats:
http://riastats.com/
It also uses Sun's Java detection script. The fact that 1.8 is showing up makes means the detection script is not accurate.

[bienator]

i already sent them a mail yesterday

[DzzD]

More stats:
http://riastats.com/
It also uses Sun's Java detection script. The fact that 1.8 is showing up makes means the detection script is not accurate.

thanks, great information, that confirm that around 25% user have still a java under 1.4 or no java.

Games published by our own members! Check 'em out!

[brackeen]

thanks, great information, that confirm that around 25% user have still a java under 1.4 or no java.

I wouldn't take that site's data at literal truth - The detection script obviously has problems.  Plus, it includes mobile browsers, game console browsers, etc.

[DzzD]

ok, but other non-java statistic as the first I have posted show approximatvly the same result, this may be not exact but I think that it is near to the reality.

[brackeen]

Well, they show the same result because they use the same java detection script. Although yes they're probably not far from reality.

For desktop-only, Adobe shows 19% desktops don't have Java, which is probably the most accurate:
http://www.adobe.com/products/player_census/flashplayer/

Although, the people who play web games is probably closer to 5-10% w/o Java.

[DzzD]

I have done some test with deployJava.js on Window XP SP3

IE6 => detected
FF2 => detected
Safari => not detected

I also have MS JVM 1.1 on IE6 and even with sun plugin disable it was not detected, like no-java even if browser run Applets fine and even if there is something in the script to detect MS JVM.

[bienator]

how does it compare to the javafx deployment toolkit?
has safari Java Script enabled?

[trembovetski]

FF2 on MacOS:
  Java Deployment toolkit: 1.5.0
  JavaFX deployment toolkit: 1.5.0_16
  mime types: Java: undefined

Safari on MacOS
  Java Deployment toolkit: 1.8.0 (wooow!)
  JavaFX deployment toolkit: 1.5.0_16
  mimetypes: Java: 1.5.0_16

=> JavaFX rulez! 

[Matzon]

IE8 on Windows - with MSVM installed:
  Java Deployment toolkit: 1.5.0_17, 1.6.0_02, 1.6.0_14
  JavaFX deployment toolkit: 0 - unknown
  mime types: Java: undefined

So none of them got it right - depressing.

If it wasn't because of the fact that we get a f* icon and a baloon every time the launch the VM (who on earth ever decided that, moronic) - it'd be best to launch an VM using the applet tag and just query the VM.

[bienator]

Ubuntu FF3 6u14b02
deployment plugin not available -> plugin2 not detected

Java Deployment toolkit: Java: 1.6.0
JavaFX deployment toolkit: 1.6.0_14-ea
mime types: Java: 1.6.0_10

Windows XP FF3:
deployment plugin available -> plugin2 detected

Java Deployment toolkit: Java: 1.6.0_14
JavaFX deployment toolkit: 1.6.0_14-ea
mime types: Java: 1.6.0_14

Windows XP IE6:
deployment plugin available -> plugin2 detected

Java Deployment toolkit: Java: 1.6.0_14
JavaFX deployment toolkit: 1.6.0
mime types: <empty>


Windows XP Google Chrome:
deployment plugin not available -> plugin2 not detected

Java Deployment toolkit: Java: 1.6.0
JavaFX deployment toolkit: 1.6.0_14
mime types: 1.6.0_14

If it wasn't because of the fact that we get a f* icon and a baloon every time the launch the VM (who on earth ever decided that, moronic) - it'd be best to launch an VM using the applet tag and just query the VM.

this is what the jfx dt actually should do as fallback mode. it even uses cookies to do it only once. I helped to fix a bug in the jfx dt but i don't know if they are already integrated in the public script.

Matzon and everyone with a not working javafx dt, could you check out this page (it uses a different version of the script):
http://jim.campanell.com/sun/JavaVersionDetector.html

@java script experts, is it possible to use two versions of the same script on the same page without conflicts?

[bienator]

installed now opera and safari on my xp box too

WinXP Opera 9.64 6u14b02
deployment plugin not available -> plugin2 not detected

Java Deployment toolkit: Java: none
JavaFX deployment toolkit: 1.6.0_14-ea
mime types: <empty>

Safari 4 Beta (528)
deployment plugin not available -> plugin2 not detected

Java Deployment toolkit: Java: none
JavaFX deployment toolkit: 0 - unknown
mime types: Java: undefined
(Safari -> FAIL, but applets work...)

[Matzon]

Matzon and everyone with a not working javafx dt, could you check out this page (it uses a different version of the script):
http://jim.campanell.com/sun/JavaVersionDetector.html

Navigator contents:
appCodeName = "Mozilla"
appName = "Microsoft Internet Explorer"
appMinorVersion = "0"
cpuClass = "x86"
platform = "Win32"
plugins = ""
opsProfile = "null"
userProfile = "null"
systemLanguage = "en-us"
userLanguage = "da"
appVersion = "4.0 (compatible; MSIE 7.0; Windows NT 5.1; .NET CLR 2.0.50727; .NET CLR 3.0.04506.648; .NET CLR 3.5.21022)"
userAgent = "Mozilla/4.0 (compatible; MSIE 7.0; Windows NT 5.1; .NET CLR 2.0.50727; .NET CLR 3.0.04506.648; .NET CLR 3.5.21022)"
onLine = "true"
cookieEnabled = "true"
mimeTypes = ""


Browser is MSIE

Java Version is 1.6.0

OS is Windows

Mac OS Version is Not Mac

Can we handle version info? Yes

This is not correct, in that it IS Internet Explorer - but the applet tag is handled with: Microsoft (R) VM for Java, 5.0 Release 5.0.0.3810.

@javascript experts, is it possible to use two versions of the same script on the same page without conflicts?

not directly, you will get method and var collisions - you will have to rename all the methods and vars.

[brackeen]

Yet another detection script:
http://www.brackeen.com/detectJava/
- Works on Safari+Win (at least on my machine
- Doesn't require external scripts

[Matzon]

Yet another detection script:
http://www.brackeen.com/detectJava/
- Works on Safari+Win (at least on my machine
- Doesn't require external scripts

I see the test for MSVM, but it still finds the java 1.6 one :/

Java: 1.6
Complete string found: 1.6.0_12
isPlugin2: true
Detected from: JavaPlugin

it didn't find the oCaps stuff ... (never heard of it before!!)
However, once I added;

<BODY BGCOLOR="#FFFFFF" STYLE="behavior:url(#default#clientCaps)" ID="oClientCaps">

it worked!!

[Matzon]

however, it will find java2 plugin first  - but if I comment that out, I get:

Java: 1.1
Complete string found: 1.1
isPlugin2: false
Detected from: MSVM

[brackeen]

Yeah that script will try to find the highest version installed. So theoretically it will find the MSVM if no Sun Java is installed

I dunno about the oClientCaps stuff... I just copied it from somewhere. Got any solution that doesn't require setting the body id?

[Matzon]

Yeah that script will try to find the highest version installed. So theoretically it will find the MSVM if no Sun Java is installed

I dunno about the oClientCaps stuff... I just copied it from somewhere. Got any solution that doesn't require setting the body id?

this seems to work:
<IE:clientCaps ID="oClientCaps" STYLE="behavior:url(#default#clientCaps)"/>

and is ignored by FF and others (opera, chrome) - all windows (all I have atm)

http://msdn.microsoft.com/en-us/library/ms531416(VS.85).aspx

[brackeen]

Awesome. Updated the script.

[bienator]

Awesome. Updated the script.

stopped working for me on safari but detects java on all other browsers i tested so far. great work!

[brackeen]

Safari's working again. Good ol' try/catch saves the day.

[DzzD]

maybe it would be cool to post an how to or something like that just a small html sample of the best working solution

[brackeen]

I dunno if this is the best working solution, but I'm trying it with Google Analytics now. It will take a day or so for the stats to show up.  Here's what I'm doing:
1. Get a copy of this script on your site:
http://www.brackeen.com/detectJava/detectJava.js
2. After your GA script do something like this:

1   2   3   4   5   6   7   8   9   10

<script src="http://mysite/js/detectJava.js" type="text/javascript"></script> <script type="text/javascript"> if (typeof pageTracker != "undefined") {     var version = detectJava.getVersion();     if (version == "1.6" && detectJava.isPlugin2()) {         version = "1.6uN";     }     pageTracker._trackEvent("java", "version", version); } </script>

It will be interesting to compare it to some of my server-side logs if there is an discrepancy. I'll re-post when I have info.