On 18 April 2019 16:56:16 BST, Mike Gilbert <[email protected]> wrote: >On Wed, Apr 17, 2019 at 10:44 PM Georgy Yakovlev <[email protected]> >wrote: >> >> On Wednesday, April 17, 2019 6:31:42 PM PDT Mike Gilbert wrote: >> > On Wed, Apr 17, 2019 at 3:35 PM Georgy Yakovlev ><[email protected]> >> wrote: >> > > # Georgy Yakovlev <[email protected]> (17 Apr 2019) >> > > # The Oracle JDK License has changed for releases starting April >16, 2019 >> > > # While it may be fine to use for some usecases it's not >comepletely clear >> > > # what is considered "personal use" and if we can legally >distribute it. >> > > # License states: >> > > # "You may not: >> > > # make the Programs available in any manner to any third party" >> > >> > I don't agree with your rationale here. >> > >> > Gentoo does not distribute the JDK due to RESTRICT="fetch mirror" >in >> > the ebuild, so Oracle's license has no relevance. >> > >> > Oracle cannot prohibit us from distributing a shell script that >moves >> > some files around. That liability is on the user who runs it. >> > >> > We cannot force you to continue maintaining this package, but I >think >> > we should have a better reason for masking/removing it. If you >cannot >> > provide one, please just drop this to maintainer-needed. >> >> I've modified the mask for now, but I still believe we should drop >it. >> I do not maintain it at all, I only work on openjdk and a bit of >icedtea. >> >> For a while[1] we've been modifying provided jar: >> >> zip -d jre/lib/rt.jar sun/misc/PostVMInitHook.class || die >> >> but license[2] states that >> >> "You may not: >> ... >> make the Programs available in any manner to any third party >> ... >> create, modify, or change the behavior of, classes, interfaces, >or >> subpackages that are in any way identified as "java", "javax", "sun", >“oracle” >> or similar convention as specified by Oracle in any naming convention >> designation. >> >> " >> >> Is it even legal? > >That does seem like it might cause some legal problems for users. > >> Java usage tracker will fail due to sandbox during builds. >> >> while writing this email I found out it's probably possible to >disable it with >> com.oracle.usagetracker.track.last.usage=false >> in >> /etc/oracle/java/usagetracker.properties >> >> need to test it > >If that does not work, a possible alternative would be to install a >file in /etc/sandbox.d to add some path to SANDBOX_PREDICT. > >Anyway, this issue does seem like grounds for removal if it is not >addressed by somebody.
IIRC, SANDBOX_PREDICT doesn't really help in this case because it triggers on any package using Java within the home directory of the build user and the HOME environment variable isn't respected. -- Sent from my Android device with K-9 Mail. Please excuse my brevity.
