Setting up the process: Better to import from or export to other projects?

[elin_decker]

We're trying to figure out the initial setup of our Flare system. We have a formal document approval process and a release team. The release team are the only people who are allowed to touch the source files. So we're thinking that the writers will work on documents locally, then move them to a staging location. The release team will add the final information in the staging location and then move the finalized files into the source location and from there publish the PDF. The point of the staging location is mainly for source control so we don't overwrite the "good" files until the release team is ready to run the new PDF. We won't ever be able to use MadCap Central, but we will be using GitHub (eventually). My questions are:

1. Is it better to export files from the local location to the staging location (push), or to import the files into the staging location (pull)? Same question for moving files from staging to source.
2. Is there a simpler way to manage all of these steps?

Thanks for any input/suggestions!

[AlexFox]

I would personally that you get into some sort of Source Control urgently, as it can save you from a lot of issues. Do you use ADO etc. at your company? How come you'll be able to use GitHub "eventually", why the delay?

[trent the thief]

Hi Elin,

You're getting confused by improper terminology. To put this in more common industry terms, The files in Flare that you're editing are Source files. When you publish PDFs, those are output files. There is no push or pull or export involved in publishing output from Flare (i.e., your PDFs). Those are source control terms.

What I've most often seen done is to have a repository location (CVS, SVN, Git, etc.) where the Flare Source Files live and a separate repository location for the published Flare output. From there, DevOps handles merging the published Flare output with the product object and packages them for clients.

As Alex said, the sooner you get some sort of source control in place, the less heartache your going to have. It doesn't matter if it's git or bitbucket right off the bat. Transferring files between different source control systems is easier than re-creating months of work if you lose a drive. You need to get those files into source control before something bad happens.

[elin_decker]

Thank you both for the feedback. I probably wasn't clear in my initial post. We're moving the entire set of files from one location (local) to another (network) and then to another (other network location). The final network location will be our source files for future updates, and the files we'll use to generate and publish the final PDF. It's a bit complicated, but it has to do with who can touch the files at what point. So the question is how best to move all of the files from place to place (export or import?).

But I do get your overall point that we should skip this file-hopping and move into Git as quickly as we can! I'll see if I can get that moving sooner, but our IT team will do the Git set-up and the writers have roughly zero influence on their schedule.