The article said that the structures could probably be mass-produced in the same way as semiconductors.
Re:Separating Interface from Implementation, Legal
on
Microsoft And The GPL/LGPL
·
· Score: 3, Informative
It seems to me after reading the MS CIFS license that MS actually has a valid concern: they want to publish the CIFS standard and retain intellectual property rights; they want to allow implementors of the standard to be able to implement and sell or give away their implementation and/or their source code; and they want to prohibit implementors from changing the licensing terms of other bundled software components.
Whether or not this conditions is included, there's nothing a licensee can do that would affect the rights of Microsoft and its other licensees to use that specification.
Microsoft is NOT trying restrict use of CIFS.
It certainly looks like it is.
They are trying to prevent a CIFS product vendor from forcing other bundled products to adhere to the same licensing model.
But the GPL and LGPL don't do that.
I and others believe that the people who came up with this aimed to imply, falsely, that the GPL and LGPL do place restrictions on bundled software, as a justification for banning their use for software based on Microsoft specifications.
I honestly can't see what these sections are supposed to protect Microsoft from, other than fair competition.
So I don't live in Japan, I'm too tall for that country anyways.
Neither do I, but since I've had to write software to work with i-mode I made sure to do my research and find the important technical specs and mailing lists.
As I understand it WAP penetration on i-mode is still relatively sparse compared to the overall number of i-mode users. As for the access control and whatnot through the WAP gateway I sort of got it right, in spirit if nothing else.
I'm sorry, but you really didn't. Japanese i-mode doesn't use WAP at all. European i-mode appears to make use of some parts of WAP 2.0, but the Japanese service is proprietary.
The point is NTT controls all access to i-mode, the WAP gateways are run by NTT and thus everyone (the katte sites) who wants to offer i-mode service has got to clear it through NTT and then the access if it is billed gets billed to the users i-mode account.
No, the katte sites are entirely unofficial - there is no need to get any kind of approval from DoCoMo to make a site available by URL entry. DoCoMo bills for access to all sites (and for email) on a per-packet basis (except that a few hundred packets per month are included in the i-mode subscription).
Was that meant to be funny? DoCoMo is the name of the company, and it really is a play on 'dokomo', though it's also supposed to be short for something like 'Do Communicate by Mobile'. The hypertext/email service is called 'i-mode', in which the 'i' doesn't have a fixed meaning but can be taken as 'information', 'Internet', or any number of other things.
It is not an internet phone although there are portals so users can get on the internet. While i-mode sites use a heavily tweaked version of HTML they are housed on a proprietary network much like online services used to be in the US before connections to the internet were the big thing.
No they aren't. NTT DoCoMo runs a gateway system, something like a WAP gateway, that fetches pages using HTTP. Most i-mode sites are hosted on ordinary web servers connected to the Internet. A site operator can set restrictions on the web server so that it will only serve i-mode pages to addresses assigned to the gateway. Site operators that wish to have more security, such as banks, can have a dedicated link to the gateway instead. (Except for the latest models, i-mode phones do not support encryption.)
DoCoMo runs an i-mode portal, the i-menu, and most users stick to the sites listed on this. However, all i-mode phones allow access to and bookmarking of unofficial 'katte' sites by URL.
Most of the pay content on the service is hosted either by NNT DoCoMo themselves or by third parties that charge fees to your phone's account.
No it isn't. DoCoMo only runs the subscription and billing system. When the i-mode gateway sends an HTTP request to an official site (one that's on the i-menu), it includes a subscriber ID which the site can use for access control and customisation.
The advice they received was that reasonable charges can be made for distribution and support (but not for licensing) without affecting the export status of the software.
Well my cable company leased me the same address for about 6 months and, er, doesn't ban servers. The main reason for using DHCP on cable is, as everywhere else, to enable central management of network configuration, reducing network administration and support costs.
I got basic cable Internet for £6 ($8) a month on top of my cable TV/phone bill, and I can run servers, and my DHCP-assigned address hasn't changed in the last 6 months. Mind you, that's the 64/32 kbps option, and the cable company (NTL) is rumoured to be going bankrupt at the end of the month...
...and the server sends over a DES session key encrypted using the server's private key.
That would be no good, as it would mean that an eavesdropper could decrypt the session key too! I'm not sure how it's actually done, but one possible way is for the client to choose a session key and encrypt it with the server's public key - then only the server can decrypt it. (However, if a man-in-the-middle can substitute a different server key without the client noticing, he can still decrypt the contents of the session. Theoretically this won't happen because the man-in-the-middle shouldn't be able to get a certificate issuer to certify a key for an address that doesn't belong to him, but the CAs don't check all that thoroughly.)
I like it because people have to get my authorization to see me on-line.
*snort* There's no server-side check for authorisation, so with unofficial client software one can add users to the watch list without their permission.
The main motivation for the development of AGP was to allow the graphics contoller to use main memory rather than its own separate memory, at a time when memory was getting more expensive.
XIP is just a way of saving RAM. When a program is loaded from file-system that is itself in RAM, there's no need to make a second copy in RAM of those parts of the program image that are read-only.
Why are files locked in the first place? In general, file locking and unlocking is done around changes that need to be atomic. Backing up a file that is in the middle of such a change is not very useful, as it may not be in a consistent state.
That assumes that you can actually measure the costs due to individual customers, and I'm not sure that that's generally the case for communications services.
Actually, right at the bottom of the article, Ballmer is quoted as saying that "money from the settlement could be used to buy software from Microsoft competitors." Of course, if Microsoft really simply have its own interests at heart here, they would simply be offering cash for the schools to spend on whatever they really needed.
The patent was granted in 1989 but applied for over 10 years earlier. The patent application seems to have been amended after initial filing so I'm not sure which of the dates listed really counts, but if you're looking for prior art then you need something further back (unless Xanadu was mostly formed by 1974).
This depends on which C library the program uses. If it's the Microsoft C run-time then you can escape a double-quote by putting a backslash before it. If there are literal backslashes before a double-quote (whether or not it is escaped) then you need to double them. Cygwin unfortunately uses a different rule and I couldn't find a totally reliable way to pass any old string of double-quotes and backslashes to it. Some DOS-compatible commands like cd have compatibility kluges to cope with lack of double-quotes, and I think the shell has a kluge to cope with lack of double-quotes around the command name, but systematic quoting/escaping should be accepted too.
This feature doesn't work perfectly. "dir FONTS:" may work, but if you try to Lock("FONTS:",SHARED_LOCK) you'll just get a lock on the first directory in the list rather than on the union of the directories.
The "Add/Remove Programs" control panel is just a pretty interface to uninstall programs (listed in HKEY_LOCAL_MACHINE\Software\Microsoft\Windows\Curr entVersion\Uninstall). There's no central tracking of file usage. Any intelligence has to be implemented by each uninstall program. Windows Installer appears to be a bit smarter, but I don't think it has anything like the all-knowing databases that rpm and dpkg use.
Slashdot Mirror
The article said that the structures could probably be mass-produced in the same way as semiconductors.
Whether or not this conditions is included, there's nothing a licensee can do that would affect the rights of Microsoft and its other licensees to use that specification.
It certainly looks like it is.
But the GPL and LGPL don't do that.
I and others believe that the people who came up with this aimed to imply, falsely, that the GPL and LGPL do place restrictions on bundled software, as a justification for banning their use for software based on Microsoft specifications.
I honestly can't see what these sections are supposed to protect Microsoft from, other than fair competition.
Ah, as I thought. That explains the not-so-random composition of some of the shots.
Neither do I, but since I've had to write software to work with i-mode I made sure to do my research and find the important technical specs and mailing lists.
I'm sorry, but you really didn't. Japanese i-mode doesn't use WAP at all. European i-mode appears to make use of some parts of WAP 2.0, but the Japanese service is proprietary.
No, the katte sites are entirely unofficial - there is no need to get any kind of approval from DoCoMo to make a site available by URL entry. DoCoMo bills for access to all sites (and for email) on a per-packet basis (except that a few hundred packets per month are included in the i-mode subscription).
"Disney Acquires Senator Chihiro, ..."?
Was that meant to be funny? DoCoMo is the name of the company, and it really is a play on 'dokomo', though it's also supposed to be short for something like 'Do Communicate by Mobile'. The hypertext/email service is called 'i-mode', in which the 'i' doesn't have a fixed meaning but can be taken as 'information', 'Internet', or any number of other things.
Which is why the European version of i-mode will be using GPRS, not PDC-P.
No they aren't. NTT DoCoMo runs a gateway system, something like a WAP gateway, that fetches pages using HTTP. Most i-mode sites are hosted on ordinary web servers connected to the Internet. A site operator can set restrictions on the web server so that it will only serve i-mode pages to addresses assigned to the gateway. Site operators that wish to have more security, such as banks, can have a dedicated link to the gateway instead. (Except for the latest models, i-mode phones do not support encryption.)
DoCoMo runs an i-mode portal, the i-menu, and most users stick to the sites listed on this. However, all i-mode phones allow access to and bookmarking of unofficial 'katte' sites by URL.
No it isn't. DoCoMo only runs the subscription and billing system. When the i-mode gateway sends an HTTP request to an official site (one that's on the i-menu), it includes a subscriber ID which the site can use for access control and customisation.
Read the press release. They do support POP3 and IMAP.
The advice they received was that reasonable charges can be made for distribution and support (but not for licensing) without affecting the export status of the software.
Well my cable company leased me the same address for about 6 months and, er, doesn't ban servers. The main reason for using DHCP on cable is, as everywhere else, to enable central management of network configuration, reducing network administration and support costs.
I got basic cable Internet for £6 ($8) a month on top of my cable TV/phone bill, and I can run servers, and my DHCP-assigned address hasn't changed in the last 6 months. Mind you, that's the 64/32 kbps option, and the cable company (NTL) is rumoured to be going bankrupt at the end of the month...
That would be no good, as it would mean that an eavesdropper could decrypt the session key too! I'm not sure how it's actually done, but one possible way is for the client to choose a session key and encrypt it with the server's public key - then only the server can decrypt it. (However, if a man-in-the-middle can substitute a different server key without the client noticing, he can still decrypt the contents of the session. Theoretically this won't happen because the man-in-the-middle shouldn't be able to get a certificate issuer to certify a key for an address that doesn't belong to him, but the CAs don't check all that thoroughly.)
*snort* There's no server-side check for authorisation, so with unofficial client software one can add users to the watch list without their permission.
The main motivation for the development of AGP was to allow the graphics contoller to use main memory rather than its own separate memory, at a time when memory was getting more expensive.
XIP is just a way of saving RAM. When a program is loaded from file-system that is itself in RAM, there's no need to make a second copy in RAM of those parts of the program image that are read-only.
Why are files locked in the first place? In general, file locking and unlocking is done around changes that need to be atomic. Backing up a file that is in the middle of such a change is not very useful, as it may not be in a consistent state.
Log in as "cypherpunks3" with password "cypherpunks".
DR-DOS is a clone of MS-DOS, and MS-DOS is a knock-off of CP/M.
That assumes that you can actually measure the costs due to individual customers, and I'm not sure that that's generally the case for communications services.
Actually, right at the bottom of the article, Ballmer is quoted as saying that "money from the settlement could be used to buy software from Microsoft competitors." Of course, if Microsoft really simply have its own interests at heart here, they would simply be offering cash for the schools to spend on whatever they really needed.
The patent was granted in 1989 but applied for over 10 years earlier. The patent application seems to have been amended after initial filing so I'm not sure which of the dates listed really counts, but if you're looking for prior art then you need something further back (unless Xanadu was mostly formed by 1974).
This depends on which C library the program uses. If it's the Microsoft C run-time then you can escape a double-quote by putting a backslash before it. If there are literal backslashes before a double-quote (whether or not it is escaped) then you need to double them. Cygwin unfortunately uses a different rule and I couldn't find a totally reliable way to pass any old string of double-quotes and backslashes to it. Some DOS-compatible commands like cd have compatibility kluges to cope with lack of double-quotes, and I think the shell has a kluge to cope with lack of double-quotes around the command name, but systematic quoting/escaping should be accepted too.
This feature doesn't work perfectly. "dir FONTS:" may work, but if you try to Lock("FONTS:",SHARED_LOCK) you'll just get a lock on the first directory in the list rather than on the union of the directories.
The "Add/Remove Programs" control panel is just a pretty interface to uninstall programs (listed in HKEY_LOCAL_MACHINE\Software\Microsoft\Windows\Curr entVersion\Uninstall). There's no central tracking of file usage. Any intelligence has to be implemented by each uninstall program. Windows Installer appears to be a bit smarter, but I don't think it has anything like the all-knowing databases that rpm and dpkg use.