I was actually there when Jordan gave that talk. He specifically mentioned `launchd', rather than `systemd', as being something to look at. In fact, people in the FreeBSD community already have `launchd' running as PID 0, though I believe it's not fully stable. Right now, it just execs `rc' so most things just work as usual; individual services will have to be migrated to get started via `launchd', but that will take time.
It used to be until tons of old classmates from elementary school showed up on my facebook account pretty much devaluating the meaning of "friend"
Then why did you accept the friend request? I limit my FB friends to pretty much only people I'd enjoy spending the whole day hanging out with.
Okay, there are also a few few relatives, whose friend requests I accepted to not start a bunch of family drama. But I largely ignore their posts, and I have groups set up such that they can't see the vast majority of my posts either. As far as those folks are concerned, I'm rarely on FB at all.
I think they were among the first to support USB in an era where PC makers were slavishly doing nothing new because nobody else had done it yet.
Apple fought USB with their own proprietary connector called Firewire. Firewire was significantly faster than USB, especially at sustained transfers, but it was more expensive to implement because of a combination of the technologies involved and Apple's license fees. Apple ended up abandoning that technology for newer versions of USB and eventually Thunderbolt, mostly due to lackluster third-party support for Firewire devices.
That is pretty fundamentally wrong. USB and FireWire were intended to be complementary connectors and protocols. USB was intended for inexpensive and low-bandwidth devices, master-slave connectivity, no DMA - the type of things that would previously have been connected with PS/2, DB9, or DB25. FireWire was intended for things that needed peer-to-peer connectivity, bandwidth and latency guarantees (like video - IEEE 1394 is part of the DV standard), and DMA - things that would previously have been done with something like SCSI or PCI. Both interface coexisted for years on Macs and PCs.
It so happens that later versions of USB have added some of FireWire's features, to the point where USB2 and USB3 can do most - but still not all - of the things that FireWire can do. Because the chipset vendors included USB in commodity chipsets, but FireWire required a discrete chip, USB has better market penetration.
Thunderbolt is another thing entirely - it's essentially 4x PCIe-Gen2, using an interface that allows for piggybacking DisplayPort as well. (For example, the Thunderbolt RAID enclosures that you see basically contain a PCIe RAID controller that shows up on the PCIe bus when you connect it.)
He never said that. The movie you think you're quoting doesn't exist. I bet you think they made sequels to The Matrix too.
When it takes off for Jupiter
is when I get interested
Tsien would have left a few years ago.
How come Christians are the only ones that are asked to suffer this unconstitutional prohibition on their free exercise?
Did you miss the part where the second item in the lawsuit was a Menorah?
Don't take it out on others just because you're imperfect and ignore all of the S.M.A.R.T. and controller warnings... Some of us tech Gurus do religiously tend to our flock of hard-drives and recognize when they are in spiritual, and physical, need of replacement....
I do hardware diagnostics for an HPC storage system vendor, including drive testing, qualification, and failure analysis. SMART has its uses, but, in my experience, if you've tripped SMART, you're already in serious trouble.
It's not helped by the fact that pieces of SMART which are actually in the ATA standards are basically tripped/not-tripped - none of the attribute structures are in the current specs, let alone which attribute IDs mean what. Heck, even getting the thresholds is no longer in the ATA standard!
Fortunately, *most* vendors implement *most* of the SMART structures the same way, and *most* use the same attribute IDs to mean the same things, and *most* still implement the sub-command for getting the thresholds. But all the really interesting data that could be used for more aggressive failure prediction (i.e. beyond the almost-always too late SMART trip) is vendor specific, and getting that information out of some vendors makes pulling teeth look like taking candy from a baby.
with only two manufacturers of 3.5in form factor hard drives (seagate, wdc)
Actually, it's Hitachi. And Toshiba has (re?)entered the 3.5" space, at least on the enterprise side of the world - I have no idea if they make consumer 3.5" drives.
... And also, does the above mean that Gnome is no longer using GCC to compile, but switching to the LLVM compiler?
LLVM is designed to be modular. It sounds like what they're doing is probably similar to what Apple did a few years back - include LLVM bit-code files for functions that aren't handled natively, then hand those off to libllvm to emit native code when needed.
I've read comments in past stories about the iOS developer program from people who own a Mac but still can't develop because the Mac is too old to run recent Xcode. So you end up having to depreciate the Mac and the iPod touch on which to test as annual expenses just like the developer fee.
Apple generally supports the current OS and developer tools on hardware going back three years, for both Macs and iPods/iPhones.
If you've determined to never pay back all your treasury bonds, how can treasury bonds be a stable investment?
You pay them all back, you just also issue new ones, which you will also pay back.
... (Only works in browsers which recognize the data: URI scheme. Tested in FF and Opera.)
Works in Safari 5.1.1 (on OS X Lion) as well.