8 ms·

I really don't understand why you are choosing to use the word "lying". Us mere humans make mistakes all the time. Typos, omissions, false starts, and so on.

I really don't understand why you are choosing to use the word "lying". Us mere humans make mistakes all the time. Typos, omissions, false starts, and so on. What is the value of throwing that raw set of events at a reviewer or complicating the understanding of the changes when viewed in retrospect from the future? What is the reason you call curating the work into a more polished form "lying"? Why do you think the time spent being intentional about changes isn't valuable when compared to the time spent by a reviewer (or your future self) to sort through the flotsam and jetsam of your intermediate work?

Nullabillity 7y ago
I'm using it to differentiate between summarizing (removing steps between A and B) and modifying (introducing new steps, reordering them, or editing them). You can do it in a way that isn't harmful (as I mentioned earlier), but good luck getting a team to actually stick to that. It also doesn't help that pretty much no tooling encourages doing it properly.
- avar 7y ago
  Are you lying to your co-workers when you draft an E-Mail to them, read it over, and decide to delete a paragraph or write it again from scratch? If your E-Mail client automatically saves drafts that's basically the equivalent of "rebase". I made a typo when writing this reply, and pressed backspace to correct it. Is use of the backspace key lying? I think you're placing a value on "history" that doesn't map onto all users of "rebase", or E-Mail client drafts. A lot of advanced users use it as the equivalent of "save" in an editor, sharing all those intermediate states is more noise than value v.s. crafting a sensible patch once you figure out what you want/what change to make.
  - Nullabillity 7y ago
    No, if you only merge changes then you're just summarizing. The true problems begin once you start creating commits that represent repository trees that you never tested or reviewed, for example by editing past commits (invalidating any testing you've done of commits after that point), deleting past commits (aside from squashing an unbroken sequence of commits, or deleting them if the squash would result in a no-op), reordering commits, or rebasing commits.
    - avar 7y ago
      You're assuming that commits are tested before they're made, and that rebase invalidates this. I don't test most of my commits, just like I don't proofread an E-Mail after every word I've written. I do that later. But yeah, the history you push to a canonical branch should generally be made up of commits that have all been tested in isolation. The rebase command doesn't make this worse, but better, e.g. with "rebase -i --exec='make test'". I also prune out history of some false steps taken. Have you never written a program and done something like "I'll use a hash here <save><compile><test>, no actually a list makes more sense <save><compile><test> ...". Those intermediate steps are commits for a lot of advanced git users. Sharing all your mistakes-as-you-go-along with the world doesn't help anyone, I'd typically be sending you a 100 patch merge request for some rather trivial change instead of 1-3 sensible commits.
    - int0x80 7y ago
      That kind of rebase is just to refresh your work against updated masters etc. Of course you have to test against the refreshed (rebased) work again!! That doesn't mean you can't rebase. It means you can't randomly push untested work. If you know what rebase means you will understand that there are very likely new interactions with your code and you have to test your updated changeset. Exactly the same as if you merge. You have to retest the resulting tree.
      - meribold 7y ago
        A merge of a branch with N unique commits creates one new, yet-to-be-tested commit/tree. A rebase creates N. I doubt that it's common that people replay all the new history after a rebase and test each new commit/tree.
fulafel 7y ago
But reviewers in most Git workflows mainly look at PRs. Then if as a reviewer you want to see how the sausage was made, you can zoom in on the commits, including all the messy reality of how the work was done. In some circumstances you might of course want to hide this, but in an open and safe collegial environment this lets the reviewer understand your thought and work process.
- raimue 7y ago
  The reviewer also won't see all the things the author tried without ever committing these states. They don't need to see all the messy steps, or they would have needed to look over the authors shoulder all the time. I rather review the final patch series with changes in logical order and not necessarily in the order the code was written or with intermediate work that was later reverted or changed again. I do look at commits, because also every commit message counts and is supposed to explain the individual change.
  - fulafel 7y ago
    Sure, it's not a perfect record. But as in most things, perfect is the enemy of good. The messy reality is valuable, when talking with your teammates about how the work was done and what kind of bumps were along the way. It's not about looking over their shoulders, it's about using data to develop together as a team, eliminating hinderances, etc - if you have the mutual trust to do that. And of course you yourself can go back and look for patterns of mistakes or problematic areas in code based on your history. Like I said, to judge the change its, the whole PR diff is usually the most useful unit of inspection when you just want to see what happens. And if it's a big pr, you can of course always merge child PR's or branches against the big PR/branch, and look at the merge diffs.
    - fulafel 7y ago
      Another formulation of the "learning as a team" idea- The science principle of publishing your experiments, including failed ones, has the same benefits in sw engineering: others can build on your failed attempts, or save time by not replicating them.

Nullabillity 7y ago

I'm using it to differentiate between summarizing (removing steps between A and B) and modifying (introducing new steps, reordering them, or editing them). You can do it in a way that isn't harmful (as I mentioned earlier), but good luck getting a team to actually stick to that. It also doesn't help that pretty much no tooling encourages doing it properly.

avar 7y ago
Are you lying to your co-workers when you draft an E-Mail to them, read it over, and decide to delete a paragraph or write it again from scratch? If your E-Mail client automatically saves drafts that's basically the equivalent of "rebase". I made a typo when writing this reply, and pressed backspace to correct it. Is use of the backspace key lying? I think you're placing a value on "history" that doesn't map onto all users of "rebase", or E-Mail client drafts. A lot of advanced users use it as the equivalent of "save" in an editor, sharing all those intermediate states is more noise than value v.s. crafting a sensible patch once you figure out what you want/what change to make.
- Nullabillity 7y ago
  No, if you only merge changes then you're just summarizing. The true problems begin once you start creating commits that represent repository trees that you never tested or reviewed, for example by editing past commits (invalidating any testing you've done of commits after that point), deleting past commits (aside from squashing an unbroken sequence of commits, or deleting them if the squash would result in a no-op), reordering commits, or rebasing commits.
  - avar 7y ago
    You're assuming that commits are tested before they're made, and that rebase invalidates this. I don't test most of my commits, just like I don't proofread an E-Mail after every word I've written. I do that later. But yeah, the history you push to a canonical branch should generally be made up of commits that have all been tested in isolation. The rebase command doesn't make this worse, but better, e.g. with "rebase -i --exec='make test'". I also prune out history of some false steps taken. Have you never written a program and done something like "I'll use a hash here <save><compile><test>, no actually a list makes more sense <save><compile><test> ...". Those intermediate steps are commits for a lot of advanced git users. Sharing all your mistakes-as-you-go-along with the world doesn't help anyone, I'd typically be sending you a 100 patch merge request for some rather trivial change instead of 1-3 sensible commits.
  - int0x80 7y ago
    That kind of rebase is just to refresh your work against updated masters etc. Of course you have to test against the refreshed (rebased) work again!! That doesn't mean you can't rebase. It means you can't randomly push untested work. If you know what rebase means you will understand that there are very likely new interactions with your code and you have to test your updated changeset. Exactly the same as if you merge. You have to retest the resulting tree.
    - meribold 7y ago
      A merge of a branch with N unique commits creates one new, yet-to-be-tested commit/tree. A rebase creates N. I doubt that it's common that people replay all the new history after a rebase and test each new commit/tree.

fulafel 7y ago

But reviewers in most Git workflows mainly look at PRs. Then if as a reviewer you want to see how the sausage was made, you can zoom in on the commits, including all the messy reality of how the work was done. In some circumstances you might of course want to hide this, but in an open and safe collegial environment this lets the reviewer understand your thought and work process.

raimue 7y ago
The reviewer also won't see all the things the author tried without ever committing these states. They don't need to see all the messy steps, or they would have needed to look over the authors shoulder all the time. I rather review the final patch series with changes in logical order and not necessarily in the order the code was written or with intermediate work that was later reverted or changed again. I do look at commits, because also every commit message counts and is supposed to explain the individual change.
- fulafel 7y ago
  Sure, it's not a perfect record. But as in most things, perfect is the enemy of good. The messy reality is valuable, when talking with your teammates about how the work was done and what kind of bumps were along the way. It's not about looking over their shoulders, it's about using data to develop together as a team, eliminating hinderances, etc - if you have the mutual trust to do that. And of course you yourself can go back and look for patterns of mistakes or problematic areas in code based on your history. Like I said, to judge the change its, the whole PR diff is usually the most useful unit of inspection when you just want to see what happens. And if it's a big pr, you can of course always merge child PR's or branches against the big PR/branch, and look at the merge diffs.
  - fulafel 7y ago
    Another formulation of the "learning as a team" idea- The science principle of publishing your experiments, including failed ones, has the same benefits in sw engineering: others can build on your failed attempts, or save time by not replicating them.