[
Date Prev][
Date Next][
Thread Prev][
Thread Next][
Date Index][
Thread Index]
[
List Home]
|
[jakarta.ee-spec] [Discuss] - Denoting which implementation was used to ratify a specification
|
Launching a separate thread as to not pollute the vote thread.
I wanted to share with the community a presentation I prepared and delivered to the Specification Committee relating to this topic.
- https://docs.google.com/presentation/d/1aYfaWvVSBRNZgge88NBkMJsgYH_tnylIP0Q4bOjDQ3w/edit?usp=sharing
For the impatient, skip to slide 9 and look through how our industry behaves when we put emphasis on one implementation as the reference implementation, ratified implementation or otherwise elevate on implementation over the others.
- https://docs.google.com/presentation/d/1aYfaWvVSBRNZgge88NBkMJsgYH_tnylIP0Q4bOjDQ3w/edit#slide=id.gbf9fe44a5a_0_238
The short version is over time there are no others.
Ultimately our goal is to create an ecosystem with multiple implementations. This is how we get freedom of choice in our industry. This is how we get vibrant spec projects filled with people. This is how we get vendors competing based on technical innovation and pride to what they've contributed to a spec.
If there aren't multiple implementations, there's no point in all the overhead of what we're doing. Our current status is that there is just one implementation of most specs, used by almost everyone, our spec projects are very quiet, we don't have huge innovations coming out and there isn't a lot of choice at the component level. Option 3 does not cause these ailments alone, but it doesn't help them either.
If we work hard we can turn this around, I am certain. However, it does not start with elevating one implementation to the top in a category of its own above a very empty list (where it is listed again at the top).
Option 3 is a vote for the status quo. I know we can be better.
--
David Blevins
http://twitter.com/dblevins
http://www.tomitribe.com
