| Re: [ide-dev] Yet another article advising to leave Eclispe |
|
On 10/16/2014 10:37 AM, Martin Lippert
wrote:
Interestingly, this is the case for products that are being built on top of Eclipse as well. Users cannot differentiate between bugs causes by Eclipse and bugs caused by the product build on top of Eclipse - and in that case they always blame the product, even if some of the bugs are caused by the underlying Eclipse components. So it depends a bit on your viewpoint… :-)Except that when shipping a product, the ones developing the product are responsible of the technology they choose and everything they ship is a piece of their product so it requires effort in validating/testing/developing those upstream parts just like they do for they own code. When using OSS components, one cannot use as excuse "My product fails because an upstream component fails" but more "My product fails because it relies on a component which wasn't good enough and that we kept without investing in improving it". Adopting big OSS "community" frameworks without getting ready to contribute is a technical suicide. (Note that this is not a remark related to how you contribute to OSS and Eclipse, I think your team is doing well; it's more that I disagree with your alternative interpretation ;) The published article puts STS in the focus and blames STS for poor quality - and it might be accurate in his case that the problems and issues that led him to leave Eclipse behind where caused by STS bugs. But the discussion should not be about high-quality Eclipse vs. poor plugins. Everybody should aim at high-quality and we all should do our best to achieve that. We all can improve. Cheers, -Martin+1 And I believe Marcel's exception reports about 3rd-party code too is very interesting and is good for helping every Eclipse project (not only ones at Eclipse.org, but also 3rd party ones). It's good to see the Foundation/Community also opening tools for 3rd-party adopters. This goes in the direction you're mentioning about higher quality for everyone. |