Furthermore there are many changes to NumPy internals, including continuing to migrate code from C to C++, that will make it easier to improve and maintain NumPy in the future.
I realise that C can be rather low level a lot of the time, but I’m not sure I’d pick C++ to help keep things easy to maintain. It opens up a Pandora’s box of possibilities.
Since Apple’s keynote this week.
You are, if you’re calling for Apple like features.
You might argue that “private cloud” is privacy preserving, but you can only implement that with the cash of Apple. I would also argue that anything leaving my machine, to a bunch of servers I don’t control, without my knowledge is NOT preserving my privacy.
I wouldn’t be so quick to write it off.
It’s a proof of concept showing the weaknesses in Microsoft’s vetting process for extensions published on the store. They then used the process to get pseudo-malicious code inside hundreds of organisations (not hundred of installs) some of which are high profile.
While I was vacating my colleagues…
I’m reminded of a scene from The Wire:
(I know. I removed a comma from the quote, but that’s how I read it)
It does matter in distributed application on LANs. The thin client model is still in operation at a lot of HPC and similar environments. Laptops and Desktops just display what is being done elsewhere.
Remote X11 is a better user experience in that environment than anything else I’ve tried. It feels like the application is local even if it’s not.
Still worth trying.
The word “support” is used in a very corporate way in the ROCm docs. They mean in terms of support contracts, sending engineers out to customer sites, having people dedicated to specific customers.
The code works on more devices than they list, but they won’t enter into a support contract with anyone using anything but their latest and greatest.
“set the magic environment variable because the tool chain will mis-detect the architecture of your unsupported card”
I don’t think it’s misdetecting it. Rather it detects it correctly, tries to use specific support for that device, but then finds that the support was switched off at compile time. The environment variable forces it to pretend to be a different (very similar) device.
Clunky, yes.
Does it have higher-order functions? Yes, therefore you can use it to do functional programming.
Everything else is syntactic sugar.
Egads! My eyes.
I’d rather it was just written in assembly. It’s the do { opening a block under the case 0, but then proceeding to have further case statements inside that block. You now have case statements in two different scopes that are part of the same switch.
In most CPU instruction sets, the only conditional instruction is branch (aka goto).
Not high. It tends to result in one of a few things.
They take the fix as offered. Probably it’s a smaller quiet project where the number of PRs is small. That, or very well run.
It remains forever open and gets lost in the masses of other out of date PRs. Maybe a bot comes along and closes it as stale. Biggest group, and these ones just tell me that I’ve got very little chance of getting the maintainer’s attention. I can see that 100s of others have experienced the same fate.
Somebody else finds it useful and adopts it, fighting for it to go in. Sometimes that someone is the maintainer. As you say, it can be inspiration for a rewrite of my contribution. That’s fine by me. Whatever works.
The maintainer makes a bunch of rework demands leading to rejection, or it gets rejected straight off. “My way or the highway” is always going to be highway. I offered a small piece of help, and if it’s not wanted I’ll happily go away.
Maybe 20% get in, but it depends on so many things.
Honestly, I run and gun. I make the change I want, and submit a merge request. I then move on. It’s then up to the maintainer to accept or reject it.
I’m not going to debate it. I’m not going to rework it over the course of months to make it perfect in the maintainer’s eyes. I don’t care enough about it. I’ve solved my problem. I’m just sharing it for others.
The things I submit are normally big fixes with the smallest possible code change, not refactorings to solve an underlying problem.
Second diagram, yes absolutely.
Short lived (1-2 day) branches, and a strong CI systems to catch regressions.
Be warned, the strength in the CI lies in its capacity to detect when some functionality that previously worked doesn’t work anymore. So, the flow must be green always, and it must evolve as the features evolve. Without good CI you’re destined for failure.
I don’t think anything works in that situation.
I keep trying catalog but it doesn’t do anything.
For me the answer is always “snapshots” and normally because of docker.
If you run a docker image store on a BTRFS drive, docker creates snapshots at various times. It never cleans them up; It has no commands that clean them up, and it means that if you delete a file it doesn’t free any space because the snapshots keep the file alive.
They’ve got problems in their tests though. There are 5 packages in core that are “unreproducible” and one is the kernel. Pretty significant package, except it fails because of a timeout.
That’s a big fucking “if”
Do you think a style guide is enough for an open source code base? Contributions could be coming from lots of directions, and the code review process to enforce a style guide is going to be a lot of work. Even rejecting something takes time.