bjackman

27 ms·

451.
by bjackman 2y ago
You can do everything in Git really. I don't see anything in jj that Git literally can't do, but as someone who spends a lot of time faffing around rebasing huge branches I can really see the appeal of something that does all the
452.
by bjackman 2y ago
Piper is an extended reimplementation of Perforce (Piper is piper, expanded recursively. It's so meta, even this acronym)
453.
by bjackman 2y ago
I have no opinion on whether JJ is good but I am pretty confident your impression about the reason for its shortcomings is wrong. Google has a basically tolerable and pretty easy-to-learn Mercurial-based frontend for its bizarre legacy Perf
454.
by bjackman 2y ago
Really glad these products are appearing! So much of my time for "reading" is in a context where I can't physically read, so audiobooks are incredibly useful. But being limited to the set of books that gets recorded by the pu
455.
by bjackman 2y ago
This isn't that, I'm not at all knowledgeable about this topic, it's just the basic level of curiosity you'd expect from an educated layman! 2 brain cells: "one hundy billy is a lotta dollars" 3rd brain cell: &
456.
by bjackman 2y ago
Yeah this is another question I had and another source of frustration with the journalists that write these pieces. Either they know more about this stuff than me and it's their job to predict that gap and explain this key detail... Or
457.
by bjackman 2y ago
This includes the actually interesting bit: > One of the thorniest questions in the conversion has been how the nonprofit will be valued. Musk’s bid sets a high bar and may mean that he, or whoever runs the nonprofit, would end up with a
458.
by bjackman 2y ago
I think a fully distributed storage system must be the way here. There must be some IPFS type system where Harvard could say "we designated a set of data that we can add to as needed but only delete from with a critical mass of storage
459.
by bjackman 2y ago
IIUC this even has logic to reconstitute TCP connections - https://criu.org/TCP_connection
460.
by bjackman 2y ago
Oh yeah I was just talking specifically about the seccomp-bpf bit. It's not comparable to nsjail as a whole.
461.
by bjackman 2y ago
It's still a reasonable comparison though. The seccomp-bpf is part of nsjail is achieving the same thing, one way to look at it is that Landlock/pledge are just a better implementation for the same approximate feature.
462.
by bjackman 2y ago
This is what the parent comment _means_ IMO. What are you are saying is: optimising for commercial success is incompatible with morality. The conclusion is that publicly traded megacorps must inevitably trend towards amorality. So yes, they
463.
by bjackman 2y ago
Regardless of whether it's faster, it's an extremely bad idea in a C program, for many of the reasons the author outlines. It's UB so you are forever gonna be worrying about what weird behaviour the compiler might create. Cla
464.
by bjackman 2y ago
CPU vendors always say this when an exploit is published before they mitigate. Sometimes they mean "no we don't think it's exploitable", sometimes the charitable reading is "we don't think anyone is exploiting
465.
by bjackman 2y ago
> I'm pretty sure both AMD[^1] and Intel[^2] have had predictive store-to-load forwarding. IIRC this was how Spectre Variant 4 worked.
466.
by bjackman 2y ago
No I'm not talking about mixed projects (although the ability to do that is very important). I'm saying that using it for separate C++ and Go projects is extremely convenient and ergonomic.
467.
by bjackman 2y ago
But this just doesn't match my experience with Blaze at all. For my internal usage with C++ & Go it's perfect. For the weird niche use case of building and packaging BPF programs (with no support from the central tooling teams
468.
by bjackman 2y ago
I always think it's a shame that these features end up getting built into ecosystem-specific build tools. Why do we need separate build systems for every language? It seems entirely possible to have build system that can do all this st
469.
by bjackman 2y ago
> The client stub is a concrete final class. It can't be mocked in tests. I believe this is deliberate, you are supposed to substitute a fake server. This is superior in theory since you have much less scope to get error reporting w
470.
by bjackman 2y ago
The fundamental issue is that GitHub thinks the artifact you are reviewing is code. But the artifact I want to review is a series of commits. I make review comments like "please split this into a separate commit", "please add
471.
by bjackman 2y ago
Not just in startups. Feels like Arm lost almost their whole software division over a few years. This was an extremely stable high-profit-margin business, and it was obvious that it was uncompetitive on the labour market. But something in t
472.
by bjackman 2y ago
I did the same, very quickly. I'm now on BlueSky and it's exactly like pre-Musk Twitter.
473.
by bjackman 2y ago
Can anyone comment on the code review experience? (I assume similar to Gitea but I haven't tried either). I recently did some moderately serious code review in GitHub and discovered it's a baby's toy version of a code review
474.
by bjackman 2y ago
Related: for anyone who speaks a bit of German, check out Texas German: https://youtu.be/1_dH403pqRU?si=KuSDR51w-_ZXmTdT Really lucky that these dialects persisted long enough that we have recordings of them.
475.
by bjackman 2y ago
Also sometimes for things that cost amounts I could easily expense I've been going through a purchase order process for a bunch of patch cables and adapters so cheap I could even expense them without a receipt. It's taking months.
476.
by bjackman 2y ago
I would even go so far as: if you see reactions like "this is shoddy work, the author should Just Do Better", you can generally surmise that the person making the comment has a poor understanding of fallibility and will make a lot
477.
by bjackman 2y ago
I am slowly inching towards trying out jj. I think the main issue will be if I find some issue in the git compatibility aspect. I do two kinds of work: 1. Work on smaller projects with simple histories. Git is perfectly satisfactory. 2. Wor
478.
by bjackman 2y ago
I have a fun exception to this! When writing BPF programs, sometimes it's tricky to get the compiler to generate code that passes the verifier. This can lead you down a path of writing bizarre C in order to try and produce the right or
479.
by bjackman 2y ago
Oh yeah I like ReligionForBreakfast!
480.
by bjackman 2y ago
I find it pretty cool how the spread of Christianity can be tracked so finely that a 50 update in earliest arrival time is exciting! I started listening to a podcast called "the history of the early church" to learn a bit more abo

More ›