There are a number of options mentioned by other posters.
And none of the free ones mentioned so far, accomplish what the author wanted.
They are posed as answers to the question, and they are no doubt useful to Engineers and researches, but they do not answer the article author's question.
So far, the tools mentioned that get close to answering the question are proprietary ones (even if free for Academics)
Do any of you know of a visual network simulator that can create an interactive network map that allows me, the instructor, to manipulate various components of a network, including the physical media, routing configuration, and which applications are being used to submit data?
Most of the responses posted so far have been "simulators" all right, but simulators that run actual virtual nodes, and the most important requirement: visual interactivity is extremely limited.
You don't use a simulation tool designed for network research as a visual aid in class, that's a road to failure.
The author was asking for simulation software useful for a visual aid for a classroom, under the (expert) instructor's control in an introductory networking class, not a research tool, to create realistic simulations for students to explore.
Most the FOSS tools mentioned are not even suitable for use by a student in an 'intro to networking' class.
Presumably people are spending some time thinking about what they look up, and following hyperlinks to read more about the subject.
Due to the inconvenience, without the Internet, they might not have looked up any information in the first place, and gone with a hunch, or whatever they vaguely remember.
So the internet allowed them to learn a bit about a subject they wouldn't have even bothered to look for information on, otherwise
That's what it sounds like when someone says they have a policy of teaching using only FOSS tools. If the FOSS tools are just as good or better for instructional purposes as the Non-FOSS tools, that's great.
However, you don't hand students an Abacus to use in a Calculus class, instead of a calculator, just because you the instructor prefer non-electronic devices.
In teaching about computer network, and making graphical simulations, however, the FOSS tools would have a lot of catching up to do.
As I mentioned, cost could be a legitimate reason.
If you can't afford an instruction tool, then you can make do with a poorer tool.
There are also legitimate reasons like unable to find or figure out how the better tool works, logistical concerns, etc.
But there are instructional tools that are available gratis for educational purposes, or that you can use without incurring more costs, that just are not FOSS.
Including things as basic as writing utensils.
Ink cartridge designs are often proprietary, and the manufacturers don't release manufacturing specifications.
Does that mean Ink pens should not be used in a classroom, I think not?
It is obvious from the description that the design of the simulator itself would not be the subject of the class.
Teaching introductory network classes does not involve teaching students anything about simulators, except, perhaps, how to use one.
The simulator is a tool, just like a physical model would be a tool.
You don't need to be choosing an inferior physical model to use in your classroom, just because the manufacturer releases full specifications that would be needed to clone the model, than the more instructive choice.
GNS3 is not an open source solution, because, yes, Cisco IOS images are required to make it work.
Of course dynamips itself and GNS contain open code, however, you would have to write a clone of IOS for them to be useful given a requirement that you use only FOSS.
There is no ready-made network simulation package that meets the I teach using FOSS only. rule.
Also, it would be highly preferable if it ran on Linux, as I teach using FOSS only."
You are sacrificing your effectiveness as a teacher, and potentially failing to help your students learn, in the name of supporting FOSS?
Look.. I like FOSS, when available and when the most suited to the task, it's great.
I believe in the classroom, you should be using the most effective instructional tools available, not ones whose licensing model you personally prefer
Of course if the Non-FOSS options' license model is so restrictive, a restriction will prevent you using it in the classroomas an instructional tool (highly doubtful), then that could be a reason to reject it.
But based on your requirements, all the decent simulation options other than you having to write some code, or put together images/mock-ups manually, are not only non-FOSS, but also not free of cost.
Perhaps you should just embrace the 'free' option, and draw up your own sketches and diagrams?
If you don't like closed solutions, that's (more or less) what you are left with.
Which is not horrible, but I assume you would not have asked the question if it was the best for your students.
That's right. No refrigerator capabilities included in the plan, and the Electricity SDK provided by the power company specifically prohibits developing Refrigerator Apps, they're automatically rejected from Energy Co. the Appliance store.
And since only digitally signed apps can be plugged into the wall....
Don't fret though, Electricity Version 4.0 is promised to deliver Refrigerating capability, costs an extra $20 a month though.
And only available using the new extremely-limited(TM) electric plans.
You don't pay by the watt hour for electricity your telephone uses, your house power could be done in a similar way.
Instead of paying per watt hour... your electricity usage is capped at any given point in time, you use as much or as little for a flat rate, but a special regulator circuit stops you from going over the cap... there are 3 plans.. the Lite plan, which includes 45 watts usage (Power a light bulb!), $60 a month; basically, the equivalent to the iPhone 2GB plan. The Bronze plan, which includes a 100 watts, "light up the whole house", for $100 a month... Gold includes 500 watts usage, "watch TV", $200 month, Platinum includes 1000 Watts usage, "perfect for cooking, and finally there is the high-power connection Energy Plus, 2000 Watts usage, "run an HVAC" system, only $500 a month.
During times when demand is higher, people trying to use too much electricity get their cap dropped.
Also, just because you have 45 Watts of usage, doesn't mean you are allowed to continuously use it all the time.
If usage is deemed excessive, power will be turned off, your account may be terminated at any time, at the sole discretion of the power company.
You may be forced to buy a more expensive plan if you continuously use 45 watts 24/7.
We keep hearing in the news of people being fired or disciplined at work, both private sector and public sector employees in that category, based on some dumb thing they posted on Facebook.
What do you think would happen if one of the president's staffers got caught posting a fake Facebook page mocking the president and using obscenities?
In the best case scenario, they would be politely asked to step down.
So if adults are subject to this, what is the justification for students to have stronger free speech protections than the rest of the public?
By doing the bare minimum they need to do to meet the legal requirements, and using a highly-paid legal team to find as many ways around as many of the requirements or mitigating as much as possible
By meeting them inconsistently. They can maintain appearances of meeting all requirements while not actually doing so..
By not putting in writing or making documentable their intent to not meet any requirement(s).
By taking full advantage of what requirements don't spell out, or anything they leave at the leasor's discretion.
By lobbying regulators for as many breaks as possible, and ensuring that they cannot be prosecuted for anything.
No set of requirements is perfect.
I would suggest that intentionally or not, (IMO) the big companies have already proven that they really are that dirty with 'lease infrastructure' requirements.
I wonder whether the court would also accept a driver's own GPS log as exculpatory evidence.
Nope, sorry, we don't trust any of that scieentifik evidence round here.
Or: So, I assume you can produce a certificate showing you were trained in the use of this GPS device?
Or... The evidence is excluded on the grounds the defendant can't prove they did not somehow manipulate the GPS device to log you as not speeding.
Or.... The GPS evidence is interesting, but does not invalidate the officer's estimate.
The officer was trained to make that estimate, therefore it is infallible, sorry, pay the fine.
Companies forced to lease infrastructure can counter that 'requirement' in several ways...
By pretending certain infrastructure doesn't exist, is all used up or reserved, or failing to document that infrastructure is leasible.
Building just slightly more than enough infrastructure to meet their own needs, so if anyone ever wants to lease, there will be very little that can be leased.
By fighting tooth and nail over what a 'fair' price is. Making sure the price is high enough to make it as least-profitable for the competition as possible.
Drawing out any dispute or disagreement as long as possible, refusing to make any effort to address the problem, until the entity trying to lease has been through thousands of CSRs and escalations.
Being as non-responsive as possible to any issues reported by the leassee in the process required to lease infrastructure
Making leasing applications as complicated as possible, with as long a time processing window as possible, with the greatest processing fees possible, and denying or sending back applications for correction if the slightest technical error is made, or if the leasing organization's people make an error (requiring applicant to lease infrastructure start over from the beginning).
Set arbitrary rules. Create new rules while the applicant's lease request is in progress, and deny the request, or bounce back, and require changes based on new rules.
Leasing company can make rules regarding the leases, that might seem attainable on their face, but have the effect of increasing costs for the competition, so the competition has to charge more than the company leasing infrastructure, to break even.
(Example: Requiring infrastructure to be leased in certain very large size bulk units. Requiring leassee to put huge deposits on file, and meet a slew of requirements. Introducing rules that limit the leassee's ability to use the infrastructure, or that artificially require them to buy more for certain functions)
Divide the infrastructure into as many tiny byte-sized pieces as possible, and nickle and dime to death. "OH... you wanted your spot on the cell tower to not be high enough to not be blocked by the tree?" "That costs extra, you'll have to pay the early termination fee to return that slot, and submit a new application, pay fees all over again."
By making infrastructure leasing requirements, and requirements to increase the amount leased, or lease more units, as complicated and difficult to deal with as possible.
By changing leasing terms as frequently as possible, applying stiff penalties for any error, claiming applicant errors were made when it was really the leasor's error.
It's not a free market, because Data service for the iPad and iPhone is an AT&T monopoly, protected by Patent and Copyrights (over the technology), and an Exclusivity deal with Apple.
In a time of emergency, the military has the right to take private property and do anything they need to do with it.
If the property is CI, that could mean making it to continue to work, or making whatever changes they need to protect it.
If you get in the way, they have the right to shoot you, or arrest you, and hold you (potentially indefinitely).
The Supreme Court has upheld the federal government's power to commandeer private property
If you refuse to help or surrender your belongings? In jurisdictions having a posse comitatus law, you could be fined and possibly even jailed.
In United States v. Russell the Supreme Court was faced with a claim for three steamers commandeered by military authorities during the Civil War. The Russell court found it obvious that "the taking of such property under such circumstances creates an obligation on the part of the government to reimburse the owner to the full value of the service." The court continued, "private rights, under such extreme and imperious circumstances, must give way for the time to the public good, but the government must make full restitution for the sacrifice." The court concluded that the obligation to make full restitution was based on an implied promise "
The Supreme Court hasn't said what happens if equipment is borrowed and returned damaged, but lower courts have been reluctant to award compensation in such cases.
Courts have refused compensation to people whose property the police damaged while executing arrest warrants or search warrants. They've also refused compensation when police intentionally damaged property in an effort to flush out a suspect.
Sixth, if I'm helping the police and someone else gets injured, can they sue me? There's no clear rule.
Er... interesting.. so what business do marketing people have picking an opinion for the company to have about the technical (security) merits of an OS, then?
If designers have built a massive, insecure system, on an insecure foundation, the expenses involved in attempting to shore up the foundation and work around security issues are inherently higher.
Designers didn't have security in mind originally, so (technically) they should re-visit every design decision you made in every aspect of the OS, everything that interacts with the user, everything that interacts with software, every data structure, every API, every remote procedure call, the core design of every algorithm, including the remote procedure call system itself.
That's a lot more expensive and requires a lot more resources, than a system that was originally designed with security as a major consideration.
And when 'making it actually secure' requires a major design change, implementing the design change will be much more expensive, then fixing minor problems.
Also, when design change requires incompatibility, to achieve some semblence of security , you have a problem with no good solution.
I believe MS' answer in these situations has been to gloss over it, try to hide it, or fudge a patch claiming they've fixed it (just by covering over one exploitable scenario that results from it).
Because more investment in security does not mean your product is more secure.
You can throw a hell of a lot of money at trying to secure something, and fail utterly, because your design and imposed requirements, inherently lead you down the path to an insecure system.
For example: mandatory compatibility with a known insecure framework, and 'compatibility' includes a lot of esoteric bugs that programs came to rely on over the years.
However, I kind of view this like an article about "Redhat reportedly ditching Windows".. they use Linux extensively, they should have done this a long time ago.
Meanwhile Redhat came out with this RHEV stuff, manager for their new virtualization platform...
WTH.. you make an OS, and your management platform from your OS needs a copy of this foreign Windows junk?:)
Isn't it amazing that Google is ahead of Redhat.
Does this mean there will be a stable Linux version of Chrome coming out soon?
Slashdot Mirror
And Apple can then add libel to the Gizmodo editors' rap sheet, and seek injunctions to take down gizmodo.com, in addition to the criminal charges?
There are plenty of people left when you rule out scumbags, and lowlifes.
However, the people left have basically no chance of getting elected, and you'd waste your vote.
The vast majority of such people are not running for office.
Who (other than a scumbag/lowlife) would want to get elected and have to work with scumbags and lowlifes, anyways?
There are a number of options mentioned by other posters.
And none of the free ones mentioned so far, accomplish what the author wanted. They are posed as answers to the question, and they are no doubt useful to Engineers and researches, but they do not answer the article author's question. So far, the tools mentioned that get close to answering the question are proprietary ones (even if free for Academics)
Most of the responses posted so far have been "simulators" all right, but simulators that run actual virtual nodes, and the most important requirement: visual interactivity is extremely limited.
You don't use a simulation tool designed for network research as a visual aid in class, that's a road to failure.
The author was asking for simulation software useful for a visual aid for a classroom, under the (expert) instructor's control in an introductory networking class, not a research tool, to create realistic simulations for students to explore.
Most the FOSS tools mentioned are not even suitable for use by a student in an 'intro to networking' class.
Presumably people are spending some time thinking about what they look up, and following hyperlinks to read more about the subject.
Due to the inconvenience, without the Internet, they might not have looked up any information in the first place, and gone with a hunch, or whatever they vaguely remember.
So the internet allowed them to learn a bit about a subject they wouldn't have even bothered to look for information on, otherwise
And you're a coward.
That's what it sounds like when someone says they have a policy of teaching using only FOSS tools. If the FOSS tools are just as good or better for instructional purposes as the Non-FOSS tools, that's great.
However, you don't hand students an Abacus to use in a Calculus class, instead of a calculator, just because you the instructor prefer non-electronic devices.
In teaching about computer network, and making graphical simulations, however, the FOSS tools would have a lot of catching up to do.
As I mentioned, cost could be a legitimate reason. If you can't afford an instruction tool, then you can make do with a poorer tool. There are also legitimate reasons like unable to find or figure out how the better tool works, logistical concerns, etc.
But there are instructional tools that are available gratis for educational purposes, or that you can use without incurring more costs, that just are not FOSS.
What do you mean by that?
Our world is full of proprietary technology.
Including things as basic as writing utensils. Ink cartridge designs are often proprietary, and the manufacturers don't release manufacturing specifications.
Does that mean Ink pens should not be used in a classroom, I think not?
It is obvious from the description that the design of the simulator itself would not be the subject of the class.
Teaching introductory network classes does not involve teaching students anything about simulators, except, perhaps, how to use one.
The simulator is a tool, just like a physical model would be a tool.
You don't need to be choosing an inferior physical model to use in your classroom, just because the manufacturer releases full specifications that would be needed to clone the model, than the more instructive choice.
Of course cost may be a legitimate factor
GNS3 is not an open source solution, because, yes, Cisco IOS images are required to make it work. Of course dynamips itself and GNS contain open code, however, you would have to write a clone of IOS for them to be useful given a requirement that you use only FOSS.
There is no ready-made network simulation package that meets the I teach using FOSS only. rule.
Also, it would be highly preferable if it ran on Linux, as I teach using FOSS only."
You are sacrificing your effectiveness as a teacher, and potentially failing to help your students learn, in the name of supporting FOSS?
Look.. I like FOSS, when available and when the most suited to the task, it's great.
I believe in the classroom, you should be using the most effective instructional tools available, not ones whose licensing model you personally prefer
Of course if the Non-FOSS options' license model is so restrictive, a restriction will prevent you using it in the classroomas an instructional tool (highly doubtful), then that could be a reason to reject it.
But based on your requirements, all the decent simulation options other than you having to write some code, or put together images/mock-ups manually, are not only non-FOSS, but also not free of cost.
Perhaps you should just embrace the 'free' option, and draw up your own sketches and diagrams?
If you don't like closed solutions, that's (more or less) what you are left with. Which is not horrible, but I assume you would not have asked the question if it was the best for your students.
That's right. No refrigerator capabilities included in the plan, and the Electricity SDK provided by the power company specifically prohibits developing Refrigerator Apps, they're automatically rejected from Energy Co. the Appliance store.
And since only digitally signed apps can be plugged into the wall....
Don't fret though, Electricity Version 4.0 is promised to deliver Refrigerating capability, costs an extra $20 a month though.
And only available using the new extremely-limited(TM) electric plans.
You don't pay by the watt hour for electricity your telephone uses, your house power could be done in a similar way.
Instead of paying per watt hour... your electricity usage is capped at any given point in time, you use as much or as little for a flat rate, but a special regulator circuit stops you from going over the cap... there are 3 plans.. the Lite plan, which includes 45 watts usage (Power a light bulb!), $60 a month; basically, the equivalent to the iPhone 2GB plan. The Bronze plan, which includes a 100 watts, "light up the whole house", for $100 a month... Gold includes 500 watts usage, "watch TV", $200 month, Platinum includes 1000 Watts usage, "perfect for cooking, and finally there is the high-power connection Energy Plus, 2000 Watts usage, "run an HVAC" system, only $500 a month.
During times when demand is higher, people trying to use too much electricity get their cap dropped.
Also, just because you have 45 Watts of usage, doesn't mean you are allowed to continuously use it all the time.
If usage is deemed excessive, power will be turned off, your account may be terminated at any time, at the sole discretion of the power company.
You may be forced to buy a more expensive plan if you continuously use 45 watts 24/7.
We keep hearing in the news of people being fired or disciplined at work, both private sector and public sector employees in that category, based on some dumb thing they posted on Facebook.
What do you think would happen if one of the president's staffers got caught posting a fake Facebook page mocking the president and using obscenities?
In the best case scenario, they would be politely asked to step down.
So if adults are subject to this, what is the justification for students to have stronger free speech protections than the rest of the public?
Oh... that risk can be addressed also
No set of requirements is perfect. I would suggest that intentionally or not, (IMO) the big companies have already proven that they really are that dirty with 'lease infrastructure' requirements.
I wonder whether the court would also accept a driver's own GPS log as exculpatory evidence.
Nope, sorry, we don't trust any of that scieentifik evidence round here.
Or: So, I assume you can produce a certificate showing you were trained in the use of this GPS device?
Or... The evidence is excluded on the grounds the defendant can't prove they did not somehow manipulate the GPS device to log you as not speeding.
Or.... The GPS evidence is interesting, but does not invalidate the officer's estimate. The officer was trained to make that estimate, therefore it is infallible, sorry, pay the fine.
Companies forced to lease infrastructure can counter that 'requirement' in several ways...
It's not a free market, because Data service for the iPad and iPhone is an AT&T monopoly, protected by Patent and Copyrights (over the technology), and an Exclusivity deal with Apple.
In a time of emergency, the military has the right to take private property and do anything they need to do with it. If the property is CI, that could mean making it to continue to work, or making whatever changes they need to protect it.
If you get in the way, they have the right to shoot you, or arrest you, and hold you (potentially indefinitely).
Hell, the police can do that sort of thing.
The only way the Feds will control anything on my network is when they do it over my cold, dead ass!
In a time of emergency, when martial law is in effect, that can be arranged.
If you want an edge... kick the Caffeine habit. Stop using it.
Wait a few years... then use it in small amounts, without getting yourself addicted again.
You could probably use Red Bull once or twice too.
But you won't be able to use it every day regardless (same as cafeeine).
If you use it every morning, or on a frequent basis, you build that tolerance, and brain comes to expect it
Could be a 'bubbles' bubble.
Due to irrational exuberance in the bubbles index, and investors massively buying up the bubbles indexes in anticipation of a bubble.
Er... interesting.. so what business do marketing people have picking an opinion for the company to have about the technical (security) merits of an OS, then?
If designers have built a massive, insecure system, on an insecure foundation, the expenses involved in attempting to shore up the foundation and work around security issues are inherently higher.
Designers didn't have security in mind originally, so (technically) they should re-visit every design decision you made in every aspect of the OS, everything that interacts with the user, everything that interacts with software, every data structure, every API, every remote procedure call, the core design of every algorithm, including the remote procedure call system itself.
That's a lot more expensive and requires a lot more resources, than a system that was originally designed with security as a major consideration.
And when 'making it actually secure' requires a major design change, implementing the design change will be much more expensive, then fixing minor problems.
Also, when design change requires incompatibility, to achieve some semblence of security , you have a problem with no good solution.
I believe MS' answer in these situations has been to gloss over it, try to hide it, or fudge a patch claiming they've fixed it (just by covering over one exploitable scenario that results from it).
Because more investment in security does not mean your product is more secure.
You can throw a hell of a lot of money at trying to secure something, and fail utterly, because your design and imposed requirements, inherently lead you down the path to an insecure system.
For example: mandatory compatibility with a known insecure framework, and 'compatibility' includes a lot of esoteric bugs that programs came to rely on over the years.
WikiWireT[r]ap.
However, I kind of view this like an article about "Redhat reportedly ditching Windows".. they use Linux extensively, they should have done this a long time ago.
Meanwhile Redhat came out with this RHEV stuff, manager for their new virtualization platform... WTH.. you make an OS, and your management platform from your OS needs a copy of this foreign Windows junk? :)
Isn't it amazing that Google is ahead of Redhat.
Does this mean there will be a stable Linux version of Chrome coming out soon?