There may be circumstances where an address appears to be valid but
cannot reasonably be verified in real time, particularly when a
server is acting as a mail exchanger for another server or domain.
"Apparent validity" in this case would normally involve at least
syntax checking and might involve verification that any domains
specified were ones to which the host expected to be able to relay
mail. In these situations, reply code 252 SHOULD be returned. These
cases parallel the discussion of RCPT verification discussed in
section 2.1. Similarly, the discussion in section 3.4 applies to the
use of reply codes 251 and 551 with VRFY (and EXPN) to indicate
addresses that are recognized but that would be forwarded or bounced
were mail received for them. Implementations generally SHOULD be
more aggressive about address verification in the case of VRFY than
in the case of RCPT, even if it takes a little longer to do so.
That's SHOULD, and there is more than one method of performing VRFY. Distributing user databases is not the most efficient way of doing this, and I know of very few, if any secondary hosts that implement it. It might happen in corporate systems with a spam firewall as a secondary exchanger but it just wouldn't work in a large scale ISP environment, particularly in one which shares secondary exchangers with other ISPs or their upstream bandwidth provider.
Spot on. Part of the reason that IBM continues to sell mainframes is that IBM understands mainframes, and that there is a generation of people, many of whom are now in management, who understand mainframes, and their way of processing and their way of billing. An aspect of IBM's On Demand campaign was that IBM would try and sell interested companies all or part of a mainframe with the option of making further processing power available when it was needed. IBM can provide processor time (and billing) down to a fairly granular level, which is appealing to companies who have variable but considerable number crunching requirements, and the current zSeries offers virtual machines that can also be managed and billed in the same way. In theory anyway. I don't think it's any surprise that despite the technologies that they are currently promoting, all the other big hardware providers also provide big iron that can be made to work in a big iron way, because they have also found that that is a good way to gain inroads into the 'mainframe mentality'.
You can. There's an emulator called Hercules which implements the S/370, ESA/390 and z/390 instruction sets. You still need a copy of the relevant OS to run on it but some of the older IBM mainframe OSes are public domain, and there is an implementation of Linux for the 390 architecture.
I would hope that the word LIKE is the important one here. It's true that representatives, be they Congressmen, Senators, MPs or whatever, have taken to blogging or MySpace or Facebook as way of 'increasing communication', but the truth that all it does is add another layer of mediated, mostly one way communication about what the honourable member would like you to know about what you are paying them to do. It's only partially how an electronically enabled government should work.
In the UK it's taken a voluntary group, MySociety, to pick up voting and speaking data from Parliament and turn it into a site that provides statistics on how our MPs perform. They are currently lobbying to get this data presented in a globally usable way. It seems to me that this is the absolute minimum data that we should get in return for our trust and the money we spend to maintain democracy. Recent concerns about how MPs use their expenses system have also caused Parliament to respond with some openness, but in the end they are our public servants and *none* of this should be secret. OK, it might not be the most rivetting read in the world, and I can understand the poster who said that they don't want to know how government works, just that it does, but I think it's the least that we are owed. Concentrate on general accountability instead of just getting 'the message' across.
There was one in StarOffice. I used it in an Exchange environment with few problems (it used POP3/IMAP and LDAP so you had to be friendly with the admin) but it disappeared from StarOffice before Sun took it over and never appeared in OpenOffice. Wonder where and why it went?
I've got three and I can't ever remember which one has money on. That's something that you can't check unless you register your card or check it on a terminal, and as I'm generally about 200 miles away, I don't get a chance to.
Most people will use an Oyster as a travelcard, and will buy a weekly or monthly card for the zones that they commute in. If you buy a weekly card for zones 1-3, and then travel outside of zones 1-3 for some reason, then the system logs that you exited the network at a station outside of the zones that your travelcard is valid for and, as I understand it, charges the full price for a ticket to your card until the end of the day, when the system recalculates it to the difference between your boundary of your card and where you left the system at the card price. It's a batch process as it would be too slow to do on the fly and also allows for incomplete journeys, where someone hasn't tapped out by passing through an open gate, for example, to be calculated at the card price of a ticket rather than the cost of a single ticket, so start points and exit points have to be stored at least until the batch processes have been run.
Go and Google Mike Corley. He has been convinced for many years that he is being persecuted by MI5 through TV and radio programmes. Follow his threads on various newsgroups from about 1995 and you will be open-mouthed with wonder and fear. If you read them for long enough to start to believe them.
Benefit fraud and importing fags and beer from the continent, which as we know is all brought in by terrorists and organised crime, so HMRC are fully justified in dismantling your car at Dover, if it looks like it's sitting a bit low on its suspension.
It's a good point. When GPL 2 was written, the Internet wasn't as mature as it is now, and the main method of physically transporting data would have been floppies. That's been superceded by CDs and now by the 'net, so it would take someone who was genuinely obstructive to break out the HD 1.44s and send out code in that way. Hmm, I really should bin that box of floppies one day...
And how do you direct browsers to their nearest local cache? Squids can't do that on their own. That's where Akamai's technology comes in. Akamai's caching server, is about the same age as squid: proxying technology was developed to cope with slow connections across university and ISP networks by serving pages from closer to the user. Akamai extended this idea to multiple locations that communicated with each other and adapted networking concepts such as DNS and BGP to direct traffic to its nearest cache and minimise data transfer across the network. So it's not wrong to say that you *could* do it with squid, but Akamai don't, and the technology to achieve what Akamai does is over and above that of just using squids in different places. For that matter squid wasn't the first caching server either.
If you ever read more than the headings of articles you'd know that/. has featured more than enough articles about this sort of thing. Slashdot doesn't 'hate copyright', it opposes copyright abuse, be it through patent stupidity or IP theft. It doesn't matter whether a piece of work is licensed at all: copyright still exists particularly for situations like this, and the right to assert copyright is the right to be protected from having your work used without attribution. The GPL isn't for photography: that's why Creative Commons exists, and CC does not preclude the right to attribution or indeed the right to make money out of intellectual property. Try and understand the issues before making yourself look stupid. It saves time.
Where the absence of Photoshop is one of the most common complaints about takeup, why not address the native solutions? Gimp needs improvements in its high end and workflow processes, so why not finance them? That applies to Google, or anyone who is interested. That's what open source is about. If your business profits from Gimp and you want it to be improved, donate money and time to the project.
I think as always Google have an interest in improving Wine, and it seems that Adobe are reticent about committing to a Linux/Unix port, perhaps over licensing issues, perhaps over the amount of code that needs to be changed (I would hazard that the Mac OS version of CS probably hasn't changed a great deal since OS9 days, with Carbonisation being the majority of the porting work), so Photoshop on Wine is a compromise that has benefits in that Google gets improvements to Wine and Adobe keeps selling Photoshop licences.
I use Pidgin on Windows and Linux and have just switched to Adium on OS X after using Mercury for a while, but getting frustrated with its closedness and the idiosyncrasies of its developer, and was surprised to see that there still isn't webcam support, which is somewhat remiss given the ubiquity of cams on laptops these days (Mercury does have it, and file transfer that works). This does strike me as a bit of a show-stopper at the moment, and it may well be that if Mozilla Messaging adopts libpurple as its IM library, that would be the place in which development should initially concentrate.
I'm old fashioned, and have always used a mail client rather than web mail where possible as I like having a rich editing environment. In recent years I have moved to IMAP mail wherever I can in order to be able to at least read my mail pretty much anywhere. Thunderbird (and Firefox)'s OS portability has made it possible to have a common mail interface on practically any computer, but what is still missing, as has been said elsewhere, is a common environment. This is where Gmail et al (but mostly GMail) has the advantage: it can be run practically anywhere that there is a web browser and Internet access. Now that Gmail has POP3 and IMAP access it can be integrated into Thunderbird. Similarly there is at least one application (for OS X at the moment - it's called Mailplane) that integrates Gmail with the desktop and hooks into iPhoto, iTunes, Growl and the other things that make OS X a cool working environment. A lot of these things are probably available as extensions for Thunderbird. The gap is physical portability: for me, the ability to open Thunderbird on any machine and retrieve your environment, or indeed, have a Thunderbird-like experience (terrible phrase) through a web browser would be very valuable. Thunderbird has a common configuration format that can be read across a WAN all ready, and it seems to me that that could be extended across the Internet in a relatively trivial way, and could be applied to a web browser interface in the same way: there could also be an option for offline configuration storage on a USB key or similar. There would be a public network and the option to have private servers (free software of course, with paid support) that could either be internal or Internet facing. The aim would be for Thunderbird to become a genuinely portable communications environment that would provide an alternative to or even a complement for Exchange or Lotus Notes while using FOSS components.
Lots of superficial lookalikes, yes, but none that approach the spec of the real thing. I seem to have grown immune to the Reality Distortion Field and have no real interest in the iPhone, but with the exception of the HTC Touch, I haven't seen one that's anything but cosmetic. And if anyone mentions the Neo 1973, I'll screw their pelvis to a cake stand.
...and the ones which anyone who wants a decent connection would never consider. It's easy to find stories about Virgin and Carphone Warehouse's peak period bandwidth throttling policies and BT's capacity issues and general commoditisation of broadband. In-session advertising is an inevitable result of the business models that these companies have adopted (in Carphone Warehouse/Talk Talk's case, they offer a free service quite aggressively, and according to reports, you get what you pay for). BT is the most worrying as they have been acquiring smaller ISPs in the last couple of years, including a couple of the larger independents, Madasafish/Brightview and Plusnet, so may roll the policy out across their brands, or indeed may create a premium market with their other brands while BT Broadband becomes the ITV of the UK Internet.
I've never been convinced by OS X as a server. Apart from in low admin environments where you can effectively set it up and walk away, it just isn't going to be useful. Most sysadmins will have their preferences for OS builds, apache builds, mail configs and so on, and applying Apple's Just Works principles to that level of management is going to lead to conflict. OS X is still good as a desktop (although I can see that my next refresh will be a decently specced small laptop running Ubuntu) but XServe would have been better with something more open.
That's actually a good way of doing it. There's an open source build server called Unattended that can provide a fully patched Windows build automatically - you provide the installer CD and it will pull down all the updates - and it recommends rebooting after each patch. That sounds like the same thing across a local network. I can recommend it as an alternative to ghost by the way: it will boot any PC that can be booted with PXE or bootp.
'Stop blogging, making videos and writing articles, and start fighting with legislation, with money, with burning tires and real 100,000 people marches'
Weeeellll, no. You need transmitters to provide that whitespace in the first place, and they ain't there in the places that the OP is talking about. A place that immediately springs to mind is Pocohontas County in WV, which is a radio and TV deadzone because of the National Radio Quiet Zone. There is a single low powered AM radio station in the area and Verizon has a monopoly on cable TV, phones and broadband. It isn't going provide any advantage there.
Slashdot Mirror
Absolute rubbish. From RFC2821:
There may be circumstances where an address appears to be valid but
cannot reasonably be verified in real time, particularly when a
server is acting as a mail exchanger for another server or domain.
"Apparent validity" in this case would normally involve at least
syntax checking and might involve verification that any domains
specified were ones to which the host expected to be able to relay
mail. In these situations, reply code 252 SHOULD be returned. These
cases parallel the discussion of RCPT verification discussed in
section 2.1. Similarly, the discussion in section 3.4 applies to the
use of reply codes 251 and 551 with VRFY (and EXPN) to indicate
addresses that are recognized but that would be forwarded or bounced
were mail received for them. Implementations generally SHOULD be
more aggressive about address verification in the case of VRFY than
in the case of RCPT, even if it takes a little longer to do so.
That's SHOULD, and there is more than one method of performing VRFY. Distributing user databases is not the most efficient way of doing this, and I know of very few, if any secondary hosts that implement it. It might happen in corporate systems with a spam firewall as a secondary exchanger but it just wouldn't work in a large scale ISP environment, particularly in one which shares secondary exchangers with other ISPs or their upstream bandwidth provider.
Spot on. Part of the reason that IBM continues to sell mainframes is that IBM understands mainframes, and that there is a generation of people, many of whom are now in management, who understand mainframes, and their way of processing and their way of billing. An aspect of IBM's On Demand campaign was that IBM would try and sell interested companies all or part of a mainframe with the option of making further processing power available when it was needed. IBM can provide processor time (and billing) down to a fairly granular level, which is appealing to companies who have variable but considerable number crunching requirements, and the current zSeries offers virtual machines that can also be managed and billed in the same way. In theory anyway. I don't think it's any surprise that despite the technologies that they are currently promoting, all the other big hardware providers also provide big iron that can be made to work in a big iron way, because they have also found that that is a good way to gain inroads into the 'mainframe mentality'.
You can. There's an emulator called Hercules which implements the S/370, ESA/390 and z/390 instruction sets. You still need a copy of the relevant OS to run on it but some of the older IBM mainframe OSes are public domain, and there is an implementation of Linux for the 390 architecture.
I would hope that the word LIKE is the important one here. It's true that representatives, be they Congressmen, Senators, MPs or whatever, have taken to blogging or MySpace or Facebook as way of 'increasing communication', but the truth that all it does is add another layer of mediated, mostly one way communication about what the honourable member would like you to know about what you are paying them to do. It's only partially how an electronically enabled government should work.
In the UK it's taken a voluntary group, MySociety, to pick up voting and speaking data from Parliament and turn it into a site that provides statistics on how our MPs perform. They are currently lobbying to get this data presented in a globally usable way. It seems to me that this is the absolute minimum data that we should get in return for our trust and the money we spend to maintain democracy. Recent concerns about how MPs use their expenses system have also caused Parliament to respond with some openness, but in the end they are our public servants and *none* of this should be secret. OK, it might not be the most rivetting read in the world, and I can understand the poster who said that they don't want to know how government works, just that it does, but I think it's the least that we are owed. Concentrate on general accountability instead of just getting 'the message' across.
There was one in StarOffice. I used it in an Exchange environment with few problems (it used POP3/IMAP and LDAP so you had to be friendly with the admin) but it disappeared from StarOffice before Sun took it over and never appeared in OpenOffice. Wonder where and why it went?
I've got three and I can't ever remember which one has money on. That's something that you can't check unless you register your card or check it on a terminal, and as I'm generally about 200 miles away, I don't get a chance to.
Most people will use an Oyster as a travelcard, and will buy a weekly or monthly card for the zones that they commute in. If you buy a weekly card for zones 1-3, and then travel outside of zones 1-3 for some reason, then the system logs that you exited the network at a station outside of the zones that your travelcard is valid for and, as I understand it, charges the full price for a ticket to your card until the end of the day, when the system recalculates it to the difference between your boundary of your card and where you left the system at the card price. It's a batch process as it would be too slow to do on the fly and also allows for incomplete journeys, where someone hasn't tapped out by passing through an open gate, for example, to be calculated at the card price of a ticket rather than the cost of a single ticket, so start points and exit points have to be stored at least until the batch processes have been run.
Go and Google Mike Corley. He has been convinced for many years that he is being persecuted by MI5 through TV and radio programmes. Follow his threads on various newsgroups from about 1995 and you will be open-mouthed with wonder and fear. If you read them for long enough to start to believe them.
Benefit fraud and importing fags and beer from the continent, which as we know is all brought in by terrorists and organised crime, so HMRC are fully justified in dismantling your car at Dover, if it looks like it's sitting a bit low on its suspension.
It's a good point. When GPL 2 was written, the Internet wasn't as mature as it is now, and the main method of physically transporting data would have been floppies. That's been superceded by CDs and now by the 'net, so it would take someone who was genuinely obstructive to break out the HD 1.44s and send out code in that way. Hmm, I really should bin that box of floppies one day...
That's quite interesting, as Microsoft used to have their own caching server system. Apple are a major client of Akamai. Hmmm...
And how do you direct browsers to their nearest local cache? Squids can't do that on their own. That's where Akamai's technology comes in. Akamai's caching server, is about the same age as squid: proxying technology was developed to cope with slow connections across university and ISP networks by serving pages from closer to the user. Akamai extended this idea to multiple locations that communicated with each other and adapted networking concepts such as DNS and BGP to direct traffic to its nearest cache and minimise data transfer across the network. So it's not wrong to say that you *could* do it with squid, but Akamai don't, and the technology to achieve what Akamai does is over and above that of just using squids in different places. For that matter squid wasn't the first caching server either.
If you ever read more than the headings of articles you'd know that /. has featured more than enough articles about this sort of thing. Slashdot doesn't 'hate copyright', it opposes copyright abuse, be it through patent stupidity or IP theft. It doesn't matter whether a piece of work is licensed at all: copyright still exists particularly for situations like this, and the right to assert copyright is the right to be protected from having your work used without attribution. The GPL isn't for photography: that's why Creative Commons exists, and CC does not preclude the right to attribution or indeed the right to make money out of intellectual property. Try and understand the issues before making yourself look stupid. It saves time.
Where the absence of Photoshop is one of the most common complaints about takeup, why not address the native solutions? Gimp needs improvements in its high end and workflow processes, so why not finance them? That applies to Google, or anyone who is interested. That's what open source is about. If your business profits from Gimp and you want it to be improved, donate money and time to the project.
I think as always Google have an interest in improving Wine, and it seems that Adobe are reticent about committing to a Linux/Unix port, perhaps over licensing issues, perhaps over the amount of code that needs to be changed (I would hazard that the Mac OS version of CS probably hasn't changed a great deal since OS9 days, with Carbonisation being the majority of the porting work), so Photoshop on Wine is a compromise that has benefits in that Google gets improvements to Wine and Adobe keeps selling Photoshop licences.
I use Pidgin on Windows and Linux and have just switched to Adium on OS X after using Mercury for a while, but getting frustrated with its closedness and the idiosyncrasies of its developer, and was surprised to see that there still isn't webcam support, which is somewhat remiss given the ubiquity of cams on laptops these days (Mercury does have it, and file transfer that works). This does strike me as a bit of a show-stopper at the moment, and it may well be that if Mozilla Messaging adopts libpurple as its IM library, that would be the place in which development should initially concentrate.
I'm old fashioned, and have always used a mail client rather than web mail where possible as I like having a rich editing environment. In recent years I have moved to IMAP mail wherever I can in order to be able to at least read my mail pretty much anywhere. Thunderbird (and Firefox)'s OS portability has made it possible to have a common mail interface on practically any computer, but what is still missing, as has been said elsewhere, is a common environment. This is where Gmail et al (but mostly GMail) has the advantage: it can be run practically anywhere that there is a web browser and Internet access. Now that Gmail has POP3 and IMAP access it can be integrated into Thunderbird. Similarly there is at least one application (for OS X at the moment - it's called Mailplane) that integrates Gmail with the desktop and hooks into iPhoto, iTunes, Growl and the other things that make OS X a cool working environment. A lot of these things are probably available as extensions for Thunderbird.
The gap is physical portability: for me, the ability to open Thunderbird on any machine and retrieve your environment, or indeed, have a Thunderbird-like experience (terrible phrase) through a web browser would be very valuable. Thunderbird has a common configuration format that can be read across a WAN all ready, and it seems to me that that could be extended across the Internet in a relatively trivial way, and could be applied to a web browser interface in the same way: there could also be an option for offline configuration storage on a USB key or similar. There would be a public network and the option to have private servers (free software of course, with paid support) that could either be internal or Internet facing. The aim would be for Thunderbird to become a genuinely portable communications environment that would provide an alternative to or even a complement for Exchange or Lotus Notes while using FOSS components.
Lots of superficial lookalikes, yes, but none that approach the spec of the real thing. I seem to have grown immune to the Reality Distortion Field and have no real interest in the iPhone, but with the exception of the HTC Touch, I haven't seen one that's anything but cosmetic. And if anyone mentions the Neo 1973, I'll screw their pelvis to a cake stand.
...and the ones which anyone who wants a decent connection would never consider. It's easy to find stories about Virgin and Carphone Warehouse's peak period bandwidth throttling policies and BT's capacity issues and general commoditisation of broadband. In-session advertising is an inevitable result of the business models that these companies have adopted (in Carphone Warehouse/Talk Talk's case, they offer a free service quite aggressively, and according to reports, you get what you pay for). BT is the most worrying as they have been acquiring smaller ISPs in the last couple of years, including a couple of the larger independents, Madasafish/Brightview and Plusnet, so may roll the policy out across their brands, or indeed may create a premium market with their other brands while BT Broadband becomes the ITV of the UK Internet.
There's an easy way to prevent it.
HITLER!
A real geek would never condescend to use .doc or .ppt. It would have to be .odt or .odp at the very least.
I've never been convinced by OS X as a server. Apart from in low admin environments where you can effectively set it up and walk away, it just isn't going to be useful. Most sysadmins will have their preferences for OS builds, apache builds, mail configs and so on, and applying Apple's Just Works principles to that level of management is going to lead to conflict. OS X is still good as a desktop (although I can see that my next refresh will be a decently specced small laptop running Ubuntu) but XServe would have been better with something more open.
That's actually a good way of doing it. There's an open source build server called Unattended that can provide a fully patched Windows build automatically - you provide the installer CD and it will pull down all the updates - and it recommends rebooting after each patch. That sounds like the same thing across a local network. I can recommend it as an alternative to ghost by the way: it will boot any PC that can be booted with PXE or bootp.
'Stop blogging, making videos and writing articles, and start fighting with legislation, with money, with burning tires and real 100,000 people marches'
'My Starcraft 2 blog [sc2blog.com] '
Gotta love it.
RMS doesn't drive. There are better reasons for this than the car's computer.
Weeeellll, no. You need transmitters to provide that whitespace in the first place, and they ain't there in the places that the OP is talking about. A place that immediately springs to mind is Pocohontas County in WV, which is a radio and TV deadzone because of the National Radio Quiet Zone. There is a single low powered AM radio station in the area and Verizon has a monopoly on cable TV, phones and broadband. It isn't going provide any advantage there.