I suspect that Apple has better than average odds of getting the 'per-application' buy-in; but we'll see how it goes. I'm inclined to be a touch pessimistic just because of the situation, as described above, where we've had the architectural underpinnings to do versioning/revision control for ages(and even fancy stuff like multiple nodes and network transparency and so forth has been there for years) and we've had applications where revision control and snapshot navigation are so overtly useful that they've been hacked in at the file format layer, also for years; but never hath the twain met(outside of IDEs/revision control systems and some very, very spendy stuff used mostly by law offices for revision control on legal documents).
Arguably, it has been "practically there now" for ages. Google Docs has it, Assorted revision-control systems have taken a bit of setup; but have had it(and boatloads of features besides) for years. VMS had it at the filesystem level back when a VAX was some pretty bitchin' hardware... NTFS volume shadow copies are in something of the same vein, as are the snapshotting features provided by most SAN vendors, Sun's ZFS, Oracle's BTRFS, etc, etc.
Obviously, the engineering challenges of doing robust versioning in an area where data-loss Just Isn't OK are nontrivial, and the amount of extra storage space required to move from a simple "here are your files" setup to a full versioning setup isn't small; but, as best I can tell, the issue isn't one of availability; but one of implementation/polish.
The revision-control stuff is great for programmers and techies(and even if you aren't programming, doing things like putting system config files under revision control allows for some neat tricks), The Google Docs stuff is super-easy; but limited to Google Docs, Shadow Copies have saved Windows shop admins countless trips to their backup tapes; but are essentially invisible to virtually all user-facing applications, same basic story with ZFS, BTRFS, and the snazzy SAN stuff.
Then you have individual programs that implement their own high-granularity revision tracking, totally separate from the FS/OS level stuff, within their own file formats(like Office with change tracking, Non-desructive edits in image manipulation programs, virtually anything that has an undo/redo stack of some kind, and so forth).
What does not seem to exist is anything where there is much cooperation across layers of the stack that makes up what the user actually faces off against: Somethin like a non-destructive image editor allows you to traverse back and forth across revisions; but its file format is just a basic blob as far as the FS is concerned. Something like Shadow Copy, Time Machine, FS snapshots, or revision control allows you to freely traverse the history of a file; but very few programs "know" about them in a useful way. I can use such a system to dredge up a snapshot of a file from 2 hours ago, and feed that to a suitable program; but the program will treat it exactly as though it were opening a file from the present. Only a few specific cases(like IDEs and revision control systems) show collaboration between the user-level program and the deeper abstraction level providing the versioning. Most everything else either does the versioning entirely within the program, with the OS/FS entirely ignorant of it, or entirely within the OS/FS/other low level mechanism with the user-level programs entirely unaware of what is going on.
Frankly, the "Recycle bin/trash" is a good(but not perfect) UI convention for a great many common computing situations. It serves as a reasonable recovery point/"in retrospect I fucked up" self help tool. It also imposes basically no requirements on the system/filesystem architecture. Would minor little additions(like having it automatically sort items by date/time of deletion) make it better? Sure. Is there anything fundamentally wrong? Not really.
Now, in high-resource environments, there is a much stronger case to be made that the "recycle bin" is obsolete and must die; but only if the system/FS/programs can be modified to accommodate the necessary changes. If, say, I mostly deal in text of some sort(writer, lawyer, programmer) even a modern laptop drive(never mind the cheap network storage) is Ludicrous Storage. Why can't I traverse a high-granularity timeline of every change made to every file I deal with?
That's the thing: For situations where you have to contend with legacy limitations, or where user space requirements are still comparatively close to the limits imposed by available technology, the "recycle bin" UI metaphor is pretty damn good. It is basically just a folder(with a tiny dollop of metadata to allow "restore this item") so it imposes minimal requirements; but still helps save users from common fuckups.
However, for environments where storage is massively ahead of traditional user requirements, and one has the liberty to re-examine certain legacy restrictions, you could arguably do much, much, better, with full versioning of all kinds of stuff.
I suspect that RIM is in for a nasty surprise if they want their tablet to succeed...
So long as their target market is "ugly; but secure and comparatively inexpensive email/cellphone appliances for suits" along with a sideline in "best keyboard for kids on cheap plans, now that sidekick is dead", they can afford to have a relatively sucky dev process. Their email/BBM stuff is 1st party, and much of the 3rd party development is by and/or for deep-pocketed corporate outfits that have hellishly complex requirements and processes anyway.
The problem arises if they want to play in the consumer market(or address the longterm/strategic threat that the more 'likeable' platforms will, either natively or in 'business' versions, harden up when it comes to security and management), they are going to have a very hard time attracting casual and indie developers to bring games and whatnot with their existing process.
The fact that they've gone and put out a tablet suggests that they do want to do this(Mr. Corporate Suit already has a laptop, and everybody in his office has the same power adapters/docking stations, that runs Outlook and all his horrid windows-only enterprise applications and a blackberry for mobile email. He will be picking up a tablet why? Tablet is a consumer move...)
B)UCMJ, Subchapter II, section 810, Article 10: "Any person subject to this chapter charged with an offense under this chapter shall be ordered into arrest or confinement, as circumstances may require; but when charged only with an offense normally tried by a summary court-martial, he shall not ordinarily be placed in confinement. When any person subject to this chapter is placed in arrest or confinement prior to trial, immediate steps shall be taken to inform him of the specific wrong of which he is accused and to try him or to dismiss the charges and release him."(emphasis mine)
UCMJ, Subchapter II, section 813, Article 13: "No person, while being held for trial, may be subjected to punishment or penalty other than arrest or confinement upon the charges pending against him, nor shall the arrest or confinement imposed upon him be any more rigorous than the circumstances require to insure his presence, but he may be subjected to minor punishment during that period for infractions of discipline."(emphasis mine, and I'm pretty sure that month after month of solid solitary doesn't count...)
Hilariously, Paypal was actually started by a libertarian as some sort of "resist the man and his fiat currency's dead hand on trade." kind of thing. Now it voluntarily licks the boots of those who would suppress the entirely legal efforts of an advocacy group to secure a man a fair trial(rather than the present detention-without-trial-of-indefinite-length...)
I'm very fond of that particular supercomputing site. Building your giant postindustrial blinkencluster in a transparent multistory chamber inside an authentic historic chapel has way more evil genius cred than any number of white cats...
At present, the only displayport features that Apple's displays handle are video and audio input. A separate USB cable is used to handle the USB hub, webcam, and microphone. There is no provision for daisy chaining. A few other monitors support displayport(generally in conjunction with other inputs, except for internal displays in laptops); but I do not know of a single one, on the market or promised, that treats displayport as other than an HDMI-equivalent video/audio cable, despite the theoretical possibility.
Apple is notoriously tight lipped about their plans; but I've never heard any mention of plans to change this(though I'm assuming that their next revision will, at least, eliminate the USB cable in favor of a thunderbolt based transport. I would certainly hope that they would support daisy-chaining as well..
I definitely find any claims of it being especially high performance deeply fishy, and the unveiling of it reeks of a PR stunt; but nothing about the (limited) hardware that is there looks definitely fake.
A bunch of 1-3Us, connected via rather scraggly patches from a few GigE switches to the on-motherboard GigE ports. Depending on the specs of the servers, we could be talking a few hundred cores, and a reasonable number of TB of storage...
It certainly is of abnormally low density, though. Each rack is 50% or more blanking plates, and I don't see any of the classic "cluster porn" shots of the Big Serious Core Switch, which they would definitely show if they could. No "Stylish rack opened enough to show all 72Us of servers exuding a sense of overwhelming power" shot or "View down the rack hallway that appears to go on forever, puny meatsack" shot either.
I'd totally believe that this is a (somewhat optimistic/growth prepared) conversion of some underused university basement to support a CS department curriculum and/or the compute needs of other disciplines; but it sure isn't what the PR says it is....
I'm aware that it can be daisy chained, I'm just concerned that that might end up being a dead letter(which would make having it block your video-out port an issue). As I noted, ordinary Displayport has supported daisy-chaining, and a general purpose AUX channel with nontrivial bandwidth, since 2009. Since then, crickets. Not even obviously-prototype FPGAs and blinking lights tech demos, never mind products that you can actually buy. Nothing. For all intents and purposes, you cannot daisy-chain displayport, and its 720Mb/s AUX channel is used entirely to carry the DDC data that an i2C bus was ample for.
Thunderbolt, as best I can tell, appears to be a displayport with a seriously buffed AUX channel. As I said, if Apple has a plan to set the industry on fire, great. If the only daisy-chaining support appears once Apple does an incremental refresh of their Cinema Displays, and eliminates the USB cable and adds a daisy-chain port, then that is going to be a pretty stunted market. Among value-conscious users, Cinema displays are a bit pricey for consideration, and among real color enthusiasts they are considered distinctly proletarian.
The fact that even the 17inch top of the line "pro" model doesn't have two ports is going to make some Eizo-wielding graphic designers/video types really unhappy...
Given the number of generically-fast-sounding names out there, it is a trifle curious; but my understanding is that trademarks only apply within limited areas of commerce(since the intent of trademark protection is to keep brands from being faked or undermined, not to make sure that a given name, particularly a generic english word, is used by only one product anywhere). Verizon would have to make the argument that cell phones and local peripheral interconnects are within sufficiently similar lines of business. It is certainly closer to being trademark infringement than would, say, someone starting a knitting show and calling it "The Fiber Channel" or a seller of large rubber bands calling their premier model "the infiniband"; but it probably isn't worth Verizon's time to push a dubious claim for a phone that they will probably have stopped selling in 18months anyway...
I'm sure that the could(possibly even one that works, unlike the mini-DP to dual-link DPI one they've been selling). I'm just wondering whether they will...
As I said, plain-old displayport has had support for display daisy-chaining and an AUX channel that(while it is only really enough to support a couple of USB ports at full steam) could easily be used to handle a variety of integrated monitor peripherals, card-readers, comparatively undemanding USB tasks, audio, 10/100 ethernet, etc. That capability has existed since 2009. Nothing that I've been able to find uses it. There are a few monitors that support displayport-in; but they treat it pretty much exactly like HDMI, and do nothing else with it. Even in the "stuff announced at tradeshows that never ships" category, I haven't seen a single displayport-based port replicator widget, monitor with daisy-chain support, or the like.
Hence my hesitancy about Apple melding light peak with their external monitor connector. Unless they have bold plans to light a fire under the ass of the industry generally, the only advantage of their fancy new interconnect will be the ability to run Firewire 800 and USB/random ancillary peripherals at full speed in their monitors. ADC2.0.
The Iranians have, somehow, discovered how to use the global trade in commodity parts to build a cluster computer(consisting of suspiciously under-filled racks, with a bunch of generic 1-3U-looking compute/storage nodes, much more empty space than I would have expected, and some pretty ragged ethernet interconnects, no visible brand IDs; but the black with reddish handles on the drive caddies looks a lot like de-branded HP...)
Other than perhaps minimally-puncturing the (always false) notion that Iran is a bunch of ignorant sand-dwellers just because it is a theocracy, I'm not seeing the big deal here. Depending on the CPU/RAM specs and how many racks there actually are(the photos are fairly cagey on the subject), it very much looks as though they've managed to put a few million dollars worth of datacenter together. News to anybody who thought that Iranians spent their time wallowing in backwardness and squalor; but pretty low-rent by cluster computer standards...
I'm pretty sure that going with Sandybridge CPUs requires going with Intel's newer chipsets. I'm not sure if the chipset TDP changed much; but the ones that support Sandy bridge are definitely different models, with slightly reworked graphics(which is presumably what allowed the 13 inchers to finally ditch the Core2+Nvidia chipset) and slightly different peripheral offerings.
Or Apple could get really crazy and issue a firmware update that allows the iPhone/iPad to optionally wirelessly sync in the background when at home/on the charger, so that "syncing" could take advantage of wireless networking and network storage capabilities(which things like the time capsule indicate that Apple can certainly handle) rather than being pretty much exactly identical to what Pilot 1000 owners were doing in 1996...
I hope Apple hasn't let their fascination with reducing port count get in the way of what might otherwise have been an interesting technology...
By amalgamating the mini-displayport and the light peak data lines, they certainly have kept another small hole in the chassis from sullying the 2001-esque purity to which they aspire. However, that means that you can't use an external monitor and a light peak device at the same time, unless you either deal with an ugly(almost certainly powered) breakout box, or buy an entirely new monitor that embeds the breakout box and/or a bunch of light peak-connected ports in itself(just like the ADC monitors of yore...).
I strongly suspect that Apple will release one or more of the latter in their next Cinema Display refresh(to bad, so sad, people who purchased the mini-displayport refresh...) which will allow them to have some USB and firewire ports, likely along with audio and webcam, on the monitor without additional cabling... How more interesting(but niche) uses of a 10gb/s interconnect, like high-speed storage or local networking, will be addressed is less clear. They are likely too niche and too expensive to make it into a mass-market monitor; so I assume that they'll be waiting on the ugly breakout box...
Lest anybody think that I am being down on Apple just for the sake of hating on Apple, consider this: Plain old Displayport has, since 2009, supported multi-display daisy chaining, along with a 720mb/s "aux" data channel for non-video peripherals in the chain. As of 2011, there are(to the best of my knowledge), zero displayport peripherals, announced or in production, that either support display daisy chaining or use the AUX channel to integrate USB ports, webcams, audio, or other peripheral functions into displayport devices without the use of additional cabling, despite 720mb/s being ample for quite a few applications. Zip, zero, nada.
Now Apple has taken light peak and, in the interests of reducing port count, basically produced a Displayport connector with an additional, high speed, AUX channel. Unless they have a clever plan in mind to make it useful for niche cases that could actually use the 10gb/s, without blocking external monitor capabilities(because is Joe Video Editor really going to want to choose between his gigantic direct-attached-RAID-array and his gigantic screen?), they've basically produced ADC2.0. Whee!
For this reason, i dont thing anything will happen in Iran unless due subversive foreign activity. A theocracy canot be overthrow for as long idiots believe that cult.>
Arguably, subversive foreign activity(at least the sort that the CIA tends to favor) is actually a boon for theocrats and would-be theocrats. When it comes to foreign puppets, we have a hard-on for repressive right wingers that just won't quit. As long as they are tractable toward us, not commies, and keep the resources flowing, they can expect more or less free reign domestically. Rule by a repressive right wing government tends to, over time, erode the institutions(rule of law, votes that aren't farces, etc.) and demographics(a middle class, the educated, political activists) on which democratic governance largely depends. However, as a pragmatic move to avoid being assassinated or deposed for being infidel American puppets(which they are), they frequently end up coddling the worst sort of nutjobs, just so long as said nutjobs will politely ignore them in exchange for some cash and/or repression of women and booze and homosexuals and stuff(see also Saudi Arabia...)
The end result is a puppet tyrant with minimal local legitimacy(and generally a taste for corruption an cronyism) who has cleaned house of all "civil society" elements; but has been paying protection, in the form of assorted domestic policies, to the worst sort of religious fanatics. When our puppet's strings eventually tangle, there is no civil society up on which to build a non-authoritarian government, and secular authoritarianism is widely discredited as corrupt, feckless, and an imperialist tool. Boom. Theocratic totalitarianism it is!
Will fix the problem of these keyboards logging you out when you leave for a quick coffee. Once again, any kind of security is thwarted by duct tape.
They've got that one figured out: If the sensor detects a range of zero, corresponding to the duct-tape trick, your workstation will play a tinny rendition of "Don't Stand so Close to Me" at earsplitting volume until one of your enraged coworkers rectifies the situation and then shoves a pen into your eye...
RFID/smartcards in the employee IDs would work. You just want to be very careful that you don't set up a system that simply encourages people to leave the card on the computer in order to stay logged in... Then you have stolen IDs floating around, and unlocked terminals. Whether or not people will do this probably depends on how tyrranical you are, and how fast they can log back in/unlock. If your horrible mess of a system image takes 5 minutes to log in, you'll need an overtly totalitarian IT security group to keep people from doing that.
If, as with the Sun Rays of old, your session(complete with state) pops up within seconds of inserting the smart card, people will be much more likely to take reasonable precautions...
Fighting known cognitive weaknesses and common patterns of poor prediction in humans really isn't worth the effort(nor, arguably, is it even in good taste) if relatively cheap technological solutions are available.
Humans forget sometimes. Some enough to describe them as "sloppy and incompetent" and fire them; but almost anyone will fuck up occasionally if they have to do it enough.(Plus, I'm guessing that nurses forget a little more often than average if their distractions include such minor items as "patient coding suddenly and dramatically in the next room'...)
Humans also get distracted fairly easily, and can't always predict when. "Just stepping away from the computer for 15 seconds" can easily become "get dragged into something and come back half an hour later".
Whether this particular(almost certainly overpriced) product is the way to go(when you could just use a $5 webcam that can also detect presence/absense and can even prevent naive attempts at presence spoofing, facial recognition is weak; but it is easier to impersonate a human-shaped lump than it is a specific face, without prior planning); but the idea that we should just buck up and strengthen our moral fiber, when a machine could easily do the job for us, is a masochistic recipe for poor results.
If we really want to make sure that we never get re-elected, we could do something crazy like making sure that forensics labs are independent entities, equally accessible to the prosecution and the defense, rather than the (quite common; but not universal) model of their being appendages of the police force...
Slashdot Mirror
I suspect that Apple has better than average odds of getting the 'per-application' buy-in; but we'll see how it goes. I'm inclined to be a touch pessimistic just because of the situation, as described above, where we've had the architectural underpinnings to do versioning/revision control for ages(and even fancy stuff like multiple nodes and network transparency and so forth has been there for years) and we've had applications where revision control and snapshot navigation are so overtly useful that they've been hacked in at the file format layer, also for years; but never hath the twain met(outside of IDEs/revision control systems and some very, very spendy stuff used mostly by law offices for revision control on legal documents).
Arguably, it has been "practically there now" for ages. Google Docs has it, Assorted revision-control systems have taken a bit of setup; but have had it(and boatloads of features besides) for years. VMS had it at the filesystem level back when a VAX was some pretty bitchin' hardware... NTFS volume shadow copies are in something of the same vein, as are the snapshotting features provided by most SAN vendors, Sun's ZFS, Oracle's BTRFS, etc, etc.
Obviously, the engineering challenges of doing robust versioning in an area where data-loss Just Isn't OK are nontrivial, and the amount of extra storage space required to move from a simple "here are your files" setup to a full versioning setup isn't small; but, as best I can tell, the issue isn't one of availability; but one of implementation/polish.
The revision-control stuff is great for programmers and techies(and even if you aren't programming, doing things like putting system config files under revision control allows for some neat tricks), The Google Docs stuff is super-easy; but limited to Google Docs, Shadow Copies have saved Windows shop admins countless trips to their backup tapes; but are essentially invisible to virtually all user-facing applications, same basic story with ZFS, BTRFS, and the snazzy SAN stuff.
Then you have individual programs that implement their own high-granularity revision tracking, totally separate from the FS/OS level stuff, within their own file formats(like Office with change tracking, Non-desructive edits in image manipulation programs, virtually anything that has an undo/redo stack of some kind, and so forth).
What does not seem to exist is anything where there is much cooperation across layers of the stack that makes up what the user actually faces off against: Somethin like a non-destructive image editor allows you to traverse back and forth across revisions; but its file format is just a basic blob as far as the FS is concerned. Something like Shadow Copy, Time Machine, FS snapshots, or revision control allows you to freely traverse the history of a file; but very few programs "know" about them in a useful way. I can use such a system to dredge up a snapshot of a file from 2 hours ago, and feed that to a suitable program; but the program will treat it exactly as though it were opening a file from the present. Only a few specific cases(like IDEs and revision control systems) show collaboration between the user-level program and the deeper abstraction level providing the versioning. Most everything else either does the versioning entirely within the program, with the OS/FS entirely ignorant of it, or entirely within the OS/FS/other low level mechanism with the user-level programs entirely unaware of what is going on.
Frankly, the "Recycle bin/trash" is a good(but not perfect) UI convention for a great many common computing situations. It serves as a reasonable recovery point/"in retrospect I fucked up" self help tool. It also imposes basically no requirements on the system/filesystem architecture. Would minor little additions(like having it automatically sort items by date/time of deletion) make it better? Sure. Is there anything fundamentally wrong? Not really.
Now, in high-resource environments, there is a much stronger case to be made that the "recycle bin" is obsolete and must die; but only if the system/FS/programs can be modified to accommodate the necessary changes. If, say, I mostly deal in text of some sort(writer, lawyer, programmer) even a modern laptop drive(never mind the cheap network storage) is Ludicrous Storage. Why can't I traverse a high-granularity timeline of every change made to every file I deal with?
That's the thing: For situations where you have to contend with legacy limitations, or where user space requirements are still comparatively close to the limits imposed by available technology, the "recycle bin" UI metaphor is pretty damn good. It is basically just a folder(with a tiny dollop of metadata to allow "restore this item") so it imposes minimal requirements; but still helps save users from common fuckups.
However, for environments where storage is massively ahead of traditional user requirements, and one has the liberty to re-examine certain legacy restrictions, you could arguably do much, much, better, with full versioning of all kinds of stuff.
I suspect that RIM is in for a nasty surprise if they want their tablet to succeed...
So long as their target market is "ugly; but secure and comparatively inexpensive email/cellphone appliances for suits" along with a sideline in "best keyboard for kids on cheap plans, now that sidekick is dead", they can afford to have a relatively sucky dev process. Their email/BBM stuff is 1st party, and much of the 3rd party development is by and/or for deep-pocketed corporate outfits that have hellishly complex requirements and processes anyway.
The problem arises if they want to play in the consumer market(or address the longterm/strategic threat that the more 'likeable' platforms will, either natively or in 'business' versions, harden up when it comes to security and management), they are going to have a very hard time attracting casual and indie developers to bring games and whatnot with their existing process.
The fact that they've gone and put out a tablet suggests that they do want to do this(Mr. Corporate Suit already has a laptop, and everybody in his office has the same power adapters/docking stations, that runs Outlook and all his horrid windows-only enterprise applications and a blackberry for mobile email. He will be picking up a tablet why? Tablet is a consumer move...)
A) Peter Thiel.
B)UCMJ, Subchapter II, section 810, Article 10: "Any person subject to this chapter charged with an offense under this chapter shall be ordered into arrest or confinement, as circumstances may require; but when charged only with an offense normally tried by a summary court-martial, he shall not ordinarily be placed in confinement. When any person subject to this chapter is placed in arrest or confinement prior to trial, immediate steps shall be taken to inform him of the specific wrong of which he is accused and to try him or to dismiss the charges and release him."(emphasis mine)
UCMJ, Subchapter II, section 813, Article 13: "No person, while being held for trial, may be subjected to punishment or penalty other than arrest or confinement upon the charges pending against him, nor shall the arrest or confinement imposed upon him be any more rigorous than the circumstances require to insure his presence, but he may be subjected to minor punishment during that period for infractions of discipline."(emphasis mine, and I'm pretty sure that month after month of solid solitary doesn't count...)
C)Worse Troll is worse.
Hilariously, Paypal was actually started by a libertarian as some sort of "resist the man and his fiat currency's dead hand on trade." kind of thing. Now it voluntarily licks the boots of those who would suppress the entirely legal efforts of an advocacy group to secure a man a fair trial(rather than the present detention-without-trial-of-indefinite-length...)
All hail the private sector, defender of liberty!
I'm very fond of that particular supercomputing site. Building your giant postindustrial blinkencluster in a transparent multistory chamber inside an authentic historic chapel has way more evil genius cred than any number of white cats...
At present, the only displayport features that Apple's displays handle are video and audio input. A separate USB cable is used to handle the USB hub, webcam, and microphone. There is no provision for daisy chaining. A few other monitors support displayport(generally in conjunction with other inputs, except for internal displays in laptops); but I do not know of a single one, on the market or promised, that treats displayport as other than an HDMI-equivalent video/audio cable, despite the theoretical possibility.
Apple is notoriously tight lipped about their plans; but I've never heard any mention of plans to change this(though I'm assuming that their next revision will, at least, eliminate the USB cable in favor of a thunderbolt based transport. I would certainly hope that they would support daisy-chaining as well..
I definitely find any claims of it being especially high performance deeply fishy, and the unveiling of it reeks of a PR stunt; but nothing about the (limited) hardware that is there looks definitely fake.
A bunch of 1-3Us, connected via rather scraggly patches from a few GigE switches to the on-motherboard GigE ports. Depending on the specs of the servers, we could be talking a few hundred cores, and a reasonable number of TB of storage...
It certainly is of abnormally low density, though. Each rack is 50% or more blanking plates, and I don't see any of the classic "cluster porn" shots of the Big Serious Core Switch, which they would definitely show if they could. No "Stylish rack opened enough to show all 72Us of servers exuding a sense of overwhelming power" shot or "View down the rack hallway that appears to go on forever, puny meatsack" shot either.
I'd totally believe that this is a (somewhat optimistic/growth prepared) conversion of some underused university basement to support a CS department curriculum and/or the compute needs of other disciplines; but it sure isn't what the PR says it is....
I'm aware that it can be daisy chained, I'm just concerned that that might end up being a dead letter(which would make having it block your video-out port an issue). As I noted, ordinary Displayport has supported daisy-chaining, and a general purpose AUX channel with nontrivial bandwidth, since 2009. Since then, crickets. Not even obviously-prototype FPGAs and blinking lights tech demos, never mind products that you can actually buy. Nothing. For all intents and purposes, you cannot daisy-chain displayport, and its 720Mb/s AUX channel is used entirely to carry the DDC data that an i2C bus was ample for.
Thunderbolt, as best I can tell, appears to be a displayport with a seriously buffed AUX channel. As I said, if Apple has a plan to set the industry on fire, great. If the only daisy-chaining support appears once Apple does an incremental refresh of their Cinema Displays, and eliminates the USB cable and adds a daisy-chain port, then that is going to be a pretty stunted market. Among value-conscious users, Cinema displays are a bit pricey for consideration, and among real color enthusiasts they are considered distinctly proletarian.
The fact that even the 17inch top of the line "pro" model doesn't have two ports is going to make some Eizo-wielding graphic designers/video types really unhappy...
They had to switch when they dropped the fiber interconnects and added the high-voltage charge pump "Thorsecure" anti-tampering feature...
Given the number of generically-fast-sounding names out there, it is a trifle curious; but my understanding is that trademarks only apply within limited areas of commerce(since the intent of trademark protection is to keep brands from being faked or undermined, not to make sure that a given name, particularly a generic english word, is used by only one product anywhere). Verizon would have to make the argument that cell phones and local peripheral interconnects are within sufficiently similar lines of business. It is certainly closer to being trademark infringement than would, say, someone starting a knitting show and calling it "The Fiber Channel" or a seller of large rubber bands calling their premier model "the infiniband"; but it probably isn't worth Verizon's time to push a dubious claim for a phone that they will probably have stopped selling in 18months anyway...
"I understand. In death, we have names..."
This the itanium team; but could they have chosen something a little less, er, pessimistic?
Is it more resistant to icebergs than the previous itanics?
I'm sure that the could(possibly even one that works, unlike the mini-DP to dual-link DPI one they've been selling). I'm just wondering whether they will...
As I said, plain-old displayport has had support for display daisy-chaining and an AUX channel that(while it is only really enough to support a couple of USB ports at full steam) could easily be used to handle a variety of integrated monitor peripherals, card-readers, comparatively undemanding USB tasks, audio, 10/100 ethernet, etc. That capability has existed since 2009. Nothing that I've been able to find uses it. There are a few monitors that support displayport-in; but they treat it pretty much exactly like HDMI, and do nothing else with it. Even in the "stuff announced at tradeshows that never ships" category, I haven't seen a single displayport-based port replicator widget, monitor with daisy-chain support, or the like.
Hence my hesitancy about Apple melding light peak with their external monitor connector. Unless they have bold plans to light a fire under the ass of the industry generally, the only advantage of their fancy new interconnect will be the ability to run Firewire 800 and USB/random ancillary peripherals at full speed in their monitors. ADC2.0.
The Iranians have, somehow, discovered how to use the global trade in commodity parts to build a cluster computer(consisting of suspiciously under-filled racks, with a bunch of generic 1-3U-looking compute/storage nodes, much more empty space than I would have expected, and some pretty ragged ethernet interconnects, no visible brand IDs; but the black with reddish handles on the drive caddies looks a lot like de-branded HP...)
Other than perhaps minimally-puncturing the (always false) notion that Iran is a bunch of ignorant sand-dwellers just because it is a theocracy, I'm not seeing the big deal here. Depending on the CPU/RAM specs and how many racks there actually are(the photos are fairly cagey on the subject), it very much looks as though they've managed to put a few million dollars worth of datacenter together. News to anybody who thought that Iranians spent their time wallowing in backwardness and squalor; but pretty low-rent by cluster computer standards...
I'm pretty sure that going with Sandybridge CPUs requires going with Intel's newer chipsets. I'm not sure if the chipset TDP changed much; but the ones that support Sandy bridge are definitely different models, with slightly reworked graphics(which is presumably what allowed the 13 inchers to finally ditch the Core2+Nvidia chipset) and slightly different peripheral offerings.
Or Apple could get really crazy and issue a firmware update that allows the iPhone/iPad to optionally wirelessly sync in the background when at home/on the charger, so that "syncing" could take advantage of wireless networking and network storage capabilities(which things like the time capsule indicate that Apple can certainly handle) rather than being pretty much exactly identical to what Pilot 1000 owners were doing in 1996...
I hope Apple hasn't let their fascination with reducing port count get in the way of what might otherwise have been an interesting technology...
By amalgamating the mini-displayport and the light peak data lines, they certainly have kept another small hole in the chassis from sullying the 2001-esque purity to which they aspire. However, that means that you can't use an external monitor and a light peak device at the same time, unless you either deal with an ugly(almost certainly powered) breakout box, or buy an entirely new monitor that embeds the breakout box and/or a bunch of light peak-connected ports in itself(just like the ADC monitors of yore...).
I strongly suspect that Apple will release one or more of the latter in their next Cinema Display refresh(to bad, so sad, people who purchased the mini-displayport refresh...) which will allow them to have some USB and firewire ports, likely along with audio and webcam, on the monitor without additional cabling... How more interesting(but niche) uses of a 10gb/s interconnect, like high-speed storage or local networking, will be addressed is less clear. They are likely too niche and too expensive to make it into a mass-market monitor; so I assume that they'll be waiting on the ugly breakout box...
Lest anybody think that I am being down on Apple just for the sake of hating on Apple, consider this: Plain old Displayport has, since 2009, supported multi-display daisy chaining, along with a 720mb/s "aux" data channel for non-video peripherals in the chain. As of 2011, there are(to the best of my knowledge), zero displayport peripherals, announced or in production, that either support display daisy chaining or use the AUX channel to integrate USB ports, webcams, audio, or other peripheral functions into displayport devices without the use of additional cabling, despite 720mb/s being ample for quite a few applications. Zip, zero, nada.
Now Apple has taken light peak and, in the interests of reducing port count, basically produced a Displayport connector with an additional, high speed, AUX channel. Unless they have a clever plan in mind to make it useful for niche cases that could actually use the 10gb/s, without blocking external monitor capabilities(because is Joe Video Editor really going to want to choose between his gigantic direct-attached-RAID-array and his gigantic screen?), they've basically produced ADC2.0. Whee!
For this reason, i dont thing anything will happen in Iran unless due subversive foreign activity. A theocracy canot be overthrow for as long idiots believe that cult.>
Arguably, subversive foreign activity(at least the sort that the CIA tends to favor) is actually a boon for theocrats and would-be theocrats. When it comes to foreign puppets, we have a hard-on for repressive right wingers that just won't quit. As long as they are tractable toward us, not commies, and keep the resources flowing, they can expect more or less free reign domestically. Rule by a repressive right wing government tends to, over time, erode the institutions(rule of law, votes that aren't farces, etc.) and demographics(a middle class, the educated, political activists) on which democratic governance largely depends. However, as a pragmatic move to avoid being assassinated or deposed for being infidel American puppets(which they are), they frequently end up coddling the worst sort of nutjobs, just so long as said nutjobs will politely ignore them in exchange for some cash and/or repression of women and booze and homosexuals and stuff(see also Saudi Arabia...)
The end result is a puppet tyrant with minimal local legitimacy(and generally a taste for corruption an cronyism) who has cleaned house of all "civil society" elements; but has been paying protection, in the form of assorted domestic policies, to the worst sort of religious fanatics. When our puppet's strings eventually tangle, there is no civil society up on which to build a non-authoritarian government, and secular authoritarianism is widely discredited as corrupt, feckless, and an imperialist tool. Boom. Theocratic totalitarianism it is!
Will fix the problem of these keyboards logging you out when you leave for a quick coffee. Once again, any kind of security is thwarted by duct tape.
They've got that one figured out: If the sensor detects a range of zero, corresponding to the duct-tape trick, your workstation will play a tinny rendition of "Don't Stand so Close to Me" at earsplitting volume until one of your enraged coworkers rectifies the situation and then shoves a pen into your eye...
RFID/smartcards in the employee IDs would work. You just want to be very careful that you don't set up a system that simply encourages people to leave the card on the computer in order to stay logged in... Then you have stolen IDs floating around, and unlocked terminals. Whether or not people will do this probably depends on how tyrranical you are, and how fast they can log back in/unlock. If your horrible mess of a system image takes 5 minutes to log in, you'll need an overtly totalitarian IT security group to keep people from doing that.
If, as with the Sun Rays of old, your session(complete with state) pops up within seconds of inserting the smart card, people will be much more likely to take reasonable precautions...
Fighting known cognitive weaknesses and common patterns of poor prediction in humans really isn't worth the effort(nor, arguably, is it even in good taste) if relatively cheap technological solutions are available.
Humans forget sometimes. Some enough to describe them as "sloppy and incompetent" and fire them; but almost anyone will fuck up occasionally if they have to do it enough.(Plus, I'm guessing that nurses forget a little more often than average if their distractions include such minor items as "patient coding suddenly and dramatically in the next room'...)
Humans also get distracted fairly easily, and can't always predict when. "Just stepping away from the computer for 15 seconds" can easily become "get dragged into something and come back half an hour later".
Whether this particular(almost certainly overpriced) product is the way to go(when you could just use a $5 webcam that can also detect presence/absense and can even prevent naive attempts at presence spoofing, facial recognition is weak; but it is easier to impersonate a human-shaped lump than it is a specific face, without prior planning); but the idea that we should just buck up and strengthen our moral fiber, when a machine could easily do the job for us, is a masochistic recipe for poor results.
If we really want to make sure that we never get re-elected, we could do something crazy like making sure that forensics labs are independent entities, equally accessible to the prosecution and the defense, rather than the (quite common; but not universal) model of their being appendages of the police force...
Being sued by Facebook is referred to as "Getting Zuckerpunched"...