I have never had a good impression about Linus, mostly after reading about his debate with professor Tannenbaun, on which he demonstrated not only his ignorance about modern operating systems design (even when this might sound as anathema to Linux fans), but also his arrogance and lack of self-criticism (let's face it, Linux is NOT innovative at all and share the same conceptual and architectural deficits than Unix)
However, some recent facts are changing my overall impression about Linus. Firs, he impressed me for the way he managed an aparently endless discussion about design decisions among kernel developers. This short except is a gem: "get on with your lives. Realize that there is no 'perfect' value for HZ". I've been myself several times in such discussions and therefore can appreciate both the need and the efficacy of his comments.
Second, he has recently confronted the developers of GNOME to tell them the awful true: user interfaces in Linux sucks and they are not able to fix them because they think end users are stupid and don¡t deserve to be considered. Will them listen? Hardly.
Finally, I read Linus' responses to a proposal to include new functionality into the kernel called syslets. His arguments are really clear and shown a deep understanding of the complexities of designing good programming interfaces, which he summarizes as " think simplicity of use along with transparency, is so important . . . It's just that I think complex interfaces that people largely won't even use is a big mistake. We should concentrate on usability first, and some excessive cleverness really isn't a big advantage."
Saturday 17 February 2007
Monday 5 February 2007
The quest for a winner open source project
Today I came across this interesting article at Information Week about the keys for a successful open source project: "How tell the open source winners from the loosers". Among other things, they offers a series of criteria to spot a successful open source project:
This article put me to consider, again, if my the project I've been developing with my colleges at the UPC, the Grid Market Middleware, has any future at all as an open source project.
Applying Information Week's test, it seams that we meet what is in my opinion the single more important criteria: Disruptive goals. Our long term goal is to offer a platform for the research of economic based grids, opening the possibility to trade computational resources among communities of users, either for profit or as a form of community collaboration. We believe this will lead to a web 2.0 like grid environment. And I think that we are in a perfect timing, as Amazon is now making grid something popular, so we could expect a grow in the demand for grid solutions.
We also have a benevolent dictator: or at least I think that, ahem, I qualify for the job.
The lack of a thriving community or commercial support is not yet a problem, as we are still building the foundations of the gmm.
What seams critical now is to improve the documentation as most of the design is still in my mind. Also we need urgently at least a full time developer to implement the core on which others could start contributing.
I hope we could tackle this resource bottleneck before we pass to enlarge the statistics of "dead on arrival" open source projects.
- A thriving community
- Disruptive goals
- A benevolent dictator
- Transparency
- Civility
- Documentation
- Employed developers
- A clear license
- Commercial support
This article put me to consider, again, if my the project I've been developing with my colleges at the UPC, the Grid Market Middleware, has any future at all as an open source project.
Applying Information Week's test, it seams that we meet what is in my opinion the single more important criteria: Disruptive goals. Our long term goal is to offer a platform for the research of economic based grids, opening the possibility to trade computational resources among communities of users, either for profit or as a form of community collaboration. We believe this will lead to a web 2.0 like grid environment. And I think that we are in a perfect timing, as Amazon is now making grid something popular, so we could expect a grow in the demand for grid solutions.
We also have a benevolent dictator: or at least I think that, ahem, I qualify for the job.
The lack of a thriving community or commercial support is not yet a problem, as we are still building the foundations of the gmm.
What seams critical now is to improve the documentation as most of the design is still in my mind. Also we need urgently at least a full time developer to implement the core on which others could start contributing.
I hope we could tackle this resource bottleneck before we pass to enlarge the statistics of "dead on arrival" open source projects.
Subscribe to:
Posts (Atom)
