Thoughts on the History of Version Control

posted in: Blog | 0

The exact methods of version control changed throughout my career, but one thing remained constant: one person needs to have final authority and control.

Back in the early 1990’s, when I was writing a document, I would get a bright idea on how to improve the the document, but I sometimes wasn’t sure if the idea would work. So I saved what I had written so far with a v1 (version 1) at the end of the file name and did a “Save As” with v2 at the end of that file name. If other bright ideas followed when working on that version, I created a v2.1, v2.2, etc. Sometimes v2 turned out to be unworkable, so I went back to v1 and did a “Save As” to create v3. In this way, I had version control. It was cumbersome, but still in my control.

When the document was done to my satisfaction, with some combination of all the versions I had worked on, I printed one copy and gave it to SMEs one at a time. I asked each SME that they write their comments so that I could tell who made each comment. Typically I started with the SMEs who knew the most detail and ended with those having only general knowledge.

When I got my manuscript back with all the comments, it was up to me which ones I accepted and which ones I did not. I did not go back to the SMEs unless I did not understand their comments. (For example, I went back if the SME crossed out an entire paragraph with a big red X — did that mean that the SME felt that the paragraph was wrong or that the paragraph was unnecessary?)

I wanted each SME to see what the others had to say and be able to indicate that they agreed or disagreed with someone else’s comments.

This all worked pretty well, with the occasional question: “How come you never take any of my comments?” (I had good reasons, and never got into trouble.)

Later in life (now the late 1990’s), I worked for a company where I was given a document to work on and was told that I was the owner. Indeed, I received emails from SMEs asking for my permission to add their comments to the document. If I gave permission, this automatically saved the document with a version number and assigned a different version number to the commented copy.

SMEs could comment on different sections of the document at the same time, but not on the same section at the same time. I could accept comments and they would be immediately incorporated into the document; but typically I waited until it was time to publish the next version of the document, at which point I emailed all SMEs to look at the document and all of the comments. Discussions (or arguments) sometimes occurred between SMEs in their additional comments. It was up to me to referee if necessary. Then I would go through the document and accept or decline each and every comment. Each and every acceptance or declination was recorded just in case there were any repercussions. This too seemed to work pretty well.

In sum, the main observation I have is that one person needs to be in charge of the communication (document, video, slide-show, etc.), otherwise chaos reigns. And if an error occurred in the document, that’s where version control helped — we determined who told us to do something that we didn’t do, and who deserved blame.

Leave a Reply