Ummmm.... Ya. I want to spend thousands to replace my completely reliable, switched gigabit or even just switched 100Mb, secure network with a shared 150 Mb, relatively unsecure network that is susceptible to interference from X boxes.. why?
|
Does Verizon's Voyager stack up to the iPhone? |
|
|
5 IT skills that won't boost your salary 
[1,407]
Women 4 times more likely than men to cough up personal info [589]
Japan's 10 funniest tech-related commercials [Videos] [407]
Throwing away a promo CD is "unauthorized distribution"? [1,265]
Adults too quick to dismiss educational video games [682]
Attack of the iPhone clones [Slideshow] [578]
10 things IT needs to know about AJAX [1,258]
This Year's 25 Geekiest 25th Anniversaries [Slideshow] [409]
|
|
If you're asking the
If you're asking the question (or at least asking THAT question in THAT way), it probably means there's no reason for you to do it.
John W. Cox
senior editor
Network World
wireless has reasons...?
Perhaps, but name a "good" reason, just one good one, to move to wireless from wired or install wireless instead of wired in a new network.
Speed? No.
Cost? No.
Reliability? No.
...and you can't use convenience. There is nothing convenient about mopping up the security debacle that awaits the wireless user. Your only hope is to be overlooked in favour of a weaker wireless hack next door because it is only a matter of time before you are hacked if you are spewing your packets into the air...
Based on the talks I've had
Based on the talks I've had for several years with enterprise networking staff, there's only ONE reason to install wireless LANs: mobility.
If your users don't need mobility, you don't need wireless. If they do, only wireless makes it possible.
I know plenty of education institutions what have switched GigE LANs, delivering gigabit to a port in lecture halls, classrooms, and dorms. That's ONE port for a class of 15, 50, or 500. The only way these mobile users can have net access is via wireless.
A well-designed WLAN (especially with the advent of 11n) can deliver all the performance, reliability, and security most users need for most applications, according to the network professionals and users that I talk with it.
John W. Cox
senior editor
Network World
So, there is only one
So, there is only one reason. You call it 'mobility' and I use broader nomenclature; 'convenience'. Mobility as a necessity is a rare thing after all. Few, if any, users really need it for anything other than cell calls.. and I could argue that the ubiquitous cell phone has now come full circle from convenience to time sucking monster.. but I digress.
Mobile convenience at the cost of speed, reliability and security is, as I have stated, no convenience at all. You carefully state that "A well-designed WLAN (especially with the advent of 11n) can deliver all the performance, reliability, and security most users need for most applications..." A politicians statement. Most users? Most applications? Define most. 75%? 51%? What do we do about the poor son of a gun that falls into the minority? Sacrificed on the altar of convenience I suppose.
Your reliance on the new "n" spec is misplaced. The history of "secure protocols" has at the very least taught us that there are no secure protocols. It is only a matter of time before someone breaks it and then convenience will turn quite horribly inconvenient for "some" users.
Yes, you are right. The only way that mobile users can have net access is via wireless. And the sky is blue and apple pie has apples in it. Give them a wire and a port and they get faster, more reliable and more secure access for the same buck. Wiring up a lecture hall is not that hard. Been there, done that. And users in a lecture who need inet access are more often than not accessing a porn site, chatting to the guy in the next row or working on an unrelated project. Heck, if the lecture really requires inet access then the whole thing could probably be done through distance learning anyway. Save some real money- stay home.
The original article mentioned that this is an upgrade from a/b/g to n. That means this is the second wireless rollout for this place. Twice the cost for less throughput than switched 100Mb, less reliability and security that is at the very least suspect. Still can't see the benefit. Oh well, at least they got some neat new toys and something to keep them occupied for a while.
There are times when wireless is the only solution. At all other times it should be avoided like the plague it is.
yes but..
to be the first to have new technology for the use of a network that doesn't use wires? because making cables takes work :)
those are not good reasons
Being first is not a good reason to do anything like this. In an educational institution the money spent to make a slower, less reliable, less secure wireless network for no other reason than to be first would be a reckless waste of money better spent on a faster, more secure, more reliable wire (or fibre) network.
Cleaning up security messes takes work also; and once your packets are floating around for anyone with a wet finger and a coat hanger to pick up it is only a matter of time.
Herb, Respectfully: It's
Herb,
Respectfully:
It's clear you have no idea what you're talking about.
Speed: GigE is certainly faster. But I still want to correct you on your '150Mbps' number. It's 150Mpbs per channel / per radio. So you're looking at closer to 600Mbps for a dual-N radio AP.
Security: Unless you are running 802.1x on your switches I would argue that wireless is more secure. Also, 802.11n has nothing to do with security. That was addressed by 802.11i.
As opposed to wired, modern wireless will require each user to authenticate with their Radius credentials (which are exchanged in a secure manner, typically through an SSL tunnel). Encryption is done via AES and each client and session has a unique key that is rekeyed every 30 minutes or so. How do you control who gets on to your wired network again?
Cost: GigE and structured wiring is not cheap. If it's already there that's one thing. But for a greenfield environment I would challenge you that wireless can make a lot of financial sense.
I'm not trying to sell you on wireless, but it was clear from your comments that you aren't looking at it with all the details.
Respectfully? Hardly.
Respectfully? Hardly. However, I am used to anonymous shills so whatever...
So, let's see... "So you're looking at closer to 600Mbps for a dual-N radio AP" Ok, theoretically you may be right. In the real world? Nope. No way. (especially if you have an xbox nearby apparently). Gig ethernet on wires will beat it every time (of course) and 100mb switched on wires will also. Sure, wireless is getting closer but we ain't there yet. Are you really telling me that you are getting 600mb over wireless of actual throughput in a working environment? Show me the study.
Security. Nothing that spews packets into the air can be considered secure. I don't care if you encrypt them, deep fry them in the Colonels seven secret herbs and spices and then dip them in ranch dressing. Some bright spark will figure out how to decrypt them and then they are free for the taking. And ya, 802.11i is the encryption standard. That's nice, but the article is about an 802.11n deployment. As part of that deployment you get 802.11i encryption built into the n gear...aren't you clever.
Next: "How do you control who gets on to your wired network again?" Hmmm.. to tell you that would be, for lack of a better word, "insecure". ;)
Structured wiring is not cheap. Right you are. Neither is "n" gear. Once you factor in the extra cost of n nics and access points it becomes, at best, a toss up in cost. If you have to move supporting structure or if you have asbestos in the walls or ceilings then maybe wireless is the way to go. Anthing other than that and I shall avoid the wireless plague. I've heard way too much brave new world hype over the years and this is just the latest salvo. If I had a nickel for every time I was told that something I was thinking about buying was secure I would own the planet.
Even if the speed is the same (and it is not) wired is still more reliable and more secure. Completely secure? No, just more secure simply because it does not put packets into the air for anyone to pickup and work on. I will take more secure, more reliable and faster over more convenient any day.
Respectfully,
me
3 good reasons
I get to speak to a lot of large companies about why they install wireless networks and here are the three best reasons I've heard:
1. Cost of building out new office space - when a company is growing and acquiring more room for the employees the "cost" of deploying a wired network is a lot more expensive than wireless. This includes cabling, switching and time (most important).
2. Security - for large organizations it is too cost prohibitive to move to 802.1x on their wired network. With wireless this is built into the standard. Enterprises see adding wireless as a way to make their networks more secure, not less.
3. Flexibility (you call it convenience). Most companies move people and assets all the time and with wired infrastructure it becomes way too cumbersome.
If you want more details http://www.airwave.com/airwaves/2007/09/25/a-real-business-case-for-the-all-wireless-office/
1. Well, the cost of adding
1. Well, the cost of adding utp to "new" office space is practically zero in comparision to the cost of construction if it is really "new". Adding wire to new construction is easy and cheap. For really new construction why not look at fibre. You may have to upgrade the photonic nics and switches in the future but the media is practically unlimited in potential throughput, slightly more secure than copper, an order of magnitude more secure than wireless, extremely reliable, has distance limits that are measured in miles (or kilometers for metric folks) and completely immune to electrical interference. Yes, it is more expensive but the payback is tangible and potentially upgradable for decades without rewiring.
Any "new" space that is not really new but simply "new to you" is likely wired already. And, if it is not wired already, I can get gigabit cards (if they are not already built into the motherboards of the pc's anyway) for $10 while wireless N cards start at close to $100 and since I can get cat6 utp installed at around $100 per drop (less in some cases) wireless has already lost 90% of it's price advantage to me. I can get gigabit switches at less than $10 per port and that kind of wraps it up for me cost wise.
2. Once again, I cannot agree that anything that pumps packets into the air to be picked up by someone in the parking lot or the floor below you is secure in any fashion. I see your point about wired encryption costing money and I am sure you see mine about wireless being inherently insecure to start with.
3. I disagree- Most companies do not move assets and people "all of the time". Most companies are fairly static in the office space. Yes, some users want and a few may even need wireless access. For those few users, if they can prove a need to access the network as they walk down the hall or ride the elevator then fine; take a chance and good luck to you. Again, my point is that wireless should be limited to be used only when absolutely neccesary. The convenience of wireless may more accurately be described as expedience and expedience will allways come back to bite you in the butt eventually.