Y
HN Search
Hacker News Search
new
|
comments
|
top
|
jobs
teacup50
searching PlanetScale…
1.
▲
2.
▲
3.
▲
4.
▲
5.
▲
6.
▲
54 ms
·
511.
▲
by
teacup50
12y ago
I don't understand why you would host your source code externally? You can buy Atlassian Stash, once, for $1800/25 users, and use it forever: https://www.atlassian.com/software/stash/pricing
512.
▲
by
teacup50
12y ago
And yet, you're paying those engineers (and every other employee) to waste their time using N different non-integrated services, all of which change at random whenever a new release is pushed out, disrupting your processes. I've b
513.
▲
by
teacup50
12y ago
Heck, there are paid , commercial software suites that are far, far, far, cheaper than this.
514.
▲
by
teacup50
12y ago
This is horrible, but not because of the Java part — it's the webview that puts your desire for a single UI technology stack ahead of user's desire for a decent UX.
515.
▲
by
teacup50
12y ago
> So honest question: what's different this time? We've rediscovered ahead-of-time compilation.
516.
▲
by
teacup50
12y ago
As you discovered, _stret/_fpret remain necessary. objc_msgSend()'s type is not expressible in C; the only time it is correct to call objc_msgSend without a cast is when calling a method IMP with the literal vararg type of id (*me
517.
▲
by
teacup50
12y ago
Why would a port of a more unstable Vim with hacky IPC-based plugin mechanism be "sprightlier"? Dropping support for operating systems and reformatting the code doesn't make it run faster.
518.
▲
by
teacup50
12y ago
Which is why Docker is an ultimately bad idea; OS-specific binary images are the wrong level of abstraction for managing the deployment of code that's otherwise completely portable.
519.
▲
by
teacup50
12y ago
This is wonderful. I would invest in something that took it a step further; turn it into self-sustaining non-profit or co-op that's focused on keeping people in their homes and able to repay the costs of repurchasing them.
520.
▲
by
teacup50
12y ago
If by "changes to the kqueue" API you mean "adding a new filter type, as kqueue was designed to support", then how is the kqueue API poorly designed? Why should the BSDs adopt a completely separate inotify hack? "10
521.
▲
by
teacup50
12y ago
That requires that you have control over all threads, which is just not something that can be guaranteed with modern operating systems and libraries that regularly spawn their own threads. Thread-interrupting delivery of signals is necessar
522.
▲
by
teacup50
12y ago
You're absolutely 100% full of crap. I already explained how kqueue's flexibility means that an EVFILT_DIR can be added easily, and work with all other event mechanisms on the system. > The Linux devs had 5 years to come up
523.
▲
by
teacup50
12y ago
The OS has to pause a thread that triggers a signal; how can it proceed if doing so would just segfault again?
524.
▲
by
teacup50
12y ago
Why do you keep ascribing EVFILT_VNODE's limitations to the the fact that the kqueue API is well designed and extensible? The inotify designers had 5 years to figure out the constraints and implications of file system notifications --
525.
▲
by
teacup50
12y ago
Would be saner to extend kqueue and hang a Linuxlator inotify compatibility shim off of that; that'd isolate the mess to sys/compat/
526.
▲
by
teacup50
12y ago
Bah. This is complete nonsense. EVFILT_VNODE is not complicated to use for the cases for which it was actually designed — monitoring a file descriptor. For more general file system monitoring, a new kqueue filter type would be ideal, but
527.
▲
by
teacup50
12y ago
I don't understand your point; kqueue is a comparatively new idea. Do you want signals written as messages across ZeroMQ? In that case, look at Mac OS X's use of Mach ports for exception handling, which is an even older idea t
528.
▲
by
teacup50
12y ago
Here's my friendly reminder. If you only pay attention to linkbait headlines and vapid blog content, you're going to have a hard time finding well-written software amidst the sea of well-marketed crap.
529.
▲
by
teacup50
12y ago
> Git and many other VCSs create noisy diffs whenever space is added and forgotten, which ultimately complicates the life of developers that want to review changes. That's because Git is stupidly opinionated about end-of-line whit
530.
▲
by
teacup50
12y ago
Who cares? Get a decent editor that doesn't give a crap if there's invisible whitespace at the end of a line.
531.
▲
by
teacup50
12y ago
> No charity is required. Infact it is profitable to hire this guy. What is required is getting rid of pointless constraints. Pointless constraints like being a reliable employee? > Part time? You get a talented developer for 30 ho
532.
▲
by
teacup50
12y ago
Really? Sickening? There are any number of candidates available that won't require special consideration to manage a disease, so why should someone solely take on the charity burden of giving this fellow a job?
533.
▲
by
teacup50
12y ago
These plates demonstrate a pretty myopic naive design view; Washington, for instance, is a lot more than just the Seattle skyline. The current Washington plates feature Mount Rainier: http://www.plateshack.com/y2k/Washi
534.
▲
by
teacup50
12y ago
> If threads aren't interesting except for parallelism, what do you think people were using them for the multiple decades when the vast majority of people were running single-cored uniprocessor systems? They largely weren't
535.
▲
by
teacup50
12y ago
This is a completely pointless semantic distinction that nobody cares about. What makes threads interesting is parallelism; if the goal is merely cooperative or pre-emptive multitasking, where there's no concurrency, then you're
536.
▲
by
teacup50
12y ago
More that Facebook's return on the amount of money they've spent on PHP is pretty poor. Facebook has had the author of jemalloc locked up working on making PHP go faster for a few years now. Surely there's a more productive p
537.
▲
by
teacup50
12y ago
D-Bus is an amazingly crap implementation of this idea.
538.
▲
by
teacup50
12y ago
> You may disagree that OSS should move fast, of course, but why should you want to slow down others? This is a nonsense false dichotomy.
539.
▲
by
teacup50
12y ago
Seeing as he explicitly referenced launchd , I'd say it was an argument for something like launchd , NOT systemd .
540.
▲
by
teacup50
12y ago
> HN don't throw stones, what boundaries are pushing to get traction right now? We're not pushing ethical boundaries.
More ›