WAP is mainly just a toy, as you say, but there are some fun/useful applications - getting stock quotes is useful, and well suited to WAP.
On the fun side, I am quite addicted to the fortune service from Excite's WAP site (sample: 'you have a talent for talking to weirdos').
So far, that's it - for me the killer app on my mobile phone (Nokia 7110) is the SMS - it has T9 predictive input to cut down on button presses, and I can use an SMS to email gateway to send short emails directly from the phone, without a lengthy WAP login sequence. There's also a gateway from email to SMS but that seems overloaded.
Of course, these email applications are not provided by my mobile service provider (Orange), who are reassuringly clueless and don't even have a suggestion box on their WAP site that I can see. All the more argument for opening up WAP services - walled gardens are only good if you have very talented gardeners, so let's open up WAP with more commonly adopted standards and direct-to-phone protocols.
Panasonic say that 'next generation' DVD-ROM drives will accept DVD-RAM - this is a read/write format with optional caddy, a Panasonic-sponsored standard but now accepted by DVD Forum. You normally use the disc in a caddy but can take it out for use in (certain other) DVD-ROM drives and DVD players.
Unfortunately you can only do 2.6 GB per side with DVD-RAM currently, but it would be quite handy. The real question is how common these so-called next-gen DVD-ROMs are becoming in practice.
There are lots of format wars, out there, e.g. RW (formerly DVD+RW). The Panasonic site, http://www.panasonic.com/industrial_oem/computer/s torage/dvd-ram/prods/tech/prove.htm, has some info but is bound to be biased, so read the DVD Faq at www.faqs.org for the full details.
Have a look at VNC - it's open source, from http://www.uk.research.att.com, and works pretty well to remotely access your desktop. It's completely stateless, so it would work fine in this scenario. The only hassle is its bandwidth usage but that could be improved with better compression.
- improving posture and body use (try Trager, Alexander technique or Feldenkrais)
- massage/physio of various types (esp. adverse mechanical tension physiotherapy, trigger point massage)
- stretching (esp AMT stretches)
- changing work patterns (taking breaks and not typing during them!)
There's probably more that I've forgotten, but right at the bottom is vitamin supplements, glucosamine - I tried the former in spades and they made absolutely no difference.
Binary-only modules are allowed in Linux (i.e. OKed by Linus) which is presumably how they've done this and still stayed GPL-compliant. Although the license weenies are probably fuelling their flamethrowers right now...
Nice to know, but the 700 series routers don't run IOS.
The cheapest routers to run IOS are the 800 series, I believe (but watch out for exactly which features are implemented, some low end feature packs are missing surprising stuff).
The 7200VXR is a very nice box - I have tested this with 38 Mbps of traffic going through it (from one fast ethernet interface to another), including half a page of access lists and route maps to mark IP Precedence. The CPU load was only 30% or so.
More realistically, it can run custom queuing with minimal CPU loads (very nice for allocating bandwidth to high priority applications, i.e. class of service/QoS), unlike some older high-end routers.
It has some backplane improvements over the older 7200s, so it's not just a matter of CPU speed and cache.
Cisco designs its own hardware and software, but it's common knowledge that it outsources quite a lot of its manufacturing, like many other high tech companies.
Multicast is available today with IPv4, and I'm not really clear why anyone would migrate to IPv6 just to get better multicast. RealPlayer already supports multicast, and it's unlikely webcasting will really hit high volumes without multicast, so Real Networks would just have to find another business model (e.g. paid-for content subscriptions, which they are already doing, though this is even more difficult in the multicast world, since authenticated access to multicast streams really needs a fairly complex encryption setup).
The main blockers for multicast IMO are that it is pretty hard to deploy, troubleshoot and manage, and is also quite prone to DoS attacks (hey, now I can DoS an *entire multicast group* from a single compromised host!!).
There are also interesting inter-provider routing and peering issues (how do you set up peering agreements between two providers that take account of some of that traffic being multicast, i.e. it will use a lot more bandwidth potentially than just the amount coming over the provider-to-provider link.)
TCP/IP has lasted about 20 years, as the Internet grew from a handful of academic sites to the backbone of e-commerce. It is an astonishingly successful set of protocols by any standards, and still has a great deal of life in it, perhaps because the first ARPANET protocols were implemented, tested and then thrown away...
Implementing WAP and other protocols is essential to seeing how well they work - there are also many other protocols, e.g. cHTML over i-mode, and so on, and soon IP over GPRS, all of which will
be implemented and tested.
Or would you rather reject protocols based on a purely paper evalation?
The original poster is probably talking about firewall administrators who have noticed more DNS traffic through their firewalls than normal - hence someone may be using this technique.
This does illustrate the need for trend-based traffic monitoring (a la Concord) and even security-driven bandwidth restrictions - e.g. only allow your DNS traffic to increase by 50% in one day, or some such heuristic. These wouldn't necessarily stop such a covert tunnel but they would make it easier to find one, and slower to use one, giving the security admins more time to trace what is happening.
Can you use repeaters in this setup, or some sort of remote stations that link back to base via microwave links? This would make it easier to cover larger areas particularly in towns.
I'm interested to see you are using ATM via the Newbridge kit - you might like to investigate MPLS, which is a way of combining ATM's fast forwarding mechanisms with IP's routing mechanisms. The result is that very large best effort networks and VPNs are very easy to set up - no need for a mesh of PVCs, you just plug them together and the routing sorts things out.
MPLS is not currently so strong on the traffic engineering side, i.e. setting up the equivalent of PVCs to steer traffic along less utilised paths, or for guaranteed QoS or fast failover paths, but Juniper and Cisco routers can already do traffic engineering and the standards are coming along.
Seems like Prairie iNet are using the 2.4 GHz band, which is generally unlicensed around the world - assuming they are using the same technology as Midcoast Wireless, a Maine ISP, which has a very useful FAQ
This is all based on IEEE 802.11 technology, which is normally used for wireless LANs with a range of a few hundred feet. The trick here seems to be using more power and directional antennae so that you can go up to 9 miles (or maybe much more).
One company making this sort of kit is Breezecom, who have an overview of wireless Internet access here.
This technology, along with the competing licensed LMDS technology, may make mincemeat of DSL and Cable - it involves no rights of way hassles, no cable laying, and can give very low latency plus bandwidths in the 1-2Mbps range. Having used Wireless LANs at conferences and trade shows, I found the latency and bandwidth very similar to a T1 line.
For info on 802.11, see the Linux Wireless LAN FAQ, which also has good links to generic WLAN info at the end. Although the technology for 802.11 long-distance (i.e. wireless local loop) is not identical, it should give you an idea of how things work.
For info on LMDS (Local Multipoint Distribution System), see the Webopedia entry for LMDS, which has links to related pages. One new European telco that is rolling out LMDS quite aggressively is FirstMark - they are also doing cool things with MPLS VPNs, which is how I know about them since my company just sold them the software to manage this:) MPLS is a way of getting the best of the ATM and IP worlds at the same time.
Low latency is important because it's a key determinant of web response time, particularly for sites with many small GIF buttons, and also because Internet routers tend to treat high-latency sessions less fairly, so they get even less bandwidth then they should. It's also essential to winning at Quake, which is clearly the critical driver here:)
This story matters more for the technology than for the particular ISP using it - it will affect most Slashdot readers in the next year or so, particularly those not covered by DSL or cable. In the UK, BT is being astonishingly slow at rolling out ADSL, and the cable companies have very little coverage, so wireless technology may be the only way to get broadband for many people...
How exactly would Nokia be hurt by competitors putting prices up? They would just walk off with the market... More likely is that there would be a price war and a few companies retire hurt, with the remaining ones putting up prices - however, price wars can't go on too long, and are hardly a reason to wish for less competition...
Convergence is happening, whether you like it or now - in other words, market boundaries are becoming much less distinct as companies find there's technology and maybe even demand for a box that can play games, show digital TV, surf the Net, etc.
Anyone who lives in Europe, Asia, or other places outside US/Canada is very likely to be paying for Internet access by the minute (local calls are not free). Although some countries are moving towards special internet access deals for flat fees, at least in the UK these ISPs are massively oversubscribed.
Travellers outside US/Canada can also make use of this feature; global roaming services such as gric.net also charge by the minute for the convenience of using any ISP in their scheme.
I just read that interview, and I think he came out of it OK - he is very focused on a single idea, i.e. a zealot (he almost has to be, by definition), and that's one big reason why the GNU movement has succeeded.
The interviewer was either playing devil's advocate or rather clueless about free software - RMS did OK in handling this, only occasionally letting rip...
I am not a huge fan of RMS and I tend to ignore the demands to refer to GNU/Linux not Linux (there are many other types of software in the typical Linux distro, e.g. Perl Artistic licensed, BSD and X11) - however, without RMS I would not now be typing this message on a Linux system compiled by GCC and running Gnome, so I think we all owe him a huge thank you, even if we don't agree with everything he says.
Courage in your convictions may sometimes be hard to distinguish from arrogance...
This slowdown occurs with raw partitions as well as VMware's virtual partitions or whatever they call them. In fact, PS/2 mice sometimes caused a slowdown, though this may have been fixed by now.
Overall it is a great product, but it's not for people who want extremely high performance without providing extra hardware to run it well.
I use VMware quite a lot - it generally runs thing at approx half the speed of the real machine. This is not usually a problem, and doesn't apply to CPU-bound work, which runs at nearly full speed, but there is a noticeable slowdown in I/O.
It's a complete waste of time to just look at the titles - these describe generally what the patent is about, but the *claims* of the patent, listed at the end, are what is actually protected.
Any details in the title, abstract, description, etc are only there to help a skilled practitioner reproduce the invention (under license), or to invent something new that doesn't infringe the claims.
VMware has at least some patents filed for - they mention this on their first press release, at http://www.vmware.com/news/pr1.html.
Also, it's meaningless to say 'countries that laugh at US patent laws' in that they are irrelevant to patents in almost every country.
If you want patents in the UK, Japan, etc, you must file them in each country. There are some short cuts, i.e. you can file European patents and International patents, but these must eventually 'fork' into country-specific patents at later stages of the process. Patents don't automatically cross national boundaries, like trademarks but unlike copyright.
Slashdot Mirror
WAP is mainly just a toy, as you say, but there are some fun/useful applications - getting stock quotes is useful, and well suited to WAP.
On the fun side, I am quite addicted to the fortune service from Excite's WAP site (sample: 'you have a talent for talking to weirdos').
So far, that's it - for me the killer app on my mobile phone (Nokia 7110) is the SMS - it has T9 predictive input to cut down on button presses, and I can use an SMS to email gateway to send short emails directly from the phone, without a lengthy WAP login sequence. There's also a gateway from email to SMS but that seems overloaded.
Of course, these email applications are not provided by my mobile service provider (Orange), who are reassuringly clueless and don't even have a suggestion box on their WAP site that I can see. All the more argument for opening up WAP services - walled gardens are only good if you have very talented gardeners, so let's open up WAP with more commonly adopted standards and direct-to-phone protocols.
Panasonic say that 'next generation' DVD-ROM drives will accept DVD-RAM - this is a read/write format with optional caddy, a Panasonic-sponsored standard but now accepted by DVD Forum. You normally use the disc in a caddy but can take it out for use in (certain other) DVD-ROM drives and DVD players.
s torage/dvd-ram/prods/tech/prove.htm, has some info but is bound to be biased, so read the DVD Faq at www.faqs.org for the full details.
Unfortunately you can only do 2.6 GB per side with DVD-RAM currently, but it would be quite handy. The real question is how common these so-called next-gen DVD-ROMs are becoming in practice.
There are lots of format wars, out there, e.g. RW (formerly DVD+RW). The Panasonic site, http://www.panasonic.com/industrial_oem/computer/
Have a look at VNC - it's open source, from http://www.uk.research.att.com, and works pretty well to remotely access your desktop. It's completely stateless, so it would work fine in this scenario. The only hassle is its bandwidth usage but that could be improved with better compression.
See the other slashdot stories for details, but there are no easy answers - recovering from RSI is a combination of:
- reducing your stress levels, and managing them with meditation/relaxation
- improving ergonomic setup (furniture, keyboard/screen setup, etc)
- improving posture and body use (try Trager, Alexander technique or Feldenkrais)
- massage/physio of various types (esp. adverse mechanical tension physiotherapy, trigger point massage)
- stretching (esp AMT stretches)
- changing work patterns (taking breaks and not typing during them!)
There's probably more that I've forgotten, but right at the bottom is vitamin supplements, glucosamine - I tried the former in spades and they made absolutely no difference.
Binary-only modules are allowed in Linux (i.e. OKed by Linus) which is presumably how they've done this and still stayed GPL-compliant. Although the license weenies are probably fuelling their flamethrowers right now...
I believe that most cockroaches are female and pregnant, so when you stamp on them you are simply scattering the eggs...
Nice to know, but the 700 series routers don't run IOS.
The cheapest routers to run IOS are the 800 series, I believe (but watch out for exactly which features are implemented, some low end feature packs are missing surprising stuff).
The 7200VXR is a very nice box - I have tested this with 38 Mbps of traffic going through it (from one fast ethernet interface to another), including half a page of access lists and route maps to mark IP Precedence. The CPU load was only 30% or so.
More realistically, it can run custom queuing with minimal CPU loads (very nice for allocating bandwidth to high priority applications, i.e. class of service/QoS), unlike some older high-end routers.
It has some backplane improvements over the older 7200s, so it's not just a matter of CPU speed and cache.
Cisco designs its own hardware and software, but it's common knowledge that it outsources quite a lot of its manufacturing, like many other high tech companies.
Interesting but a bit too paranoid perhaps...
Multicast is available today with IPv4, and I'm not really clear why anyone would migrate to IPv6 just to get better multicast. RealPlayer already supports multicast, and it's unlikely webcasting will really hit high volumes without multicast, so Real Networks would just have to find another business model (e.g. paid-for content subscriptions, which they are already doing, though this is even more difficult in the multicast world, since authenticated access to multicast streams really needs a fairly complex encryption setup).
The main blockers for multicast IMO are that it is pretty hard to deploy, troubleshoot and manage, and is also quite prone to DoS attacks (hey, now I can DoS an *entire multicast group* from a single compromised host!!).
There are also interesting inter-provider routing and peering issues (how do you set up peering agreements between two providers that take account of some of that traffic being multicast, i.e. it will use a lot more bandwidth potentially than just the amount coming over the provider-to-provider link.)
TCP/IP has lasted about 20 years, as the Internet grew from a handful of academic sites to the backbone of e-commerce. It is an astonishingly successful set of protocols by any standards, and still has a great deal of life in it, perhaps because the first ARPANET protocols were implemented, tested and then thrown away...
Implementing WAP and other protocols is essential to seeing how well they work - there are also many other protocols, e.g. cHTML over i-mode, and so on, and soon IP over GPRS, all of which will
be implemented and tested.
Or would you rather reject protocols based on a purely paper evalation?
The original poster is probably talking about firewall administrators who have noticed more DNS traffic through their firewalls than normal - hence someone may be using this technique.
This does illustrate the need for trend-based traffic monitoring (a la Concord) and even security-driven bandwidth restrictions - e.g. only allow your DNS traffic to increase by 50% in one day, or some such heuristic. These wouldn't necessarily stop such a covert tunnel but they would make it easier to find one, and slower to use one, giving the security admins more time to trace what is happening.
Can you use repeaters in this setup, or some sort of remote stations that link back to base via microwave links? This would make it easier to cover larger areas particularly in towns.
I'm interested to see you are using ATM via the Newbridge kit - you might like to investigate MPLS, which is a way of combining ATM's fast forwarding mechanisms with IP's routing mechanisms. The result is that very large best effort networks and VPNs are very easy to set up - no need for a mesh of PVCs, you just plug them together and the routing sorts things out.
MPLS is not currently so strong on the traffic engineering side, i.e. setting up the equivalent of PVCs to steer traffic along less utilised paths, or for guaranteed QoS or fast failover paths, but Juniper and Cisco routers can already do traffic engineering and the standards are coming along.
More MPLS info is at http://www.mplsrc.com/
Seems like Prairie iNet are using the 2.4 GHz band, which is generally unlicensed around the world - assuming they are using the same technology as Midcoast Wireless, a Maine ISP, which has a very useful FAQ
This is all based on IEEE 802.11 technology, which is normally used for wireless LANs with a range of a few hundred feet. The trick here seems to be using more power and directional antennae so that you can go up to 9 miles (or maybe much more).
One company making this sort of kit is Breezecom, who have an overview of wireless Internet access here.
This technology, along with the competing licensed LMDS technology, may make mincemeat of DSL and Cable - it involves no rights of way hassles, no cable laying, and can give very low latency plus bandwidths in the 1-2Mbps range. Having used Wireless LANs at conferences and trade shows, I found the latency and bandwidth very similar to a T1 line.
For info on 802.11, see the Linux Wireless LAN FAQ, which also has good links to generic WLAN info at the end. Although the technology for 802.11 long-distance (i.e. wireless local loop) is not identical, it should give you an idea of how things work.
For info on LMDS (Local Multipoint Distribution System), see the Webopedia entry for LMDS, which has links to related pages. One new European telco that is rolling out LMDS quite aggressively is FirstMark - they are also doing cool things with MPLS VPNs, which is how I know about them since my company just sold them the software to manage this
Low latency is important because it's a key determinant of web response time, particularly for sites with many small GIF buttons, and also because Internet routers tend to treat high-latency sessions less fairly, so they get even less bandwidth then they should. It's also essential to winning at Quake, which is clearly the critical driver here
This story matters more for the technology than for the particular ISP using it - it will affect most Slashdot readers in the next year or so, particularly those not covered by DSL or cable. In the UK, BT is being astonishingly slow at rolling out ADSL, and the cable companies have very little coverage, so wireless technology may be the only way to get broadband for many people...
How exactly would Nokia be hurt by competitors putting prices up? They would just walk off with the market... More likely is that there would be a price war and a few companies retire hurt, with the remaining ones putting up prices - however, price wars can't go on too long, and are hardly a reason to wish for less competition...
Convergence is happening, whether you like it or now - in other words, market boundaries are becoming much less distinct as companies find there's technology and maybe even demand for a box that can play games, show digital TV, surf the Net, etc.
Anyone who lives in Europe, Asia, or other places outside US/Canada is very likely to be paying for Internet access by the minute (local calls are not free). Although some countries are moving towards special internet access deals for flat fees, at least in the UK these ISPs are massively oversubscribed.
Travellers outside US/Canada can also make use of this feature; global roaming services such as gric.net also charge by the minute for the convenience of using any ISP in their scheme.
I just read that interview, and I think he came out of it OK - he is very focused on a single idea, i.e. a zealot (he almost has to be, by definition), and that's one big reason why the GNU movement has succeeded.
The interviewer was either playing devil's advocate or rather clueless about free software - RMS did OK in handling this, only occasionally letting rip...
I am not a huge fan of RMS and I tend to ignore the demands to refer to GNU/Linux not Linux (there are many other types of software in the typical Linux distro, e.g. Perl Artistic licensed, BSD and X11) - however, without RMS I would not now be typing this message on a Linux system compiled by GCC and running Gnome, so I think we all owe him a huge thank you, even if we don't agree with everything he says.
Courage in your convictions may sometimes be hard to distinguish from arrogance...
Sorry to reply to my own message *again*, but where I said logitech.com I meant legato.com...
This slowdown occurs with raw partitions as well as VMware's virtual partitions or whatever they call them. In fact, PS/2 mice sometimes caused a slowdown, though this may have been fixed by now.
Overall it is a great product, but it's not for people who want extremely high performance without providing extra hardware to run it well.
I use VMware quite a lot - it generally runs thing at approx half the speed of the real machine. This is not usually a problem, and doesn't apply to CPU-bound work, which runs at nearly full speed, but there is a noticeable slowdown in I/O.
VMware has patents too, see http://www.vmware.com/news/pr1.html where it mentions 'patent-pending'.
And it actually has a product, not that this means anything in patent wars...
It's a complete waste of time to just look at the titles - these describe generally what the patent is about, but the *claims* of the patent, listed at the end, are what is actually protected.
Any details in the title, abstract, description, etc are only there to help a skilled practitioner reproduce the invention (under license), or to invent something new that doesn't infringe the claims.
I don't see how the fourth one is any different to VMware - it does everything listed.
VMware has at least some patents filed for - they mention this on their first press release, at http://www.vmware.com/news/pr1.html.
Also, it's meaningless to say 'countries that laugh at US patent laws' in that they are irrelevant to patents in almost every country.
If you want patents in the UK, Japan, etc, you must file them in each country. There are some short cuts, i.e. you can file European patents and International patents, but these must eventually 'fork' into country-specific patents at later stages of the process. Patents don't automatically cross national boundaries, like trademarks but unlike copyright.
I think you mean MVS - VMS runs on Vaxes and Alpha's.