RFC: a new try for an official community approved certification - Mailing list pgsql-advocacy
From | Jaime Casanova |
---|---|
Subject | RFC: a new try for an official community approved certification |
Date | |
Msg-id | CAJKUy5jy5fzrWZ9MuYkNWC73APX0F6rV-OrCXPuBQE7mOczxWQ@mail.gmail.com Whole thread Raw |
Responses |
Re: RFC: a new try for an official community approved certification
Re: RFC: a new try for an official community approved certification Re: RFC: a new try for an official community approved certification Re: RFC: a new try for an official community approved certification |
List | pgsql-advocacy |
Hi everyone, At PgCon I mentioned a was going to create this thread so here we go: == why do we need one? The fact that there is no official certification is a real problem because anyone can offer theirs and have bad answers or confuse people about how things work. You can think that people looking for a certification should ensure first but what if the company has a good reputation because other certifications they have are good? And this is an example of this happening, recently I got a certification from a less-known company, and I found at least one question in which there weren't any correct answers so I had to choose the answer that was at least partially correct. And of course this could happen to anyone. == we have tried… and failed Several years ago there was a project for creating a pensum that all companies that wanted to offer the "official" certification should comply with. But being postgres developed and improved so fast we ended up with half the pensum and we already had to adapt part of it because of the changes in the new version. But of course there was the problem of not really prepared trainers, a know a guy that was certified by a well known postgres company but he clearly understood wrong somethings because he was teaching that you can restore a pg_dump'ed database and then apply wals from the original server and have PITR (doh). == so the problems - mantaining/updating a pensum would be difficult - is not only a matter of covering the pensum, it should be done right - other problems are: -- current certifications are not neutral and cover also tools from the company doing the certification -- even if they have the same name, certifications cover normally different subjects in different companies == so, my idea - Let's focus on creating just one certification covering only core/contrib functionalities of postgres, and let companies have their own certifications for third party products. - don't try to force the pensum, just let's create a database of validated questions and validated answers. If we focus on only one certification (something like associate postgresql or something like that) we can let companies decide if they want to have 1, 2 or 3 modules to prepare people to give the exam. - Who creates the questions? And who validates them? The answer to both questions is "trainers" from companies that want to offer the exam. So, a trainer creates a question and another trainer (hopefully from another company) validates the question and the answers for the question. Questions and answers should also be validated for postgres version and operating system. If a company cannot successfully prepare people for the exam they could check if they are teaching right (I mean that they are not saying old/not exact things). So, having a certification made from a pool of validated questions/answers don't only will help companies choosing trainings but will in effect improve the quality of trainers. There are a lot of open questions yet, like where the exams will be taken. In the community infrastructure? in the company's? but I guess we have still enough for start commenting. -- Jaime Casanova Director de Servicios Profesionales SYSTEMGUARDS - Consultores de PostgreSQL
pgsql-advocacy by date: