Hey,
I'd like to clarify two points about my previous email:
1. it is purely my own personal opinion. It is loose from any of the MediaWiki related projects I am "involved" with.
2. The questions I asked are not meant as criticism on the concept of having a core platform team, treating MediaWiki as a product, or improving the design and quality of the MediaWiki codebase. I think all of these are good ideas and hope WMF is very successful in pursuing them. This is why I'm asking hard questions about problems that I perceive to seriously undermine the chances of meeting the stated goals.
Cheers
-- Jeroen De Dauw | https://entropywins.wtf | https://keybase.io/jeroendedauw Software craftsmanship advocate ~=[,,_,,]:3
On 3 April 2017 at 09:34, Jeroen De Dauw jeroendedauw@gmail.com wrote:
Hey,
Like any significant codebase with a long development history, there are
remnants of design choices and experiments that are no longer in use, and some areas of code are in need of modernization. However, at its core is a large amount of highly functional, secure, performant code, capable of supporting a robust platform through the use of extensions and hooks. There is also a great amount of flexibility to adapt to new requirements.
This makes it sound like the MediaWiki codebase is pretty well designed. That is in stark contrast to my view, which is that it is a typical big ball of mud with serious pervasive issues too numerous to list. So I'm curious how you arrived at your view.
...
The stated goals and lineup strike me as very WMF, and something I've seen often enough before. What makes you think the results will not be poor leadership and disastrous technical results (in my estimation costing WMF many millions of USD) like in the past? What will be different this time?
Cheers
-- Jeroen De Dauw | https://entropywins.wtf | https://keybase.io/jeroendedauw Software craftsmanship advocate ~=[,,_,,]:3