Date: Wed, 18 Oct 2000 13:44:57 -0700 (PDT)
|From:||Eugene Eric Kim|
|Subject:||OHS Meeting at SRI on October 17|
I've posted minutes from yesterday's meeting on my web site:
Thanks to Howard for recording and transcribing these for us.
In summary, our efforts to find funding, led by Doug and Pat Lincoln, are starting to bear fruit. It looks like we have a real good chance of getting funding from NIH and possibly DARPA, probably by next summer but possibly as early as December.
What this means is that we need to start getting organized, and we need to start making some of the hard decisions that we've been avoiding thus far. One of the things we did was to bring in Mark Mitchell as the OHS project manager. Mark is the CTO of CodeSourcery, an open source company, and one of the leaders of the Software Carpentry project, which is developing next generation open source development tools. He will make an outstanding lead for the project.
Many of the details regarding the development process are still to be determined, and I'm sure Mark will start making some announcements and soliciting feedback over the next few months. One of the things that is sure to happen, however, is that Mark, Doug, and Pat will select a small group of core developers.
What does this mean for this group in particular? In reality, not much. Nothing that any of us have done has ever been officially sanctioned as part of the OHS, but we have all obviously made some very important contributions both in context and in code. I hope that discussion and maybe even a little coding will continue on this list, until some of the official processes get ironed out. There are still certainly a lot of important things that need to be done, and your contributions will not go unappreciated.
In summary, some good things are happening around us, which bodes well for the project as a whole. As always, comments and questions are welcome.
Eugene Eric Kim
By Howard Liu <email@example.com>
Doug brought the group up-to-date on recent news. NIH has expressed strong interest in funding our work, possibly up to $5 million by next year. Doug and Pat are working hard on two whitepapers as a precursor to this funding. Because of the funding possibility, we need to start getting serious about how this project will get off the ground, including identifying who will be in charge of the various projects, and who will work on them. This needs to be worked out to the satisfaction of the funders, who will be auditing our work.
SRI will be the home for the project. Pat Lincoln will be in charge of the overall program, and will work on business development issues, such as making proposals. Mark Mitchell will be Project Development Manager of the OHS project.
OHS development is the first project of the bootstrap "program." Other programs to be launched soon are Application Support, headed by SRI's Carla Woodworth and Reg Ford, and a Fellows Program, headed by SRI's Marcelo Hoffmann. The Application Support project will support serious users, providing training and other services. The Fellows Program will host experienced people sent to SRI and paid and supported by their home organizations. There is already some interest from Singapore and Switzerland business schools in participating in such a program.
Pat Lincoln described the NIH proposal. Collaboration will increase the effectiveness of NIH's efforts. For example, one of the challenges posed by the National Cancer Institute's director was that researchers are often unknowingly competing with each other, without knowing what the others are working on. Doug noted that similar problems exist in a number of environments, including DARPA, NIH, and the Software Productivity Consortium.
There is imminent support for our project coming from the government. At the earliest, we might have funding by December, although a conservative estimate is by next summer.
Mark Mitchell explained his role in the OHS project. Mark, President and CTO of CodeSourcery, LLC., will be handling detail work specifically regarding OHS development. CodeSourcery will provide engineering suport for the project and will outsource other work to subcontractors. Mark solicited resumes from the group for this and other projects. He can be reached at firstname.lastname@example.org.
Jack Park asked whether the OHS will be required to use Python, as is the Software Carpentry project. Mark said that there are no such mandates, and implementation-specific decisions still need to be made.
Marcelo Hoffmann briefly described the Fellows Program. He noted some challenges, such as making sure that fellows actually contribute to the overall goals of the program, rather than just work on something narrow that doesn't relate to anything else. We need to have an overall body of work ready before the fellows arrive.
John Deneen asked whether there could be some synergy between the OHS effort and some internal SRI projects. Lee Iverson discussed the Digital Earth project, and explained that they needed standards development, which requires collaboration. The OHS can improve the standard development process, not only software development process. Standards development need consensus. The OHS can provide mechanism to build consensus.
Doug commented on the Genoa project. He also explained that he has a new term for the OHS viewer, which he formerly called a "lens." Because Xerox PARC had a Magic Lens project, Doug's new term is "hyperscope."
Pat described Tom Gruber's system, "Introspect," which is closed source, as is Rod Welch's system. The OHS project is committed to open source. Doug commented that the OHS project has the potential of showing that open source, evolutionary projects are better.
Carla Woodworth described her background and the role of her project. Carla fielded systems at SRI, including the Command and Control system. Lived and worked directly with military clients. Regarding the OHS, she will work with producing user documentation, training, support. Her group just opened an office in San Luis Obispo, and they have a staff already. Doug commented that Carla's area may eventually dwarf the development effort.
Eugene Kim explained what all of these new changes mean to our previous efforts. Basically, this is a move to get more serious, and to implement more structure. The first step that was needed was to bring in a project manager, and we have brought in an outstanding one in Mark. Now we have to make some hard decisions. We need to identify the values and directions of the core group, in order to facilitate evolution. This also means deciding who will form the core development group, which will ultimately be the decision of the project manager. Those who are not or should not be included in that group can still contribute; this is part of the evolutionary strategy that we have set out to accomplish. Mark added that we need to redefine the expectations of each participant.
Eugene said that the weekly meetings in its current form will be suspended. The structure of future, OHS-specific meetings are to be determined. However, Pat last week suggested the possibility of doing a monthly seminar, where people could come in and talk about their work of general interest. People seemed enthused about this idea. Doug also added that he'll be hosting a get-together at his house on Wednesday, November 1, when Henry van Eyken, Bootstrap Institute's webmaster, will be in town.
Questions and answers. Mark explained that by late November, early December, he hopes to have put together an initial development team, and that work can get started in earnest by the beginning of 2001. Doug added that he wants to put together an advisory board with people such as Andy Poggio.
Doug explained that Grant Bowman, Nicholas Carroll, and Sheldon Brahms have been playing around with both the DOS and Smalltalk Augment clients. He cautioned that this does not mean that Augment will be widely available, or that the OHS project is an Augment port. Nicholas added that we do not want to be marketing Augment. Doug will be dumping old Augment documentation on the web site, using Eugene's Augment-to-HTML converter and gateway.
Mark added that we may want to present the project differently to different groups, e.g. focus on collective IQ, hyperscope. Perhaps, we can have different websites with mutual links.
Sandy Klausner had some architectural questions about OHS. Eugene asked that these questions be discussed outside of this meeting, as it is somewhat beyond the scope of this meeting's agenda, and since we don't yet have a unified vision for this yet. This is one of the important pending decisions that Mark and the group will have to make.