Development

Before Sun's release of the VM source code, Strongtalk development has been quiescent, since little could be done without the source code, athough many people have expressed interest in it.

Since the VM source has become available, we are now organizing an open source development effort. The goal is to propagate the innovations that Strongtalk contains into the hands of developers and end users, in whatever way works.

What form that takes depends to a large extent on you, the developer community. This could take the form of a productized version of Strongtalk, or variants of it. It could also take the form of adaptation of various parts of the technology to other languages or Smalltalk implementations. Any or all of these are good- we just want to get these technologies out there.

If you are a VM wizard, and you want to help get Strongtalk into a truly usable state, or if you are interested in borrowing technology for another Smalltalk implementation; now is the time to get involved, since things are just getting organized! Discussion takes place at http://groups.google.com/group/strongtalk-general. We especially need volunteers with good virtual-machine and C++ expertise, as well as top-flight Smalltalk programmers.

Check back here over time for information as the development effort is fleshed out.

Development Status

Strongtalk is currently in basically a pre-alpha release state. It runs, and is moderately stable although there are still crashes. The basic technology is fully working, but there are a variety of things needed in a production version that are not there yet. It is not yet ready for production use

Development Guidelines

Development is currently hosted at Google Code at Strongtalk, which currently holds the source code repository (under Subversion) and issue/defect database.

How to Contribute

We welcome contributors with expertise in virtual-machine design, Smalltalk expertise, or knowledge of type-systems. To contribute, you should volunteer on the discussion group. The issue database can give you ideas for what needs to be worked on.

Before we can accept contributions, you will have to agree to the legal boilerplate, which just ensures that any contributions are covered by the BSD open source license, and that the contribution won't be encumbered by your patents or any patents you know of. It is a simple process, simply involving adding a line to the end of the contributor's license to indicate your agreement.

At the moment, the way to submit the file is to download a copy, add your name and email address at the end (and your company name only if you are contributing on behalf of a company), and post the file to the strongtalk-commit discussion group, where a committer can pick it up and commit it to the database. Once that is done your code can be considered for integration into the system.

Once you have added your license for contributions, you contributions can be submitted. This is done (at least for now) by attaching the code file to the issue database item with which it is associated, with a status of New so it can be reviewed.