Well, it sounded like their main Israel facility is in Haifa. That's in the north of the country, along the coast, and not as desert-like as the southern parts we always think of.
Also, the location would put them near Technion University, so they'd have the academic research base to support their operation as well.
Having been through the CCNA and CCNP certifications, however, I can say that while Cisco does put a lot of their own stuff into those certs, they also put far more generally useful internetworking information.
True, if you handed me a non-Cisco router or managed switch, a CCNx might not know "the exact commands to type" to make it useful. But chances are that they would know exactly how to configure it, given some basic reference material. Protocols are standard and universal enough (ok, some are proprietary, but not most of them).
I feel the exact same way. When you stick something usable in front of someone, they tend to not have any desire to reach out, expand their horizons, and learn about other options.
When I was growing up, a computer didn't do much unless you could program it. And what commercial software you could buy, wasn't so dramatically better than what one could self-code, as to remove the motivation to do so.
You can do absolutely anything with Plone - except have decent performance;-)
Yeah, it's slow, sluggish, etc... Sure, you can improve speed by doing a whole hunk of distributed stuff with Zope's ZEO, front-end caching, 3 gazillion tweaks and such, but at the end of the day, it is still slow as crap. (Unless you run it on a bunch of really fast servers and cache the hell out of it.)
I'm running a Zope/Plone site myself, and I'm desperate to migrate to something else. I've looked at Mambo and Drupal before, but they seem to generally only offer my a 90% solution and are always missing those little minor things (like intermediate index pages that aren't "auto-generated blog format" or "news article listings") that Plone has.
I wonder how good/bad it is for hard drives to be shutting off and spinning up so often... Laptop drives may be designed with that in mind, but not sure about desktop drives.
When I first got a desktop motherboard with such "power management" features, I suspect that the constant stop/start cycles on the hard drives eventually lead to one of the drives failing on me. (of course this was many years ago)
I never heard that one before, but it sort-of makes sense. Of course uninstalling/deleting extra programs may or may not actually cure registry bloat. Then again, with intelligent design, it should be possible to make registry bloat not cause such access delays. (of course I have no idea how MS implments registry access)
Instant sleep, instant wake, but the battery still dies after a couple days to a week (depending on age of the battery). I find it quite annoying, as I'm the type who only uses my PowerBook when I'm out-and-about and rarely at home. However, I would say that the instant suspend/resume is my absolute favorate feature of the machine.
I really wish Apple did implement a hibernation mode, though, which didn't use any battery at all. My old IBM ThinkPad 600E had it, and I was using it for a while. Sure, the suspend/resume cycle takes longer, but you can leave it in hibernation indefinitely. The downside, of course, is that hibernation becomes MUCH less practical as RAM sizes increase (TP only had 128MB and even that took time to hibernate/resume), and laptop hard drives aren't particularly fast.
You know, I've heard people say Windows XP is "fast" because it boots quickly.;-) Furthermore, I've found it commonplace for people to want to think of boot-up time as a measure of their computer's performance. This, of course, is only among those people who actually shut off their computers;) (excluding laptops here, because we all shut those off, or the battery dies in suspend mode after a week) Still remember back when even I would shut off my computer when not using it.
Of course now I never shut off most of my actively-used non-laptop computers. In fact, I think the only times I've ever shut down any of my main machines was to add/change hardware. (likewise, they only get rebooted for hardware changes or OS patches that require a reboot)
Many people in the Windows world seem convinced that a major cause of a slow machine is "I've got a lot of things installed on it".
Now those of is in the here see that as nonsense, since filling up the hard drive with not-currently-executing code does not have any impact on memory usage or CPU usage.
However, a lot of Windows programs have this tendency to install things that "always run in the background", and that does eat memory and impact CPU load. Back in the day when I actually used Windows a lot (and when RAM was still expensive), it was commonplace for people like us to spend the time digging through the Start Menu and the Registry to disable all those little side-processes.
As a result, people like me had machines that were MUCH faster and more responsive than most normal people with their storebought machines with specs usually much better than mine on-paper.
I suspect the same may be true today, between store pre-loaded crap and resident bits of installed software, even if cheap RAM has averted some of the issue.
Back in 1992, AOL wasn't an ISP (asside from maybe e-mail). They were an "online service" providing exclusive content to their members, just like Prodigy and CompuServe. Of course this was back in the day when "the internet" as we now know it didn't provide as much to the average home user, and such independent online services did have their place.
Sure, maybe they do that now. But from the mid-to-late 90s until recently, AOL was just a huge floodgate unleashing scores of totally clueless newbies upon the internet, laying waste to Usenet, and cluttering the web with flashy and ugly crap-sites. Heck, even chain e-mails that once cluttered my in-box from family members who didn't know any better had a tendency to originate somewhere in the AOL-verse.
Hehe... the Federal government (DoD-wise, anyways) outsources damn near everything tech related. Sure, they hire technical people, to park them in a cubicle until they forget everything but the buzzword-of-the-week, and ultimately retire when they get old.
(Of course they outsource to US companies hiring US citizens, at least.)
First of all, what is it with all you "now we can have drivers ported!" people? Nothing ever stopped anyone from porting/writing drivers for commercial/closed Solaris, and in fact people have independently written drivers for it (though not many people). Sun has had freely available documentation and example code for Solaris device drivers for a very long time.
Secondly, from what I've seen of device driver code in Solaris, I'd say its very structurally similar to what I've seen in the *BSDs, so porting would likely not be very difficult. (Also, Sun's driver docs talk a lot about writing drivers that work on both endians, use mutexes for everything, and are generally designed to work with multiple instances on SMP systems.)
Of course there are also two parts to "speed" for an interactive multi-user machine. There's response-time, where AIX really shined on those machines, and actual processing power. Even if it runs an intensive task slowly, the simple fact that it has snappy response to an interactive user's input (i.e. UI speed, even if command-line) tasks makes the machine feel a lot faster.
I also do agree that Solaris is the friendliest, or at least the "cleanest feeling" commercial UNIX. Never messed much with HP-UX boxes, mainly since I've not been around them much and never went out of my way to get any.
Interesting, because the AIX-based RS/6000s my school used for remote access (they're still running, though less used than when I was there) were multi-processor 112MHz PowerPC machines (7013-J30, and 7013-J40).
These handled the load VERY well, and even could handle MULTIPLE remote-X sessions of Pro/Engineer (serious CAD software) across the campus network to the dorms. Maybe hardware or maybe just AIX, but they could certanly handle a beating and keep on ticking responsively.
You know, nothing ever prevented anyone from writing Solaris drivers back when the OS was closed. Sun had plenty of publicly available documentation and sample code, freely available on their website, for driver development.
Solaris has very good cross-version device driver compatability. A binary driver module for Solaris 7 may work on something as new as Solaris 9 (maybe even 10), provided that it was built for the same architecture (i.e. 32-bit vs. 64-bit).
Meanwhile, Linux kernel modules are seldom compatable (once compiled into a binary) across minor sub-version variants.
Which makes me wonder... Which PPC systems did it run on? There have been several made, from several companies (IBM RS/6000 and Apple Macintosh being the two most popular), and they all tend to use different chipsets/firmware/etc.
Yeah, it would almost make more sense to push for commodity-style PowerPC based motherboards, that fit an ATX chassis, and took common PC-type components.
However, given that Apple is a hardware company right now, it probably would hurt them more than it could help them.
Then again, with Microsoft's push to make themselves independent of x86 (maybe, if we're lucky), it could be very interesting if Windows/PPC ever emerged again (yeah, heard rumors that NT once ran on PPC, but I don't think I've ever heard of actual products, unlike NT/Alpha).
I agree with you 100%. On a very similar note, an even bigger trick Microsoft managed to pull was to convince users that computers being crash-happy, quirky, and otherwise prone to non-deterministic failures was normal.
In fact, it has become quite socially acceptable for people to talk about how their computer is "broken," "slow," or otherwise "hates them." In reality, the computer is usually just perfectly fine and running well. The problem is some Microsoft software is either flaky, or so gummed up with 3rd party software that it becomes slow and/or flaky. The most annoying thing of all is that this is a natural side-effect of just actively using the darn machine for a while without any wipe-and-reinstall processes.
I'm sorry, but computer OSes should not be prone to getting "full of cruft" from active use, and should not get "slower and quirkier over time" from any reason other than particular newer software getting more bloated/slower.
And no, it's not that Linux/MacOSX/etc. being "stable" is anything special to write home about. Pretty much every other OS I've used is stable by default. It's Windows being "unstable" that's the unique thing.
Before anyone tries to say that "newer versions don't crash all over the place anymore," please keep in mind that the issue I'm discussing isn't OS stability, but rather how MS has convinced the users that OS instability is the norm, and stability is some new thing to be cherished, if they're really good.
Another problem I'd like to note is that the gap between what you can code yourself, and what is considered "commercial-quality software" is ever widening. Back in the days of the Apple ][, you could easily learn to "roll your own" games/software that was useful enough, in relation to what you could easily get from the outside, that it felt worth the effort to learn and tinker.
As time went on, it seemed like what you could "roll your own" on became continually less and less impressive compared to what everyone else "could just get at the store", that the "wow factor" of homebrew coding just evaporated.
I could say exactly the same think for homebrew electronics tinkering, by the way.
It seems like the F/OSS movement has breathed a whole new life into homebrew development, and has shown that you can make a difference outside a corporate development team. This is a wonderful thing, but... It usually does not extend to games, and it isn't very visible to the average home user. (i.e. not as much of a "show-off" capability)
Those of us who started in the Apple ][ and earlier days have a real edge on the new kids, in that we were able to build our knowledge in a day when there were greater personal and social rewards, and when there seemed like more of a "point" to home coding.
I've always found myself having an interest in network administration myself, and also have a CS background.
I find it interesting how many CS students don't know jack about networking.
On the flip side, IT students are supposed to know networking, and talk about how they want to learn networking, but you KNOW that they'll NEVER actually know jack about the subject.
Of course these both describe the common case I've run into, and I've known good exceptions in both categories (ok, more the former than the latter, at least at my old school and not the one where I live).
Yeah, we've been getting a slew of articles these days on Solaris 10 that "review" the product by simply reading the marketing materials. Also, yet another article on Solaris 10 that tries to only look at the x86 version, and then complains when it doesn't measure up. Well guess what? The x86 version of Solaris has NEVER measured up. Sol 10 is Sun's first attempt at changing that, and it truely won't go anywhere (beyond their approved-compatable hardware) until 3rd parties get more invested in development.
Solaris 10 is first and foremost an UltraSPARC-based OS. That's where it runs best, supports almost all the hardware, and is all around a good thing. (Though the x86-64 version should be interesting down the line, as I hear Sun is now working on Opteron servers entirely of their own motherboard design.)
I just wish, for once, someone would review the OS by actually USING IT on the proper hardware, and talk about new and interesting features that aren't blabbed about on the shiney sheets thrown around by marketing.
For example, one of the biggest and most obvious new features of Solaris 10 (that doesn't make the list of "Zones! Self-healing! ZFS, when we finish it!" would have to be the Service Management Facility. They've completely redone the entire framework of how services are managed (i.e. "init.d", "inetd", etc.), to even include service dependency tracking and allow non-dependent services to start in parallel (making big systems boot a lot faster).
At least all of the MacOS X articles by journals like this were the result of actually trying to use and explore the OS itself. (Even if they were formulaic, and pretty much involved saying "this is cool", "hey, the/etc files don't do anything," "oh, they use something called NetInfo," "back to babbling")
Slashdot Mirror
Well, it sounded like their main Israel facility is in Haifa. That's in the north of the country, along the coast, and not as desert-like as the southern parts we always think of.
Also, the location would put them near Technion University, so they'd have the academic research base to support their operation as well.
Having been through the CCNA and CCNP certifications, however, I can say that while Cisco does put a lot of their own stuff into those certs, they also put far more generally useful internetworking information.
True, if you handed me a non-Cisco router or managed switch, a CCNx might not know "the exact commands to type" to make it useful. But chances are that they would know exactly how to configure it, given some basic reference material. Protocols are standard and universal enough (ok, some are proprietary, but not most of them).
I feel the exact same way. When you stick something usable in front of someone, they tend to not have any desire to reach out, expand their horizons, and learn about other options.
When I was growing up, a computer didn't do much unless you could program it. And what commercial software you could buy, wasn't so dramatically better than what one could self-code, as to remove the motivation to do so.
You can do absolutely anything with Plone - except have decent performance ;-)
Yeah, it's slow, sluggish, etc... Sure, you can improve speed by doing a whole hunk of distributed stuff with Zope's ZEO, front-end caching, 3 gazillion tweaks and such, but at the end of the day, it is still slow as crap. (Unless you run it on a bunch of really fast servers and cache the hell out of it.)
I'm running a Zope/Plone site myself, and I'm desperate to migrate to something else. I've looked at Mambo and Drupal before, but they seem to generally only offer my a 90% solution and are always missing those little minor things (like intermediate index pages that aren't "auto-generated blog format" or "news article listings") that Plone has.
I wonder how good/bad it is for hard drives to be shutting off and spinning up so often... Laptop drives may be designed with that in mind, but not sure about desktop drives.
When I first got a desktop motherboard with such "power management" features, I suspect that the constant stop/start cycles on the hard drives eventually lead to one of the drives failing on me. (of course this was many years ago)
I never heard that one before, but it sort-of makes sense. Of course uninstalling/deleting extra programs may or may not actually cure registry bloat. Then again, with intelligent design, it should be possible to make registry bloat not cause such access delays. (of course I have no idea how MS implments registry access)
Instant sleep, instant wake, but the battery still dies after a couple days to a week (depending on age of the battery). I find it quite annoying, as I'm the type who only uses my PowerBook when I'm out-and-about and rarely at home. However, I would say that the instant suspend/resume is my absolute favorate feature of the machine.
I really wish Apple did implement a hibernation mode, though, which didn't use any battery at all. My old IBM ThinkPad 600E had it, and I was using it for a while. Sure, the suspend/resume cycle takes longer, but you can leave it in hibernation indefinitely. The downside, of course, is that hibernation becomes MUCH less practical as RAM sizes increase (TP only had 128MB and even that took time to hibernate/resume), and laptop hard drives aren't particularly fast.
You know, I've heard people say Windows XP is "fast" because it boots quickly. ;-) Furthermore, I've found it commonplace for people to want to think of boot-up time as a measure of their computer's performance. This, of course, is only among those people who actually shut off their computers ;) (excluding laptops here, because we all shut those off, or the battery dies in suspend mode after a week) Still remember back when even I would shut off my computer when not using it.
Of course now I never shut off most of my actively-used non-laptop computers. In fact, I think the only times I've ever shut down any of my main machines was to add/change hardware. (likewise, they only get rebooted for hardware changes or OS patches that require a reboot)
Many people in the Windows world seem convinced that a major cause of a slow machine is "I've got a lot of things installed on it".
Now those of is in the here see that as nonsense, since filling up the hard drive with not-currently-executing code does not have any impact on memory usage or CPU usage.
However, a lot of Windows programs have this tendency to install things that "always run in the background", and that does eat memory and impact CPU load. Back in the day when I actually used Windows a lot (and when RAM was still expensive), it was commonplace for people like us to spend the time digging through the Start Menu and the Registry to disable all those little side-processes.
As a result, people like me had machines that were MUCH faster and more responsive than most normal people with their storebought machines with specs usually much better than mine on-paper.
I suspect the same may be true today, between store pre-loaded crap and resident bits of installed software, even if cheap RAM has averted some of the issue.
Back in 1992, AOL wasn't an ISP (asside from maybe e-mail). They were an "online service" providing exclusive content to their members, just like Prodigy and CompuServe. Of course this was back in the day when "the internet" as we now know it didn't provide as much to the average home user, and such independent online services did have their place.
Sure, maybe they do that now. But from the mid-to-late 90s until recently, AOL was just a huge floodgate unleashing scores of totally clueless newbies upon the internet, laying waste to Usenet, and cluttering the web with flashy and ugly crap-sites. Heck, even chain e-mails that once cluttered my in-box from family members who didn't know any better had a tendency to originate somewhere in the AOL-verse.
Hehe... the Federal government (DoD-wise, anyways) outsources damn near everything tech related. Sure, they hire technical people, to park them in a cubicle until they forget everything but the buzzword-of-the-week, and ultimately retire when they get old.
(Of course they outsource to US companies hiring US citizens, at least.)
First of all, what is it with all you "now we can have drivers ported!" people? Nothing ever stopped anyone from porting/writing drivers for commercial/closed Solaris, and in fact people have independently written drivers for it (though not many people). Sun has had freely available documentation and example code for Solaris device drivers for a very long time.
Secondly, from what I've seen of device driver code in Solaris, I'd say its very structurally similar to what I've seen in the *BSDs, so porting would likely not be very difficult. (Also, Sun's driver docs talk a lot about writing drivers that work on both endians, use mutexes for everything, and are generally designed to work with multiple instances on SMP systems.)
Of course there are also two parts to "speed" for an interactive multi-user machine. There's response-time, where AIX really shined on those machines, and actual processing power. Even if it runs an intensive task slowly, the simple fact that it has snappy response to an interactive user's input (i.e. UI speed, even if command-line) tasks makes the machine feel a lot faster.
I also do agree that Solaris is the friendliest, or at least the "cleanest feeling" commercial UNIX. Never messed much with HP-UX boxes, mainly since I've not been around them much and never went out of my way to get any.
Interesting, because the AIX-based RS/6000s my school used for remote access (they're still running, though less used than when I was there) were multi-processor 112MHz PowerPC machines (7013-J30, and 7013-J40).
These handled the load VERY well, and even could handle MULTIPLE remote-X sessions of Pro/Engineer (serious CAD software) across the campus network to the dorms. Maybe hardware or maybe just AIX, but they could certanly handle a beating and keep on ticking responsively.
You know, nothing ever prevented anyone from writing Solaris drivers back when the OS was closed. Sun had plenty of publicly available documentation and sample code, freely available on their website, for driver development.
Solaris has very good cross-version device driver compatability. A binary driver module for Solaris 7 may work on something as new as Solaris 9 (maybe even 10), provided that it was built for the same architecture (i.e. 32-bit vs. 64-bit).
Meanwhile, Linux kernel modules are seldom compatable (once compiled into a binary) across minor sub-version variants.
No, but you may find outdated Hebrew or Greek phrases.
The cool thing with NetMeeting, actually, is that it isn't "proprietary MS-only," but rather uses more standard protocols.
I've been able to talk to NetMeeting from SunForum on a Solaris machine. (wouldn't surprise me if GnomeMeeting also worked for this)
Which makes me wonder... Which PPC systems did it run on? There have been several made, from several companies (IBM RS/6000 and Apple Macintosh being the two most popular), and they all tend to use different chipsets/firmware/etc.
Yeah, it would almost make more sense to push for commodity-style PowerPC based motherboards, that fit an ATX chassis, and took common PC-type components.
However, given that Apple is a hardware company right now, it probably would hurt them more than it could help them.
Then again, with Microsoft's push to make themselves independent of x86 (maybe, if we're lucky), it could be very interesting if Windows/PPC ever emerged again (yeah, heard rumors that NT once ran on PPC, but I don't think I've ever heard of actual products, unlike NT/Alpha).
I agree with you 100%. On a very similar note, an even bigger trick Microsoft managed to pull was to convince users that computers being crash-happy, quirky, and otherwise prone to non-deterministic failures was normal.
In fact, it has become quite socially acceptable for people to talk about how their computer is "broken," "slow," or otherwise "hates them." In reality, the computer is usually just perfectly fine and running well. The problem is some Microsoft software is either flaky, or so gummed up with 3rd party software that it becomes slow and/or flaky. The most annoying thing of all is that this is a natural side-effect of just actively using the darn machine for a while without any wipe-and-reinstall processes.
I'm sorry, but computer OSes should not be prone to getting "full of cruft" from active use, and should not get "slower and quirkier over time" from any reason other than particular newer software getting more bloated/slower.
And no, it's not that Linux/MacOSX/etc. being "stable" is anything special to write home about. Pretty much every other OS I've used is stable by default. It's Windows being "unstable" that's the unique thing.
Before anyone tries to say that "newer versions don't crash all over the place anymore," please keep in mind that the issue I'm discussing isn't OS stability, but rather how MS has convinced the users that OS instability is the norm, and stability is some new thing to be cherished, if they're really good.
Another problem I'd like to note is that the gap between what you can code yourself, and what is considered "commercial-quality software" is ever widening. Back in the days of the Apple ][, you could easily learn to "roll your own" games/software that was useful enough, in relation to what you could easily get from the outside, that it felt worth the effort to learn and tinker.
As time went on, it seemed like what you could "roll your own" on became continually less and less impressive compared to what everyone else "could just get at the store", that the "wow factor" of homebrew coding just evaporated.
I could say exactly the same think for homebrew electronics tinkering, by the way.
It seems like the F/OSS movement has breathed a whole new life into homebrew development, and has shown that you can make a difference outside a corporate development team. This is a wonderful thing, but... It usually does not extend to games, and it isn't very visible to the average home user. (i.e. not as much of a "show-off" capability)
Those of us who started in the Apple ][ and earlier days have a real edge on the new kids, in that we were able to build our knowledge in a day when there were greater personal and social rewards, and when there seemed like more of a "point" to home coding.
I've always found myself having an interest in network administration myself, and also have a CS background.
I find it interesting how many CS students don't know jack about networking.
On the flip side, IT students are supposed to know networking, and talk about how they want to learn networking, but you KNOW that they'll NEVER actually know jack about the subject.
Of course these both describe the common case I've run into, and I've known good exceptions in both categories (ok, more the former than the latter, at least at my old school and not the one where I live).
Yeah, we've been getting a slew of articles these days on Solaris 10 that "review" the product by simply reading the marketing materials. Also, yet another article on Solaris 10 that tries to only look at the x86 version, and then complains when it doesn't measure up. Well guess what? The x86 version of Solaris has NEVER measured up. Sol 10 is Sun's first attempt at changing that, and it truely won't go anywhere (beyond their approved-compatable hardware) until 3rd parties get more invested in development.
/etc files don't do anything," "oh, they use something called NetInfo," "back to babbling")
Solaris 10 is first and foremost an UltraSPARC-based OS. That's where it runs best, supports almost all the hardware, and is all around a good thing. (Though the x86-64 version should be interesting down the line, as I hear Sun is now working on Opteron servers entirely of their own motherboard design.)
I just wish, for once, someone would review the OS by actually USING IT on the proper hardware, and talk about new and interesting features that aren't blabbed about on the shiney sheets thrown around by marketing.
For example, one of the biggest and most obvious new features of Solaris 10 (that doesn't make the list of "Zones! Self-healing! ZFS, when we finish it!" would have to be the Service Management Facility. They've completely redone the entire framework of how services are managed (i.e. "init.d", "inetd", etc.), to even include service dependency tracking and allow non-dependent services to start in parallel (making big systems boot a lot faster).
At least all of the MacOS X articles by journals like this were the result of actually trying to use and explore the OS itself. (Even if they were formulaic, and pretty much involved saying "this is cool", "hey, the
And here's a link with info on NCW.
I read most of the book mentioned in this site. Interesting concepts, even if the book is 3x longer than it needs to be to get the point across.