_Might_? Is that the right word? I'm cynical about the legal system, but surely it boils down to how much money you have? I don't know if the FSF is rich enough to defend the term "Open Source" in a court room...
If you, along with everyone else, actually READ what you are talking about you will find that the customer (ie. calling party) is the person responsible for paying for the call to the ISP. Jim Anderton (as is usual) has got all his facts arse-about face.
I wasn't sure if Jim Anderton was right about this one or not; that's why I said "I'm not sure about this" in my original post. Keep your pants on:)
You're probably right in implying Jim Anderton may not be the world's most reliable/inpartial source of information.
I'm forced to disagree with some of your points...
If anyone cares to actually look at the FACTS regarding this case you will find that there are good reasons behind the Telecom move to charging for data to non IP-Net numbers.
Yes, good reasons for Telecom:) I think it is fairly obvious that the main reason they want to move to this system is so they can get around Kiwi Share. Data calls are costing them (both financially [if the ISP is not on Telecom's network due to interconnect agreements] and in opportunity costs); they can only recoup costs by making local data calls illegal/impossible.
1. Is is to reduced the load on VOICE circuits so that emergency services etc. can be easily reached. There have been recorded instances of difficulty contacting these services during peek Internet usage hours.
Those reported difficulties were due to software problems in the call centres; nothing to do with the voice exchanges. Exchanges will dump normal calls in favour of 111 calls.
2. When dialing into an IP-Net number the caller is on a voice circuit only as far as their exchange. From this point onwards they are on DATA QUALITY circuits thereby increasing their 'quality of connection' and dropping error rates.
That's no different from normal phone connections. How else would 56k connection speeds be possible?
3. It reduces some of the costs for the ISP as the 'POP' (dial in number ie. MODEM) is run by Telecom and tey are responsible for all upgrades etc. to the firmware of said POP. The circuit from the POP to the ISP is data quality and merely plugs in to their exisiting network.
My understanding is that this is just a routing issue... the modems, etc. remain with the ISPs.
4. A large number of ISPs have had no trouble with this concept. Paradise.net (once of the largest growing ISPs in NZ) has recently moved to offering their clients access via. IP-Net, this gives them the advantage that they now have Nationwide access and that the connections their clients make to the ISP are more reliable (reducing need for packet re-sends and thus their costs).
Paradise.net doesn't care because it isn't on Telecom's network; it is with Saturn. And saying that Paradise is one of the "largest growing" ISPs is misleading; they are growing quickly, but their coverage is quite limited and I doubt they even have 10,000 customers yet.
5. For the 'local' IP-Net option there is NO ADDITIONAL charge levied on the ISP and the caller also gets around the 2c per. minute charge that would otherwise be levied if calling on a normal voice number. IF an ISP chooses to take the 'nationwide' option they will pay an increased fee but this will give them the advantage that they will immediately have notionwide conerage, this increasing their customer base.
The original plan leaked by Jim Anderton late last year specified that Telecom would pass on the 2c charge from the consumer to the destination ISP. I'm not sure if this measure is in the proposal Telecom are mooting now.
Eventually, I foresee bidirectional wireless high-speed service offering ways around all the constant roadblocks to full-time access around the world.
Actually, Wellington NZ has already taken a few steps down this path. Some enterprising ISPs have opened up some non-traditional ways of accessing the web in Wellington (mainly to get around Telecom's _absurd_ ISDN pricing [$200 per month plus 4c per minute]):
(a) Microwave radio: Netlink offer a wireless service to anyone with line-of-sight to Wellington's University.
(b) Satellite: Ihug offer a service made up of down-link satellite and up-link modem called Starnet.
(c) Not wireless, but still interesting... CityLink is a 100 Mb/s fibre-optic circuit linking the entire Wellington central business district with about 7 or 8 ISPs, giving up to 100 Mb/s internet access (although the pipes out of NZ can't currently handle that much bandwidth...;)
But, isn't it about time that carriers and governments see that they should be moving towards better service to extract higher fees
I think part of the problem is that currently Telecom can't make (much) money on data calls. When Telecom was sold off to Ameritech a condition of the sale was that Telecom could never charge for local telephone calls (part of a sale agreement called the "Kiwi Share"). That was probably viewed as perfectly ok by Telecom at the time, because most local phone calls don't last too long and Telecom can still turn a profit by charging for long-distance and monthly line rental. But when you've got 400,000 internet users on the phone for several hours at a time, that's (a) local phone calls that Telecom can't charge for and (b) an _opportunity cost_ (in that, if the phone line wasn't being used for a local data call, then maybe somebody would be using it for a long-distance call, and thus paying Telecom money).
Telecom only make about ~$NZ1 billion profit each year (a large portion of which goes to overseas investors), so I suppose they feel they need to get a bit more blood out of the stone by applying this "net tax";)
And they are also not just trying to restrict access to the net, they are trying to get out of the long overdue upgrading needed.
Yes, Telecom are beginning to roll out ADSL services to Auckland and Wellington in June/July (Auckland and Wellington are the two largest cities in NZ, combined population ~1.5 million).
What I find funny is that Telecom's announcement specifically mentions that "Internet calls are now considered lower priority than voice - and calls to your ISP may be blocked if the exchange is overloaded". But if I understand ADSL correctly (and if I don't, please post a correction, I don't want to spread FUD!) ADSL goes via copper to the exchange too - if the exchanges can't handle modem calls, aren't they going to melt under ADSL?:)
I have to laugh at any country that tries to somehow restrict access to the 'net via taxes or censorship.
I don't think Telecom want to restrict net access; they just want their ISP, Xtra, to become the dominant ISP, and the easiest way of acheiving that is to force all the other ISPs out of the market via net taxes and predatory pricing. (If you read this you'll see that Telecom's ISP, Xtra, is exempt from the charges).
It's basically _way_ unfair competition... there are a number of issues like this in the NZ telecommunications industry at the moment, and a major problem is that the Minister of Communications and the Commerce Commission are completely ball-less and won't doing anything to keep Telecom in check.
Note that NT Server is purchased just as frequently as an application server (e.g. Exchange, IIS [yes, I know most people don't like it, but hey], SQL Server) as it is for file and print services.
The article said you'd need a pilot's license to fly one, but I don't get the impression you need to take off and/or land from an airport... if there are too many of these things whizzing around, aren't there going to be a lot of fatalities?:)
. I have to hand it to Microsoft, they did an ace job butchering the platform...although it didn't take much work - Java demonstrates that VMs are still a dumb idea in 1999.
I don't know that Microsoft really had to do very much to kill Java... it seems to me that one of Java's primary goals was to realise the "write once, run anywhere" dream. Since that never really worked the way it was supposed to, Java had/has a credibility problem.
It's a real shame - Java as a language has some way nice features. But until that primary goal of "write once, run anywhere" is realised, Java _on the client-side_ seems likely to remain limited mainly to applets...
Micros~1 understood the developers like nobody else. Linux needs to focus on administrators, not so much end users (IMHO) or _even_ developers.
I think it's possible to strongly argue that one of the reasons Win32 is dominant at the moment is precisely _because_ Microsoft has focused on its developers. By giving them powerful tools, "lowering the bar" with regard to quickly creating professional applications (e.g. VB), and generally "looking after" their developers, Microsoft has managed to win the battle for developer mindshare. Indeed, Microsoft don't hesistate to admit that keeping developers on board is a key part of their strategy.
The key is in administration of networks and web networks.
I agree that simple, accessible (i.e. graphical, as much as I hate to admit it;) and remotely-available administration tools are a compelling way of administering a network. But the _value_ of a network to a business is governed by more factors than just its administrability. The types of applications and functions the network supports is surely a key component of the network's value. Linux needs to offer compelling features and functions _in addition to_ simple yet powerful administration features before it will be ready to take on Microsoft and win on the server.
Looking back it's obvious how Microsoft managed to gain such a substantial foothold in the medium- to low-end server market: they owned the desktop. If Linux is to compete similarly (I'm not necessarily saying it _should_, but let's say it does), then offering compelling alternatives to Win32 applications on the desktop needs to be a priority. Already substantial impressive work has been done, but there is a ways to go yet...
So I think it is possible to make a strong argument that obtaining and nurturing developer mindshare is just as important as obtaining and nurturing administrator mindshare.
Paradox? If that's the Borland/Inprise DB to which you refer, I wasn't aware it was available for Linux... heck, it's hardly available for Windows anymore...
It also lacks transactions, a primary building block of enterprise database applications, and seems to have some severe performance problems with left and right joins.
When some idiot^Wmanager sends a document in a Microsoft format
What's "idiotic" about using the world's de facto document format? You may not like the fact that Microsoft Office is the de facto standard, but that doesn't mean it isn't so.
Cheers Alastair
Re:NT Resource Kit : GNU Inside
on
GNU Inside?
·
· Score: 1
There are other gnu utilities included in the resource kit; the source code is in there, even the GPL. The whole lot is in a "gnu" directory on the CD:)
Right from the "excuse the words -- (GASP!) OPEN SOURCE!".
If Ballmer was at a trade show or some other Microsoft-centric event, then it's possible his use of the phrase "excuse the words" was actually a gentle piss-take on Microsoft's non-OSS-centric development mindset;)
We weren't there, we don't know the exact context of his comments, so we probably shouldn't try to read _too_ far between the lines...
Windows 2000 has *how* many billion lines of code?
NT 4 has about 5, maybe 6 million lines. W2K is slated to to 30 million.
My guess is that they have reached a point where MS VC++ simply looks at the code, and says "no way, man"
Erm, I don't think it's likely to all be in one file, you know.
IMHO, Open Source is the only way for MS-Windows to survive in the long run.
Why?
Surely any product can survive as long as it's better than the competition? For most users, the (sad?) truth is that Windows is a better bet than Linux because it's more accessible and offers a wider range of applications... it's those two points that the OSS community need to address in order to make Linux more palatable on the desktop...
It would seem to be a purely commerical incentive... if they go OSS they basically lose a healthy portion of their revenue stream...
I would be interested to see if they consider realising the source to some of their older products (e.g. DOS, Windows 3.1). If they released these products as is into the community it could spark some interest in OSS in general, and at the same time Microsoft wouldn't really be giving away too many secrets.
I'm sorry -- but no one wants to dig through any Microsoft code, other than to laugh at some of the awful programming techniques.
I'm not sure it's safe to jump to that conclusion... some of the tricks in Microsoft's OSes and applications are quite funky, and the code behind 'em might actually teach us a thing or two...
Slashdot Mirror
the combined might of the FSF and, well, everyone
_Might_? Is that the right word? I'm cynical about the legal system, but surely it boils down to how much money you have? I don't know if the FSF is rich enough to defend the term "Open Source" in a court room...
Cheers
Alastair
Yeah. WRT net access: they link the schools up for free, then hit them monthly for business charges.
Uptime of greater then 40 days is impossiable with a MS OS 95-NT-?2000?. There is a counte overflow that crashes the system.
That bug does not affect Windows NT. It will not affect Windows 2000.
Our NT and 3.1 systems were alway having software problems or OS coruption for no apparent reson!
Maybe they weren't being administered correctly. A well-maintained NT box can stay up for months without difficulty.
Cheers
Alastair
Oops, GST, I forgot.
:)
It gets funnier... NZ has GST too
Cheers
Alastair
If you, along with everyone else, actually READ what you are talking about you will find that the customer (ie. calling party) is the person responsible for paying for the call to the ISP. Jim Anderton (as is usual) has got all his facts arse-about face.
:)
I wasn't sure if Jim Anderton was right about this one or not; that's why I said "I'm not sure about this" in my original post. Keep your pants on
You're probably right in implying Jim Anderton may not be the world's most reliable/inpartial source of information.
Are you condoning Telecom's actions in your post?
Cheers
Alastair
I'm forced to disagree with some of your points...
:) I think it is fairly obvious that the main reason they want to move to this system is so they can get around Kiwi Share. Data calls are costing them (both financially [if the ISP is not on Telecom's network due to interconnect agreements] and in opportunity costs); they can only recoup costs by making local data calls illegal/impossible.
If anyone cares to actually look at the FACTS regarding this case you will find that there are good reasons behind the Telecom move to charging for data to non IP-Net numbers.
Yes, good reasons for Telecom
1. Is is to reduced the load on VOICE circuits so that emergency services etc. can be easily reached. There have been recorded instances of difficulty contacting these services during peek Internet usage hours.
Those reported difficulties were due to software problems in the call centres; nothing to do with the voice exchanges. Exchanges will dump normal calls in favour of 111 calls.
2. When dialing into an IP-Net number the caller is on a voice circuit only as far as their exchange. From this point onwards they are on DATA QUALITY circuits thereby increasing their 'quality of connection' and dropping error rates.
That's no different from normal phone connections. How else would 56k connection speeds be possible?
3. It reduces some of the costs for the ISP as the 'POP' (dial in number ie. MODEM) is run by Telecom and tey are responsible for all upgrades etc. to
the firmware of said POP. The circuit from the POP to the ISP is data quality and merely plugs in to their exisiting network.
My understanding is that this is just a routing issue... the modems, etc. remain with the ISPs.
4. A large number of ISPs have had no trouble with this concept. Paradise.net (once of the largest growing ISPs in NZ) has recently moved to
offering their clients access via. IP-Net, this gives them the advantage that they now have Nationwide access and that the connections their clients make to the ISP are more reliable (reducing need for packet re-sends and thus their costs).
Paradise.net doesn't care because it isn't on Telecom's network; it is with Saturn. And saying that Paradise is one of the "largest growing" ISPs is misleading; they are growing quickly, but their coverage is quite limited and I doubt they even have 10,000 customers yet.
5. For the 'local' IP-Net option there is NO ADDITIONAL charge levied on the ISP and the caller also gets around the 2c per. minute charge that would otherwise be levied if calling on a normal voice number. IF an ISP chooses to take the 'nationwide' option they will pay an increased fee but this will give them the advantage that they will immediately have notionwide conerage, this increasing their customer base.
The original plan leaked by Jim Anderton late last year specified that Telecom would pass on the 2c charge from the consumer to the destination ISP. I'm not sure if this measure is in the proposal Telecom are mooting now.
Cheers
Alastair
Eventually, I foresee bidirectional wireless high-speed service offering ways around all the constant roadblocks to full-time access around the world.
;)
Actually, Wellington NZ has already taken a few steps down this path. Some enterprising ISPs have opened up some non-traditional ways of accessing the web in Wellington (mainly to get around Telecom's _absurd_ ISDN pricing [$200 per month plus 4c per minute]):
(a) Microwave radio: Netlink offer a wireless service to anyone with line-of-sight to Wellington's University.
(b) Satellite: Ihug offer a service made up of down-link satellite and up-link modem called Starnet.
(c) Not wireless, but still interesting... CityLink is a 100 Mb/s fibre-optic circuit linking the entire Wellington central business district with about 7 or 8 ISPs, giving up to 100 Mb/s internet access (although the pipes out of NZ can't currently handle that much bandwidth...
Cheers
Alastair
But, isn't it about time that carriers and governments see that they should be moving towards better service to extract higher fees
;)
I think part of the problem is that currently Telecom can't make (much) money on data calls. When Telecom was sold off to Ameritech a condition of the sale was that Telecom could never charge for local telephone calls (part of a sale agreement called the "Kiwi Share"). That was probably viewed as perfectly ok by Telecom at the time, because most local phone calls don't last too long and Telecom can still turn a profit by charging for long-distance and monthly line rental. But when you've got 400,000 internet users on the phone for several hours at a time, that's (a) local phone calls that Telecom can't charge for and (b) an _opportunity cost_ (in that, if the phone line wasn't being used for a local data call, then maybe somebody would be using it for a long-distance call, and thus paying Telecom money).
Telecom only make about ~$NZ1 billion profit each year (a large portion of which goes to overseas investors), so I suppose they feel they need to get a bit more blood out of the stone by applying this "net tax"
Cheers
Alastair
And they are also not just trying to restrict access to the net, they are trying to get out of the long overdue upgrading needed.
:)
Yes, Telecom are beginning to roll out ADSL services to Auckland and Wellington in June/July (Auckland and Wellington are the two largest cities in NZ, combined population ~1.5 million).
What I find funny is that Telecom's announcement specifically mentions that "Internet calls are now considered lower priority than voice - and calls to your ISP may be blocked if the exchange is overloaded". But if I understand ADSL correctly (and if I don't, please post a correction, I don't want to spread FUD!) ADSL goes via copper to the exchange too - if the exchanges can't handle modem calls, aren't they going to melt under ADSL?
Cheers
Alastair
I have to laugh at any country that tries to somehow restrict access to the 'net via taxes or censorship.
I don't think Telecom want to restrict net access; they just want their ISP, Xtra, to become the dominant ISP, and the easiest way of acheiving that is to force all the other ISPs out of the market via net taxes and predatory pricing. (If you read this you'll see that Telecom's ISP, Xtra, is exempt from the charges).
It's basically _way_ unfair competition... there are a number of issues like this in the NZ telecommunications industry at the moment, and a major problem is that the Minister of Communications and the Commerce Commission are completely ball-less and won't doing anything to keep Telecom in check.
Cheers,
Alastair
(Getting a touch off-topic, but I can't let that FUD go unchallenged... ;)
:)
Um, NT didn't crash on the Yorktown. The app it was running did. The OS kept on going (doing absolutely nothing useful, admittedly...
Cheers
Alastair
Note that NT Server is purchased just as frequently as an application server (e.g. Exchange, IIS [yes, I know most people don't like it, but hey], SQL Server) as it is for file and print services.
Cheers
Alastair
NT's maximum file size = NT's maximum partition size = 16 exobytes. (1 exobyte = 1000 terabytes, I believe... can anybody confirm?)
;)
So not a big problem for a while, hopefully
Cheers
Alastair
... that stay on the ground.
:)
The article said you'd need a pilot's license to fly one, but I don't get the impression you need to take off and/or land from an airport... if there are too many of these things whizzing around, aren't there going to be a lot of fatalities?
Cheers
Alastair
. I have to hand it to Microsoft, they did an ace job butchering the platform...although it didn't take much work - Java demonstrates that VMs are still a dumb idea in 1999.
I don't know that Microsoft really had to do very much to kill Java... it seems to me that one of Java's primary goals was to realise the "write once, run anywhere" dream. Since that never really worked the way it was supposed to, Java had/has a credibility problem.
It's a real shame - Java as a language has some way nice features. But until that primary goal of "write once, run anywhere" is realised, Java _on the client-side_ seems likely to remain limited mainly to applets...
Cheers
Alastair
Good post. I have a couple of comments...
;) and remotely-available administration tools are a compelling way of administering a network. But the _value_ of a network to a business is governed by more factors than just its administrability. The types of applications and functions the network supports is surely a key component of the network's value. Linux needs to offer compelling features and functions _in addition to_ simple yet powerful administration features before it will be ready to take on Microsoft and win on the server.
Micros~1 understood the developers like nobody else. Linux needs to focus on administrators, not so much end users (IMHO) or _even_ developers.
I think it's possible to strongly argue that one of the reasons Win32 is dominant at the moment is precisely _because_ Microsoft has focused on its developers. By giving them powerful tools, "lowering the bar" with regard to quickly creating professional applications (e.g. VB), and generally "looking after" their developers, Microsoft has managed to win the battle for developer mindshare. Indeed, Microsoft don't hesistate to admit that keeping developers on board is a key part of their strategy.
The key is in administration of networks and web networks.
I agree that simple, accessible (i.e. graphical, as much as I hate to admit it
Looking back it's obvious how Microsoft managed to gain such a substantial foothold in the medium- to low-end server market: they owned the desktop. If Linux is to compete similarly (I'm not necessarily saying it _should_, but let's say it does), then offering compelling alternatives to Win32 applications on the desktop needs to be a priority. Already substantial impressive work has been done, but there is a ways to go yet...
So I think it is possible to make a strong argument that obtaining and nurturing developer mindshare is just as important as obtaining and nurturing administrator mindshare.
Cheers
Alastair
Borlands DB (forget the name)
Paradox? If that's the Borland/Inprise DB to which you refer, I wasn't aware it was available for Linux... heck, it's hardly available for Windows anymore...
Cheers
Alastair
It also lacks transactions, a primary building block of enterprise database applications, and seems to have some severe performance problems with left and right joins.
Cheers
Alastair
When some idiot^Wmanager sends a document in a Microsoft format
What's "idiotic" about using the world's de facto document format? You may not like the fact that Microsoft Office is the de facto standard, but that doesn't mean it isn't so.
Cheers
Alastair
There are other gnu utilities included in the resource kit; the source code is in there, even the GPL. The whole lot is in a "gnu" directory on the CD :)
Cheers
Alastair
Right from the "excuse the words -- (GASP!) OPEN SOURCE!".
;)
If Ballmer was at a trade show or some other Microsoft-centric event, then it's possible his use of the phrase "excuse the words" was actually a gentle piss-take on Microsoft's non-OSS-centric development mindset
We weren't there, we don't know the exact context of his comments, so we probably shouldn't try to read _too_ far between the lines...
Cheers
Alastair
Windows 2000 has *how* many billion lines of code?
;)
NT 4 has about 5, maybe 6 million lines. W2K is slated to to 30 million.
My guess is that they have reached a point where MS VC++ simply looks at the code, and says "no way, man"
Erm, I don't think it's likely to all be in one file, you know.
IMHO, Open Source is the only way for MS-Windows to survive in the long run.
Why?
Surely any product can survive as long as it's better than the competition? For most users, the (sad?) truth is that Windows is a better bet than Linux because it's more accessible and offers a wider range of applications... it's those two points that the OSS community need to address in order to make Linux more palatable on the desktop...
(As always, my _very_ humble opinion
Cheers
Alastair
Well, since most of Windows 9x is written in Ada it wouldn't matter much.
Why would the fact that 95 is written in Ada (it isn't, but let's not let the facts blind us here) reduce its worth in the eyes of the OSS community?
And NT is mostly C, but has hundreds of millions lines of code (250 million I believe maybe it was 350).
NT 4 is about 5, maybe 6 million lines of code. W2K is slated to top 30 million.
Cheers
Alastair
It would seem to be a purely commerical incentive... if they go OSS they basically lose a healthy portion of their revenue stream...
I would be interested to see if they consider realising the source to some of their older products (e.g. DOS, Windows 3.1). If they released these products as is into the community it could spark some interest in OSS in general, and at the same time Microsoft wouldn't really be giving away too many secrets.
Cheers
Alastair
I'm sorry -- but no one wants to dig through any Microsoft code, other than to laugh at some of the awful programming techniques.
I'm not sure it's safe to jump to that conclusion... some of the tricks in Microsoft's OSes and applications are quite funky, and the code behind 'em might actually teach us a thing or two...
Cheers
Alastair