This is my experiments to see where http updates are not seen on the client side.  
See bug http://sourceforge.net/p/autoplot/bugs/1393/

Note alwaysChanging.jy is running as a cron job in my crontab file:

* * * * * java -cp  /usr/local/bin/autoplot -Djava.awt.headless=true org.virbo.autoplot.AutoplotUI --script=/home/jbf/public_html/autoplot/changing/alwaysChanging.jy

= Results =

== Normal HTTP ==
When I point Autoplot to the file 
file:///home/jbf/public_html/autoplot/changing/temp.htm?column=1&table=0&filePollUpdates=60
it updates as expected.

When I point Autoplot to the URL
http://www-pw.physics.uiowa.edu/~jbf/autoplot/changing/temp.htm?column=1&table=0&filePollUpdates=60

Here are the headers:
HTTP/1.1 200 OK
Date: Fri, 15 May 2015 17:16:05 GMT
Server: Apache/2.0.63
Last-Modified: Fri, 15 May 2015 17:16:03 GMT
ETag: "51dc74f-20f-ffafdac0"
Accept-Ranges: bytes
Content-Length: 527
Keep-Alive: timeout=15, max=100
Connection: Keep-Alive
Content-Type: text/html

It WORKS over http as well.  Shoot.  Now to see what SVN does differently.

== Dan's SVN ==

Here's the URL: http://emfisis.physics.uiowa.edu/pub/jy/filters/smoothGUI.jy

HTTP/1.1 200 OK
Date: Fri, 15 May 2015 17:18:03 GMT
Server: Apache/2.4.7 (Ubuntu)
Last-Modified: Thu, 14 May 2015 19:52:02 GMT
ETag: "2fc-516100fb5e7f5"
Accept-Ranges: bytes
Content-Length: 764
Keep-Alive: timeout=5, max=100
Connection: Keep-Alive

no change here:
HTTP/1.1 200 OK
Date: Fri, 15 May 2015 17:19:48 GMT
Server: Apache/2.4.7 (Ubuntu)
Last-Modified: Thu, 14 May 2015 19:52:02 GMT
ETag: "2fc-516100fb5e7f5"
Accept-Ranges: bytes
Content-Length: 764
Keep-Alive: timeout=5, max=100
Connection: Keep-Alive

and after a change:
HTTP/1.1 200 OK
Date: Fri, 15 May 2015 17:22:09 GMT
Server: Apache/2.4.7 (Ubuntu)
Last-Modified: Fri, 15 May 2015 17:22:02 GMT
ETag: "2fc-51622151f73aa"
Accept-Ranges: bytes
Content-Length: 764
Keep-Alive: timeout=5, max=99
Connection: Keep-Alive

and a bit later, without any changes:
HTTP/1.1 200 OK
Date: Fri, 15 May 2015 17:49:52 GMT
Server: Apache/2.4.7 (Ubuntu)
Last-Modified: Fri, 15 May 2015 17:22:02 GMT
ETag: "2fc-51622151f73aa"
Accept-Ranges: bytes
Content-Length: 764
Keep-Alive: timeout=5, max=99
Connection: Keep-Alive

= Ed's Comments =
* Ed thinks I should do this same test without webstart.  It might be some webstart system that is not doing this properly.


See also: http://fishbowl.pastiche.org/2002/10/21/http_conditional_get_for_rss_hackers/

And I asked the webstart group:
https://community.oracle.com/message/13071527