4 ms·

Whew this thread is rife with elitism. About git? To the folks sharing your difficulties with git, bravo. To everyone who says they "don't get" how people coul

Whew this thread is rife with elitism. About git? To the folks sharing your difficulties with git, bravo. To everyone who says they "don't get" how people could be confused or lost while using git, you are the reason why those folks are lost, and you're perpetuating the culture of elitism by playing dumb. Git is hard as shit to understand unless you came from development before it existed, or you have had the luxury of trying other VCS professionally. Most people just know git, and it's confusing as hell if you have no reference. Don't even get me started with how this has poisoned the industry on the whole: Proper VCS implementations with CI/CD are incredibly rare, even at places with world class engineering talent.

notdonspaulding 4y ago
My hypothesis about git apologists is that they fall into one of 2 groups: 1. People who have an organizational development model and workflow which is equivalent in complexity to the Linux kernel development model and workflow (i.e. thousands of developers loosely coordinating the release of systems or mission-critical software on a regular cadence). For them git fits their needs better than almost anything else, and it makes sense because it was made for exactly that use case. 2. People who learned their first VCS after Github had reached critical mass and if you had to pick one VCS to start with, you picked git by default. For them their brain fits git, and they think "the git way" is synonymous with "the right way". Many folks in group 2 never stop to realize that they are on a development team of 5 (or 50, or 500)...and they always cut their release from `master`...and all of their branches are always pushed/pulled from the same Gitlab remote...and within 5 minutes they can chat with anyone who's made a commit in the last 2 years. I don't take issue with group 1 people who understand the git model and love it and adopt it in their workflow because they need its complexity and they're willing to pay the UX tax to get it. I take issue with the folks who don't understand the git model or realize they don't need any of the complexity and aren't aware of the existence of solutions like fossil and mercurial.

notdonspaulding 4y ago

My hypothesis about git apologists is that they fall into one of 2 groups: 1. People who have an organizational development model and workflow which is equivalent in complexity to the Linux kernel development model and workflow (i.e. thousands of developers loosely coordinating the release of systems or mission-critical software on a regular cadence). For them git fits their needs better than almost anything else, and it makes sense because it was made for exactly that use case. 2. People who learned their first VCS after Github had reached critical mass and if you had to pick one VCS to start with, you picked git by default. For them their brain fits git, and they think "the git way" is synonymous with "the right way". Many folks in group 2 never stop to realize that they are on a development team of 5 (or 50, or 500)...and they always cut their release from `master`...and all of their branches are always pushed/pulled from the same Gitlab remote...and within 5 minutes they can chat with anyone who's made a commit in the last 2 years. I don't take issue with group 1 people who understand the git model and love it and adopt it in their workflow because they need its complexity and they're willing to pay the UX tax to get it. I take issue with the folks who don't understand the git model or realize they don't need any of the complexity and aren't aware of the existence of solutions like fossil and mercurial.