There is no direct linkage from the client to the server, simply an XML file that provides hints to the caller as to what structure to send messages. Hints that may or may not be ignored.
WSDL files that are hand-written may be released under a license, but what if the WSDL is machine generated by a runtime, such as the Apache one?
Nor is it mandatory to use WSDL to talk to a SOAP service. With any written documentation as to request structure I could perhaps rewrite my own WSDL/compose XML messages without any direct importing of the WSDL File. Or I could use the WSDL from a non GPL server and then rebind the client to a GPL instance. If they shared the same WSDL (and consistent behavour), I should not have to care what the licensing of the endpoint was.
It says you can run your server with all OSS software (bsd, GPL, and other OSI certified license), or you pay for the closed-source version.
This means that OSS apps can use the framework for hosting their systems, but if you want to stay closed source, you need to pay.
It's primarily that way for business reasons -to make companies who dont want OSS code to pay- but it's an interesting idea.
vmware images are how we (regularly) deploy stuff without caring where it is run. the uk grid images are done differently, by having a bit of dedicated file space on a SAN providing the sys images -and having the blades boot of that bit of the image.
essentially our blades can reconfigure to boot what looks like a different physical image as fast as you can restart them; vmware is something we run on top for extra dynamicness and load sharing.
I have access to some of the machines; we donate idle systems to the project in exchange for low cost (read free) access to the superjanet network. When they arent doing UK NeSC grid stuff I can bring up vmware images of whatever distro I feel like, run whatever stuff we need -in my case usually distributed testing of distributed software.
That is how the grid works -it uses spare cycles on machines in the network. Unlike Seti@home, they are very fussy about bandwidth; you need a serious link to play. Most of the tier-1 sites are UK academic sites -rutherford labs, oxford, the london universities, etc. Tier-2 sites are us industrial sites with machines we let them play on.
Like you say, data storage is a big issue, and something the grid needs to work on. A lot of grid forum work is on data.
Incidentally, when the LHC comes on line, then the serious data starts to collect. Any collisions -events, they call them- generate vast amounts of data. There is some logic in the system to immediately split dull events from potentially interesting ones, but those interesting ones happen often enough you need to buffer it all up, then do more rigorous analysis to see if it is something new or something known about. The uk grid will be used for analysis of stuff the front end thinks could be interesting. Right now it can test the data generated by simulations (people can get phds writing good event simulations), so they can verify that the analysis code works.
One of the great fear in high energy physics is that the filter and detection logic is buggy and that nobel-grade events are missed because the data is simply too bizarre for the analysis code, and so gets missed out,..
The real attack on portability was not IE itself, but IE's 'quirky' handling of standards, and, worst of all, ActiveX.
If you go to a site which relies on AX to download some binary junk to your box to work (the way so many MS/MSN sites do), the chances of getting it to work cross platform is pretty low (though crossover do support it now).
Fortunately, code download has been so widely abused by the Phishermen that Windows XP SP2 effectively turns it off. There is a little bar to tell you off for not trusting AX, but no irritating popups telling you are unworthy,
Having moved back to the UK, I now own a set of DECT phones which are the dogs bollocks.
Interference: zero. Range: Excellent. Scalability: superb. The DECT standard includes an open spec for adding new phones (from any vendor) to the same base station. So we have 3 phones off one one station, one per floor.
I dont know what frequency it runs on, and frankly dont care. It works. the WLAN works. that is all I need to know.
Agreed. When monitors break (which is rare, they outlive PCs) you chuck them.
on the other hand, I think there is a lot to be said for having a license to use regedit. There is nothing that scares anyone providing informal windows support is helping someone who ran regedit to 'clean things up'.
but the moment you do that, the RIAA will come down against licensed using regedit to turn off autorun of DRM-protected CDs...you could have your license revoked for that.
There are two benefits. One, you can be sure everything has drivers. Linux on laptops is still so hit and mix. The old 1999 laptop now hooked up the hifi stack runs suse9.1pro fine, my current laptop has buggy acpi and refuses to boot it.
Two, you avoid paying the MS tax from the outset. That denies them cash, and it shows to the customer how much of the price of a PC is the OS.
There was something by Andy "ange-ftp" Norman of HP Laboratories and some colleagues that did that only more aggressively. Anything with the code red hole had IIS turned off.
This stuff was not a worm (it was centrally managed), and deployed inside the HP class A network (15.*.*.*), where all machines are owned by the IT department. Thus they could get away with it.
Inside the firewall you can do this thing -companies and universities are obvious places. Bu t I wonder about the big broadband ISPs, I dont think I want them scanning my box, as before long the MPAA will want them shutting down BitTorrent services.
My ISP (plus.net) blocked me off the net for attempting to make outbound port 135 connections; these netbios over TCP calls were seen as a sign of a virus on my machine.
It was actually just Outlook 2003's new "auto-reconnect" facility trying to rejoin to a corporate site whose DNS address resolved but which was firewalled off (sensibly).
I suppose you could actually view plus.net blocking all network access of outlook 2003 users a benefit for their own safety -especially if extended to IE use in general.
yes -the one thing that we can be sure of that OSS code will be properly indented. Nobody need fear that military systems built on linux will have illegally tabulated code:)
As wikipedia puts it, "temporary" is reassuring, like "The emergency" will soon be over, Eurasia and Oceania will be back at peace again, etc. etc.
One of the interesting features of this thirty year old UK legislation is that it gives the government the right to exclude people from northern ireland (i.e. part of the UK) from the rest of the country. It is as if suspicious people from, say, Alaska, were prevented from visiting the rest of the US.
If you are curious about the UK "troubles", I would suggest you read "Bandit Country". Though not from a library, obviously, as they would get suspicous.
I agree with you mostly; my wife was born in Nairobi, and it is *scary* there nowadays.
But there is an growth in 'appropriate' technologies. What failed in the past were these grandiose infrastructure schemes -big dams, etc,that were meant to help the countries by giving everyone western style infrastructure.
One of the most interesting technologies turns out to be mobile phones. There are examples of farmers in bangalore using the village cellphone to get the current prices of rice in London, so as to get less ripped off by traders buying their wares. And fisherman in india using the phones to work out which port is giving the best prices for which fish.
I think in one place (Philippines), you can even go into business reselling airtime on your own phone, through something managed by the telco.
So: appropriate technology. Go for it. That doesn't necessarily mean that 4-head PCs are appropriate; we will have to see the results.
My ISP (plus.net) blocks port 135 outbound. or to be precise, if you keep trying to open port 135, your network access gets taken away. They try and do it to control windows-based worms, but outlook 2003 has exactly the same signature of the worms.
I dont know about SPARC, but Java 1.4 on x86 can hand off FP maths to the SSE registers, rather than the classic "shite" x86 FPU, the one Patterson and Hennessey cover as a "what not to do" example in their Computer Architecture Book.
And that is one of the nice things of Java: you get a boost from whatever accelleration is in the platform, if the compiler handles it.
I think to make best use of the SIMD features in modern CPUs we need better array operations with compilers/runtimest that bind to the new SIMD FPUS. I think Fortran still has the edge there, for historical reasons.
We use IPSec for back-to-work VPN. Our WLAN is configured as Jean-Tourhilles told them to: it looks like a public WLAN, so guests get online. To get back into the corporate network you set up an IPSec link to the (local) entry point.
One problem: roaming doesnt work. If you move APs, you get disconnected.
Problem two: driver support. Windows, linux, its all a pain. Its not that these OSs dont come with drivers, its that the corporates want ones that work with their authentication tokens (like smartcards), and want the box to be secure when dialled in. So when the IPSec link goes up, the laptop seals itself from the outside world.
problem three: not all APs route IPSec. Sometimes you are at hotels and conferences with WLAN, Internet, but no IPSec. That hurts.
So it is painful, and support people suffer. SSH is a lot simpler to set up. But IPSec is powerful, and I like it a lot (when it is working)
I did that not so long back with an AV-receiver; I wanted to make sure it would handle the outputs of the laptop right. They were bemused but happy to help -and didnt try and sell me winXP.
I have a netgear PrismII based PC card: slow but stable -worked out the box in SuSE 9.1
My work laptop has an ActionTec mini-PCI card and that was trouble indeed. It kept on locking on a session (RH.9), and now that I am running WinXP SP2 on the laptop, it wont hibernate while the card is in use.
So: open source -incomplete drivers you'd have to fix by hand. Closed source -shit drivers you cant fix. Either way -no out-the-box networking.
I actually got JJT to get RH9 working on my laptop; I was in HPLabs in Palo Alto, cornered him and refused to leave till it was going. It only took half an hour.
Not his fault though -RH9.0's defaults were towards PCMCIA cards, not mini-PCI, but it still shows the problems with mainstream linux.
Of course, that was last year. This weekend I stuck SuSE 9.1 on an old laptop, it found the netgear PC card, bonded to the (open) WLAN and was on the net, no network config dialog boxes at all.
we in the UK have had the joy of two politicians (Jeffrery Archer and Jonathon Aitken) put away for being lying bastards, and the country rejoiced widely.
yes, corporates are stuck in Windows land. One of their goals is to run fancy app server stuff, and for that -be it ASP, ASP.net or Java based- means windows, and historically a commercial unix (Sun, HP, IBM), with Linux a late entrant.
now that BSD does Java, things may change.
But outside the corporate, big sites like IMDB and Apache run FreeBSD, as far as I know.
yes, my email is up and down too. hosting issues that I cannot fix while on the road. Next week everything will move to home hosted on a laptop running suse 9.1. the availability wont increase, but at least I can blame myself and not the incompetence of others.
I must disagree, somewhat.
There is no direct linkage from the client to the server, simply an XML file that provides hints to the caller as to what structure to send messages. Hints that may or may not be ignored.
WSDL files that are hand-written may be released under a license, but what if the WSDL is machine generated by a runtime, such as the Apache one?
Nor is it mandatory to use WSDL to talk to a SOAP service. With any written documentation as to request structure I could perhaps rewrite my own WSDL/compose XML messages without any direct importing of the WSDL File. Or I could use the WSDL from a non GPL server and then rebind the client to a GPL instance. If they shared the same WSDL (and consistent behavour), I should not have to care what the licensing of the endpoint was.
-Steve Loughran, Apache Axis SOAP stack team.
It says you can run your server with all OSS software (bsd, GPL, and other OSI certified license), or you pay for the closed-source version.
This means that OSS apps can use the framework for hosting their systems, but if you want to stay closed source, you need to pay. It's primarily that way for business reasons -to make companies who dont want OSS code to pay- but it's an interesting idea.
vmware images are how we (regularly) deploy stuff without caring where it is run. the uk grid images are done differently, by having a bit of dedicated file space on a SAN providing the sys images -and having the blades boot of that bit of the image.
essentially our blades can reconfigure to boot what looks like a different physical image as fast as you can restart them; vmware is something we run on top for extra dynamicness and load sharing.
I have access to some of the machines; we donate idle systems to the project in exchange for low cost (read free) access to the superjanet network. When they arent doing UK NeSC grid stuff I can bring up vmware images of whatever distro I feel like, run whatever stuff we need -in my case usually distributed testing of distributed software.
That is how the grid works -it uses spare cycles on machines in the network. Unlike Seti@home, they are very fussy about bandwidth; you need a serious link to play. Most of the tier-1 sites are UK academic sites -rutherford labs, oxford, the london universities, etc. Tier-2 sites are us industrial sites with machines we let them play on.
Like you say, data storage is a big issue, and something the grid needs to work on. A lot of grid forum work is on data.
Incidentally, when the LHC comes on line, then the serious data starts to collect. Any collisions -events, they call them- generate vast amounts of data. There is some logic in the system to immediately split dull events from potentially interesting ones, but those interesting ones happen often enough you need to buffer it all up, then do more rigorous analysis to see if it is something new or something known about. The uk grid will be used for analysis of stuff the front end thinks could be interesting. Right now it can test the data generated by simulations (people can get phds writing good event simulations), so they can verify that the analysis code works.
One of the great fear in high energy physics is that the filter and detection logic is buggy and that nobel-grade events are missed because the data is simply too bizarre for the analysis code, and so gets missed out,..
What if the CTO is allergic to peanuts? Is there a fallback option?
The real attack on portability was not IE itself, but IE's 'quirky' handling of standards, and, worst of all, ActiveX.
If you go to a site which relies on AX to download some binary junk to your box to work (the way so many MS/MSN sites do), the chances of getting it to work cross platform is pretty low (though crossover do support it now).
Fortunately, code download has been so widely abused by the Phishermen that Windows XP SP2 effectively turns it off. There is a little bar to tell you off for not trusting AX, but no irritating popups telling you are unworthy,
yeah, the panasonic one I had was death to wlan.
Having moved back to the UK, I now own a set of DECT phones which are the dogs bollocks.
Interference: zero. Range: Excellent. Scalability: superb. The DECT standard includes an open spec for adding new phones (from any vendor) to the same base station. So we have 3 phones off one one station, one per floor.
I dont know what frequency it runs on, and frankly dont care. It works. the WLAN works. that is all I need to know.
Agreed. When monitors break (which is rare, they outlive PCs) you chuck them.
on the other hand, I think there is a lot to be said for having a license to use regedit. There is nothing that scares anyone providing informal windows support is helping someone who ran regedit to 'clean things up'.
but the moment you do that, the RIAA will come down against licensed using regedit to turn off autorun of DRM-protected CDs...you could have your license revoked for that.
There are two benefits. One, you can be sure everything has drivers. Linux on laptops is still so hit and mix. The old 1999 laptop now hooked up the hifi stack runs suse9.1pro fine, my current laptop has buggy acpi and refuses to boot it.
Two, you avoid paying the MS tax from the outset. That denies them cash, and it shows to the customer how much of the price of a PC is the OS.
Nice to see they chose Suse BTW.
This stuff was not a worm (it was centrally managed), and deployed inside the HP class A network (15.*.*.*), where all machines are owned by the IT department. Thus they could get away with it. Inside the firewall you can do this thing -companies and universities are obvious places. Bu t I wonder about the big broadband ISPs, I dont think I want them scanning my box, as before long the MPAA will want them shutting down BitTorrent services.
My ISP (plus.net) blocked me off the net for attempting to make outbound port 135 connections; these netbios over TCP calls were seen as a sign of a virus on my machine.
It was actually just Outlook 2003's new "auto-reconnect" facility trying to rejoin to a corporate site whose DNS address resolved but which was firewalled off (sensibly).
I suppose you could actually view plus.net blocking all network access of outlook 2003 users a benefit for their own safety -especially if extended to IE use in general.
yes -the one thing that we can be sure of that OSS code will be properly indented. Nobody need fear that military systems built on linux will have illegally tabulated code :)
As wikipedia puts it, "temporary" is reassuring, like "The emergency" will soon be over, Eurasia and Oceania will be back at peace again, etc. etc.
One of the interesting features of this thirty year old UK legislation is that it gives the government the right to exclude people from northern ireland (i.e. part of the UK) from the rest of the country. It is as if suspicious people from, say, Alaska, were prevented from visiting the rest of the US.
If you are curious about the UK "troubles", I would suggest you read "Bandit Country". Though not from a library, obviously, as they would get suspicous.
I agree with you mostly; my wife was born in Nairobi, and it is *scary* there nowadays.
But there is an growth in 'appropriate' technologies. What failed in the past were these grandiose infrastructure schemes -big dams, etc,that were meant to help the countries by giving everyone western style infrastructure.
One of the most interesting technologies turns out to be mobile phones. There are examples of farmers in bangalore using the village cellphone to get the current prices of rice in London, so as to get less ripped off by traders buying their wares. And fisherman in india using the phones to work out which port is giving the best prices for which fish.
I think in one place (Philippines), you can even go into business reselling airtime on your own phone, through something managed by the telco.
So: appropriate technology. Go for it. That doesn't necessarily mean that 4-head PCs are appropriate; we will have to see the results.
Maybe someone is secretly laying the groundwork for the Daleks when they come.
Maybe they are already here...
That is kind of brutal.
My ISP (plus.net) blocks port 135 outbound. or to be precise, if you keep trying to open port 135, your network access gets taken away. They try and do it to control windows-based worms, but outlook 2003 has exactly the same signature of the worms.
I dont know about SPARC, but Java 1.4 on x86 can hand off FP maths to the SSE registers, rather than the classic "shite" x86 FPU, the one Patterson and Hennessey cover as a "what not to do" example in their Computer Architecture Book.
And that is one of the nice things of Java: you get a boost from whatever accelleration is in the platform, if the compiler handles it.
I think to make best use of the SIMD features in modern CPUs we need better array operations with compilers/runtimest that bind to the new SIMD FPUS. I think Fortran still has the edge there, for historical reasons.
Agreed.
But at least they didnt call the RIAA about the 20GB of music on the drive.
Good question.
We use IPSec for back-to-work VPN. Our WLAN is configured as Jean-Tourhilles told them to: it looks like a public WLAN, so guests get online. To get back into the corporate network you set up an IPSec link to the (local) entry point.
One problem: roaming doesnt work. If you move APs, you get disconnected.
Problem two: driver support. Windows, linux, its all a pain. Its not that these OSs dont come with drivers, its that the corporates want ones that work with their authentication tokens (like smartcards), and want the box to be secure when dialled in. So when the IPSec link goes up, the laptop seals itself from the outside world.
problem three: not all APs route IPSec. Sometimes you are at hotels and conferences with WLAN, Internet, but no IPSec. That hurts.
So it is painful, and support people suffer. SSH is a lot simpler to set up. But IPSec is powerful, and I like it a lot (when it is working)
go in there. take your laptop and test it.
I did that not so long back with an AV-receiver; I wanted to make sure it would handle the outputs of the laptop right. They were bemused but happy to help -and didnt try and sell me winXP.
I have a netgear PrismII based PC card: slow but
stable -worked out the box in SuSE 9.1
My work laptop has an ActionTec mini-PCI card and that was trouble indeed. It kept on locking on a session (RH.9), and now that I am running WinXP SP2 on the laptop, it wont hibernate while the card is in use.
So: open source -incomplete drivers you'd have to fix by hand. Closed source -shit drivers you cant fix. Either way -no out-the-box networking.
I actually got JJT to get RH9 working on my laptop; I was in HPLabs in Palo Alto, cornered him and refused to leave till it was going. It only took half an hour.
Not his fault though -RH9.0's defaults were towards PCMCIA cards, not mini-PCI, but it still shows the problems with mainstream linux.
Of course, that was last year. This weekend I stuck SuSE 9.1 on an old laptop, it found the netgear PC card, bonded to the (open) WLAN and was on the net, no network config dialog boxes at all.
Perjury is a wonderful piece of justice.
we in the UK have had the joy of two politicians (Jeffrery Archer and Jonathon Aitken) put away for being lying bastards, and the country rejoiced widely.
yes, corporates are stuck in Windows land. One of their goals is to run fancy app server stuff, and for that -be it ASP, ASP.net or Java based- means windows, and historically a commercial unix (Sun, HP, IBM), with Linux a late entrant.
now that BSD does Java, things may change.
But outside the corporate, big sites like IMDB and Apache run FreeBSD, as far as I know.
yes, my email is up and down too. hosting issues that I cannot fix while on the road. Next week everything will move to home hosted on a laptop running suse 9.1. the availability wont increase, but at least I can blame myself and not the incompetence of others.