Rant - I hate GIT
-
kelton5020 wrote:
I'd imagine you're kind of a green horn to the technology, and you haven't quite figured out how all of these things actually benefit you, if you take a moment to learn them.
Oh, I'm certainly green when it comes to Git. As to version control, I've used it since the days of RCS in DOS. Marc
Latest Article: Intertexti - Resurrecting Apple's HyperCard
My BlogWell if you really have, you wouldn't be complaining about not knowing/understanding the terminology :p
-
Source Control? We don't need stinkin' source control... ;P Seriously, it's my understanding that git is meant to manage multiple branches easily, probably that's why is overly complex for simpler tasks like working in a one shared source tree.
CEO at: - Rafaga Systems - Para Facturas - Modern Components for the moment...
RafagaX wrote:
it's my understanding that git is meant to manage multiple branches easily, probably that's why is overly complex for simpler tasks like working in a one shared source tree.
Yes, I would agree. I can see where my idea of working with revision control is different. I rarely would branch my SVN tree, and the idea of keeping local a local repository with all of the local branches and changes on my computer (until they are pushed onto the remote repository) I can see as possibly beneficial but overly complex, and frankly, scares me a bit with regards to losing my work due to a disk failure or the complexity of merging my work if I fall too far behind everyone else's branches. However, having read something recently, I am beginning to change the mental picture I have been holding - namely, that there is also a local repository as well as a remote repository. That helps me understand why a commit doesn't actually get seen on the remote server until a push happens. Marc
Latest Article: Intertexti - Resurrecting Apple's HyperCard
My Blog -
This is one of the most pathetically complex things I've ever had to work with. It CONSTANTLY gets in the way of actually getting work done - I spend more time f***ing around with pushes and adds and branches and checkouts that I do actually making code changes!!! What a P.O.S. This diagram[^] near the bottom of the post says it all. Marc
Reverse Engineering Legacy Applications
How To Think Like a Functional Programmer
My Blog
Computational Types in C# and F#It looks like you just do not need it. JIT has just 2 strong points - distributed architecture and conflict resolution. You use "I" and not "we" in the post, so it is unlikely that you can benefit from either. Besides these 2 advantages, GIT still retains its nature of a quick hack. I guess this happens because the focus is still on making things possible as opposed to making them easy or forgiving.
-
It looks like you just do not need it. JIT has just 2 strong points - distributed architecture and conflict resolution. You use "I" and not "we" in the post, so it is unlikely that you can benefit from either. Besides these 2 advantages, GIT still retains its nature of a quick hack. I guess this happens because the focus is still on making things possible as opposed to making them easy or forgiving.
77465 wrote:
You use "I" and not "we" in the post, so it is unlikely that you can benefit from either.
Well, this IS for a project with a couple other devs on it.
77465 wrote:
I guess this happens because the focus is still on making things possible as opposed to making them easy or forgiving.
Those are not mutually exclusive. My client is an expert at Git and whenever he explains something, it all makes sense and I "get it", but then when I try to do something at home, it blows up for me. It's a learning curve, and I've read through books and blog posts, but there is still a disconnect between what I do, that I think is doing it right, and what ends up happening. However, my rant was at the fact that I tried to do everything possible to kill my changes and revert back to the remote repository, to absolutely no avail. I ended up deleting the entire folder and re-cloning it. When I read in the documentation that doing a "reset --hard" (or whatever the syntax is) should kill my local changes, and doing a fetch / pull / checkout should get me to the latest version on the remote, and it obviously didn't work because "git status" still returned all the changes I wanted to throw away, then yes, I rapidly come to the conclusion that it's a pile of s***. Marc
Latest Article: Intertexti - Resurrecting Apple's HyperCard
My Blog -
77465 wrote:
You use "I" and not "we" in the post, so it is unlikely that you can benefit from either.
Well, this IS for a project with a couple other devs on it.
77465 wrote:
I guess this happens because the focus is still on making things possible as opposed to making them easy or forgiving.
Those are not mutually exclusive. My client is an expert at Git and whenever he explains something, it all makes sense and I "get it", but then when I try to do something at home, it blows up for me. It's a learning curve, and I've read through books and blog posts, but there is still a disconnect between what I do, that I think is doing it right, and what ends up happening. However, my rant was at the fact that I tried to do everything possible to kill my changes and revert back to the remote repository, to absolutely no avail. I ended up deleting the entire folder and re-cloning it. When I read in the documentation that doing a "reset --hard" (or whatever the syntax is) should kill my local changes, and doing a fetch / pull / checkout should get me to the latest version on the remote, and it obviously didn't work because "git status" still returned all the changes I wanted to throw away, then yes, I rapidly come to the conclusion that it's a pile of s***. Marc
Latest Article: Intertexti - Resurrecting Apple's HyperCard
My BlogIn a sense, yes, it is that pile. When I started using GIT I created a 3 repository model of the real project and tried commands there first. We also had few developers, but thanks to the lack of coordination and frequent changes in specs and overall project direction GIT payed for itself. I guess that was what a much larger and better organized team would experience.
-
This is one of the most pathetically complex things I've ever had to work with. It CONSTANTLY gets in the way of actually getting work done - I spend more time f***ing around with pushes and adds and branches and checkouts that I do actually making code changes!!! What a P.O.S. This diagram[^] near the bottom of the post says it all. Marc
Reverse Engineering Legacy Applications
How To Think Like a Functional Programmer
My Blog
Computational Types in C# and F#From the perspective of someone at the other end of the experience curve, I found Git almost impenetrable - I'm learning French right now (and I'm not a young whipper snapper), Git is harder than an entirely new language as far as I'm concerned. I would guess that if I broke through that outer shell and was already experienced with VCS concepts then Git would be the best choice - it's an assumption since why else would so many people and companies flock to it? I am guessing that the complexity pays back later when you have lots of contributors, branches etc. and maybe Bazaar, Mercurial etc. aren't so capable at that stage..., I have no idea, but there must be some reason! However, the most frequent comment I see and which tallies with my own experience, is "If it was my choice I'd use Mercurial". Mercurial is so much easier to get your head around, hugely helped by TortoiseHg, from total novice to functional in minutes not years. Small plug, not aimed at your level of knowledge, Choosing a Version Control System - A Beginners Tour of the Options[^] What I can't work out is why, if so many folk are saying "I'd choose Mercurial", does everyone choose Git?
-
This is one of the most pathetically complex things I've ever had to work with. It CONSTANTLY gets in the way of actually getting work done - I spend more time f***ing around with pushes and adds and branches and checkouts that I do actually making code changes!!! What a P.O.S. This diagram[^] near the bottom of the post says it all. Marc
Reverse Engineering Legacy Applications
How To Think Like a Functional Programmer
My Blog
Computational Types in C# and F#I hate TFS, but that's only because I'm a real noob, and because of the way the repo can get so totally out of sync structure wise with your workspace.
-
This is one of the most pathetically complex things I've ever had to work with. It CONSTANTLY gets in the way of actually getting work done - I spend more time f***ing around with pushes and adds and branches and checkouts that I do actually making code changes!!! What a P.O.S. This diagram[^] near the bottom of the post says it all. Marc
Reverse Engineering Legacy Applications
How To Think Like a Functional Programmer
My Blog
Computational Types in C# and F#Marc - a couple of suggestions...
- Mercurial is much nicer from a *users* point of view than Git (IMO)
- I found hg init[^] and Version Control by Example[^] very useful for comparing/contrasting how SVN and DVCSs work
And one other datapoint - one of my colleagues, who was just not getting on with SVN, picked up Mercurial in a couple of days, and has had no problems with it - it's just worked...
Java, Basic, who cares - it's all a bunch of tree-hugging hippy cr*p
-
Amrykid wrote:
By the way, I want to thank you for pointing out SmartGit - that has made life a LOT easier, and my client is very impressed with the application as well! Marc
Latest Article: Intertexti - Resurrecting Apple's HyperCard
My Blog -
Amrykid wrote:
By the way, I want to thank you for pointing out SmartGit - that has made life a LOT easier, and my client is very impressed with the application as well! Marc
Latest Article: Intertexti - Resurrecting Apple's HyperCard
My Blog