Why? apple have PCI slots but I can't use every piece of PCI hardware known to man in there.
If Apple were making x86-based machines I'm sure it'll be apple-certified style hardware (rather than "dell" certified home PC hardware, the cheapest batch they could buy that day) with apple-certified peripherals... as long as I can throw my GeForce 4 or newer graphics card in.:)
If the IP world standardises on interception technologies then we'll have some idea of how to thwart it.
Bring it on. I know you're doing it anyway. Bring it on, let people see what you're doing, let privacy advocates explain to the general public that yes, major internet equipment supports sniffing their traffic, look here for the standard and bewm! Maybe you'll get some sympathy.
I've tried explaining to lay people (non-technical friends) what can be done with todays technology and they look at me dumbfounded. Track your position by your cell phone? Huge databases to analyse the spending patterns of people? What about communication interception? Heck, I've shown a few friends pictures of the golf balls in the UK and they still refused to accept it. sigh!
Well, its what would happen if you asked, really nicely, to multihome some upstream-allocated network block.
Ie, ISP A has a/16. THey give you, B a/23. B decides to multihome with it and ISP grants them permission to - it _can_ happen, it _does_ happen albeit infrequently (thank universe!).
Try aussie-isp (majordomo@aussie.net), but still I'd hang out on nanog to catch the network-related stuff. The US _is_ the internet still regardless of what the rest of us think.
You _can_ get lucky if you're _near_ the provider in question with the superblock you're in.
Example: Say you've got x.x.x.0/24 out of x.x.0.0/16. Now, if people ignore you're announcement they're going to send traffic towards the provider announcing x.x.0.0/16. Somewhere along the way a network in the path might actually be paying attention to your routes, and your traffic gets shuffled towards you.
(But then, somewhere between THERE and you might be a network which doesn't pay attention and it heads back towards the/16 announcement.)
In short - remember, routing is hop-by-hop. Just because n-1 nodes in the path are listening to the announcement, things don't have to work. Similarly things might be working even if a node in the path isn't listening.
Now, some more facts - do some googling to determine meanings behind some terms/acronyms:
* the whole internet isn't populated with/20s and larger. In fact, there's still a lot of historical "swamp space" - see 203.0.0.0/10 (Australia). Its full of/24s. They're still globally visible because when the "nazi" filters were making the rounds at NANOG a while back.) If you're resourceful you might find the filters Randy Bush made up whilst working at Verio (i think!) which limited netmask lengths based on prefixes. So, fe, large chunks of space had a/19 limitation but the swamp space didn't. It was copied, verbatim, into many Cisco routers.
* Mass BGP filtering isn't to protect memory usage, its also to protect update times. Those CPUs can only _talk_ to neighbouring routers at speeds much below the linerates of cards (even today!:) and so taking 20 minutes to pull in a full BGP table would be 20 minutes where most routers performed in a degraded state. (yes, routers today are increasingly using seperate lookup, forwarding and data paths, but..)
* For a fun bit of historical information do some google searching for the AS7007 incident (or the mass deaggregation/redistribution incident.) Basically someone confed up a router, deaggregated large chunks of IP space into/24s and started locking up parts of the internet. Unfortunately due to bugs in software and non-instant propagation times these announcements just kept going round and round. Eventually netadmins had to coordinate with each other to shut down large parts of the internet "backbone" (there was a definable one, mostly, back then) to purge the announcements and then bring stuff back up again.
Phew. I drifted a bit there. I find it interesting to listen and learn about things like this so one doesn't make the mistake in other fields.
Thats bizarre. I have a Z505HE (12inch PIII-450ish, 192meg RAM, 10gig hard disk) and the _case_ is wearing away after 3 years of constant use. It hasn't missed a beat and I've carried the thing in my bags around the world many times.
I also have a picturebook of about the same vintage (pii-300ish, 64meg RAM).
The Z505 is my FreeBSD travel laptop and the Picturebook is my Win98 travel laptop. Different situations call for different laptops - the software modems didn't work at all under FreeBSD/Linux at the time.
Here's the crunch - the batteries are both dead. They give me about 5 minutes of power - enough to change rooms. New batteries will set me back about AUD$800 a piece for 2ish hour jobs. $1600 is 2/3 the cost of.. my 12inch iBook.
Actually, you'll generally find that a slashdotted box on 10megabits of bandwidth dies due to server load, not a lack of bandwidth.
And thats generally because people don't tune their boxes to run with many files/sockets, they keep running apache when they could be running something like boa or thttpd..
Case in point: a couple of years ago a good friend of mine decided to host some, lewd pictures of an actress. His machine was immediately swamped as the URL leaked out over the internet. 30 minutes and an install of boa later the load average dropped down to under 0.5 whilst serving out a good 10 mbits.
(I'm just as guilty - I run apache on my website. But then I don't host my cute stuff there.:)
I initially thought this - but after doing a bit of research found that this is a tad more powerful.
For example, if you have a system with certain cards they may be able to DMA directly into the full memory space. Some drivers (most?:) will require 'bounce-buffer' tricks.
(bounce-buffering is where you DMA into low-memory and then copy the buffers into the required location. For example, think ISA cards in a system with >16meg of RAM. You can't simply require all the system buffers to be in the 16meg ISA address range so you DMA into a set of buffers in low-memory and then copy the data into place.)
Err, they can already triangulate your position relatively accurately using GSM (and I guess GPRS, PCS, etc.)
Remember - if your handset can measure the signal strength to certain basestations to choose optimal ones to continue calls on, so can the base stations.
Over-sampling happens everywhere when you're actually processing, well, anything digitally.
An example - you wouldn't work with video at, say, 700x525 if you're going to be doing any scaling or rotation work. You'll end up with scaling/rotation artifacts (where pixels, after a translation, end up taking up partial pixels everywhere.)
Now I'm sure a similar thing could happen with frame rates. Although, that said, there's lots of whackiness inherent in frame/score timings - take a look at this for an excellent introduction.
(Disclaimer: I'm not a graphic artist - I'm a coder. I just helped a desperate friend finish a couple of her short-film projects in Final Cut Pro. Its one of the only times I read the manual beforehand.)
Not necessarily - it just has to do well in Korea or a couple other asian countries.
I'm sure there are a lot of geek products made out there - I've seen cell phones in Japan and Korea that I'm sure I'll never see in Australia or the US - which are just damned nifty.
Not once did you acknowledge that your lifestyle is a direct consequence of those actions.
America's foreign and local policies, as blunt, unfair and unbalanced as they are, have opened the floodgates for the large middle-class lifestyle most of you (me, in AU) enjoy.
As an example, I may not agree with how the US handled the construction of the Panama Canal by supporting the overthrow of the government, furthering its interests in the area.
(Admittedly that article doesn't cover the actual uprising much, but do some research - its quite interesting reading.)
Now, think about it little. The canal did benefit America - both the rich and the poor. It opened up trade avenues previously closed.
Big business might have its pockets in the government but that is simply an expression of human nature. And its not the only one causing much of/.'s problems.
There has to be a balance between government interests, corporate interests and the interests of average citizens. Each of them requires and spawns the other two in the hybrid capitalism/socialism/democracy that the western world embodies.
Right now corporations and the government are pushing against the average citizen, grabbing more and more headway because the average citizen is content. Until that changes or something emotionally-uprising appears you're not going to see many of them pushing back.
Plot a few customer satisfaction points based on employee costs/employee usefulness.
Now, imagine where the H1B employees sit. Imagine where the local workers sit. Draw a great big glob around what your customers are willing to accept.
Then, in this economy, take the least cost route.
I've seen managers do exactly this. Its all about balance. Unless elitism is your selling point a company is more likely to tend towards bottom lines than building long-term employee relations.
I don't blame the companies. They're simply trying to stay financially viable.
Re:Pah, cann't be bothered reading the article
on
A Better Finder?
·
· Score: 1
Regardless of its origins the Chooser interface is sorely missed by just about any MacOS user I've come across.
That aside, there is gobs of room for improvement in mounting SMB shares. The interface is.. cubersome feeling. That big empty dialog box with a small drop-down box to select a share - eww. Those error _CODES_ when an SMB mount fails - double eww.
It feels like someone just slapped it on as an afterthought.;/
Apple, if you're listening, please improve your SMB mount interface!
I don't think you need a visible frequency to blind someone - I'm sure you could do it with an intense pulse of ultraviolet. You just need something which will be absorbed by the retina (rather than passing straight through it, eg radio or x-rays.)
Slashdot Mirror
Only static apps. You could, eww, put some alias-understanding code in the libc open() function call. That'd trap most of the abuse cases.
Thats old-school X86. New-School X86 has this wonderful thing called APIC which is a bit nicer to program for.
Rumour has it that all x86-64 machines will come with APICs - including single-CPU machines.
Why? apple have PCI slots but I can't use every piece of PCI hardware known to man in there.
.. as long as I can throw my GeForce 4 or newer graphics card in. :)
If Apple were making x86-based machines I'm sure it'll be apple-certified style hardware (rather than "dell" certified home PC hardware, the cheapest batch they could buy that day) with apple-certified peripherals.
If the IP world standardises on interception technologies then we'll have some idea of how to thwart it.
Bring it on. I know you're doing it anyway. Bring it on, let people see what you're doing, let privacy advocates explain to the general public that yes, major internet equipment supports sniffing their traffic, look here for the standard and bewm! Maybe you'll get some sympathy.
I've tried explaining to lay people (non-technical friends) what can be done with todays technology and they look at me dumbfounded. Track your position by your cell phone? Huge databases to analyse the spending patterns of people? What about communication interception? Heck, I've shown a few friends pictures of the golf balls in the UK and they still refused to accept it. sigh!
Whoa there, cowboy. What, you think a single machine can't handle 2000 concurrent TCP connections without dying?
Err, I think my term abuse was a bit wonky.
/16 provider - since you've grabbed IP space off an ISP (B) who has grabbed something larger from its upstream provider (C).
/16 since D is filtering. So, /16.
If you have an AS path A -> B -> C -> D -> E
C has a link to this
Now, E and D only see the
from E's perspective the network looks:
E -> D -> C -> F, F being this provider with the
ok, now D sees the network as:
D -> C -> F
but C sees the network as
C -> B
and, since B has subdivided this network block up,
B -> A
Used to happen quite frequently out here in Australia.
Gotta love swamp space.
Well, its what would happen if you asked, really nicely, to multihome some upstream-allocated network block.
/16. THey give you, B a /23. B decides to multihome with it and ISP grants them permission to - it _can_ happen, it _does_ happen albeit infrequently (thank universe!).
Ie, ISP A has a
Try aussie-isp (majordomo@aussie.net), but still I'd hang out on nanog to catch the network-related stuff. The US _is_ the internet still regardless of what the rest of us think.
You _can_ get lucky if you're _near_ the provider in question with the superblock you're in.
/16 announcement.)
/20s and larger. In fact, there's still a lot of historical "swamp space" - see 203.0.0.0/10 (Australia). Its full of /24s. They're still globally visible because when the "nazi" filters were making the rounds at NANOG a while back.) If you're resourceful you might find the filters Randy Bush made up whilst working at Verio (i think!) which limited netmask lengths based on prefixes. So, fe, large chunks of space had a /19 limitation but the swamp space didn't. It was copied, verbatim, into many Cisco routers.
:) and so taking 20 minutes to pull in a full BGP table would be 20 minutes where most routers performed in a degraded state. (yes, routers today are increasingly using seperate lookup, forwarding and data paths, but..)
/24s and started locking up parts of the internet. Unfortunately due to bugs in software and non-instant propagation times these announcements just kept going round and round. Eventually netadmins had to coordinate with each other to shut down large parts of the internet "backbone" (there was a definable one, mostly, back then) to purge the announcements and then bring stuff back up again.
Example: Say you've got x.x.x.0/24 out of x.x.0.0/16.
Now, if people ignore you're announcement they're going to send traffic towards the provider announcing x.x.0.0/16. Somewhere along the way a network in the path might actually be paying attention to your routes, and your traffic gets shuffled towards you.
(But then, somewhere between THERE and you might be a network which doesn't pay attention and it heads back towards the
In short - remember, routing is hop-by-hop. Just because n-1 nodes in the path are listening to the announcement, things don't have to work. Similarly things might be working even if a node in the path isn't listening.
Now, some more facts - do some googling to determine meanings behind some terms/acronyms:
* the whole internet isn't populated with
* Mass BGP filtering isn't to protect memory usage, its also to protect update times. Those CPUs can only _talk_ to neighbouring routers at speeds much below the linerates of cards (even today!
* For a fun bit of historical information do some google searching for the AS7007 incident (or the mass deaggregation/redistribution incident.) Basically someone confed up a router, deaggregated large chunks of IP space into
Phew. I drifted a bit there. I find it interesting to listen and learn about things like this so one doesn't make the mistake in other fields.
.. and turn on query logging, do some explain/analyze select stuff and add indexes.
Thats bizarre. I have a Z505HE (12inch PIII-450ish, 192meg RAM, 10gig hard disk) and the _case_ is wearing away after 3 years of constant use. It hasn't missed a beat and I've carried the thing in my bags around the world many times.
.. my 12inch iBook.
I also have a picturebook of about the same vintage (pii-300ish, 64meg RAM).
The Z505 is my FreeBSD travel laptop and the Picturebook is my Win98 travel laptop. Different situations call for different laptops - the software modems didn't work at all under FreeBSD/Linux at the time.
Here's the crunch - the batteries are both dead. They give me about 5 minutes of power - enough to change rooms. New batteries will set me back about AUD$800 a piece for 2ish hour jobs. $1600 is 2/3 the cost of
Having them. You can then build up a defence against similar situations.
Awareness or experience is always preferable over ignorance.
Actually, you'll generally find that a slashdotted box on 10megabits of bandwidth dies due to server load, not a lack of bandwidth.
:)
And thats generally because people don't tune their boxes to run with many files/sockets, they keep running apache when they could be running something like boa or thttpd..
Case in point: a couple of years ago a good friend of mine decided to host some, lewd pictures of an actress. His machine was immediately swamped as the URL leaked out over the internet. 30 minutes and an install of boa later the load average dropped down to under 0.5 whilst serving out a good 10 mbits.
(I'm just as guilty - I run apache on my website. But then I don't host my cute stuff there.
I initially thought this - but after doing a bit of research found that this is a tad more powerful.
For example, if you have a system with certain cards they may be able to DMA directly into the full memory space. Some drivers (most?
(bounce-buffering is where you DMA into low-memory and then copy the buffers into the required location. For example, think ISA cards in a system with >16meg of RAM. You can't simply require all the system buffers to be in the 16meg ISA address range so you DMA into a set of buffers in low-memory and then copy the data into place.)
Thats why one would put the actual power enabling logic in _hardware_.
What I'd like to see is a microwave which can sense when something is going bad and _turn itself off_.
You know, like those "smart irons" people have been selling for a while which turn themselves off (mostly) before they start a fire.
Err, they can already triangulate your position relatively accurately using GSM (and I guess GPRS, PCS, etc.)
Remember - if your handset can measure the signal strength to certain basestations to choose optimal ones to continue calls on, so can the base stations.
Over-sampling happens everywhere when you're actually processing, well, anything digitally.
An example - you wouldn't work with video at, say, 700x525 if you're going to be doing any scaling or rotation work. You'll end up with scaling/rotation artifacts (where pixels, after a translation, end up taking up partial pixels everywhere.)
Now I'm sure a similar thing could happen with frame rates. Although, that said, there's lots of whackiness inherent in frame/score timings - take a look at this for an excellent introduction.
(Disclaimer: I'm not a graphic artist - I'm a coder. I just helped a desperate friend finish a couple of her short-film projects in Final Cut Pro. Its one of the only times I read the manual beforehand.)
Unless you "read" stileproject!
What will you do when computers start returning different results? :-)
Not necessarily - it just has to do well in Korea or a couple other asian countries.
I'm sure there are a lot of geek products made out there - I've seen cell phones in Japan and Korea that I'm sure I'll never see in Australia or the US - which are just damned nifty.
Yes there is..
Because you can.
Not once did you acknowledge that your lifestyle is a direct consequence of those actions.
/.'s problems.
America's foreign and local policies, as blunt, unfair and unbalanced as they are, have opened the floodgates for the large middle-class lifestyle most of you (me, in AU) enjoy.
As an example, I may not agree with how the US handled the construction of the Panama Canal by
supporting the overthrow of the government, furthering its interests in the area.
(Admittedly that article doesn't cover the actual uprising much, but do some research - its quite interesting reading.)
Now, think about it little. The canal did benefit America - both the rich and the poor. It opened up trade avenues previously closed.
Big business might have its pockets in the government but that is simply an expression of human nature. And its not the only one causing much of
There has to be a balance between government interests, corporate interests and the interests of average citizens. Each of them requires and spawns the other two in the hybrid capitalism/socialism/democracy that the western world embodies.
Right now corporations and the government are pushing against the average citizen, grabbing more and more headway because the average citizen is content. Until that changes or something emotionally-uprising appears you're not going to see many of them pushing back.
Imagine a 3-dimensional space of:
* employee costs
* employee usefulness
* customer satisfaction
Plot a few customer satisfaction points based on employee costs/employee usefulness.
Now, imagine where the H1B employees sit. Imagine where the local workers sit. Draw a great big glob around what your customers are willing to accept.
Then, in this economy, take the least cost route.
I've seen managers do exactly this. Its all about balance. Unless elitism is your selling point a company is more likely to tend towards bottom lines than building long-term employee relations.
I don't blame the companies. They're simply trying to stay financially viable.
Regardless of its origins the Chooser interface is sorely missed by just about any MacOS user I've come across.
.. cubersome feeling. That big empty dialog box with a small drop-down box to select a share - eww. Those error _CODES_ when an SMB mount fails - double eww.
;/
That aside, there is gobs of room for improvement in mounting SMB shares. The interface is
It feels like someone just slapped it on as an afterthought.
Apple, if you're listening, please improve your SMB mount interface!
I don't think you need a visible frequency to blind someone - I'm sure you could do it with an intense pulse of ultraviolet. You just need something which will be absorbed by the retina (rather than passing straight through it, eg radio or x-rays.)