I am absolutely positive to reuse existing solutions and to not redo efforts that have been done elsewhere. That is why I did not insist on the clustering solution being an apache project. Though as you for sure know there are two kinds of open source projects. One kind is driven and "owned" by a community and the other kind by a company. The difference is in freedom for users as well as for contributors.

For example in the case of config management and deployment Talend will also like to help drive the solution. In case of a community driven project this will be possible. In case of a fuse product it may or may not be possible depending on the current position of the Fuse management. So while in open source it is always possible to fork a project only community driven and owned projects allow collaboration of several vendors in the same project in the long run.

So for this reason I think that strategic dependencies of apache projects should be community driven if possible.

Just to name one case where this is not the case is spring. While the product is great and helps the apache projects a lot it is not possible to really take part in the spring development. So while we would sur elike to fix or change many things in spring we could only do this by creating a fork which would be a baad choice. So it iss no wonder that many projects try to be more independent of spring.

Christian


Am 15.04.2011 11:11, schrieb Rob Davies:
We use other open source projects all the time in our apache projects - even 
core functionality. What should matter is the right solution - I hate to think 
any apache project suffers from not invented here syndrome.
On 15 Apr 2011, at 09:57, Christian Schneider wrote:


--
----
http://www.liquid-reality.de

Reply via email to