Hacker News new | past | comments | ask | show | jobs | submit | setheron's comments login

I really enjoyed Snippets at Google which was a company wide way to share your changelog.

There were also some nice integrations to pull in commits, doc edits , etc..


The jj community on discord has been great also. It took a while for it to sync in but when I use git now it feels "wrong"

At University of Waterloo we used scheme

Long classpaths with unnecessary direct dependencies (i.e. not runtime) cause surprising slowdowns due to lookup performance.


I forgot to mention, I wrote about it https://fzakaria.com/2024/11/08/jvm-boot-optimization-via-ja...

I'm kind of intrigued by Mill but I've fallen into the same trap I've observed in others. I'm over indexed in mental capacity in having wasted learning Bazel and it's equivalent systems.

The lift to another system has to be enough to surpass that loss.


Been using jujutsu which does stacked diffs. Just the use of GitHub UI still a shame for stacked diffs and I'm unwilling to use graphite if I can't convince the rest of my team.


> I'm unwilling to use graphite if I can't convince the rest of my team

OOC, what are the main objections against?


Maybe try Aviator! (full disclosure I work here)


Every company has lore like this; I remember when I joined Amazon hearing how someone would create SEV-1 tickets because their phone didn't work (Jeff B. would get personally paged for all SEV-1 at the time)


Yea what's the citation? Nothing is stopping any company that "commits" to do a 180; they only need to cite changing conditions.

"Committing" something as a public trading company has little weight behind it. At the end of the day, they have a fiduciary responsibility to their shareholders and can do any changes in service to that.


Isn't it working here: https://cs.opensource.google/bazel/bazel/+/master:src/main/s...

I remember there were docs how to onboard a repo to that list.


Yea it's still there, that is backed by Kythe.


I remember at Oracle they built systems to shut down the previous presumed leader to definitively know it wasn't ghosting.


Yep, the "STONITH" technique [1]. But programmatically resetting one node over a network/RPC call might not work, if internode-network comms are down for that node, but it can still access shared storage via other networks... The Oracle's HA fencing doc mentions other methods too, like IPMI LAN fencing and SCSI persistent reservations [2].

[1] https://en.wikipedia.org/wiki/STONITH

[2] https://docs.oracle.com/en/operating-systems/oracle-linux/8/...


They had access to the ILOM and had some much more durable way to STONITH. Of course every link can "technically" fail but it brought it to some unreasonable amount of 9s that it felt unwarranted to consider.


Yep and ILOM access probably happens over the management network and can hardware-reset the machine, so the dataplane internode network issues and any OS level brownouts won't get in the way.


So for us normies does that mean skip TLA?


None of the replacements I've seen so far are mature enough for me to recommend them over TLA+.


I believe this is really the tragedy of formal verification tools. Everybody wants a tool as robust as a compiler. At the same time, nobody wants to invest into development of such tools. Microsoft Research 20 years ago was probably an exception to that. The other companies wish to immediately hide these tools and the benchmarks behind the IP and closed source. As a result, we have early stage MVPs that are developed by 1-3 people.


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

Search: