If 128MB version costs $25, why they didn't go with 2GB for $30 instead?
That would be a *great* idea! I'm stuck with a BSD machine with 128 MB. If I load up a page full of images, *Xorg* uses 600 MB !! I'm running firefox remotely over X11, and just displaying that takes up 5x system memory (I happen to have 2 GB of swap, can't remember why).
Anyway, I agree with the GP. Plus: calling anything coming from Chine "The Chinese" just leads to cultural ignorance. If they were from US, they would be {University name} researchers, from the EU they would be {country or big university} researchers, but from China they are just "The Chinese". I understand that the level of detail in reporting should decrease with geographical distance. Still, writing "The Chinese" as the origin when a chinese company makes a self-driving car, or a researcher has a wacky idea is just purposeful ignorance. It could lead to xenophobia or animosity, but I think ignorance is bad enough by itself.
(not related to my previous post:) How does this work with DNS caches? Say Ebay implemented this and gave back A records depending on the client IP address. If user A was using Google DNS in Norway and requested DNS records for Ebay, then Google could
a) Cache the reply and return those for all other users (B,C,D...). This would mean that all users of that Google DNS server, maybe over all of Europe, would get the A records for Ebay's Norway server (I don't think Ebay has a server in Norway, but never mind that)
b) Perform a new request to Ebay for each user of its DNS service, but this would slow down DNS requests!
So which one is it? Does google have a separate cache for each/16? I'm interested to know how they deal with this. I guess it's not a dealbreaker, but all options involve some trade-offs.
I was preparing to hate this (I don't trust Google that much & OpenDNS do some questionable things), but I can't find anything wrong with this from a privacy or openness perspective. I think there has already been huge DNS pools that return a random set of records, so the idea of DNS being universal and reproducible is gone, if that idea ever existed (and I see no reason why that would be a problem).
Isn't this what anycast is supposed to solve? For example when using 6to4, one can specify a single, globally known address, but it takes you to a local 6to4 gateway (if it happens to work at a given ISP..). Would anycast support commercial entities setting up a CDNs with a few anycast addresses, that route to the nearest server farm? I don't know if I prefer this, since it requires putting some "intelligence" in the routers. Going by the adoption rate of IPv6, this wouldn't be supported by ISP before 2030 anyway...
Sorry for researching this, but if anycast is better supported in IPv6, then that could be why the explanation page lacks any mention of IPv6. Sites would have to supply a few normal addresses in addition to the anycast address in DNS for redundancy, but browsers could try the anycast one first for speed. I think this would be a good solution, but I don't know if it's better than Google and OpenDNS's suggestion.
I'm typing this on a FreeBSD machine with an R128 graphics card and 128 MB of system RAM (waiting for my laptop to come back from repairs). It can run firefox locally, but I'm using it over networked X11 because it's a bit faster (it's running on my gateway/server machine). It is my main machine at the moment, and it's working very well. Things like xterm, gvim and irssi and ssh are perfectly fine!
Many people have the idea that computers get slower just by being old, but I think/. readers are aware that it's just the software that gets more complex. I understand that MESA would make this decision, and I hope that it was motivated by some worthwhile changes in the APIs, maybe linked with supporting new and exciting devices.
The problem with dropping support is that many organisation take away the old versions from their web or ftp sites. I experienced this recently, trying to get Python 2.6 for windows, because some software depended on it. I would have to build Python myself, but found somewhere to copy it from . If these people don't make the old version available, then years of coding effort will be lost, code which could still have benefited many people.
"Plenty of people work on "old" cars and trucks for utilitarian reasons. (In my area a forty-year-old truck is a common work vehicle.) It saves a shitload of money and will in future."
True indeed, here I was complaining about generalisation, and I was making a huge one myself. It's also interesting to hear about the car computers, there has been some talk about how these are "black boxes" in the tech press, but the people writing that are probably just as clueless as I am.
There is so much *great* tech in IT which can be used by anyone. Web servers, encryption, file sharing (both local and on the internet), and these are just some random examples... This will all go away, and I hate it.
Before electricity became ubiquitous, there were lots of great products that used hand-cranks and other genious mechanical systems. Now, living three days without electrical power would be a disaster. People will choose a electrical ("automatic") product over the mechanical one, even if the mechanical product is in all ways better.
The cloud (and web tech as well, I would argue) is going to be what everyone uses. It is generally easier to use, but there are lots of exceptional programs that aren't "cloud based". Sadly, the mainstream doesn't seem to be able to handle exceptions well, and will discard these. The cloud also makes more business sense, which always helps. If people depend on you (a business) 24/7, that's much better than if they interact with you 1000 times a day, and could easily find a replacement (think hosted e-mail vs. POP3 or running your own server).
I currently run a few linux/BSD machines at home, providing some very nice services for me.I recently realised that I will be like the people who make furniture in their garage, fix up old cars (I won't go into the new computerised cars), or even make simple electronic circuits. It can't be defended as an utilitarian thing any more, it's just a hobby. The cloud providers will be like huge chain stores, modern car makers and HP/Dell/Apple in comparison.
So in conclusion IT will be concentrated in large boring companies that know which products normal consumers and enterprises want, and provide them cheaply and easily. I'm actually depressed about this -- I thought that IT was something special: it has to do with how we learn, communicate and even think. Not only that, it can make millions of impossible tasks seem like a 1 hour job. When looking more closely however, telephone services, plumbing and automotive transportation were just as revolutionary. I can only hope that a new field of such great innovation will appear soon.
You don't *need* a good firewall. A firewall is meant to give an extra layer of good security, but it is used in 90 % of the cases as a quick fix for configuration errors. OSes need to close off unnecessary services anyway, because people take laptops to all kinds of scary places. The IP stack is not inherently insecure. It doesn't harm you if your machine sends a TCP RST packet or an ICMP unreachable when something tries to connect. If only people could make transport mode IPSec easy, we could have a "local" net that worked from anywhere, the "firewall" would be strong encryption.
When copying from a hard drive to another location on the same disk, the new Ubuntu is a few seconds slower than the previous LTS, both with Unity and using the old GNOME 2 shell of Ubuntu Classic. Ubuntu 11.04 Classic finishes a fraction of a second before Unity.
Desktop users don't care about +/- a few percent in file copy speed. What they should have tested was: does the desktop grind to a slow, unusable halt when copying files? I know 10.10 did, and also Windows to some extent, but not as bad. This would be a huge win if it worked better on the new Ubuntu.
Because let's not mince words about this. Infrastructure is fairly expensive, but once it is in place bandwidth across it is extremely cheap (often approaching as low as 3 cents per gigabyte, according to several studies).
What studies, exactly? 3 cents seems WAY too expensive. Users are already paying tiered prices for the speed (5mbit, 20mbit etc), is that even considered? Probably not.. the price per GB could be an interesting measure anyway, keeping in mind that speed and transfer volume are highly correlated.
If we are excluding infrastructure, I can't think of any reason why it would be that much. The marginal cost of transfering 1GB would come from increased power usage of the switches, and possibly increased maintenance costs (32 bit counters rolling over?). Maybe the copper wires get "worn out" from shooting too many electrons through them.. Seriously, 3 cents/GB??
I still prefer the official MSN client. I actually keep a Windows VM going just for that, and for opening office docs and DRM'ed/proprietary videos. The client has some features that Gaim (it's not called Pidgin on my Ubuntu system yet) lacks, like webcam, and it *doesn't pop up windows in the foreground when I'm doing something else*. I can't believe that people put up with that.
What makes you think that the music labels would allow this feature any more than the Zune squirting? Additionally you portray it to be more permissive than the Zune version. So: is it more restricted, or will it be blocked by the RIAA?
This problem was true for the iPod too, and the iPhone launch changed nothing. The argument would be (I think) that Apple has a monopoly on MP3 players, and is abusing that position to do better in the music sales business. This would be valid if Apple was a monopoly.
The new thing about the iPhone is that it supports bluetooth and 802.11g WLAN. This leaves a potential for transferring music directly between phones, og between a PC (e.g. iTunes) and an iPhone. There is no indication that Apple will officially support it, and they do have 100% control over the platform.
One could say that the iPhone is even more resticted than the Zune because it does not allow music sharing AT ALL, even though it would be relatively easy to implement given the power of the platform.
There are a lot of protability that one could want from wireless, some would require 3G to be REALLY cool:
Syncing with iTunes according to some algorithm, or just adding new bought songs, all while the 'phone is just lying there, maybe charging. (WIFI/BT)
Squirting (WIFI/BT)
Streaming from iTunes (WIFI/BT/3G) (I have that, minus the iTunes part, over 3G using a custom Java app and HTTP)
Buying music directly from iTMS (WIFI/3G)
Internet radio, podcasts, vidcasts (WIFI)
Apple may chose ignore these options, and it may work out fine. In the future, though, people will want these products, because wireless is more convenient, and other companies will make them.
So there is no rush, but Apple will have to deal with the over-the-air DRM issue. And it's gonna be ugly. Like the Zune;)
In particular the statement that he was able to determine there was no wireless router in use at the time cannot be substantiated. It is possible to have a wireless router that NATs you right back to your public IP. In fact, I've done it (with out the wireless part) at least twice for different reasons.
There is also the possibility that a simple access point was used, without any NATing being involved at all. This, of course, requires that the ISP offers multiple public IPs via DHCP, and that a record from the ISP, if any, shows that multiple IPs were in use. Given that no trace of file sharing software was found on the hard drive, it is quite probable that the P2P activity was performed by someone else connected to the AP.
If only a single IP was used by the subscriber, the parent's explanation would be the only possible way, and I assume the plaintiff would need to show to the court some equipment that could do the double NAT'ing, for it to be a good defence.
That picture reminds me of my old 333MHz laptop. My company shipped a bunch of them to Africa, because they weren't being used anymore. I got one too, and my mom keeps it under the couch for web surfing. It runs FreeBSD, and starts x with no WM and only Opera running by default. That runs totally smooth on 64MB of RAM. Of course, it was only mine that had this software, but the other computer were shipped without an OS, and I'm guessing that Linux will be used. The only problem is with net connectivity. I had a spare PCMCIA WLAN card laying around, and I don't think the same is true for the local shaman. If the cost of the OLPC is sufficiently low, it will be a much better alternative, partly due to the SW/HW integration.
preepmtive multi-tasking which is the feature of both OS/2 and NT that distinguishes them from Window 3.x/9x that used co-operative multi-tasking.
Windows 9x did support preemptive multitasking to an extent. The NT-flag/IRET point still is correct, but I must assume that the NT flag was used in Windows 9x too.
Windows 95 and 98 only pre-emptively multitask if the running software is capable of dealing with pre-emptive multitasking. Much of it is not. Even tasks which are part of the operating system (eg updating graphics and window displays) are often adversely affected by programs which "busy wait".
If you wanted to netboot or boot from a LiveCD, you wouldn't need a hard drive either (for liveCD one could store permanent files on a USB drive). Should we sue them for net selling HDD-less desktops?
The hard drive has now become the symbol of the computer working. My desktop 'feels' more responsive when it makes that scratching sound when I click the Firefox icon, although I suspect my silent laptop is almost just as fast.
Also, that sound is the first thing to check when computers lock up, etc.
Why replace something that works well (PS/2)?
on
Leopard Vs. Vista
·
· Score: 2, Insightful
Why do we need USB for connecting a kayboard and a mouse. I don't exactly know how it works on the hardware level, but it seems that the PS/2 connectors do a good job as an interface for keyboards/mice. The PC industry has gotten to the point where every new technology does not have to be better than what it replaces. It's like saying that everyone should use WLANs because they are easier and newer than wired Ethernet. Great, but moving this 4GB file doesn't feel quite as.. "snappy". And is it really as stabl***CARRIER LOST*** Same goes for EFI. There is no reason to rush it, the BIOS works great! No-one cares about the BIOS anymore, the OS (drivers) just talk directly to the HW, so who cares if the system needs partitioning and boots in 16-bit mode , I don't notice that, and neither do 99% of the developers.
This reasoning assumes that the ultimate goal of the Chinese gov't is to arrest its citizens. People using Wikipedia are not necessarily free-thinking or critical off the government, so this would be almost random people (although not the poor farmers who don't have Internet access). I do think the Chinese regime is bad, and it is oppressive, but I don't think it is evil for the sake of being evil. Even if it was, there are thousands of other ways to prosecute random citizens. The un-banning of WP allowed many people to get lots of 'bad' information, probably making this whole venture a net loss for the Chinese government.
If it was trying to arrest people who use WP for finding info which conflicts with their views (i.e. being bad for keeping its perverse amounts of power), it should rather monitor the proxy connections and other work-arounds, for finding people who, for whatever reason, really need to get on the 'wiki.
In summary, as others also have concluded, this may be 1) a technical glitch, 2) a policy change which had unforseen consequences, and was reverted, 3) a political struggle between different factions in the government or 4) a plot to find out what people might like WP, log them, prosecute them, and maybe scare others from using 'open' websites carelessly. The fourth option seems unlikely to me.
Slashdot Mirror
If 128MB version costs $25, why they didn't go with 2GB for $30 instead?
That would be a *great* idea! I'm stuck with a BSD machine with 128 MB. If I load up a page full of images, *Xorg* uses 600 MB !! I'm running firefox remotely over X11, and just displaying that takes up 5x system memory (I happen to have 2 GB of swap, can't remember why).
"Microsoft wants to emulate the success of the iPhone"
So does Apple "want to emulate the success of the PC"?
Sorry, I forgot: Signed "The Norwegian" :)
^^ yes
Anyway, I agree with the GP. Plus: calling anything coming from Chine "The Chinese" just leads to cultural ignorance. If they were from US, they would be {University name} researchers, from the EU they would be {country or big university} researchers, but from China they are just "The Chinese". I understand that the level of detail in reporting should decrease with geographical distance. Still, writing "The Chinese" as the origin when a chinese company makes a self-driving car, or a researcher has a wacky idea is just purposeful ignorance. It could lead to xenophobia or animosity, but I think ignorance is bad enough by itself.
Hi, I asked this below, but I'll try again: How is DNS caching handled at Google's and OpenDNS's servers?
So which one is it? Does google have a separate cache for each /16? I'm interested to know how they deal with this. I guess it's not a dealbreaker, but all options involve some trade-offs.
I was preparing to hate this (I don't trust Google that much & OpenDNS do some questionable things), but I can't find anything wrong with this from a privacy or openness perspective. I think there has already been huge DNS pools that return a random set of records, so the idea of DNS being universal and reproducible is gone, if that idea ever existed (and I see no reason why that would be a problem).
Isn't this what anycast is supposed to solve? For example when using 6to4, one can specify a single, globally known address, but it takes you to a local 6to4 gateway (if it happens to work at a given ISP..). Would anycast support commercial entities setting up a CDNs with a few anycast addresses, that route to the nearest server farm? I don't know if I prefer this, since it requires putting some "intelligence" in the routers. Going by the adoption rate of IPv6, this wouldn't be supported by ISP before 2030 anyway...
Sorry for researching this, but if anycast is better supported in IPv6, then that could be why the explanation page lacks any mention of IPv6. Sites would have to supply a few normal addresses in addition to the anycast address in DNS for redundancy, but browsers could try the anycast one first for speed. I think this would be a good solution, but I don't know if it's better than Google and OpenDNS's suggestion.
Here you go, just not linked from the downloads page: http://python.org/ftp/python/
Thanks! I'll see if I can get that application going (PyROOT) when I get my laptop back :)
I'm typing this on a FreeBSD machine with an R128 graphics card and 128 MB of system RAM (waiting for my laptop to come back from repairs). It can run firefox locally, but I'm using it over networked X11 because it's a bit faster (it's running on my gateway/server machine). It is my main machine at the moment, and it's working very well. Things like xterm, gvim and irssi and ssh are perfectly fine!
Many people have the idea that computers get slower just by being old, but I think /. readers are aware that it's just the software that gets more complex. I understand that MESA would make this decision, and I hope that it was motivated by some worthwhile changes in the APIs, maybe linked with supporting new and exciting devices.
The problem with dropping support is that many organisation take away the old versions from their web or ftp sites. I experienced this recently, trying to get Python 2.6 for windows, because some software depended on it. I would have to build Python myself, but found somewhere to copy it from . If these people don't make the old version available, then years of coding effort will be lost, code which could still have benefited many people.
"Plenty of people work on "old" cars and trucks for utilitarian reasons. (In my area a forty-year-old truck is a common work vehicle.) It saves a shitload of money and will in future." True indeed, here I was complaining about generalisation, and I was making a huge one myself. It's also interesting to hear about the car computers, there has been some talk about how these are "black boxes" in the tech press, but the people writing that are probably just as clueless as I am.
There is so much *great* tech in IT which can be used by anyone. Web servers, encryption, file sharing (both local and on the internet), and these are just some random examples... This will all go away, and I hate it.
Before electricity became ubiquitous, there were lots of great products that used hand-cranks and other genious mechanical systems. Now, living three days without electrical power would be a disaster. People will choose a electrical ("automatic") product over the mechanical one, even if the mechanical product is in all ways better.
The cloud (and web tech as well, I would argue) is going to be what everyone uses. It is generally easier to use, but there are lots of exceptional programs that aren't "cloud based". Sadly, the mainstream doesn't seem to be able to handle exceptions well, and will discard these. The cloud also makes more business sense, which always helps. If people depend on you (a business) 24/7, that's much better than if they interact with you 1000 times a day, and could easily find a replacement (think hosted e-mail vs. POP3 or running your own server).
I currently run a few linux/BSD machines at home, providing some very nice services for me.I recently realised that I will be like the people who make furniture in their garage, fix up old cars (I won't go into the new computerised cars), or even make simple electronic circuits. It can't be defended as an utilitarian thing any more, it's just a hobby. The cloud providers will be like huge chain stores, modern car makers and HP/Dell/Apple in comparison.
So in conclusion IT will be concentrated in large boring companies that know which products normal consumers and enterprises want, and provide them cheaply and easily. I'm actually depressed about this -- I thought that IT was something special: it has to do with how we learn, communicate and even think. Not only that, it can make millions of impossible tasks seem like a 1 hour job. When looking more closely however, telephone services, plumbing and automotive transportation were just as revolutionary. I can only hope that a new field of such great innovation will appear soon.
You don't *need* a good firewall. A firewall is meant to give an extra layer of good security, but it is used in 90 % of the cases as a quick fix for configuration errors. OSes need to close off unnecessary services anyway, because people take laptops to all kinds of scary places. The IP stack is not inherently insecure. It doesn't harm you if your machine sends a TCP RST packet or an ICMP unreachable when something tries to connect. If only people could make transport mode IPSec easy, we could have a "local" net that worked from anywhere, the "firewall" would be strong encryption.
Desktop users don't care about +/- a few percent in file copy speed. What they should have tested was: does the desktop grind to a slow, unusable halt when copying files? I know 10.10 did, and also Windows to some extent, but not as bad. This would be a huge win if it worked better on the new Ubuntu.
Because let's not mince words about this. Infrastructure is fairly expensive, but once it is in place bandwidth across it is extremely cheap (often approaching as low as 3 cents per gigabyte, according to several studies).
What studies, exactly? 3 cents seems WAY too expensive. Users are already paying tiered prices for the speed (5mbit, 20mbit etc), is that even considered? Probably not.. the price per GB could be an interesting measure anyway, keeping in mind that speed and transfer volume are highly correlated. If we are excluding infrastructure, I can't think of any reason why it would be that much. The marginal cost of transfering 1GB would come from increased power usage of the switches, and possibly increased maintenance costs (32 bit counters rolling over?). Maybe the copper wires get "worn out" from shooting too many electrons through them.. Seriously, 3 cents/GB??
I still prefer the official MSN client. I actually keep a Windows VM going just for that, and for opening office docs and DRM'ed/proprietary videos. The client has some features that Gaim (it's not called Pidgin on my Ubuntu system yet) lacks, like webcam, and it *doesn't pop up windows in the foreground when I'm doing something else*. I can't believe that people put up with that.
What makes you think that the music labels would allow this feature any more than the Zune squirting? Additionally you portray it to be more permissive than the Zune version. So: is it more restricted, or will it be blocked by the RIAA?
This problem was true for the iPod too, and the iPhone launch changed nothing. The argument would be (I think) that Apple has a monopoly on MP3 players, and is abusing that position to do better in the music sales business. This would be valid if Apple was a monopoly.
The new thing about the iPhone is that it supports bluetooth and 802.11g WLAN. This leaves a potential for transferring music directly between phones, og between a PC (e.g. iTunes) and an iPhone. There is no indication that Apple will officially support it, and they do have 100% control over the platform.
One could say that the iPhone is even more resticted than the Zune because it does not allow music sharing AT ALL, even though it would be relatively easy to implement given the power of the platform.
There are a lot of protability that one could want from wireless, some would require 3G to be REALLY cool:
Apple may chose ignore these options, and it may work out fine. In the future, though, people will want these products, because wireless is more convenient, and other companies will make them.
So there is no rush, but Apple will have to deal with the over-the-air DRM issue. And it's gonna be ugly. Like the Zune;)
In particular the statement that he was able to determine there was no wireless router in use at the time cannot be substantiated. It is possible to have a wireless router that NATs you right back to your public IP. In fact, I've done it (with out the wireless part) at least twice for different reasons. There is also the possibility that a simple access point was used, without any NATing being involved at all. This, of course, requires that the ISP offers multiple public IPs via DHCP, and that a record from the ISP, if any, shows that multiple IPs were in use. Given that no trace of file sharing software was found on the hard drive, it is quite probable that the P2P activity was performed by someone else connected to the AP. If only a single IP was used by the subscriber, the parent's explanation would be the only possible way, and I assume the plaintiff would need to show to the court some equipment that could do the double NAT'ing, for it to be a good defence.
That picture reminds me of my old 333MHz laptop. My company shipped a bunch of them to Africa, because they weren't being used anymore. I got one too, and my mom keeps it under the couch for web surfing. It runs FreeBSD, and starts x with no WM and only Opera running by default. That runs totally smooth on 64MB of RAM. Of course, it was only mine that had this software, but the other computer were shipped without an OS, and I'm guessing that Linux will be used. The only problem is with net connectivity. I had a spare PCMCIA WLAN card laying around, and I don't think the same is true for the local shaman. If the cost of the OLPC is sufficiently low, it will be a much better alternative, partly due to the SW/HW integration.
preepmtive multi-tasking which is the feature of both OS/2 and NT that distinguishes them from Window 3.x/9x that used co-operative multi-tasking.
Windows 9x did support preemptive multitasking to an extent. The NT-flag/IRET point still is correct, but I must assume that the NT flag was used in Windows 9x too.
From http://en.wikipedia.org/wiki/Pre-emptive_multitasIf you wanted to netboot or boot from a LiveCD, you wouldn't need a hard drive either (for liveCD one could store permanent files on a USB drive). Should we sue them for net selling HDD-less desktops?
The hard drive has now become the symbol of the computer working. My desktop 'feels' more responsive when it makes that scratching sound when I click the Firefox icon, although I suspect my silent laptop is almost just as fast.
Also, that sound is the first thing to check when computers lock up, etc.
Why do we need USB for connecting a kayboard and a mouse. I don't exactly know how it works on the hardware level, but it seems that the PS/2 connectors do a good job as an interface for keyboards/mice. The PC industry has gotten to the point where every new technology does not have to be better than what it replaces. It's like saying that everyone should use WLANs because they are easier and newer than wired Ethernet. Great, but moving this 4GB file doesn't feel quite as.. "snappy". And is it really as stabl***CARRIER LOST*** Same goes for EFI. There is no reason to rush it, the BIOS works great! No-one cares about the BIOS anymore, the OS (drivers) just talk directly to the HW, so who cares if the system needs partitioning and boots in 16-bit mode , I don't notice that, and neither do 99% of the developers.
This reasoning assumes that the ultimate goal of the Chinese gov't is to arrest its citizens. People using Wikipedia are not necessarily free-thinking or critical off the government, so this would be almost random people (although not the poor farmers who don't have Internet access). I do think the Chinese regime is bad, and it is oppressive, but I don't think it is evil for the sake of being evil. Even if it was, there are thousands of other ways to prosecute random citizens. The un-banning of WP allowed many people to get lots of 'bad' information, probably making this whole venture a net loss for the Chinese government.
If it was trying to arrest people who use WP for finding info which conflicts with their views (i.e. being bad for keeping its perverse amounts of power), it should rather monitor the proxy connections and other work-arounds, for finding people who, for whatever reason, really need to get on the 'wiki.
In summary, as others also have concluded, this may be 1) a technical glitch, 2) a policy change which had unforseen consequences, and was reverted, 3) a political struggle between different factions in the government or 4) a plot to find out what people might like WP, log them, prosecute them, and maybe scare others from using 'open' websites carelessly. The fourth option seems unlikely to me.
Kind of like how the internets are routed today..:D