Open sourcing herb@home?

[Tom Humphrey]

Several people have me about the possibility of open-sourcing the herb@home software, (most recently in this post), so I’m starting this thread to outline my thoughts on this and to hopefully elicit discussion. This is all speculative and to avoid any misunderstanding herb@home is not currently open source and currently I don’t have any plans to any release code – but this is something I’d like to investigate.

The strongest argument that I can see for partially or completely open sourcing the system would be the potential to get more programmers working on the code. Open software might also help reassure some potential institutional collaborators. More generally, the software represents hundreds of hours of programming work and I would personally like to see it more widely used.

My main reservations about open sourcing are that it might entail substantial time demands to provide technical support for the program, time which could otherwise be spent on software improvements, end-user support and botanical documentation. However, this additional support commitment would definitely be worthwhile if it resulted in significant contributions to the project from other programmers or substantially increased use of the software. So really the question is whether there are other programmers who would choose to work on this project?

A secondary concern is that it might be problematic to end up with a series of competing projects – all vying for the same limited pool of users. I don’t know that this would be an issue in reality. (for example Distributed Proofreaders has made it’s software available and does not appear to suffer from competition as a consequence). I think it would be better to have multiple collections collaborating with herb@home rather than all running independent volunteer projects.

Isolated parts of the herb@home system (notably the doc form web interface) may also be useful as discreet components. So an open-source release of just parts of the system might also be worthwhile.

Given the herb@home software, it would be relatively straightforward for a museum or university to set a derivative of the project. The server-side software is written in PHP and uses a MySql database. The project is currently running on a LAMP server. Hardware-wise the system is fairly undemanding, apart from needing lots of storage, the current set up is an AMD Athlon64 X2 2600 with 2 GB ram and ~1.2 TB hard disk storage (split between four drives). Amazon S3 is used for offsite backup of the database.

The documentation form makes use of a fairly substantial javascript AJAX web app. For the edification and enlightenment of anyone who wants to delve slightly further into the workings of the software, the javascript web app, with fairly skeletal documentation is viewable here (auto-generated jsdoc documentation).

Regards,

Tom
[email protected]