Spoken like a true bellhead, You are missing the point. VoIPs network is free (as in freedom) IP based, as in dirt cheap, if I pee it will land on an equipment vendor. Not stuck on a few media. Ever run SS-7 over cable-TV networks? How about a metropolitain gigabit ethernet lan? Youre going to say SONET, well gee, that only costs 10x of a gigabit ethernet... where can I get SONET termination? can I run packets on that? oh.. need to encapsulate it in IP... hmm... why?
How much is that PBX in the window? ok, so Id like an SS-7 switching network, and I aint a phone company, oh? cant have one? have to run my own wires? hmm...
Separating control from data only makes sense if the network is smart. Smart networks only make sense if the manager of the network is your friend. Usually, that is not the case for anyone except the phone company. The whole point of IP is to make the intermediate network a non-issue. make it stupid so that there isnt any value there, and it can be replaced by any number of technologies or providers. That is always going to be cheaper for end users, but not the phone company.
Problem: lame, underprovisioned ISP. You can dial-in at 56 kbps, but your downloads always go at 14kbps because hes got a million clients on a single T1.
Solution: Dial your ISP, Use VoIP to open a connection with a demarkation point somewhere on the net (maybe very cheap because you can do the demarkation in software.) now you have guaranteed 56 kbps line coming out of your ISP (if the lame ISP claims to support VoIP, then this has to work.)
OK, maybe DOA for minidisk is a bit harsh, but in the long run, any single vendor standard is going to fail. The only reason vendors put out a proprietary standard is because they have some technical advantage that they think is killer. Eventually (within 5 years, worst case) the market catches up with them and the format dies. If it pays off within the five years, then they made a sound business decision, but for the clients, who will be stuck with a medium that they cannot read after some lapse of time, it is probably a bad idea.
I didnt want to pick on Sony... I like their stuff. I have a digital camera with a memory stick in it, but it is clear, with their new gear, they are dropping their proprietary media for compact flash & friends. MiniDisk had a good run, but it is good as dead now, I dont believe there is a followon format, so in a few years the old media will be unreadable. OK, heres another vendor example: How many folks have been thrilled with their IOmega Zip drives in the past few years... how many people can still read them !? so CF, normal CD-RW & DVD, usb key drives, these are all relatively open versions of things that were formerly the realm of vendor specific media.
formats with broad industry support will always smother single vendor ones, given time. So when a single vendor format comes out, it is pretty much guaranteed to be DOA, but folks might not know it yet (because they fall for its nifty special features.)
People to be polite, and not mislabel traffic for their own advantage. What happens when some bit-torrent users figure out how to double their download speeds by setting the QoS bits on their traffic.
People to agree on priorities... if there is a late breaking virus, maybe it is more important to download the patch than for fifteen teenagers to share a rave in quadrophonic sound.
People to be reasonable... I dont care if you do drive a mercedes 500, all your traffic is NOT high priority. (that is, can people buy QoS? If you have QoS then the whole billing question becomes very interesting, and the price of the data will shoot back up to voice network levels, because every intervening hop will, quite reasonably, want their cut.
QoS is a DOA technology on the Internet. The technology makes a lot of sense on corporate networks, where there is somebody in charge, but in the wide world, it just is fundamentally not going to happen because the interested parties have no incentive to make it work.
IP telephony will happen because the bandwidth will rise to the point that voice traffic becomes noise to everyone but the last mile. The last mile will have to take care of their own problems (perhaps using a cheapo version of QoS, such as preferring packets on a certain port, but it will not require any action of the network.)
oh... folks were complaining about acronyms, so.. DOA -- Dead On Arrival, the status of unfortunate patients on reception in the Emergency ward of a hospital. Also applies to technologies, ie. MiniDisc, MemoryStick, (oh.. stop picking on Sony...) DAT, Video Disk, (technologies that arrived and died without garnering much market share.)
film the local content, and make it available without DRM on the net. Doesn't cost the performers a thing, since they would not have gotten a distribution deal anyways, and gives them exposure to boot. You get your time shifting, the world gets a far richer entertainment menu, big media loses ratings/eyeballs/marketshare/money.
I agree with your exceptions. The original poster's proposition was: "the public paid for it, so it should be free." The long list of justifiable exceptions only re-enforces how hollow the basic argument is. The correct. if unsatisfying, statement is: If the public paid for it, the government (as in those bums you put into office) must do the "right" thing, taking into consideration, all the conflicting interests, the amount of funding that people were able to justify throwing at any given project, and make some choices. reasonable people will disagree on what the "right" thing is.
There are posts claiming "we didn't mean real-time." hmm... Show me where there is an exclusion from the FOI act for real-time data. If aerial photography is available, then why not have a site with a town's video surveillance data in real-time, much like the traffic cameras that are common around most cities. It's video of public places, anybody can watch anyone else on the street, so there should be no expectation of privacy. I would love to have such a thing for my daycare, and have a look at what my children are doing when I have a break at work without disturbing anyone.
I don't know whether this information should be public or not, but to blithely claim that unless national security or citizen's privacy is at stake, all information should be public is, well, rash.
Completely agree that people need to know, and that the information has to get out there, and that it has to get out there ASAP. The only question is, how soon is ASAP, and how much are you willing to pay to answer the requirement. "The shipment of nuclear warheads will be on I-95 mile 35 at 2:15 pm." Many people will want to be out of the area when that shipment goes past, given a choice. Others will might want to excercise their free speech rights to picket it, still others might elect to put a rocket through the truck cab, and walk off with the warheads. Think of the security you need to put in place if you broadcast all such movements to everyone. It is massively cheaper to simply not tell people what is being done until after it has occurred.
Free speech / free access to information is about being able to review government actions and hold it responsible. but we elect folks once every few years. That audit process is not a real-time one, and has to be based on a balance of facts.
"required" in practice doesn't mean much. "funded" means a lot more. and keep in mind that this type of tech is fairly new. If you just throw money at it, you can easily end up with a billion dollar "solution" that everyone will be pleased to finger as "your tax dollars at work." Risk exists in all projects. If it happens in private enterprise, the company goes under or writes off a loss. In the government, it is a scandal. Putting a big solution in place, without building from previous experience, is asking for trouble. These answers are hard, and take time. so the government is going to move slowly where there is such a perceived risk.
ok... well two years ago, it was worth 66 cents, now we are at around 75 US cents. so it has gone up about 18%. thirty years ago, as any beer drinking Canadian (closest analogy to "red-blooded American" I could think of:-) will tell you, the Canadian dollar was as high as $1.10 US. so the national lament goes... personally, it's a load of bull.
as for making goods cheaper... hmm... if they are natural resources, those are all costed in US$ anyways, makes no difference. if it manufactured goods, then most foreign components are going to be purchased in US$. so won't make much difference either.
xchange rates are just trade friction. when rates change, prices slowly adjust to reflect the new cost structure. There is not really a long term benefit. the argument would make sense if high value items were manufactured directly from Canadian natural resources. I don't think that is too common a case.
my guess is that costs in Canada are lower because there is a public health care system, which controls costs better than the american system, and many other sorts of organizations, like workman's compensation which reduce liabilities, so that insurance costs are lower across the board. The un-employment insurance programs reduce social diparity and unrest, and make the country cheaper to police, again reducing costs. That corporations use the same programs to smooth over low-demand periods by having workers on those programs then, and available when demand picks up. So they don't have to spend as much on hiring, since the skilled people remain in the industry through the dips. Again, this reduces costs for industry.
You sure about that logic? some other data collection paid for by public money:
Social insurance databases
Driver's license db's
all police investigations, regardless of whether charges are laid.
medicare payment treatment and payment records
nuclear missile plans.
the approved architectural plans for that nice, bombable Hoover Dam.
tax records of all sorts
how the governement recognises you, as opposed to someone pretending to be you, and gives you access to your own information...
military supply orders and troop movements.
Basing the argument on the government having paid for the collection is a iffy at best. The basis should rather be based on maximizing the public good,which is, in the general case, harder to figure out. One has to weigh: privacy concerns vs. defence (against Terrorists domestic and foreign) vs. public benefit. The answer will come out different depending on what the data is, what technology is in place/reasonable, and how much the organisation is willing to spend to make the information public. How soon to make it public is also going to have a big effect on how much it costs. folks on the internet want information upto the second.
You have a chemical spill in Seattle. You have a real-time information system for exchange among first responders who are doing their work. It hits the news and their site gets slashdotted. It's a dynamically built site, so caching by google is of no use whatever. The firemen and coast guard can no longer get information from aerial reconnaisance being done by a Canadian survey plane that happenned to be available. So they don't know where in the harbour the spill has gotten to.
Wall it off? OK, you need a separate network accessible by city, province, state, and many branches of two national governments, as well transportation (railways, airlines) in the area, and any specialized contractors that might be called in. And it has to be setup ahead of time, and managed and funded so that it is up when a crisis happens.
What is the cost of making that site public?
Does the public need to know where there is a chemical spill? Of course they do! Should they get same information the government does on their first responder systems? Would be nice, but if the architecture/technology in place cannot answering that sort of demand, what do you do? Most people would accept as reasonable that you have a first responder system that is only available to a few, then have other systems which are used for public dissemination (aka. press conferences, other web sites, etc...)
Sure, each city goes out and buys ArcView or whatever, and they have a heck of a time doing anything cheaply with it, but check out:
http://www.atlas.gc.ca
This is built on Chameleon, a GPL frontend for the GPL UMN mapserver whose development were partially funded by Canadian and American governments, respectively, for purely selfish reasons (reducing the costs of producing GIS servers, and being able to provide more information to more groups more cheaply.)
The biggest selling point of Debian is that it is not controlled by a single entity with a single agenda. It will not have it's pricing policy change, by random fiat of venture capitalists. Nor will it make inexplicably odd decisions, such as the de-integration/massacre of KDE in redhat 8 (a.k.a. bluecurve). Debian does not do religion it does not attempt to reduce choice, but only supports the choices your organization makes. It just asks as a tool box, and lets people install what they want, easily, and without fuss.
As someone in an organization which has several hundred installed servers, we looked things over in the past year, and are choosing Debian as the next platform for us. Debian stable is the best choice in organizations that can control their application environment. We are lucky, in that most of our apps are in-house, and we have reasonable corporate memory and support in place.
The main attraction of redhat used to be that their software was newer than debian, and that their installation was easier. if you assume a reasonable level of expertise, the installation is a non-issue. Since we were standardized on redhat 7.x, we were having to backstitch things to just install the latest hardware, so the latest and greatest was not helping us at all.
We started to look longingly at Debian stable... 3 years since the last release, only updates in that time... free... decent kde via backports...
Remember what redhat did last year? Any clue how many corporations were running thousands of instances of free redhat 7.x, getting only patches from RH, who were suddenly SOL as of last December? Did you perchance notice a passing similarity between the plans for RHAS (5 year life cycle) and Debian stable (3 years so far)
apt-get wonderful not because of the ease of use of apt-get itself (which is wonderful nonetheless) but because there are tens of thousands of packages which are in the repositories, ready to go, far more packages than are available from any combination of dependency-hell + freshrpms.net + google. "redhat xx rpm" and far more simply installable.
Debian stable is what Redhat enterprise can only hope to become, but will never be, because they have priced themselves, slowly but surely, out of the market. Three things have happenned in the past year or two which will fundamentally alter things:
Debian sarge release with a new installer real soon now, so the stable version will not look so ancient, making it more attractive, and supporting 2.6 as an option (which redhat still doesn't), and with an installer that should reduce the major pain of the current installer down to minor grumbles.
KNOPPIX has made debian unstable for easy enough for anybody to try out and even install, so testdrives show that it ain't so bad
redhat's radical changes: discontinue free, re-continue as fedora, free as in beer, but controlled as in messed up KDE distribution has successfully rattled enough cages that folks are looking elsewhere and finding that debian "just works."
It is just the right answer. All that said, if your shop is committed to binary-ware, and you favourite bit-vendor won't support your chosen environment, you are toast. Do not go there.
Talk, cajole, encourage, convince, or switch vendor or plain drop the binary ware if you can afford to do it, but do not use the commercial software on an unsupported platform. That is the worst of both worlds: The free people don't use your package X so they can't help you, and the paid people go through their menus and hit "we don't support that, click!"
Forget cruise missiles. Remotely Piloted vehicles (RPV) are a field with tremendous potential. With an RPV, you do not need overly complex guidance systems, because you have a human pilot sitting on the ground. You save costs by building the aircraft without the safety factors needed for a vehicle that will carry people. If this guy could help a company build these things for less $, I am sure he would pay for himself rapidly.
Sample weaponless applications:
drones for target practice by the military (sure, all militaries that need them have them already, but how much do they cost?)
drones for lighting up enemy radar systems (see the last Israeli/Syrian war)
cargo aircraft for parcel delivery.
Used as Remotely piloted vehicles. As long as they're small enough, even if it crashes the damage done is likely to be minimal, and you do not have to worry about losing a pilot. How about delivering mail in the high artic or ant-arctiv.
If you can build an RPV with enough range, you could reduce costs of supply to those places quit a bit, and get it to them far more p often. think about it? How do you get fresh vegetables to an antarctic base, or island airport in the pacific ? Use a piloted plane, you want to have a large amount of cargo to offset the costs. Smaller bases in remote areas are supplied more rarely today because of that.
With an RPV, you could ferry medecines and samples back and forth far more frequently. who cares if it is a 20 hour flight, after 8 hours, the relief pilot comes in, after 16, the night shift guy comes in. They can all live in their home town, and kiss their children at the end of the work day, even if their plan goes down in flames into the pacific ocean.
delivery of small amounts of whatever to whomever via air. the longer the range of the thing, the more flexible it becomes.
while satellite tech is used a lot for surveillance, a good camera on a cruise missile could provide higher res imagery of an area of interest (hence the huge US RPV program.)
My guess is that the main thing holding this field back is telecommunications. You want to use something like IRIDIUM, a cell phone like network with global coverage (last I heard they were still in the air), to relay signals from the planes to the pilot bases. That way you don't have to worry about having transmitters near the field of operations, which reduces the cost of operations again. You need enough bandwidth to provide a live video feed of some quality, but the better the automated systems the less quality you need (aside from the reconaissance applications.) I don't know if any of the civilian satellite constellations in orbit have sufficient bandwidth for that, but that's probably the biggest cost, and biggest barrier to this market exploding.
(P.S. once this is working, you can also outsource cargo piloting to India... )
Do it in reserved time for the first couple of runs. Have them involved in evaluating the impact of the scans, show them the results. After it is done a few times with no impact, theyll get bored^h^h^h^h^h comfortable with scans and you may be able to schedule them during the day.
They should be scheduled, so that if something does go wrong they can at least ask you to scan again and reproduce the problem, or eliminate your work from the list of suspects. If you do it without telling them and it causes problems that they spend hours or days figuring out, you can bet you will be confined to 3am forever thereafter.
Folks who run real sites want the hits, but when slashdot (or Google) points at a hobby site, they have a responsibility to set up their own site as a caching proxy, and link the proxy, so as not to kill the target. After the rush, switch to a re-direct and it will sort itself out over time.
Of course, that would cost slashdot big bucks, and slow down slashdot... on second throught lets nuke those hobby sites real good.
might be what you are looking for. http://www.gnuenterprise.org
Unfortunately, it doesn't look like enough of it is done yet to be useful. They are working on frameworks on top of which all the applications can be built, but nothing "ready to go" is available yet.
we should help SCO & MS any way we can
on
SCO Offline
·
· Score: 1
Law and order are more important than any disagreement over copyright. People who willfully violate copyrights by circulating warez & Mp3's against the copyright owners' express permissions, spammers, and crackers creating malware, none of them are of any use to the open source community.
Open source is about respect for copyright, our choice of the terms of the copyright (ie. GPL) is the disagreement with those companies, and we have an legal disagreement about some ownership issues. That does not give anyone the right to behave in a criminal fashion.
To be fair, the script I started with was around 800 lines. Then I took out 90% of the comments & examples that were irrelevant. There is one script in shorewall which has some meat, for the rest of the directory the config files are 80% comments. So I took it out, and repeated the analysis. that is: wc -l * --> 1108 lines. grep -v '^#' * | wc -l --> 217
The comments are so huge that it is impossible to understand the actual configuration. That's my real gripe: ease of auditing is an important characteristic to me, and that is generally poorly supported in GUI's.
I've got DSL, two interfaces, a web, email, ssh, dns services and a squid & dansguardian running on the router machine doing NAT for my home LAN. It should be straight-forward. No IP-SEC, no plug-to's, no proxies. Nothing I've seen does it right, nor are they any simpler to understand than just plowing through the HOWTO's, and futzing with a sample from there. One really needs to understand all of a firewall configuration. These GUI's try to shield you, but it doesn't work, and the configs are really complicated to understand. This is just a default configuration:
/etc/shorewall# wc -l *...
4051 total basquette:/etc/shorewall# wc -l/etc/init.d/rc.firewall
145/etc/init.d/rc.firewall
I can understand 145 lines. 4051 lines is a pain to audit. Don't tell me I should audit by pointing and clicking, that is too painful for words.
Yesterday's posting described issues with current AMD 64-bit linux distros. Can one just use a 32 bit one for now, and wait a while for the 64 bit ones to mature?
If not, it doesn't sound reasonable (as in, what?!!? Getting X to work is a challenge?)
Free Software is about not stealing commercial code. A significant proportion of linux users use it because it is legal to copy for free, and the legal part is important to them. There is a significant proportion of windows users who have cracked OS or application licenses. It isn't as if cracked software is not easily available. For those people, there is no advantage to the freedom of Linux.
So people who use linux value the copyright which gives them free software and so are more likely to respect copyright in general, and pay up.
It's pretty impressive. 2 megawatts and 3000 sq. ft. is quite good.
frightning, but given the power, quite good. Imagine the UPS.
on the other hand... "100 hour MTBI is desirable" Ack! It is hoped that it won't crash more than once every three days? That is up from 40 hours on the current one. oh. They're putting in lots of RAS features, and they still can't target higher than that. depressing.
custom interconnect. That is the exciting part. It looks like a lot of fun. The directions are good and make sense. connectionless api (mcast & bcast only?) is going to take some getting used to...
"Restrictions apply to all telemarketers, although they may differ depending on whether they use a fax or a telephone. As a minimum, telemarketers must maintain "Do not call/fax lists" and provide customers with a fax or telephone number where a responsible person can be reached. Specific rules are included at the end of this document."
This is the best part:
"What are the consequences if telemarketers don't follow the rules?
Telephone companies can notify these telemarketers that telephone service to the lines used in connection with placing calls (telephone or fax) may be suspended or disconnected within two business days."
um... phone company may cut off their most lucrative paying customers... sure...
I'm sure their quivering in their boots now...
Not only do you have to chase the buggers,
then you have to pester the phone company,
and then they just change the company name
and keep going. Not even a fine.
Canadians are so... wimpy^H^H^H^H^H understanding and sympathetic.
Slashdot Mirror
How much is that PBX in the window? ok, so Id like an SS-7 switching network, and I aint a phone company, oh? cant have one? have to run my own wires? hmm...
Separating control from data only makes sense if the network is smart. Smart networks only make sense if the manager of the network is your friend. Usually, that is not the case for anyone except the phone company. The whole point of IP is to make the intermediate network a non-issue. make it stupid so that there isnt any value there, and it can be replaced by any number of technologies or providers. That is always going to be cheaper for end users, but not the phone company.
backgrounders:
Problem: lame, underprovisioned ISP. You can dial-in at 56 kbps, but your downloads always go at 14kbps because hes got a million clients on a single T1.
Solution: Dial your ISP, Use VoIP to open a connection with a demarkation point somewhere on the net (maybe very cheap because you can do the demarkation in software.) now you have guaranteed 56 kbps line coming out of your ISP (if the lame ISP claims to support VoIP, then this has to work.)
uh... why wont everyone do that?
I didnt want to pick on Sony... I like their stuff. I have a digital camera with a memory stick in it, but it is clear, with their new gear, they are dropping their proprietary media for compact flash & friends. MiniDisk had a good run, but it is good as dead now, I dont believe there is a followon format, so in a few years the old media will be unreadable. OK, heres another vendor example: How many folks have been thrilled with their IOmega Zip drives in the past few years... how many people can still read them !? so CF, normal CD-RW & DVD, usb key drives, these are all relatively open versions of things that were formerly the realm of vendor specific media.
formats with broad industry support will always smother single vendor ones, given time. So when a single vendor format comes out, it is pretty much guaranteed to be DOA, but folks might not know it yet (because they fall for its nifty special features.)
- People to be polite, and not mislabel traffic for their own advantage. What happens when some bit-torrent users figure out how to double their download speeds by setting the QoS bits on their traffic.
- People to agree on priorities... if there is a late breaking virus, maybe it is more important to download the patch than for fifteen teenagers to share a rave in quadrophonic sound.
- People to be reasonable... I dont care if you do drive a mercedes 500, all your traffic is NOT high priority. (that is, can people buy QoS? If you have QoS then the whole billing question becomes very interesting, and the price of the data will shoot back up to voice network levels, because every intervening hop will, quite reasonably, want their cut.
QoS is a DOA technology on the Internet. The technology makes a lot of sense on corporate networks, where there is somebody in charge, but in the wide world, it just is fundamentally not going to happen because the interested parties have no incentive to make it work.IP telephony will happen because the bandwidth will rise to the point that voice traffic becomes noise to everyone but the last mile. The last mile will have to take care of their own problems (perhaps using a cheapo version of QoS, such as preferring packets on a certain port, but it will not require any action of the network.)
oh... folks were complaining about acronyms, so.. DOA -- Dead On Arrival, the status of unfortunate patients on reception in the Emergency ward of a hospital. Also applies to technologies, ie. MiniDisc, MemoryStick, (oh.. stop picking on Sony...) DAT, Video Disk, (technologies that arrived and died without garnering much market share.)
film the local content, and make it available without DRM on the net. Doesn't cost the performers a thing, since they would not have gotten a distribution deal anyways, and gives them exposure to boot. You get your time shifting, the world gets a far richer entertainment menu, big media loses ratings/eyeballs/marketshare/money.
There are posts claiming "we didn't mean real-time." hmm... Show me where there is an exclusion from the FOI act for real-time data. If aerial photography is available, then why not have a site with a town's video surveillance data in real-time, much like the traffic cameras that are common around most cities. It's video of public places, anybody can watch anyone else on the street, so there should be no expectation of privacy. I would love to have such a thing for my daycare, and have a look at what my children are doing when I have a break at work without disturbing anyone.
I don't know whether this information should be public or not, but to blithely claim that unless national security or citizen's privacy is at stake, all information should be public is, well, rash.
Think of the security you need to put in place if you broadcast all such movements to everyone.
It is massively cheaper to simply not tell people what is being done until after it has occurred.
Free speech / free access to information is about being able to review government actions and hold it responsible. but we elect folks once every few years. That audit process is not a real-time one, and has to be based on a balance of facts.
"required" in practice doesn't mean much. "funded" means a lot more. and keep in mind that this type of tech is fairly new. If you just throw money at it, you can easily end up with a billion dollar "solution" that everyone will be pleased to finger as "your tax dollars at work."
Risk exists in all projects. If it happens in private enterprise, the company goes under or writes off a loss. In the government, it is a scandal. Putting a big solution in place, without building from previous experience, is asking for trouble. These answers are hard, and take time.
so the government is going to move slowly where there is such a perceived risk.
now we are at around 75 US cents. so it has gone up about 18%. thirty years ago, as any beer drinking Canadian (closest analogy to "red-blooded American" I could think of
as for making goods cheaper... hmm... if they are natural resources, those are all costed in US$ anyways, makes no difference. if it manufactured goods, then most foreign components are going to be purchased in US$. so won't make much difference
either.
xchange rates are just trade friction. when rates change, prices slowly adjust to reflect the new cost structure. There is not really a long term benefit. the argument would make sense if high value items were manufactured directly from Canadian natural resources. I don't think that is too common a case.
my guess is that costs in Canada are lower because
there is a public health care system, which controls costs better than the american system, and many other sorts of organizations, like workman's compensation which reduce liabilities, so that insurance costs are lower across the board. The un-employment insurance programs reduce social diparity and unrest, and make the country cheaper to police, again reducing costs. That corporations use the same programs to smooth over low-demand periods by having workers on those programs then, and available when demand picks up.
So they don't have to spend as much on hiring, since the skilled people remain in the industry through the dips. Again, this reduces costs for industry.
Basing the argument on the government having paid for the collection is a iffy at best. The basis should rather be based on maximizing the public good,which is, in the general case, harder to figure out. One has to weigh: privacy concerns vs. defence (against Terrorists domestic and foreign) vs. public benefit. The answer will come out different depending on what the data is, what technology is in place/reasonable, and how much the organisation is willing to spend to make the information public. How soon to make it public is also going to have a big effect on how much it costs. folks on the internet want information upto the second.
You have a chemical spill in Seattle. You have a real-time information system for exchange among first responders who are doing their work. It hits the news and their site gets slashdotted. It's a dynamically built site, so caching by google is of no use whatever. The firemen and coast guard can no longer get information from aerial reconnaisance being done by a Canadian survey plane that happenned to be available. So they don't know where in the harbour the spill has gotten to.
Wall it off? OK, you need a separate network accessible by city, province, state, and many branches of two national governments, as well transportation (railways, airlines) in the area, and any specialized contractors that might be called in. And it has to be setup ahead of time, and managed and funded so that it is up when a crisis happens.
What is the cost of making that site public? Does the public need to know where there is a chemical spill? Of course they do! Should they get same information the government does on their first responder systems? Would be nice, but if the architecture/technology in place cannot answering that sort of demand, what do you do? Most people would accept as reasonable that you have a first responder system that is only available to a few, then have other systems which are used for public dissemination (aka. press conferences, other web sites, etc...)
uh... perhaps because
86% of Canadian foreign trade is with the US , so everybody in Canada notices
the differences in that particular exchange rate.
http://www.atlas.gc.ca
This is built on Chameleon, a GPL frontend for the GPL UMN mapserver whose development were partially funded by Canadian and American governments, respectively, for purely selfish reasons (reducing the costs of producing GIS servers, and being able to provide more information to more groups more cheaply.)
As someone in an organization which has several hundred installed servers, we looked things over in the past year, and are choosing Debian as the next platform for us. Debian stable is the best choice in organizations that can control their application environment. We are lucky, in that most of our apps are in-house, and we have reasonable corporate memory and support in place.
The main attraction of redhat used to be that their software was newer than debian, and that their installation was easier. if you assume a reasonable level of expertise, the installation is a non-issue. Since we were standardized on redhat 7.x, we were having to backstitch things to just install the latest hardware, so the latest and greatest was not helping us at all.
We started to look longingly at Debian stable... 3 years since the last release, only updates in that time... free... decent kde via backports...
Remember what redhat did last year? Any clue how many corporations were running thousands of instances of free redhat 7.x, getting only patches from RH, who were suddenly SOL as of last December? Did you perchance notice a passing similarity between the plans for RHAS (5 year life cycle) and Debian stable (3 years so far)
apt-get wonderful not because of the ease of use of apt-get itself (which is wonderful nonetheless) but because there are tens of thousands of packages which are in the repositories, ready to go, far more packages than are available from any combination of dependency-hell + freshrpms.net + google. "redhat xx rpm" and far more simply installable.
Debian stable is what Redhat enterprise can only hope to become, but will never be, because they have priced themselves, slowly but surely, out of the market. Three things have happenned in the past year or two which will fundamentally alter things:
Within a year or two major ISV's will support debian (stable, at least), because the customers are going there.
And it ain't just me saying it and HP already does it from some telco's
It is just the right answer. All that said, if your shop is committed to binary-ware, and you favourite bit-vendor won't support your chosen environment, you are toast. Do not go there.
Talk, cajole, encourage, convince, or switch vendor or plain drop the binary ware if you can afford to do it, but do not use the commercial software on an unsupported platform. That is the worst of both worlds: The free people don't use your package X so they can't help you, and the paid people go through their menus and hit "we don't support that, click!"
- drones for target practice by the military (sure, all militaries that need them have them already, but how much do they cost?)
- drones for lighting up enemy radar systems (see the last Israeli/Syrian war)
- cargo aircraft for parcel delivery.
Used as Remotely piloted vehicles. As long as they're small enough, even if it crashes the damage done is likely to be minimal, and you do not have to worry about losing a pilot. How about delivering mail in the high artic or ant-arctiv.
- delivery of small amounts of whatever to whomever via air. the longer the range of the thing, the more flexible it becomes.
- while satellite tech is used a lot for surveillance, a good camera on a cruise missile could provide higher res imagery of an area of interest (hence the huge US RPV program.)
My guess is that the main thing holding this field back is telecommunications. You want to use something like IRIDIUM, a cell phone like network with global coverage (last I heard they were still in the air), to relay signals from the planes to the pilot bases. That way you don't have to worry about having transmitters near the field of operations, which reduces the cost of operations again. You need enough bandwidth to provide a live video feed of some quality, but the better the automated systems the less quality you need (aside from the reconaissance applications.) I don't know if any of the civilian satellite constellations in orbit have sufficient bandwidth for that, but that's probably the biggest cost, and biggest barrier to this market exploding.If you can build an RPV with enough range, you could reduce costs of supply to those places quit a bit, and get it to them far more p often. think about it? How do you get fresh vegetables to an antarctic base, or island airport in the pacific ? Use a piloted plane, you want to have a large amount of cargo to offset the costs. Smaller bases in remote areas are supplied more rarely today because of that.
With an RPV, you could ferry medecines and samples back and forth far more frequently. who cares if it is a 20 hour flight, after 8 hours, the relief pilot comes in, after 16, the night shift guy comes in. They can all live in their home town, and kiss their children at the end of the work day, even if their plan goes down in flames into the pacific ocean.
(P.S. once this is working, you can also outsource cargo piloting to India... )
Do it in reserved time for the first couple of runs. Have them involved in evaluating the impact of the scans, show them the results. After it is done a few times with no impact, theyll get bored^h^h^h^h^h comfortable with scans and you may be able to schedule them during the day.
They should be scheduled, so that if something does go wrong they can at least ask you to scan again and reproduce the problem, or eliminate your work from the list of suspects. If you do it without telling them and it causes problems that they spend hours or days figuring out, you can bet you will be confined to 3am forever thereafter.
Folks who run real sites want the hits, but
when slashdot (or Google) points at a hobby site,
they have a responsibility to set up their own site as a caching proxy, and link the proxy, so as not to kill the target. After the rush, switch to a re-direct and
it will sort itself out over time.
Of course, that would cost slashdot big bucks,
and slow down slashdot... on second throught
lets nuke those hobby sites real good.
might be what you are looking for. http://www.gnuenterprise.org
Unfortunately, it doesn't look like enough of it is done yet to be useful. They are working on frameworks on top of which all the applications can be built, but nothing "ready to go" is available yet.
Law and order are more important than any disagreement over copyright. People who willfully violate copyrights by circulating warez & Mp3's against the copyright owners' express permissions, spammers, and crackers creating malware, none of them are of any use to the open source community.
Open source is about respect for copyright, our choice of the terms of the copyright (ie. GPL) is the disagreement with those companies, and we have an legal disagreement about some ownership issues. That does not give anyone the right to behave in a criminal fashion.
"Vigilante Justice" is an oxymoron.
To be fair, the script I started with was around 800 lines. Then I took out 90% of the comments & examples that were irrelevant. There is one script in shorewall which has some meat, for the rest of the directory the config files are 80% comments. So I took it out, and repeated the analysis.
that is:
wc -l * --> 1108 lines.
grep -v '^#' * | wc -l --> 217
The comments are so huge that it is impossible to understand the actual configuration. That's my real gripe: ease of auditing is an important characteristic to me, and that is generally poorly supported in GUI's.
Where most Nouns are Capitalized by Everone. It is a good Thing. I always liked being able to find Nouns easily in a sentence.
Verbs at the end is confusing though. You no Idea Phrase what about is, until Pile of Verbs at the end get to. It Holistic Sentence parsing make is.
4051 total
basquette:/etc/shorewall# wc -l
145
I can understand 145 lines.
4051 lines is a pain to audit. Don't tell me I should audit by pointing and clicking, that is too painful for words.
Yesterday's posting described issues with current
AMD 64-bit linux distros. Can one just use
a 32 bit one for now, and wait a while for the
64 bit ones to mature?
If not, it doesn't sound reasonable (as in, what?!!?
Getting X to work is a challenge?)
'nuf said....
Free Software is about not stealing commercial code. A significant proportion of linux users use it because it is legal to copy for free, and the legal part is important to them. There is a significant proportion of windows users who have cracked OS or application licenses. It isn't as if cracked software is not easily available. For those people, there is no advantage to the freedom of Linux.
So people who use linux value the copyright which gives them free software and so are more likely to respect copyright in general, and pay up.
It's pretty impressive.
2 megawatts and 3000 sq. ft. is quite good.
frightning, but given the power, quite good.
Imagine the UPS.
on the other hand... "100 hour MTBI is desirable"
Ack! It is hoped that it won't crash more than
once every three days? That is up from 40 hours
on the current one. oh. They're putting in lots of
RAS features, and they still can't target higher than that. depressing.
custom interconnect. That is the exciting part.
It looks like a lot of fun. The directions are good
and make sense. connectionless api (mcast & bcast only?) is going to take some getting used to...
Hope they can pull it off.
"Restrictions apply to all telemarketers, although they may differ depending on whether they use a fax or a telephone. As a minimum, telemarketers must maintain "Do not call/fax lists" and provide customers with a fax or telephone number where a responsible person can be reached. Specific rules are included at the end of this document."
This is the best part:
"What are the consequences if telemarketers don't follow the rules? Telephone companies can notify these telemarketers that telephone service to the lines used in connection with placing calls (telephone or fax) may be suspended or disconnected within two business days."
um... phone company may cut off their most lucrative paying customers... sure... I'm sure their quivering in their boots now... Not only do you have to chase the buggers, then you have to pester the phone company, and then they just change the company name and keep going. Not even a fine.
Canadians are so ... wimpy^H^H^H^H^H understanding and sympathetic.