Spending $17 million to replace git

[drhowarddrfine]

I'll do it for $9 million.

 

[OpenFreeNet]

I didn't understand what the real problem is with Git?

Modern Git replacements/extenders like Sapling, Jujutsu, Gitbutler usally add to Git these features: you have a stack of commits and you can improve them, iteratively, adding things to commits, splitting or folding them; the way you fix merge conflicts is saved, so when you change the commit history again or receive changes upstream, many merges can be reapplied automatically; the history of changes is saved in a log and you can undo/redo and review what you have changed in the repo, not only in the files.

It is easier having a polite history, because commits (expecially not pushed commits) are more "fluid".

Sapling has very few commands, so the CLI is a lot simplified respect Git.

 

[elephant]

The Atlassian stack works well for me. Most of my interaction is a shell prompt with git while Atlassian's hooks into the git repo do the rest. I wish these people well but don't expect to be shopping for a replacement soon.

 

[doul]

OpenFreeNet ok, thanks for your explanations, I started learning git a year ago, today I use about ten commands, it's enough for me alone. Big teamwork should have more issues.

 

[aphor]

It just seems like programmers keep trying to follow some bus somewhere. Like greener pastures when they don't exist.
There is always some better scheme to sell.

VersionControl - FreeBSD Wiki

The history of git is interesting. There are ancillary lessons to learn too.

 

[thogard]

I have always considered git to be a pain in the rear.

I find the terms in git just wrong. I think the verb tense direction is often wrong.

I've had Swedish and Finnish flatmates use the English phase "Will you borrow me 20 dollars?" enough times to lock it in my head as a verb direction issue. Git seems to have taken it to a new level.

As far as someone spending millions to fix it... Can we find a way to pay the linguist Larry Wall $1m to fix the terms using only sed s/oldterm/newterm/g to create the patches plus fix the core docs?

 

[drhowarddrfine]

I've had Swedish and Finnish flatmates use the English phase "Will you borrow me 20 dollars?"

I've heard Americans say that. Another phrase that drives me nuts is, "Me and my friend are going to the store". I learned in my first years of schooling that you put the others first in a sentence. "My friend and..." You can then check the phrasing by leaving the 'friend' part out like, "Me going to the store." You wouldn't say it that way.

Of course, you can't correct people cause they would be offended.

 

[thogard]

Another phrase that drives me nuts is, "Me and my friend are going to the store".

That is from dating where "my friends" is a secondary noun group with the "me" and the unspoken noun "you" are the primary. We could fix this with stuffing suffixes on worlds like latin. Maybe we can get the author of Perligata to look at the git syntax too.

 

[OpenFreeNet]

I've heard Americans say that. Another phrase that drives me nuts is, "Me and my friend are going to the store". I learned in my first years of schooling that you put the others first in a sentence. "My friend and..." You can then check the phrasing by leaving the 'friend' part out like, "Me going to the store." You wouldn't say it that way.

Language is fascinating. I'm Italian, and I would say something like "I'm going to the store with my friend", but "Me and my friend are going to the store" implies the fact that both me and my friend are interested into buying something, while the former not necessarily.

 

[mer]

If everyone spoke american english, it would be so much easier.
Maybe not as much fun, but easier.

And before anyone gets upset, that is a joke.

 

[JohnK]

If everyone spoke american english, it would be so much easier.
Maybe not as much fun, but easier.

And before anyone gets upset, that is a joke.

I agree with mer, everyone should write computer programs!

...because programming languages are based on english.

 

[scottro]

Of course, you can't correct people cause they would be offended.

Your experience is a bit different than mine. I've found that most non native speakers prefer to be corrected. (Though privately, if possible).

 

[mer]

I agree with mer, everyone should write computer programs!

...because programming languages are based on english.

Rust, Algol, Cobol, PL/1, LIsp, Prolog, Befunge, Piet?

scottro
Correcting: I think most engineers wind up working with lots of different non-native English speakers and our brain does the automatic translation stuff on odd syntax. My experience has been correcting, done respectfully, perhaps as a question, is well received.

Me, I've never had a problem asking someone to repeat themselves or correct my pronunciation because I'd rather get a name correct (or as close as my mouth allows) than assume an easy for me to say nick name.
Heck, at the local really good family owned Mexican place we eat at, I point at the Conchinita Pibil on the menu try to say it and accept laughs and corrections.

 

[blackbird9]

I agree with mer, everyone should write computer programs!

...because programming languages are based on english.

You haven't worked with any Chinese Perl programmers, have you...

 

[drhowarddrfine]

scottro I was talking about American populated sites where the educational opportunities passed them by, apparently.

 

[scottro]

Ah, I could see that. To comment about it would get into the realm of politics. Was it Barnum who said no one ever went broke overestimating the stupidity of the American public. (Meaning USA).

 

[rbranco]

It's a fancy wrapper around git branch, revert, worktree, stash, etc. You can have "unlimited undo" just by reverting.

 

[Espionage724]

I've heard Americans say that. Another phrase that drives me nuts is, "Me and my friend are going to the store". I learned in my first years of schooling that you put the others first in a sentence. "My friend and..." You can then check the phrasing by leaving the 'friend' part out like, "Me going to the store." You wouldn't say it that way.

Of course, you can't correct people cause they would be offended.

I know the rules, but saying it like that casually sounds odd

I might phrase it "Me and some friends are headed to the store" (implying they're my friends but not double-subjecting)

 

[scottro]

For non native speakers, a lot of grammar rules seem to be ignored these days, as language becomes more casual and more stupid people get into positions of power, mocking those who know English grammar as being intellectual and saying how they like the uneducated. But, that could be a whole separate thread. As drhowarddrfine says, it was taught, at least when I was in school, that one put themselves last, friends and I or friends and me, depending upon case, and I was taught the same rules drhowarddrfine mentions.

The putting me or I first is usually a sign that a person hasn't been taught correctly, or these days, wants to show that they're not educated. If they're non-native speakers of course, it's less important in my opinion.

 

[OpenFreeNet]

The putting me or I first is usually a sign that a person hasn't been taught correctly, or these days, wants to show that they're not educated. If they're non-native speakers of course, it's less important in my opinion.

Many of us (me for sure) are lazy. For learning and remembering subtle details of a foreign language, I had to make exercises.

When I open a random page from "English Grammar In Use", I (re)discover always something of interesting. For example "I like washing my hair" means "I enjoy washing my hair", while "I like to wash my hair twice a week" doesn't mean that I enjoy it, but it means that I like the outcome of the habit. If I didn't read this page right now, I would write "I like washing my hair twice a week", that is absurd, because it seems that "two is a magic number"

 

[scottro]

Oh, and by uneducated, I don't mean no college, this level of English is learned, or should be learned, well before then. And college doesn't guarantee education or brains. Lots of stupid people with college degrees are posting their ignorance for all to see. And many of them, if wealthy, went to supposedly good colleges.

 

[bakul]

That's what coding needs. It needs to be social.

Likely this is more about being aware of who (in your team) is working on what and being able to look at their work in progress. I take at least a cursory look at most every FreeBSD commit. It would be nice to know what changes are upcoming in some areas of interest, associated tests, PRs, documentation etc. Not sure how well this can scale when the "team" consists of 100+ people working remotely.

The underlying idea being building any complex structure is a group effort. As an example, consider how many different professions are involved in building a house, a skyscraper, a car or an airplane etc. Many professionals have to work in parallel, coordinate with each other but also get out of each other's way. Better tools & practices have evolved over time in other areas but software has a long way to go. So I see this effort as a step in the right direction (or at least, where I think the investors should push them!).

 

[Jose]

Modern Git replacements/extenders like Sapling, Jujutsu, Gitbutler usally add to Git these features: you have a stack of commits and you can improve them, iteratively, adding things to commits, splitting or folding them; the way you fix merge conflicts is saved, so when you change the commit history again or receive changes upstream, many merges can be reapplied automatically; the history of changes is saved in a log and you can undo/redo and review what you have changed in the repo, not only in the files.

I've been doing this with plain Git for at least ten years.

It is easier having a polite history, because commits (expecially not pushed commits) are more "fluid".

Sigh, nobody appreciates the finely curated commits in my merge requests. I try to group changes thematically so they're easier to review. I'm literally the only person who does that where I work.