[ Date Index ][
Thread Index ]
[ <= Previous by date / thread ] [ Next by date / thread => ]
Neil Williams wrote: > On Wednesday 19 Mar 2003 5:32 pm, Bill Wilson wrote: > >> From bill@xxxxxxxx >> >>We would love to make our product open source and suggested that LUG >>members help us write the code for free. Unfortunately nobody was > > > We are a small group at DCLUG really so I think it's unlikely that you'd find > enough members with both the right skills and the requisite time available. Motivation is the big one, the only people who care upfront are customers and potential customers, i.e. those people using your software, or likely to use it. You can open the source to existing customers under restrictive licences (that could include you actually incorporating the changes into live systems, no one wants to create a support headache), see if that generates interest, useful contributions. I think the component route is an interesting one, do you have a component that is compelling value added, but not necessarily directly competitive to your core. Maybe for a 4GL, a screen designer, and database layer. The actual logic and your own screens stay with you. > The costs are only there because it began and is currently a closed source > project. The costs are there because software is expensive to write, the question is who pays. > I'd say the best route > to Open Source is from the beginning - that way certain costs aren't incurred > in the first place. Netscape4 couldn't be made into an open source project, > work started from a new base. As far as I am aware the Mozilla project was roundly criticised for starting again. Not because it was the wrong technical decision, it was probably needed, but because it doesn't always work well for the distributed development model. Distributed development is great for maintainance, enhancement, features, but it is very hard to bring large numbers of developers on to immature projects, that requires a clear plan and well thought-out modularization. Interestingly the Mozilla project has a very strong infrastructure to manage software components, ensure ownership of components, track bugs, far more structured than any other effort (except perhaps CPAN?!, not sure on KDE, GNOME, and the like but I don't think they are anywhere near as structured). I suspect Mozilla didn't start everything from scratch given the similar look and feel, they presumably kept either a lot of code, or a lot of plans and designs. A thousand people can write kernel hardware drivers, but no one will want to until, Linus had made a working OS, that at least offers some functionality of interest.
Attachment:
pgp00012.pgp
Description: PGP signature