Prev: gcj compiled executable performance
Next: byte array
From: Alan Malloy on 27 Mar 2010 20:30 David Lamb wrote: > Alan Malloy wrote: >> David Lamb wrote: >>> You seem to be suggesting it is wise to plan for an intermediate >>> level of permissions between just sandbox and allpermissions. Is it >>> common to do that, or do most people just go for allpremissions if >>> sandbox is unacceptable? >> >> I think many people do go for allpermissions, but it's not the best >> strategy if you're hoping for people who don't implicitly trust you or >> your company to download and run your program. > > The only documentation I could find on the <security> element in .jnlp > files (for Java 1.6)only mentions allpermissions. > http://java.sun.com/javase/6/docs/technotes/guides/javaws/developersguide/syntax.html#security > > Are there other elements that can go there, and, if so, where do I find > out what they are? > > > I confess I haven't done this since college, and while I remember writing a JNLP file that had specific permissions, I can't find any evidence online that such a thing exists, so I could be wrong. At any rate, it sounds like all-permissions still asks the user for confirmation before each "dangerous" operation, so there doesn't seem to be any harm in using it. -- Cheers, Alan (San Jose, California, USA)
From: Andrew Thompson on 28 Mar 2010 01:47 On Mar 28, 11:01 am, David Lamb <dal...(a)cs.queensu.ca> wrote: > Alan Malloy wrote: > > David Lamb wrote: > >> You seem to be suggesting it is wise to plan for an intermediate level > >> of permissions between just sandbox and allpermissions. Is it common > >> to do that, or do most people just go for allpremissions if sandbox is > >> unacceptable? > > > I think many people do go for allpermissions, but it's not the best > > strategy if you're hoping for people who don't implicitly trust you or > > your company to download and run your program. > > The only documentation I could find on the <security> element in .jnlp > files (for Java 1.6)only mentions allpermissions.http://java.sun.com/javase/6/docs/technotes/guides/javaws/developersg... The developers guide is a good overview, but for the low down details on JNLP, download the spec. I asked a member of the Sun deployment team if they could make the entire spec. web browsable on their site, but the reply was WTE "No, because of 'licensing'". <http://java.sun.com/javase/technologies/desktop/javawebstart/download- spec.html> > Are there other elements that can go there, .. The oddly named 'j2ee-application-client-permissions'. That gets unprompted access to the JNLP API services for the FileOpen/SaveService, the ExtendedService, the PrintService.. (I think that is it). And the window warning/banner is removed. The JACP permissions are rarely used, but handy for those cases where they are the only extended permissions that might be of use to an app. Note that I offer JaNeLA* to validate the launch files and other resources of JWS based launches. I mention this because you used the term 'allpermissions' twice above (one time incorrectly spelt). There is no such security level. * <http://pscode.org/janela> Alternately, if you can 'read' an XSD, you might go directly to the XSD used by JaNeLA to find what is allowable. <http://pscode.org/JNLP-6.0.xsd> -- Andrew T. pscode.org
From: Roedy Green on 28 Mar 2010 18:47
On Sat, 27 Mar 2010 15:11:17 -0700, Alan Malloy <alan.NO.SPAM(a)malloys.org> wrote, quoted or indirectly quoted someone who said : >I think many people do go for allpermissions, but it's not the best >strategy if you're hoping for people who don't implicitly trust you or >your company to download and run your program. Imagine your end user >loads up your JNLP file and is presented with a dialog. Is he more >likely to use your program if it says "this program needs permission to >connect to the Internet, and read/write to one particular file" or if it >says "this program needs permission to do anything it wants with your >computer"? Oddly the answer to that question depends on your audience. Some people won't give permission to something they cannot understand. It sounds worse that "anything". The author is clearly trying to snow them. -- Roedy Green Canadian Mind Products http://mindprod.com If you tell a computer the same fact in more than one place, unless you have an automated mechanism to ensure they stay in sync, the versions of the fact will eventually get out of sync. |