Yeah, that’s the final version. Writing the applet out is a last-ditch effort if all other detection methods fail.
Yeah, that’s a more accurate solution instead of the LiveConnect way I was doing it. I was just trying to do it without any external classes.
Yes make the website slow for everyone who doesn’t have java installed ;D
Stats from Spanish site, with very non-technically-inclined visitors
Major versions
hits share java_version
538 66% 1.6
158 19% Unknown
73 9% 1.5
25 3% 1.1
11 1% 1.4
2 0% 1.7
Minor versions:
hits share java_version
182 22% 1.6.0_11
160 19% 1.6.0
158 19% Unknown
85 10% 1.6.0_10
53 6% 1.6
38 4% 1.6.0_12
28 3% 1.5.0
25 3% 1.1
18 2% 1.5.0_16
13 1% 1.5
7 0% 1.6.0_03
7 0% 1.5.0_13
7 0% 1.6.0_02
6 0% 1.4.2
5 0% 1.4
5 0% 1.5.0_09
5 0% 1.6.0_07
2 0% 1.7.0
1 0% 1.5.0_10
1 0% 1.5.0_01
1 0% 1.6.0_06
[quote]158 19% Unknown
[/quote]
lot of users, any idea ? no java ??
It’s Spain, so 19% no Java is not a surprise for me.
Typical for USA is 15% or so. IIRC
arround 20% without java appear to be lot for me ! we are very far of the “everybody have java”
good to know about france : since a cupple of years now, you can pay “gouvernment charge” using internet and you must have java on your desktop to do it (at least it was using an Applets past years ), and you can pay later if you do it using internet rather then sending letter, I think this has probably diffused a lot the Java plugin in france as all workers must pay those charges.
I think the javascript isn’t perfect yet.
There’s no mention of 1.2.x or 1.3.x, and 1.1.8 (SUN JVM, not MSVM) also doesn’t appear in the stats.
I forget to say : no it wont because Applet wont start where there is no Java :P, it will only make slow on computer with Java working but for wich the JS cannot find it. But the most important is that it will give combos that are not detected OS/browser/Java version and we then will be able to patch the JS script.
What about mac visitors?
Cas 
I picked up a few 1.3 VM hits.
1.2 never worked with the tag AFAIK.
And for a 1.1 Sun VM, was that the one in Netscape? That would probably show up via the LiveConnect method.
I was able to get 1.2 working in netscape, not 1.1.8 tho
this thread needs an update !
http://www.statowl.com/java.php
nothing really new…
1.6 56.5%
1.5 17.5%
1.4 6%
1.3 0%
other or undetected : 20 % (1.1 / 1.2, IMO, are not detected)
we were just discussing java plugin 2 uptake over at #LWJGL (freenode.net), seems like the situation is pretty good now
http://www.statowl.com/java.php
Java 1.6 uptake is at 62.10%
Java 1.5 is at 15.18%
and Java 1.4 + all jre versions below that are at 4.59%
Java 1.6 has seen a 2% rise month on month
Now that Java also has a semi decent update mechanism, it looks like its happy days for java applet programming.
heres a second source to confirm the above http://riastats.com
nice to ear that
some precisions (in case of) :
java 1.6 is still around 54% (62% if only for client that have Java detected) EDIT : sorry … forget what I said I may need some rest…
[quote]and Java 1.4 + all jre versions below that are at 4.59%
[/quote]
this does not include version <=1.2 ( that can probably be find in several enterprise (except computer/graphics one) that usually keep computer for a while before changing them)
[quote]Now that Java also has a semi decent update mechanism, it looks like its happy days for java applet programming
[/quote]
I found it very nice, this times this a real improvment for Java good work Sun 
somethign interresting is to compare (on RIASTAT) win XP (45% 1.6), win Vista (77% 1.6) & win 7 (34% 1.6)
another really nice feature that’ll be coming soon to firefox and no doubt other browsers will follow is plugin check
http://dailyapps.net/2009/11/tool-mozilla-plugin-check/
so double nagging for users, the browser and by the java updater
again this’ll keep more users on the latest and greatest plugin
[quote]another really nice feature that’ll be coming soon to firefox and no doubt other browsers will follow is plugin check
[/quote]
Ah thats great.
I’ll see if I can get my Java stats from google stats, once I find the details again…
anything below 1.4 can safely be ignore as they probably total less then 1% and besides the plugin was super horrid on those anyway (well it was horrid until 1.6.0_10), better to consider those computers as not having java then to still try support them in some way.
Just had a look at the latest java stats (yes I was bored :))
However there is some good news this month, from the latest April 2010 stats the Java Plugin2 is now for the first time the dominant applet plugin and from here on will continue to rise.
Plugin2 is now used by 52% of all Java Installs.
The Java plugin is still installed on about 80% of all computers.
All versions of java below 1.6 are now dying fast and have combined market share of below 15% of all installed java installs. Java 1.5 is 12.69% of that 15%, so aiming for any java version below 1.5 now is pretty insignificant and those versions can probably be ignored.
The rate at which java was loosing market share has also seemed to slow massively compared to what it was last year, only dropped by 0.07% last month. Java was loosing about 0.5% a month late last year. This is probably attributed to the improved java client side experience and the much nicer update process java now has. Judging from the current trend I’m predicting that java will probably now start stabilising at about 80% market share and may even start to rise a little by the end of 2010.
first all the following is said really without any offense against you Kapta but…
hum… hum… I am far to aggree
Java 1.6 => 57%…
Java => 80%
this still make 23% difference … (even 10% diff would still be significant)
as far as I know it was already the case one year earlier ( nearly same stats last year), IMO : this means that half of computers are updating while another half do not, those stats seems to not change a lot with time
23% is NOT "insignificant ", your claim sounds a little demagocic to me