A fast router typically will not use DRAM due to the high latency involved in the lookups. There's also more to it than just looking up the destination address and forwarding. There's also access control list (ACLs), multicast routing, and so on which do not work in your scenerio. Also, how long does it take to populate a class A route into the table? There's also overlapping routes and source routing as well.
I'm sorry, but routing is often not as simple as just looking up the destination address and forwarding the packet, especially when you're trying to do this to 10+ million packets/second.
I'm working on a product now that handles well over a million packets per second and has to perform some rather complex routing, besides handling many different encapsulations and mapping each source to a potentially different routing table (there can be multiple routing tables internally).
PC133 RAM is only part of it. As a developer of routers I know of at least several methods for storing the routing tables in the data path.
High-end routers do not use DRAM due to its high latency. DRAM works well for localized access, but is terrible for random access. In this case, SRAM is used. Now for routers, it is likely that Content Addressable Memory will be used. The largest CAMs I am aware of are 9Mb (megaBITS) and the largest ZBT SRAMS currently available are 36Mb (4MB). Granted, multiple chips can be used, but only so many chips can be placed on a memory bus before loading becomes a factor. A 133MHz bus can only have so many chips connected to it.
A high-end router today should be able to store upwards of 1,000,000 routes. With IPv6 this becomes far more difficult due to the 128 bit length of the IP addresses.
Also note that a router does a longest match lookup. Some use a hash table and populate entries when there's a miss by using the slow path, but even that becomes difficult since you don't want too many collisions.
The service is obtained over the Ethernet connection, so it should go through your server and automatically download the information. I would guess that since the data is going over the Ethernet it will be reverse engineered (unless they encrypt the data). If you use your server as a firewall (or even if you don't) it is a simple matter to look at all the data.
As for the required "service", I don't understand your complaint, because unlike TiVo the service is included with the product with no service fee.
Also, if it is like my current Replay box (which I hacked by adding a second hard drive:) the TV listings also include various categories, and you can update your settings and selections via a web portal. I can see what my Replay is set to record and can change it via the web from work and the next time my box dials in it will get the updated settings. I would guess that this box will be even more powerful since it uses a broadband connection, where you should be able to access it directly.
I think people might have problems with LGPL software due to the changes that may be required for various embedded operating systems. Take a look at zlib for how to make a library embedded friendly. In zLib, all of the OS related code is in a single file which makes porting very easy. Remember, in embedded systems, functions like malloc and printf might not be available.
I worked on a project where I had to change the boot loader to use zLib so the image would fit into the flash memory. zLib was very simple to use due to the excellent abstraction layer. I had *no* operating system to rely on.
Now zLib is under a BSD-style license instead of LGPL. For a LGPL library to work, just make the OS wrapper files BSD but the core LGPL. If it is written properly, anyone embedding the code should only have to change the OS wrappers, or if any changes are made to the core they would not reveal the internal workings of their product (unless they want to add some feature they want to remain proprietary to the core).
A site like this could be very useful for those of us designing stacks and equipment.
For example, where I work we are writing software for routing, PPP, PPPoE, RIP, OSPF, and so on with all sorts of encapsulations. In an early field test we discovered a lot of cases where our PPP stack would fall over but we couldn't reproduce the problems with our lab equipment. Since then we wrote a bunch of test tools to purposely corrupt packets in any way imaginable to test our software.
BTW, does anyone know of a good site displaying various exploit packets? My firwall seems to catch a lot of them (I'm on a broadband connection).
I have been using the SuSE 7.2 crypto file system for some time without any problems. It's no speed demon, but it should be secure. When installed you are prompted for the password while booting up. It should be easy to change so you can enter the password only when you want to.
I have had very good luck so far with my Netgear fr314. It has excellent logging capabilities and periodically sends all logs and alerts by email. It was easy to set up and allowed me to set up a web server behind the firewall. My main reason for getting it was that I have several computers and don't want to dedicate a computer to just being a firewall.
The Netgear allows me to block all Active X, java, and many cookies (I have Active X blocked for most sites for my roommate's windows computer).
Performance wise it seems pretty good. I havn't noticed any degredation in performance, often downloading at over 400KBps (Kbytes/sec).
It has the option of content filtering, but that's not something I want (except for things like doubleclick.net).
It has many common services already configured and allows for more to be added quite easily.
I wish it allowed some more complicated rules, however. For example, I want to allow some ports to only be accessed from certain IP addresses. I can configure the ports allowed or denied and the IP addresses allowed or denied, but not combinations of both. To handle that I run a secondary firewall on the server which allows more options.
Also, the Netgear is limited to 8 clients without buying an upgrade.
In terms of logging, I am quite impressed. It logs all port scans, attempted accesses to known trojans like netbus, pings of death, and other malicious behavior. It also classifies port scans as either possible or probable.
It also draws only around 10 watts, and here in CA where my electric rate is hitting upwards of 0.20$/kwh,
@Home blocked SMB a couple of years ago at my end. I used to be able to see all of my neighbor's computers, some of which had enabled full sharing. I reported this problem to @Home several times, but they didn't care about this major security breach. They finally fixed it after several articles appeared describing the huge hole. I think IIS is a much bigger hole. IIS should be banned.
There are two good swap meets I know of in or around Silicon Valley. The first is at Foothill College on the second Saturday of the month. It's impossible to miss, just show up early.
The other one is on the first Sunday of the month in Livermore at a college off of Airport Blvd (I forget the name).
Of the two, the Foothill College one is the biggest. There's a lot of trailing-edge technology there as well. I've seen all sorts of things there. The last time I saw some HUGE CCDs for astronomy. A couple of times I've even seen electron microscopes there.
Re:Desperately hoping for non-Linux support
on
KDE 2.2 Tagged
·
· Score: 3, Interesting
I have been installing KDE on Solaris for a couple of years now. I've added patches for features like audio support for Solaris and I tracked down a nasty konsole bug that would cause it to constantly crash on Solaris.
KDE 2.1.1 compiles and installs with a little bit of work on Solaris. You first must download and install the latest Solaris patches due to some bugs in X. There arn't too many add on packages required for Solaris, unlike Gnome.
You need GCC, some libraries like libjpeg, and QT, of course.
It would be nice if a standard package were available for Solaris, though.
I havn't tried 2.2 yet.
-Aaron
Re:I'm designing an "intelligent router"
on
Smart Routers
·
· Score: 2
I might add that the box I'm working on does this with standard IPv4 and clients running standard software. No custom software on the subscriber's computer is needed. No custom protocols are used. In fact, the specifications are supposedly free (see http://www.net.com/products/broadband/new.index.sh tml).
The article on Caspian is rather sparse on information and mostly marketing fluff, however, the service creation model described is definitely the way things are going.
As for being able to sniff traffic and analyze packets, the box I'm working on doesn't go beyond layer 4 except for handling L2TP tunnels.
Another nice thing about the box I'm working on is adding new protocols and interfaces is a piece of cake. My code already handles all of the various PPPoE, PPP, Ethernet, and ATM encapsulations, but adding new ones like POS (Packet over Sonet) is also a piece of cake. This is due to the fact that it's based on a network processor. No, it doesn't run Linux nor is it capable of running it due to its highly specialized design.
Down the road expect for these routers to look deeper and deeper into the traffic as the network processors become more powerful.
-Aaron
I'm designing an "intelligent router"
on
Smart Routers
·
· Score: 2
At my current job at Net.com I'm currently implementing what Caspian is talking about. I'm working on a BRAS - Broadband Remote Access Server which is able to interface with a portal and various services to allow the subscriber to do what they want to do.
For example, say it's Friday night and you want to watch that new movie that just came out. You log into your ISP's portal and go to the video selection, click on the movie you want and go watch it. Behind the scenes, the portal tells our box that you need, say, 5Mbps of bandwidth to the video server. Our box will guarantee that you have the bandwidth needed for the video, even if your roomate starts downloading a bunch of porn in the middle of a really exciting action scene.
This has other applications, for example gaming, video conferencing, or anything where a certain quality of service is required.
Now the BRAS needs to identify various flows and be able to individually shape flows as needed. Usually all that is needed is to look at either the layer 3 or layer 4 information, but unlike a traditional router, both the source and destination are important.
The product I am working on can guarantee bandwidth on a per-flow basis, where a subscriber might have multiple flows. That way traffic from a video server, or packets going to other gamers, will have the bandwidth and/or latency needed.
Our product is controlled via an open API, which is based on Corba and XML. This allows our box to be easily integrated into existing infrastructure (i.e. web portals and billing packages).
Having worked at a company that manufactured and designed PCI devices, there are some serious problems with PCI. First of all, there is no way to properly prioritize traffic. For example, a sound card *must* be able to get the data it needs at low latency, or a NIC card must be able to read an entire packet at 100Mbps or 1Gbps or you will get underruns, whereas this is less critical for disk I/O or graphics.
PCI was a major improvement over the VESA localbus, but things have improved since.
In the networking world, I am seeing a lot of development around AMD's HyperChannel. I don't know the specifics of how it works, but I am starting to see a lot of products utilizing it. For networking, I'm dealing with bandwidths starting at 2.5Gbps and scaling upwards from there.
PCI has outgrown its usefulness in the server world. It's not difficult to saturate 133MB/sec of the standard 32-bit 33MHz PCI. 64-bit 66MHz is an improvement, but the rate things are moving, it won't be long before that's a big bottleneck.
I'm sure that today's big raid servers can probably saturate even a 66MHz 64-bit PCI interface.
Konq has great history, click on Window->Show Sidebar.
I also am unable to get HTTPS to work on Sparc Sun Solaris. Perhapse there is an endian issue?
As for speed and stability, it feels much faster than Netscape, loads much faster, and usually renders better. It's also more stable. I havn't played around with Mozilla on Solaris.
I am a Replay Tv owner and also have a relative who works as an engineer at Replay. Replay added the described "feature" in an attempt to raise much-needed money. People complained so they removed it.
As a software developer, I definitely see the advantages to automatic upgrades, especially when the hardware is a closed system. Without automatic upgrades, technical support can become a nightmare, since everyone could have different revs of the firmware, and new versions should fix old problems. Theoretically, the product will improve with age. Automatic upgrades of non-closed systems (i.e. a PC) are not a good thing, since there is a good chance that the upgrade will break user-installed software and/or hardware.
Now both Replay and Tivo are hungry for money. Both companies are losing money. Accordign to Freeedgar, Tivo lost over 200 million dollars, last quarter losing 89 million dollars. Replay has also lost a lot of money, but never went IPO. Replay is currently being acquired by Sonic Blue, formerly Diamond Multimedia and S3.
I can safely say that they are not a hoax, they just have a sense of humour. Inside their lobby they have Robby the Robot, which they now own. They had an open house a few months ago where they had several cast members from various Star Trek shows sign autographs. It made the local news media.
A company called Alien Technology has developed a very low power display technology that looks very promising. There was a write-up in the February Scientific American about it.
Automatic upgrade support needed
on
New Linux Worm
·
· Score: 2
What is really needed in Linux is the ability for various distributions to automatically install security updates. I realize that many admins have written scripts to do this, but this should be a default option. For dialup users, it should check for updates every time the user connects to the Internet, and for 24/7 connections it should check for updates once a day.
I read some articles describing massive amounts of methane at the ocean bottom in a gel-form. Apparently there's quite a lot of it and it doesn't take much warming of the ocean water for the gel to break-down and release the methane into the atmosphere. Methane is a much more serious greenhouse gas than CO2, so it is possible for positive reinforcement here. This could make a little global warming due to other factors suddenly increase significantly. Other positive reinforcement methods are the reduction of ice at the poles, since ice reflects IR back into space quite nicely.
All in all, there's a lot we don't know about global warming, but that doesn't mean that we should be reckless in how we treat the environment.
I would beg to differ with you on on-line vs traditional retail. There are many expenses for traditional retail that just don't apply to online sales. For example, a traditional store must maintain a showroom and have inventory in the showroom. An Internet Retailer (IR) does not need this and can thus reduce the real-estate, utilities, and manpower required for this.
Also, a brick-and-mortar store needs to hire cashiers to be able to handle peak demand. An IR can get by with fewer cashiers since most of the work is automated and the transactions don't necessarily need to be processed in real-time.
A traditional store needs to hire employees to maintain the showroom and to keep track of inventory, check for shoplifting, install survalence cameras, etc. A IR doesn't need to worry about shoplifting, and the displays don't need to be maintained (take a photograph and you're done).
Now, there's some additional cost for on-line, such as hiring a web designer and for web hosting (which is cheap these days), but compared to the traditional store the cost is far less.
If you're counting catalog companies as traditional, they have the expense of publishing their catalogs, printing them, and mailing them. A IR doesn't have this expense, only the publishing expense.
I can say from experience that a line-rate OC-12 ACL list is quite feasable, and in fact OC-48 (2.4Gbps) is quite feasable with today's technology.
Some of the new Network Processors are absolutely astounding in terms of what they can accomplish. Take for example the Agere network processor. It has no problems doing ACL at OC-48. Or the Sibyte network processor, with dual 1GHz MIPS cores running Linux, which should be more than fast enough to handle OC-12.
At work we do all of our development on Solaris and since CDE sucks I tried to compile GNOME. It required too many different packages from all over the net and a good number of them wouldn't compile.
I then tried KDE, which only requires 4 files to get started, QT, Support, Libs, and Base. I have been compiling and running KDE on Solaris ever since. KDE 2.1 will be a nice upgrade, especially now that I have fixed the damned bug where konsole would crash periodically. I'm hoping my patch makes it into 2.1.
As for web browsers, the only reason I still run Netscape is because our corporate mail and calender are based on it and KDE mail doesn't support IMAP yet. Konqueror is an awesom browser, especially in 2.1 since it has most of the Netscape features I liked (like dragging URLs between windows).
I havn't tried Gnome for a while, but looking at the list of required libraries and whatnot it looks like the same mess it was before.
As someone with a lot of hands-on experience with network processors (althouth not the Intel one), I must add that all of the network processors I have investigated have many coprocessors for doing operations like lookups, queueing, and so forth. The idea is that the actual processor core does not have to do a whole lot.
One cannot compare a network processor to a general purpose processor since all of the NPs I've looked at are very specific to one application, networking.
For example, one could not run SETI@Home calculations on a network processor, nor could they run Linux, as their memory architectures are often limited with most of the program memory residing on-chip and/or in very fast SRAM. Right now the largest high-speed SRAM chip available is around 4MB. It becomes impractical to add more than 16 MB of SRAM due to loading of the bus (assuming 64-bit). At 166MHz it is even worse.
As for multiple contexts, many of the network processors can switch between contexts very quickly, but also remember that NP cores do not have many of the things a general purpose processor has. There's no paging or fancy memory management, nor is there floating point.
I see several reasons for this. For one thing, there's this great emphasis on reading, yet I see little emphasis on math and science in our schools. Science in many public schools (at least in California) is a joke. Over the years, science has been dumbed down to the point to where those of us who find science interesting are bored to death in the classroom where they rehash the same topics year after year in a heavily watered down format.
The other problem I see is that (at least here in CA) they have changed the math and science cirriculum such that much of the homework would fall under art or possibly social science. Have you seen many of the homework assignments today? No calculations, just draw pretty pictures.
I blame this on the lack of qualified teachers. Let's face it, with today's economy, no sane person with a science education would waste it as a teacher in public schools since it won't pay the bills. Most of our teachers have a social science background.
After this last election I now know how dumb our population has become and it's sad. We have the exact opposite problem as the Gulgofringens (sp?) in HHGTTG. Our useless third of the population has taken over, leaving humanities future in doubt when in 2018 the giant space goat eats the Earth because we're too busy philosophizing about the nature of our navals to see it coming.
Slashdot Mirror
Damned, there goes all the promotions I receive for teen sex, viagra, spy software, herbal remedies, and all these great ways to make money at home.
This will hurt, as I've had the same email address for over 4 years.
-Aaron
A fast router typically will not use DRAM due to the high latency involved in the lookups. There's also more to it than just looking up the destination address and forwarding. There's also access control list (ACLs), multicast routing, and so on which do not work in your scenerio. Also, how long does it take to populate a class A route into the table? There's also overlapping routes and source routing as well.
I'm sorry, but routing is often not as simple as just looking up the destination address and forwarding the packet, especially when you're trying to do this to 10+ million packets/second.
I'm working on a product now that handles well over a million packets per second and has to perform some rather complex routing, besides handling many different encapsulations and mapping each source to a potentially different routing table (there can be multiple routing tables internally).
PC133 RAM is only part of it. As a developer of routers I know of at least several methods for storing the routing tables in the data path.
High-end routers do not use DRAM due to its high latency. DRAM works well for localized access, but is terrible for random access. In this case, SRAM is used. Now for routers, it is likely that Content Addressable Memory will be used. The largest CAMs I am aware of are 9Mb (megaBITS) and the largest ZBT SRAMS currently available are 36Mb (4MB). Granted, multiple chips can be used, but only so many chips can be placed on a memory bus before loading becomes a factor. A 133MHz bus can only have so many chips connected to it.
A high-end router today should be able to store upwards of 1,000,000 routes. With IPv6 this becomes far more difficult due to the 128 bit length of the IP addresses.
Also note that a router does a longest match lookup. Some use a hash table and populate entries when there's a miss by using the slow path, but even that becomes difficult since you don't want too many collisions.
The service is obtained over the Ethernet connection, so it should go through your server and automatically download the information. I would guess that since the data is going over the Ethernet it will be reverse engineered (unless they encrypt the data). If you use your server as a firewall (or even if you don't) it is a simple matter to look at all the data.
:) the TV listings also include various categories, and you can update your settings and selections via a web portal. I can see what my Replay is set to record and can change it via the web from work and the next time my box dials in it will get the updated settings. I would guess that this box will be even more powerful since it uses a broadband connection, where you should be able to access it directly.
As for the required "service", I don't understand your complaint, because unlike TiVo the service is included with the product with no service fee.
Also, if it is like my current Replay box (which I hacked by adding a second hard drive
-Aaron
I think people might have problems with LGPL software due to the changes that may be required for various embedded operating systems. Take a look at zlib for how to make a library embedded friendly. In zLib, all of the OS related code is in a single file which makes porting very easy. Remember, in embedded systems, functions like malloc and printf might not be available.
I worked on a project where I had to change the boot loader to use zLib so the image would fit into the flash memory. zLib was very simple to use due to the excellent abstraction layer. I had *no* operating system to rely on.
Now zLib is under a BSD-style license instead of LGPL. For a LGPL library to work, just make the OS wrapper files BSD but the core LGPL. If it is written properly, anyone embedding the code should only have to change the OS wrappers, or if any changes are made to the core they would not reveal the internal workings of their product (unless they want to add some feature they want to remain proprietary to the core).
A site like this could be very useful for those of us designing stacks and equipment.
For example, where I work we are writing software for routing, PPP, PPPoE, RIP, OSPF, and so on with all sorts of encapsulations. In an early field test we discovered a lot of cases where our PPP stack would fall over but we couldn't reproduce the problems with our lab equipment. Since then we wrote a bunch of test tools to purposely corrupt packets in any way imaginable to test our software.
BTW, does anyone know of a good site displaying various exploit packets? My firwall seems to catch a lot of them (I'm on a broadband connection).
I have been using the SuSE 7.2 crypto file system for some time without any problems. It's no speed demon, but it should be secure. When installed you are prompted for the password while booting up. It should be easy to change so you can enter the password only when you want to.
I have had very good luck so far with my Netgear fr314. It has excellent logging capabilities and periodically sends all logs and alerts by email. It was easy to set up and allowed me to set up a web server behind the firewall. My main reason for getting it was that I have several computers and don't want to dedicate a computer to just being a firewall.
The Netgear allows me to block all Active X, java, and many cookies (I have Active X blocked for most sites for my roommate's windows computer).
Performance wise it seems pretty good. I havn't noticed any degredation in performance, often downloading at over 400KBps (Kbytes/sec).
It has the option of content filtering, but that's not something I want (except for things like doubleclick.net).
It has many common services already configured and allows for more to be added quite easily.
I wish it allowed some more complicated rules, however. For example, I want to allow some ports to only be accessed from certain IP addresses. I can configure the ports allowed or denied and the IP addresses allowed or denied, but not combinations of both. To handle that I run a secondary firewall on the server which allows more options.
Also, the Netgear is limited to 8 clients without buying an upgrade.
In terms of logging, I am quite impressed. It logs all port scans, attempted accesses to known trojans like netbus, pings of death, and other malicious behavior. It also classifies port scans as either possible or probable.
It also draws only around 10 watts, and here in CA where my electric rate is hitting upwards of 0.20$/kwh,
@Home blocked SMB a couple of years ago at my end. I used to be able to see all of my neighbor's computers, some of which had enabled full sharing. I reported this problem to @Home several times, but they didn't care about this major security breach. They finally fixed it after several articles appeared describing the huge hole. I think IIS is a much bigger hole. IIS should be banned.
There are two good swap meets I know of in or around Silicon Valley. The first is at Foothill College on the second Saturday of the month. It's impossible to miss, just show up early.
The other one is on the first Sunday of the month in Livermore at a college off of Airport Blvd (I forget the name).
Of the two, the Foothill College one is the biggest. There's a lot of trailing-edge technology there as well. I've seen all sorts of things there. The last time I saw some HUGE CCDs for astronomy. A couple of times I've even seen electron microscopes there.
I have been installing KDE on Solaris for a couple of years now. I've added patches for features like audio support for Solaris and I tracked down a nasty konsole bug that would cause it to constantly crash on Solaris.
KDE 2.1.1 compiles and installs with a little bit of work on Solaris. You first must download and install the latest Solaris patches due to some bugs in X. There arn't too many add on packages required for Solaris, unlike Gnome.
You need GCC, some libraries like libjpeg, and QT, of course.
It would be nice if a standard package were available for Solaris, though.
I havn't tried 2.2 yet.
-Aaron
I might add that the box I'm working on does this with standard IPv4 and clients running standard software. No custom software on the subscriber's computer is needed. No custom protocols are used. In fact, the specifications are supposedly free (see http://www.net.com/products/broadband/new.index.sh tml).
The article on Caspian is rather sparse on information and mostly marketing fluff, however, the service creation model described is definitely the way things are going.
As for being able to sniff traffic and analyze packets, the box I'm working on doesn't go beyond layer 4 except for handling L2TP tunnels.
Another nice thing about the box I'm working on is adding new protocols and interfaces is a piece of cake. My code already handles all of the various PPPoE, PPP, Ethernet, and ATM encapsulations, but adding new ones like POS (Packet over Sonet) is also a piece of cake. This is due to the fact that it's based on a network processor. No, it doesn't run Linux nor is it capable of running it due to its highly specialized design.
Down the road expect for these routers to look deeper and deeper into the traffic as the network processors become more powerful.
-Aaron
At my current job at Net.com I'm currently implementing what Caspian is talking about. I'm working on a BRAS - Broadband Remote Access Server which is able to interface with a portal and various services to allow the subscriber to do what they want to do.
For example, say it's Friday night and you want to watch that new movie that just came out. You log into your ISP's portal and go to the video selection, click on the movie you want and go watch it. Behind the scenes, the portal tells our box that you need, say, 5Mbps of bandwidth to the video server. Our box will guarantee that you have the bandwidth needed for the video, even if your roomate starts downloading a bunch of porn in the middle of a really exciting action scene.
This has other applications, for example gaming, video conferencing, or anything where a certain quality of service is required.
Now the BRAS needs to identify various flows and be able to individually shape flows as needed. Usually all that is needed is to look at either the layer 3 or layer 4 information, but unlike a traditional router, both the source and destination are important.
The product I am working on can guarantee bandwidth on a per-flow basis, where a subscriber might have multiple flows. That way traffic from a video server, or packets going to other gamers, will have the bandwidth and/or latency needed.
Our product is controlled via an open API, which is based on Corba and XML. This allows our box to be easily integrated into existing infrastructure (i.e. web portals and billing packages).
-Aaron
Having worked at a company that manufactured and designed PCI devices, there are some serious problems with PCI. First of all, there is no way to properly prioritize traffic. For example, a sound card *must* be able to get the data it needs at low latency, or a NIC card must be able to read an entire packet at 100Mbps or 1Gbps or you will get underruns, whereas this is less critical for disk I/O or graphics.
PCI was a major improvement over the VESA localbus, but things have improved since.
In the networking world, I am seeing a lot of development around AMD's HyperChannel. I don't know the specifics of how it works, but I am starting to see a lot of products utilizing it. For networking, I'm dealing with bandwidths starting at 2.5Gbps and scaling upwards from there.
PCI has outgrown its usefulness in the server world. It's not difficult to saturate 133MB/sec of the standard 32-bit 33MHz PCI. 64-bit 66MHz is an improvement, but the rate things are moving, it won't be long before that's a big bottleneck.
I'm sure that today's big raid servers can probably saturate even a 66MHz 64-bit PCI interface.
-Aaron
Konq has great history, click on Window->Show Sidebar.
I also am unable to get HTTPS to work on Sparc Sun Solaris. Perhapse there is an endian issue?
As for speed and stability, it feels much faster than Netscape, loads much faster, and usually renders better. It's also more stable. I havn't played around with Mozilla on Solaris.
This was posted with Konqueror
I am a Replay Tv owner and also have a relative who works as an engineer at Replay. Replay added the described "feature" in an attempt to raise much-needed money. People complained so they removed it.
As a software developer, I definitely see the advantages to automatic upgrades, especially when the hardware is a closed system. Without automatic upgrades, technical support can become a nightmare, since everyone could have different revs of the firmware, and new versions should fix old problems. Theoretically, the product will improve with age. Automatic upgrades of non-closed systems (i.e. a PC) are not a good thing, since there is a good chance that the upgrade will break user-installed software and/or hardware.
Now both Replay and Tivo are hungry for money. Both companies are losing money. Accordign to Freeedgar, Tivo lost over 200 million dollars, last quarter losing 89 million dollars. Replay has also lost a lot of money, but never went IPO. Replay is currently being acquired by Sonic Blue, formerly Diamond Multimedia and S3.
I can safely say that they are not a hoax, they just have a sense of humour. Inside their lobby they have Robby the Robot, which they now own. They had an open house a few months ago where they had several cast members from various Star Trek shows sign autographs. It made the local news media.
A company called Alien Technology has developed a very low power display technology that looks very promising. There was a write-up in the February Scientific American about it.
What is really needed in Linux is the ability for various distributions to automatically install security updates. I realize that many admins have written scripts to do this, but this should be a default option. For dialup users, it should check for updates every time the user connects to the Internet, and for 24/7 connections it should check for updates once a day.
I read some articles describing massive amounts of methane at the ocean bottom in a gel-form. Apparently there's quite a lot of it and it doesn't take much warming of the ocean water for the gel to break-down and release the methane into the atmosphere. Methane is a much more serious greenhouse gas than CO2, so it is possible for positive reinforcement here. This could make a little global warming due to other factors suddenly increase significantly. Other positive reinforcement methods are the reduction of ice at the poles, since ice reflects IR back into space quite nicely.
All in all, there's a lot we don't know about global warming, but that doesn't mean that we should be reckless in how we treat the environment.
I would beg to differ with you on on-line vs traditional retail. There are many expenses for traditional retail that just don't apply to online sales. For example, a traditional store must maintain a showroom and have inventory in the showroom. An Internet Retailer (IR) does not need this and can thus reduce the real-estate, utilities, and manpower required for this.
Also, a brick-and-mortar store needs to hire cashiers to be able to handle peak demand. An IR can get by with fewer cashiers since most of the work is automated and the transactions don't necessarily need to be processed in real-time.
A traditional store needs to hire employees to maintain the showroom and to keep track of inventory, check for shoplifting, install survalence cameras, etc. A IR doesn't need to worry about shoplifting, and the displays don't need to be maintained (take a photograph and you're done).
Now, there's some additional cost for on-line, such as hiring a web designer and for web hosting (which is cheap these days), but compared to the traditional store the cost is far less.
If you're counting catalog companies as traditional, they have the expense of publishing their catalogs, printing them, and mailing them. A IR doesn't have this expense, only the publishing expense.
I can say from experience that a line-rate OC-12 ACL list is quite feasable, and in fact OC-48 (2.4Gbps) is quite feasable with today's technology.
Some of the new Network Processors are absolutely astounding in terms of what they can accomplish. Take for example the Agere network processor. It has no problems doing ACL at OC-48. Or the Sibyte network processor, with dual 1GHz MIPS cores running Linux, which should be more than fast enough to handle OC-12.
At work we do all of our development on Solaris and since CDE sucks I tried to compile GNOME. It required too many different packages from all over the net and a good number of them wouldn't compile.
I then tried KDE, which only requires 4 files to get started, QT, Support, Libs, and Base. I have been compiling and running KDE on Solaris ever since. KDE 2.1 will be a nice upgrade, especially now that I have fixed the damned bug where konsole would crash periodically. I'm hoping my patch makes it into 2.1.
As for web browsers, the only reason I still run Netscape is because our corporate mail and calender are based on it and KDE mail doesn't support IMAP yet. Konqueror is an awesom browser, especially in 2.1 since it has most of the Netscape features I liked (like dragging URLs between windows).
I havn't tried Gnome for a while, but looking at the list of required libraries and whatnot it looks like the same mess it was before.
-Aaron
As someone with a lot of hands-on experience with network processors (althouth not the Intel one), I must add that all of the network processors I have investigated have many coprocessors for doing operations like lookups, queueing, and so forth. The idea is that the actual processor core does not have to do a whole lot.
One cannot compare a network processor to a general purpose processor since all of the NPs I've looked at are very specific to one application, networking.
For example, one could not run SETI@Home calculations on a network processor, nor could they run Linux, as their memory architectures are often limited with most of the program memory residing on-chip and/or in very fast SRAM. Right now the largest high-speed SRAM chip available is around 4MB. It becomes impractical to add more than 16 MB of SRAM due to loading of the bus (assuming 64-bit). At 166MHz it is even worse.
As for multiple contexts, many of the network processors can switch between contexts very quickly, but also remember that NP cores do not have many of the things a general purpose processor has. There's no paging or fancy memory management, nor is there floating point.
I see several reasons for this. For one thing, there's this great emphasis on reading, yet I see little emphasis on math and science in our schools. Science in many public schools (at least in California) is a joke. Over the years, science has been dumbed down to the point to where those of us who find science interesting are bored to death in the classroom where they rehash the same topics year after year in a heavily watered down format.
The other problem I see is that (at least here in CA) they have changed the math and science cirriculum such that much of the homework would fall under art or possibly social science. Have you seen many of the homework assignments today? No calculations, just draw pretty pictures.
I blame this on the lack of qualified teachers. Let's face it, with today's economy, no sane person with a science education would waste it as a teacher in public schools since it won't pay the bills. Most of our teachers have a social science background.
After this last election I now know how dumb our population has become and it's sad. We have the exact opposite problem as the Gulgofringens (sp?) in HHGTTG. Our useless third of the population has taken over, leaving humanities future in doubt when in 2018 the giant space goat eats the Earth because we're too busy philosophizing about the nature of our navals to see it coming.