riobard

38 ms·

421.
by riobard 13y ago
So the implication is that it doesn't consume local disk space if connected as network drive, right? That would be a very useful feature!
422.
by riobard 13y ago
Any study proving this?
423.
by riobard 13y ago
We definitely have very different ideas of “complicated”.
424.
by riobard 13y ago
Like finding a USB floppy drive and a floppy disk in 2013?
425.
by riobard 13y ago
None of the above has anything unique to do with testing on a floppy drive. You need a good way to simulate different network conditions, and a good inspector to check the various states on the client.
426.
by riobard 13y ago
Ha, I didn't know that! Where is the Developer panel on iPhone?
427.
by riobard 13y ago
It's 2013 and we are using a floppy drive to simulate high network latency. If that's not convoluted, I'm not sure what else could be.
428.
by riobard 13y ago
Define “performance“ and “cacheability”.
429.
by riobard 13y ago
That's really convoluted to simulate bad network. Here is an easy route if you are on OS X: get Apple's Network Link Conditioner [1], a System Preferences panel plugin. Once installed, tune your network to 3G Lossy and test again. [1]: htt
430.
by riobard 13y ago
Totally. I read the whole manual when I did my driving test, but I still got some a copy of the questions they'll ask to review just before the test. This is definitely a much better choice than some source-unknown papers :D
431.
by riobard 13y ago
My biggest worry is that people may rely too much on this instead of, you know, RTFM, by which they'll learn a lot of useful stuff, especially those concerning safety.
432.
by riobard 14y ago
Nor PowerPC, which apparently many NAS units are using. Mine is a WD My Book Live Duo running Debian/PowerPC. Additionally, AeroFS relies on Java, which is pretty big dependency (read: pain) to have on resource-restricted environments like
433.
by riobard 14y ago
Really hope someone could write a similar introduction to buildout ( http://www.buildout.org/ )
434.
by riobard 14y ago
Don't underestimate the desire of Wall Street for lower network latency. A 33% reduction in latency is a BIG DEAL when you do high-frequency trading.
435.
by riobard 14y ago
Someone should definitely do a Kickstarter project just to make such a screen.
436.
by riobard 14y ago
Seriously? NUDT (National University of Defense Technology) still names it Kylin? How shameless… A little bit of the old story for those who don't know: Around 2001~2002, NUDT, Lenovo Group, and a few other organizations took a government g
437.
by riobard 14y ago
“the entire process started up with only a few hundred KB's of memory (on startup)” There is no way this could be true: last time I checked, a bare minimal Go HTTP server requires at least 2.8MB of memory on 64-bit machine. Are you using th
438.
by riobard 14y ago
“it feels much more comfortable to actually have control over the resources you are using than relying on the magic the JRE does for you.” That's exactly the problem Java/JRE was trying to solve. People make mistakes. Given the large quanti
439.
by riobard 14y ago
Completely agree. APIs are so important for many optimizations to pull off. I'd really like to use a lot more buffer()/memoryview() objects in Python. Unfortunately many APIs (e.g. sockets) won't work well with them (at least in Python 2.x.
440.
by riobard 14y ago
Cool! This looks very interesting!
441.
by riobard 14y ago
Correct me if I'm wrong, but doubling the performance of the simulation from a dual-core to a quad-core processor doesn't necessarily mean you have ”linear scalability” unless you can demonstrate the trend continues as more cores are used (
442.
by riobard 14y ago
VM's IO performance is usually pretty bad. Try an SSD.
443.
by riobard 14y ago
“If you take the time and patience to understand your system and sift through man pages, configuration time will obviously shrink. If it's growing it's because you've never sat down to truly understand the cogs and screws of your system.” T
444.
Multiplexing on top of TCP(250bpm.com)
2 points by riobard 14y ago|0 comments
445.
by riobard 14y ago
Does it make any difference for append-only writes vs. in-place modifications?
446.
by riobard 14y ago
Does this mean that I have to msync multiple mmap'ed chunks in order to batch write? For example msync(ptr1, len1) msync(ptr2, len2) msync(ptr3, len3) Where [ptr1, ptr1+len1], [ptr2, ptr2+len2], ... are the chunks within a
447.
by riobard 14y ago
How do you batch writes to mmap-ed region?
448.
by riobard 14y ago
Honest question: for $430 why would anyone want this crap instead of a decent tablet?
449.
by riobard 14y ago
True, it's a 2560x1440 display, but it's not “retina“ yet. There is still a big difference with and without subpixel rendering, especially for smaller texts. Well, at least to my “retina” spoiled eyes. YMMV. iOS does not use subpixel render
450.
by riobard 14y ago
Unfortunately vertical subpixel rendering is much less effective than horizontal rendering due to the nature of latin scripts.

More ›