I re­cently pro­posed to defer Pro­ject Jig­saw from Java 8 to Java 9.

Feed­back on the pro­posal was about evenly di­vided as to whether Java 8 should be de­layed for Jig­saw, Jig­saw should be de­ferred to Java 9, or some other, usu­ally less-re­al­is­tic, op­tion should be taken.

The ul­ti­mate de­ci­sion rested, of course, with the Java SE 8 (JSR 337) Ex­pert Group. After due con­sid­er­a­tion, a strong ma­jor­ity of the EG agreed to my pro­posal.

In light of this de­ci­sion we can still make progress in Java 8 to­ward the con­ver­gence of the higher-end Java ME Plat­forms with Java SE. I pre­vi­ously sug­gested that we con­sider defin­ing a small num­ber of Pro­files which would allow com­pact con­fig­u­ra­tions of the SE Plat­form to be built and de­ployed. JEP 161 lays out a spe­cific ini­tial pro­posal for such Pro­files.

There is also much use­ful work to be done in Java 8 to­ward the fully-mod­u­lar plat­form in Java 9. Alan Bate­man has sub­mit­ted JEP 162, which pro­poses some changes in Java 8 to smooth the even­tual tran­si­tion to mod­ules, to pro­vide new tools to help de­vel­op­ers pre­pare for mod­u­lar­ity, and to dep­re­cate and then, in Java 9, ac­tu­ally re­move cer­tain API el­e­ments that are a sig­nif­i­cant im­ped­i­ment to mod­u­lar­iza­tion.

Thanks to every­one who re­sponded to the pro­posal with com­ments and ques­tions. As I wrote ini­tially, de­fer­ring Jig­saw to a Java 9 re­lease in 2015 is by no means a pleas­ant de­ci­sion. It does, how­ever, still ap­pear to be the best avail­able op­tion, and it is now the plan of record.