Hacker Newsnew | past | comments | ask | show | jobs | submit | more tuatoru's commentslogin

> It’s a bit like trying to drive faster on a congested motorway.

> The solution is Continuous Delivery – being able to reliably and safely ship to production in tiny chunks, daily or even multiple times a day – one feature, one bug fix at a time. Not having to batch work up because testing and deployment have such a large overhead.

"Just one more lane, bro; that'll fix it."


Yeah, no. It'll have vanished next year.


Jujutsu’s changelog (https://jj-vcs.github.io/jj/latest/changelog/) goes all the way back to 2022 and shows there was a release as recently as two weeks ago. I don’t see why the maintainers would stop at this point.

Also, from Jujutsu’s README (https://github.com/jj-vcs/jj#mandatory-google-disclaimer):

> I (Martin von Zweigbergk, martinvonz@google.com) started Jujutsu as a hobby project in late 2019, and it has evolved into my full-time project at Google, with several other Googlers (now) assisting development in various capacities.


Wow, 2022? Such age, many longevity.


Why would you think that?


This seems like one of those "hmmm, that's odd" moments that could lead to a scientific breakthrough.


It surprises me that economists are not advocating that Europe set up special economic zones, as have been tried in less-developed countries to some success. Hong Kong could be considered the Ur-SEZ.

On reflection it is not surprising that SEZs have not been tried. They would be too embarassing to Europe's political class.


This seems like an argument from incredulity, and also static-world fallacy.

People are not thinking about the trend in AI. How good were AIs three years ago? How good are they now? How good will they be in another three years?

Kelsey Piper's analogy is good. How much smarter is a teacher than a room full of kindergartners? Who gets whom to do their bidding?


A future we didn't quite make it into.


PIC is short for PICTURE, which was the original form of the keyword. Picture the number printed this way.

Edit: the numbers in parentheses are repeats. PIC 9(9) means up to 9 digits in the printed representation of the value stored in the variable that has this PIC.


Microsoft's moat and cash cow is Active Directory.

If corps ever stop needing that, MS will be in trouble. Until then, few worries. They can play at following fads (not that I think AI is a fad) and not worry about execution.


Active Directory comes free with Windows Server, it was never a "cash cow", that would be the Office suite.

Microsoft is actively killing of Active Directory and they're replacing it with Entra ID, which is "just" OAuth and hence is relatively easily replaced by a competing product.


Windows Server 2025 included many AD changes that were non-cloud related. They're not actively killing it.

https://learn.microsoft.com/en-us/windows-server/get-started...


Most of those are scalability / security improvements that are only relevant to fortune 500 sized megacorps.

These are the kind of release notes you'd expect to see for a dying product on life support, used only by those few remaining customers that are too big and too invested in it to migrate away.

Also, the Server 2025 release is the first one to change anything in Active Directory, the 2022 and 2019 versions had essentially nothing in the release notes for AD.

A few performance tweaks and a long-overdue security catch-up or two in a decade is not a vibrant, living and breathing product. It's a shambling zombie.


That's a weird take but you're entitled to it, of course; normally we see products die with zero improvements for many years. I'm well aware of the rel history of AD, but thanks for the re-education.


Well to be fair, Entra is still backed by AD behind the scenes as the database. But yeah, I’m also not sure a directory server is still required to store domain objects; tree structures are pretty much solved in all kinds of databases and don’t warrant LDAP anymore anyway.


It takes companies a long time to reorganise processes. Especially in ways that reduce headount, because manager status is tied to count of reports.

This study is far too soon. In 2030, if nothing changes meanwhile, we might know the effect of AI as we have it now in 2025 on employment.


If AI takes a fraction α of jobs, for it to be employment increasing, elasticity of demand for labor needs to be 1/(1-α).


Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: