I use Graphite at {{ day job }} and it's pretty good. I strongly dislike having to use their git wrapper CLI `gt` though. git does already support this out of the box.
So instead, this uses git primitives and just drops that handy comment in Github to visualize the stack.
Hey! Mostly just rebase out of habit actually, but I've been exploring --update-refs recently.
Two things come to mind that I don't love about `gt`:
- the philosophy of "every commit is a PR" falls apart sometimes. Sometimes I want to logically separate commits in a PR—but not every commit passes CI. This makes it easier to review. Or call out optional changes that can easily be dropped.
- It broke my workflow in a few ways. The one thing I notice the most is that I like to "pop" a commit into staged changes and make edits. So I can easily see a diff of what I'm editing. I expected `gt modify` to do this. So instead I git reset --soft, commit, and `gt submit`
Here is total employment for "Computer Systems Design and Related Services" going back all the way to 1987. This is the best category I can find that includes data over a decade+ time frame.
This category is defined by the BLS as "establishments primarily engaged in providing expertise in the field of information technologies through one or more of the following activities: (1) writing, modifying, testing, and supporting software to meet the needs of a particular customer; (2) planning and designing computer systems that integrate computer hardware, software, and communication technologies; (3) on-site management and operation of clients' computer systems and/or data processing facilities; and (4) other professional and technical computer related advice and services."
Thanks for reading everyone. I’ve gotten some feedback over on Reddit as well that the example is not effectively showing the benefits of SIMD. I plan on revising this.
One of my goals of writing these articles is to learn so feedback is more than welcome!
What's fun is that, as the use of SIMD in your example is useless, LLVM correctly completely removes it, and makes your "neon" and "fallback" versions exactly the same - without any SIMD (compiler explorer: https://godbolt.org/z/YWoMGoaxT).
As an additional note, aarch64 always has NEON (similar to how x86-64 always has SSE2; extensions useful to dispatch would be SVE on aarch64 and AVX/AVX2/AVX-512 on x86-64), so no point dynamically checking for it.
> One of my goals of writing these articles is to learn so feedback is more than welcome!
When I went into the Rust playground to see the assembly output for the Cumulative Sum example, I could only get it to show the compiler warnings, not the actual assembly. I'm probably doing something wrong, but for me this was a barrier that detracted from the article. I'd suggest incorporating the assembly directly into the article, although keeping the playground link for people who are more dedicated / competent than I am.
That link is what are normally called vanity nameservers. That allows "branding" them so that "dig NS foo.com" says ns1.foo.com and ns2.foo.com instead of pinky.ns.cloudflare.com and brain.ns.cloudflare.com. But CloudFlare is still the provider.
What you cannot do is set your domain's nameservers to some other other provider, e.g. Route53. There's just no option to tell the registry you want to use non-CloudFlare backed values (outside of "call us" enterprise plans).
This is particularly nefarious when combined with buying a new domain. New domains can't be transferred to another registrar for 60 days, so if you need a DNS feature or config CloudFlare can't provide on a domain you just bought, you're just totally stuck for 2 months.
I was similarly confused, I recall seeing this in the UI -- looking again, it seems I need to upgrade to the business plan.
That's disheartening. Pay more for us to do less.
The document you've provided extends my confusion.
Both the 'Primary (Full setup)' and 'Secondary DNS' pages it calls out... seem to indicate that CloudFlare has to stay in the mix. Either the authoritative nameserver (defeating the point), or as a child receiving transfers.
This feels deliberately obtuse.
Edit: The peer comment from V99 helped me understand. This is their vanity solution - still CF. Reportedly cannot place the SOA elsewhere
So instead, this uses git primitives and just drops that handy comment in Github to visualize the stack.
reply