Alexboy wrote:I'm not completely familiar with the repository platform
Repositories are commonly used to store data that gets changed often, by different people.
Suppose you have to write some text with several other persons. Different people make changes at different times, so you quite soon get a problem of "what is the latest" version, or you made changes, and then you get new changes from someone else by email, and you have to merge your changes with the newly received text somehow. Doing that manually gets very messy very soon.
A repository handles all that stuff automagically, everybody copies a revision from the repository, makes changes, and submits the changes (thus creating a new revision). The repository handles merging of changes (and complains when it cannot do it automatically). You can also ask it for updates, and you get the changes submitted since you asked for a copy into your text.
In other words, it acts as a central point in managing changes in data. Changes from different people get merged by the system, and it never misses a subtle text change at some unexpected place.
It also tracks who changed what at which time. Since you can add a short description when you submit changes, you get a quick overview of what happened in the last revisions, and why.
These properties make repositories pretty much defacto-standard in collaborative projects.
What "data" it stores doesn't really matter. It can be as small as a report you write, the source code of a towns newgrf, or something more big like the source code of openttd. I am sure there exist much larger repositories in the world though
Alexboy wrote:the differences between dev.openttdcoop and bundles.openttdcoop
dev.openttdcoop is the development site, where the project 'lives'. The project owners, and the users get an overview of what happened in the project, the issues are a list of things that need looking into. bundles.openttdcoop is a computer that takes the data from projects, assembles it into a program or a newgrf, or a game script, or so, and publishes the result. It's a quite separate thing, and it may eg rebuild the same revision more than once due to internal changes or upgrades, for example.