...being able to trade off two companies against each other gets me a better price.
Indeed. From the OEMs on down to consumers, the competition is the important thing here. The real story here is the competition going on here. What isn't really being mentioned is that there are great market forces at work here, AND those forces are producing better products and lowering prices for consumers.
And considering how teeny AMD was compared to Intel just a few short years ago it doesn't really seem to matter who the fastest, fastest, fastest processor is. At this point CPU speed has become SO good that there are few businesses that even *look* closely at CPU speed/performance/price. It's taken as a whole. "We're getting X type of machine for Y price." Sure, some firms want to get the latest-and-greatest, but for most, it's about the whole package.
Competition in the marketplace is responsible for that.
Your dual/single insights are so right on the money--I just had to learn the hard way.:-)
About 9 months ago I went from a 22" CRT primary and a 17" CRT secondary to a single 30" LCD primary with no secondary. My belief was that since it was such a large display, I'd get some of my desk real estate back and still have the screen room to work. No dice.
What I found was all-too-often I was fussing around having move things hither-and-yonder because I'd grown so accustomed to the benefits of having dual displays for HTML / SQL / coding stuff and efficiently using 'em for that. The single display just wasn't cuttin' it.
So... another trip to Ye Olde Apple store for a LCD (a smaller 20"), and all is well again. One display for the task at hand... one display for things to keep an eye on.
Aside from the joy of it for those that are so inclined, the main reason for working on other architectures is because it often brings to light subtle errors in code--particularly in the compiler--because of the differences in the hardware's instructions and such.
In the case of the VAX and Alphas, both out-dated platforms to many people, they've both been quite good at making coding errors surface, so they're very useful for that if nothing else.
If memory serves in fact, one of the OpenBSD devs, Miod, fixed such an error in the compiler that was picked up because the VAX puked in building X on the same compiler instructions that other platforms were perfectly willing to tolerate.
In the end it produces a better product for all of us since it can often help developers find and fix bugs--especially the hard-to-find and hard-to-duplicate varities. That's pretty cool.
Whether or not we as a society are keen to admit it, the fact of the matter is that credit checks are a fairly major indicator of an employee's likelihood to steal. There are a *lot* of examples of people being put--and putting their employers--into ugly, compromising situations because of the employee's debt. Put simply: Increase the degree of that person's likelihood to need "a way out," and you WILL increase the risk of corporate theft and embezzlement.
This may not represent YOU as a person, but it does represent people generally speaking.
Doing background checks on individuals--especially those with access to your company's till--should quite often include seeing if they're in the position personally to be more likely than others to steal if given the chance.
Let me put it this way: At least at a minimum, at least *do* the credit check on the prospective new hire. That way you as the employer can have a candid discussion about it with the candidate and decide if you're at risk or not.
When someone has started a company and grown it from an idea and a seedling into something real, protecting it is rational. Heck, it's rational to want to protect a company you *didn't* start if only because you want to protect the company to protect your own place there. Let's face it: there is a LOT of trust given to employees in most companies. That trust is indeed (like it or not) room for Very Bad Things(tm) to happen.
Re:Apple opted for poor quality when they chose In
on
Apple's Growing Pains
·
· Score: 1
AMD cooler? What alternate dimension are you from?
In our server racks, the AMD machines are *noticably* cooler both actual CPU temps and the heat they throw out the back. (Yes, they're otherwise identical cases, too.) They also pull quite a bit less power, too. If memory serves it's something like 1/3 less. I ran a meter on them each for a week, and it was a measurable difference. YMMV of course, but that's what I'm seein'
I haven't measured the heat out the fans from the machines with a thermometer, but it is a *substantial* difference. The Xeons are definitely hotter.
I *still* have to wonder about the AMD chips though--sure they might have been a few bucks more (as another poster commented), but let's face it, the PowerPC chips weren't coming in boxes of cereal. I'm sure when the chips were down (pun intended) the PowerPCs were more expensive that Intel procs. And in the overall scheme of things, we're talking very minor price differences in that volume.
Oh, and since you're asking, I hail from Zoron, but let's do be hush-hush about that, eh?;-)
Re:Apple opted for poor quality when they chose In
on
Apple's Growing Pains
·
· Score: 1
I'm not completely sure I buy your logic, but I'll admit, you might be on to something.
I have two Macs myself a 17" G4 PowerBook and a dual 2.3 G5, and even between the G4 and the G5, there are little things that aren't quite right on the G5. Yet (knock wood) nothing weird ever happens on the G4. On the G5 for instance my IMAP mail client truncates emails (yet the same emails are fine in the G4--even days later), a couple of applications randomly decide they don't want to open the associated files... just weird stuff like that.
My theory isn't that it's x86 per se, but that it's a time-from-launch related thing (i.e. generations). I wonder if the x86 swap will indeed magnify these types of issues.
A couple of things have always puzzled me with the x86 selection: why not use AMD or Sun 64 bit processors if you're moving off the PowerPC? They're cooler, lower power, and arguably better performing. Aside from the comments of another poster about the qualty of those Sun machines running x86 chips, Sun's reputation for quality is *very* high. Very.
I have many Sun machines (maybe 20)--even old SPARCstations--and they're all still chugging away. Those that are on the shelf still power on and run when asked to; others still serve as day-to-day machines in a variety of roles.
In contrast my x86 machines from the same era have all died and gone to PC recyclers. Hmmm... maybe your *are* onto something with the x86 thing....
"I just got our quarterty inventory report--we've got TEN extra pallets of 1,000-watt power supplies! I'm steamin' mad about this--I want them priced to MOVE!!"
"Well... uh... I don't think there's much dema.."
"I don't give a rats tail about what you think. Do whatever it takes to move those power supplies!!"
In news today, Supermicro announces a quad Operton motherboard for 1U chassis...."
Microsoft sysprep for XP allows you to create a "MASTER" image and from that master, to automagically regenerate the identical config onto a new hard drive again and again (assuming you have legit license keys, of course).
The beauty is that you can re-load not just one machine, but quite literally *thousands* from your original hard drive, reboot once, put in the license key, reboot again, and you have a fully working machine, patched to the most recent level of your MASTER.
I've used this technique since Windows 2000, and I simply do not reload Windows manually anymore.
We believe consumers are already paying to support broadband access to the Internet through subscription fees and, as a result, consumers should have the freedom to use this connection without limitations
This is the same argument that Level3 tried to play recently with Cogent. We carry *your* traffic, so you have to pay us.
WRONG!
You're carrying the consumer's traffic.
# The web site pays their colo for hosting their site/servers;
# The consumers pay their ISP for their connectivity to the Internet.
The end. Done. Finished. End of story.
Somehow though greed is setting in and the ISPs are trying to charge the web sites for the same thing they're already being paid for once by their own customers. Phooey.
but it's THREE RU tall. Where's this "space savings" they're talking about?
I thought my eyes were deceiving me. 3u. Huh?
In a standard 42 cabinet, that's just 14 machines--assuming you have no switches, patch panels, power strips, UPSes or the like eating into your usable space.
Hmmm... even the quiet-factor doesn't matter much--I don't know of too many desks that sit *in* a server room/colo.
Normally, I'm very appreciative of datacenter goodies... am I just missing something with this?
You might not take personal offense, but I doubt you felt like respecting the judgement of the people who were enforcing such a stupid system.
I totally see your point, and to some extent it's natural to feel that way, but actually, (while you may not believe me) I found it quite heartening. I was quite glad to see it was enforced with me, too, to be honest. It sent a message to the other staff and managers (in a company of ~1,000 people) that there was NO favoritism.
The beauty is that later with potentially problematic people on down the road, it makes it impossible for that person to say to management, "What... don't you trust me?!"
The easy answer is, "It's not about trust. This is policy," and it's easy to see that that policy has been enforced with everyone, everywhere, even-handedly, without bias.
The opposite of this is when one person announces his/her departure and the security credentials are revoked, but when someone else leaves, they're allowed to retain them. To me, that's folly.
I guess at the end of the day the courtesy most people want most is: consistency. Just make everyone else play by the same rules. That's fair.
Really - this is SOP in many, if not most places. At my company, anyone with "sensitive" access is immediately revoked upon receipt of written resignation. Period.
Bear in mind also that this isn't likely a reflection on you per se, but more something procedural. It's there so that when someone with malice in their heart leaves they have less chance of doing it; likewise, it's also there so that if someone *else* is trying to pin something on you--heck you're leaving anyway--there's less chance of that succeeding.
I worked as a retail / service manager in my college job and regularly had a *lot* of cash in my hands by the end-of-day on my closing shifts and even after YEARS with the company as clearly one of the most trusted and respected people in the organization, within a couple of hours of my meeting with the VP to announce my leaving, they'd already changed the combinations and keys and moved me away from shifts where I would be responsible for handling money.
It was no reflection on me. It was the system. I took no offence, harbored no grudge. It was what it was.
Microsoft--and the two staffers shown in this video--deserve strong praise for the *unedited* candor, the self-depricating humor, and the absense of spin on this video.
Maybe I've missed the comments, but what no one seems to mention here is that these guys--clearly both geeks at heart (in a good way)--really are peeling back a lot of the layers of MS's site. The candor about their security problems, the 2gb memory issues, and a variety of other things was refreshing.
Personally, I don't see why there's all the kerfuffle. You're spot on: there are only ~200 these domains available and most of those don't even matter.
So you've got s.pro
Great. Um. Name.
With the billions of people on the globe and ~200--indeed fewer than 100 if you count only the.net/.com/.org--this seems to be not even worth discussing. Further, I'd venture that the majority of them with end up being parked domains anyway, so it's an even smaller issue still.
Ya know... I used to use the MSN Messenger client exclusively, then because of a need to communicate with one person who used Yahoo!, I tried Trillian. HOLY HECK.
With the choice of skins and what not, what a relief. For me the MSN interface has just become this weirdly bloated thing.
Trillian though... what a great piece of software IMHO. And best of all (unlike Messenger), it doesn't create those competely bogus hooks into OE that prevent you from closing the app because it's dependent on the other. Bah.
A big block of IP addresses does not a major ISP make.
Perhaps not on its surface, though this really has nothing to do with blocks reserved per se--especially for other uses like HAM readio. It's about IP block size and what's being done with the IP block.
In this case Cogent has:
1) the entire/8 for its use as an ISP and a common carrier--to say nothing of:
2) THEIR OWN FIBER under the ocean
3) one of only 13 ROOT DNS servers globally (C.ROOT-SERVERS.NET is in Cogent's 192.33.4.0/24 IP space)
as Cogent does, you're surely, surely not a small ISP!
The point is: as most of us are non-multi-homed end users of ISPs--even major ones like Cogent--we're now all subject to the whims of *other* ISPs as to whether or not we can see customers who aren't even hosted by them?!?! Grrrr.
For instance: right now no one on Comcast, Road Runner, or Verizon can see our sites or those of our customers. How does L3 get off doing that?
As a cogent customer, it's *really* not their fault, IMHO. I gather L3 pulled the same stunt with XO last week.
As to the notion by another poster of not expecting peering with someone bigger for free forever: 38.0.0.0/8 Class A is Cogent/PSI... how much bigger than being an entire Class A (and then some?!) does one have to be to be considered [ahem] "equal"?
It was a mutual arrangement: they both allowed transit for one another's packets... pretty fair given the size and stature of them both, I'd say.
Why do people keep saying that FOSS products don't have licenses?
I've got three letters for ya:
F
U
D
If a foggy haze can form over FOSS licensing being a scary question mark, it might just turn one PHB away from the FOSS and towards the proprietary and/or closed source.
Which would you rather do? Write a single application that would run on all Windows systems, or six different ones, each with its own unique quality assurance and support problems?
Ok, once again for the benefit folks in the cheap seats, let's review:
Windows 3.1
Windows 95
Windows 95B
Windows 95B
Windows 98
Windows Me
Windows NT Workstation
Windows NT Server
Windows NT Terminal Server
Windows 2000 Pro
Windows 2000 Server
Windows 2000 Advanced Server
Windows 2000 Server Datacenter Edition
Windows XP Home
Windows XP Pro
Windows XP Pro SP2
Windows XP Pro 64-bit
Windows XP Media Center Edition
Windows 2003 Server
Windows 2003 Server Small Business Server
blah... blah... blah...
OK, now, let's combine that with the various versions of IE
4
4.01
5
5.5
6
...many/all of which have slightly / entirely different APIs, names, usage conventions and you have a Royal Mess(tm). Just look at the IE toolbar market--most companies gave up supporting anything older than XP.
As one Windows C++ developer friend of mine described the process of working with these many versions: "Lions and Tigers and Bears, OH MY!"
Making most any reasonably complex app work on multiple versions of Windows is difficult at best and impossible at worst. That Windows is a panacea is jut plain wrong.
As any low-level hardware developer will tell you, there are a couple of EXTREMELY dangerous assumptions here:
1.) That the original work done is understandable. Writing drivers is H A R D. Sometimes--often in fact--there are VERY subtle, extremely easily overlooked things happening in drivers. This is exacerbated by the fact that in many cases, the work done in developing a driver by one group is garbage and basing code on garbage becomes REAL garbage. No sir. Not interested.
This is to say nothing of: what happens when it breaks in subtle, hard-to-reproduce fashions? Where do you go for your documentation on it? Another driver? [shudders]
2.) That the original work doesn't have encumbered binaries or other equally troublesome hooks.
To get to the real heart of the matter it's this: if Sun will provide Linux developers documentation, why won't they provide BSD developers that same documentation?
I have to second the motion here. Sun's pricing of late is jaw-dropping. My *only* complaint--and the primary reason a great deal of the Sun hardware I have in service is still in service (aside from the fact that it's EXTREMELY reliable)--is that Sun patently refuses to supply OpenBSD with the necessary documentation for supporting the newest Sun hardware.
Grrrr. If I could run OpenBSD on the latest Sun offerings, I'd snap up Sun gear like it was going out of style. (No comments from the peanut gallery on that one, please.;-)
In the mean time the little ol' SPARCstations and Netras are chugging along doing their various tasks without a peep.
the ISPs adequately keep their eyes peeled for infestations in their network. It wouldn't take but a handful of infected machines on 100mb networks to DoS even the healthiest of networks.
Given the HORRRRRRIBLE track record of even some VERY large, notable ISPs in cutting off members who have spam-bot / zombie / worm-infested machines, this increase in bandwith makes me both excited at the good possibilites and shudder at the possible bad, too.
Slashdot Mirror
Your dual/single insights are so right on the money--I just had to learn the hard way. :-)
About 9 months ago I went from a 22" CRT primary and a 17" CRT secondary to a single 30" LCD primary with no secondary. My belief was that since it was such a large display, I'd get some of my desk real estate back and still have the screen room to work. No dice.
What I found was all-too-often I was fussing around having move things hither-and-yonder because I'd grown so accustomed to the benefits of having dual displays for HTML / SQL / coding stuff and efficiently using 'em for that. The single display just wasn't cuttin' it.
So... another trip to Ye Olde Apple store for a LCD (a smaller 20"), and all is well again. One display for the task at hand... one display for things to keep an eye on.
Aside from the joy of it for those that are so inclined, the main reason for working on other architectures is because it often brings to light subtle errors in code--particularly in the compiler--because of the differences in the hardware's instructions and such.
In the case of the VAX and Alphas, both out-dated platforms to many people, they've both been quite good at making coding errors surface, so they're very useful for that if nothing else.
If memory serves in fact, one of the OpenBSD devs, Miod, fixed such an error in the compiler that was picked up because the VAX puked in building X on the same compiler instructions that other platforms were perfectly willing to tolerate.
In the end it produces a better product for all of us since it can often help developers find and fix bugs--especially the hard-to-find and hard-to-duplicate varities. That's pretty cool.
Whether or not we as a society are keen to admit it, the fact of the matter is that credit checks are a fairly major indicator of an employee's likelihood to steal. There are a *lot* of examples of people being put--and putting their employers--into ugly, compromising situations because of the employee's debt. Put simply: Increase the degree of that person's likelihood to need "a way out," and you WILL increase the risk of corporate theft and embezzlement.
This may not represent YOU as a person, but it does represent people generally speaking.
Doing background checks on individuals--especially those with access to your company's till--should quite often include seeing if they're in the position personally to be more likely than others to steal if given the chance.
Let me put it this way: At least at a minimum, at least *do* the credit check on the prospective new hire. That way you as the employer can have a candid discussion about it with the candidate and decide if you're at risk or not.
When someone has started a company and grown it from an idea and a seedling into something real, protecting it is rational. Heck, it's rational to want to protect a company you *didn't* start if only because you want to protect the company to protect your own place there. Let's face it: there is a LOT of trust given to employees in most companies. That trust is indeed (like it or not) room for Very Bad Things(tm) to happen.
I haven't measured the heat out the fans from the machines with a thermometer, but it is a *substantial* difference. The Xeons are definitely hotter.
I *still* have to wonder about the AMD chips though--sure they might have been a few bucks more (as another poster commented), but let's face it, the PowerPC chips weren't coming in boxes of cereal. I'm sure when the chips were down (pun intended) the PowerPCs were more expensive that Intel procs. And in the overall scheme of things, we're talking very minor price differences in that volume.
Oh, and since you're asking, I hail from Zoron, but let's do be hush-hush about that, eh?
I'm not completely sure I buy your logic, but I'll admit, you might be on to something.
I have two Macs myself a 17" G4 PowerBook and a dual 2.3 G5, and even between the G4 and the G5, there are little things that aren't quite right on the G5. Yet (knock wood) nothing weird ever happens on the G4. On the G5 for instance my IMAP mail client truncates emails (yet the same emails are fine in the G4--even days later), a couple of applications randomly decide they don't want to open the associated files... just weird stuff like that.
My theory isn't that it's x86 per se, but that it's a time-from-launch related thing (i.e. generations). I wonder if the x86 swap will indeed magnify these types of issues.
A couple of things have always puzzled me with the x86 selection: why not use AMD or Sun 64 bit processors if you're moving off the PowerPC? They're cooler, lower power, and arguably better performing. Aside from the comments of another poster about the qualty of those Sun machines running x86 chips, Sun's reputation for quality is *very* high. Very.
I have many Sun machines (maybe 20)--even old SPARCstations--and they're all still chugging away. Those that are on the shelf still power on and run when asked to; others still serve as day-to-day machines in a variety of roles.
In contrast my x86 machines from the same era have all died and gone to PC recyclers. Hmmm... maybe your *are* onto something with the x86 thing....
"Yah, George?"
"I just got our quarterty inventory report--we've got TEN extra pallets of 1,000-watt power supplies! I'm steamin' mad about this--I want them priced to MOVE!!"
"Well... uh... I don't think there's much dema.."
"I don't give a rats tail about what you think. Do whatever it takes to move those power supplies!!"
Microsoft sysprep for XP allows you to create a "MASTER" image and from that master, to automagically regenerate the identical config onto a new hard drive again and again (assuming you have legit license keys, of course).
The beauty is that you can re-load not just one machine, but quite literally *thousands* from your original hard drive, reboot once, put in the license key, reboot again, and you have a fully working machine, patched to the most recent level of your MASTER.
I've used this technique since Windows 2000, and I simply do not reload Windows manually anymore.
WRONG!
You're carrying the consumer's traffic.
# The web site pays their colo for hosting their site/servers;
# The consumers pay their ISP for their connectivity to the Internet.
The end. Done. Finished. End of story.
Somehow though greed is setting in and the ISPs are trying to charge the web sites for the same thing they're already being paid for once by their own customers. Phooey.
Three cheers for Google.
In a standard 42 cabinet, that's just 14 machines--assuming you have no switches, patch panels, power strips, UPSes or the like eating into your usable space.
Hmmm... even the quiet-factor doesn't matter much--I don't know of too many desks that sit *in* a server room/colo.
Normally, I'm very appreciative of datacenter goodies... am I just missing something with this?
The beauty is that later with potentially problematic people on down the road, it makes it impossible for that person to say to management, "What... don't you trust me?!"
The easy answer is, "It's not about trust. This is policy," and it's easy to see that that policy has been enforced with everyone, everywhere, even-handedly, without bias.
The opposite of this is when one person announces his/her departure and the security credentials are revoked, but when someone else leaves, they're allowed to retain them. To me, that's folly.
I guess at the end of the day the courtesy most people want most is: consistency. Just make everyone else play by the same rules. That's fair.
Bear in mind also that this isn't likely a reflection on you per se, but more something procedural. It's there so that when someone with malice in their heart leaves they have less chance of doing it; likewise, it's also there so that if someone *else* is trying to pin something on you--heck you're leaving anyway--there's less chance of that succeeding.
I worked as a retail / service manager in my college job and regularly had a *lot* of cash in my hands by the end-of-day on my closing shifts and even after YEARS with the company as clearly one of the most trusted and respected people in the organization, within a couple of hours of my meeting with the VP to announce my leaving, they'd already changed the combinations and keys and moved me away from shifts where I would be responsible for handling money.
It was no reflection on me. It was the system. I took no offence, harbored no grudge. It was what it was.
Microsoft--and the two staffers shown in this video--deserve strong praise for the *unedited* candor, the self-depricating humor, and the absense of spin on this video.
:-)
Maybe I've missed the comments, but what no one seems to mention here is that these guys--clearly both geeks at heart (in a good way)--really are peeling back a lot of the layers of MS's site. The candor about their security problems, the 2gb memory issues, and a variety of other things was refreshing.
Heck, they even mention firefox.
Good work all. Good work.
Personally, I don't see why there's all the kerfuffle. You're spot on: there are only ~200 these domains available and most of those don't even matter.
.net/.com/.org--this seems to be not even worth discussing. Further, I'd venture that the majority of them with end up being parked domains anyway, so it's an even smaller issue still.
So you've got s.pro
Great. Um. Name.
With the billions of people on the globe and ~200--indeed fewer than 100 if you count only the
Ya know... I used to use the MSN Messenger client exclusively, then because of a need to communicate with one person who used Yahoo!, I tried Trillian. HOLY HECK.
With the choice of skins and what not, what a relief. For me the MSN interface has just become this weirdly bloated thing.
Trillian though... what a great piece of software IMHO. And best of all (unlike Messenger), it doesn't create those competely bogus hooks into OE that prevent you from closing the app because it's dependent on the other. Bah.
In this case Cogent has: as Cogent does, you're surely, surely not a small ISP!
The point is: as most of us are non-multi-homed end users of ISPs--even major ones like Cogent--we're now all subject to the whims of *other* ISPs as to whether or not we can see customers who aren't even hosted by them?!?! Grrrr.
For instance: right now no one on Comcast, Road Runner, or Verizon can see our sites or those of our customers. How does L3 get off doing that?
As a cogent customer, it's *really* not their fault, IMHO. I gather L3 pulled the same stunt with XO last week.
As to the notion by another poster of not expecting peering with someone bigger for free forever: 38.0.0.0/8 Class A is Cogent/PSI... how much bigger than being an entire Class A (and then some?!) does one have to be to be considered [ahem] "equal"?
It was a mutual arrangement: they both allowed transit for one another's packets... pretty fair given the size and stature of them both, I'd say.
Ok, once again for the benefit folks in the cheap seats, let's review:
Windows 3.1
Windows 95
Windows 95B
Windows 95B
Windows 98
Windows Me
Windows NT Workstation
Windows NT Server
Windows NT Terminal Server
Windows 2000 Pro
Windows 2000 Server
Windows 2000 Advanced Server
Windows 2000 Server Datacenter Edition
Windows XP Home
Windows XP Pro
Windows XP Pro SP2
Windows XP Pro 64-bit
Windows XP Media Center Edition
Windows 2003 Server
Windows 2003 Server Small Business Server
blah... blah... blah...
OK, now, let's combine that with the various versions of IE
4
4.01
5
5.5
6
As one Windows C++ developer friend of mine described the process of working with these many versions: "Lions and Tigers and Bears, OH MY!"
Making most any reasonably complex app work on multiple versions of Windows is difficult at best and impossible at worst. That Windows is a panacea is jut plain wrong.
Er, aren't the Sun drivers for Linux open-source?
No.
Can't the OpenBSD folk just port them?
No.
As any low-level hardware developer will tell you, there are a couple of EXTREMELY dangerous assumptions here:
1.) That the original work done is understandable. Writing drivers is H A R D. Sometimes--often in fact--there are VERY subtle, extremely easily overlooked things happening in drivers. This is exacerbated by the fact that in many cases, the work done in developing a driver by one group is garbage and basing code on garbage becomes REAL garbage. No sir. Not interested.
This is to say nothing of: what happens when it breaks in subtle, hard-to-reproduce fashions? Where do you go for your documentation on it? Another driver? [shudders]
2.) That the original work doesn't have encumbered binaries or other equally troublesome hooks.
To get to the real heart of the matter it's this: if Sun will provide Linux developers documentation, why won't they provide BSD developers that same documentation?
If you really want the scoop on this, here's a good starting point: Sun OpenBSD documentation article
I have to second the motion here. Sun's pricing of late is jaw-dropping. My *only* complaint--and the primary reason a great deal of the Sun hardware I have in service is still in service (aside from the fact that it's EXTREMELY reliable)--is that Sun patently refuses to supply OpenBSD with the necessary documentation for supporting the newest Sun hardware.
;-)
Grrrr. If I could run OpenBSD on the latest Sun offerings, I'd snap up Sun gear like it was going out of style. (No comments from the peanut gallery on that one, please.
In the mean time the little ol' SPARCstations and Netras are chugging along doing their various tasks without a peep.
This seems akin to taking the body of a Ferrari and putting it atop a '78 Maverick!
To say nothing about adding many, many layers of complexities into the mix....
In short this strikes me as a classic example of a Very Bad Idea (tm).
The mods must be on the pipe. How did your post possibly get an "interesting" rating?!
:-|
Maybe if you wrap your post next time in [sarcasm] [/sarcasm] tags it will be clearer that THIS IS A JOKE.
I'm all for this AS LONG AS...
the ISPs adequately keep their eyes peeled for infestations in their network. It wouldn't take but a handful of infected machines on 100mb networks to DoS even the healthiest of networks.
Given the HORRRRRRIBLE track record of even some VERY large, notable ISPs in cutting off members who have spam-bot / zombie / worm-infested machines, this increase in bandwith makes me both excited at the good possibilites and shudder at the possible bad, too.