Cisco's IP Phones - Seven Digits And Cat5

kevin922 writes: "Check it out! IDEO, a small close-knit development firm (they are responsible for products such as the Visor, Transmeta Webslate, and a variety of other cool things) has developed Cisco's new IP Phone. This device (which looks like a normal phone) plugs directly into a 10-baseT connection and has a phone # associated with it. When you plug the phone in somewhere else it takes the # with it. I'm planning on getting some eval units to try out. Should the PBX guys start reading up on TCP/IP? :)" Doesn't look like these things are available just now, but the concept is long, long overdue. Bypass the phone company -- in fact, just ignore the phone company.

Comments

Make it 12 digits and open up a world of...

[yerricde]

I should be able to dial 064 028 067 061 and get a prerecorded message with today's News for Nerds headlines. Or dial 192 215 176 126 to listen to my playlist of unsigned bands on my hands-free phone. Or dial 216 033 238 007 and have rsynth read me my spam.

Numbers look familiar? They're the IP addresses of the respective web sites. A "phone number" for voice over IP would be a static IP address.

Re:Make it 12 digits and open up a world of...

[TheGreek]

Yeah, but people keep making noises about wanting to move to IPv6 RSN, so that'll break in a few years.

12 digits.

[yerricde]

But I picked seven, since those others would be even harder to explain. Some finite number of digits which would specify an IP-connected phone is all I meant.

Wouldn't it be obvious that you meant twelve digits? If it's IP connected, you might as well be able to dial an IP address (news sites such as /. and NYTimes might have Dial-A-Headline, etc.)

Wow, I'm impressed with IDEA

[Hanno]

It's funny - their project list contains a number of projects that I like for their cool design.

I wasn't aware that all these products were designed by the same company. Designers should be mentioned in a product's manual... Keep up your work, guys.

------------------

Re:Already done

[eyeball]

Even better, SIP (Session Initiation Protocol), is an IETF standard (rfc 2543) which (in some respects) replaces some of the functionality of H.323. Here's a nice comparison: http://www.cs.columbia.edu /~hgs/sip/h323-comparison.html. Also check out the SIP faq and columbia.edu's SIP Main Site. It defines the protocols for locating and setting up any type of multimedia call, be it voice, video, virtual presence, whatever...

Re:Bypass the phone company my tuckus.

[EQ]

Not true any more.

Look at comapnies like Williams Communications, Qwest - and the newest 800 lb gorilla in the world: Level 3 communications.

They are going to flood the market with cheap bandwidth and knock the old line guys like AT&T and Sprint off at the knees.

Try reading the mini-essay on Silicon Economics at LEvel-3. They have my vote - any company whose CEO lists Snow Crash as one of his favorite novels is definitely out to do things differently.

But it's still got a telephone number

[Anonymous Coward]

I was worried to read that the phone "retains its number wherever it is."

To me, this continues the existing paradigm that the telephone number specifies the telephone rather than the person you are contacting. So I'm gonna lug one of those beasts around to plug it in wherever I go? So when I move offices at work, we're gonna swap two physically identical objects?

Far better to have a process by which I can tell the phone that it now has my telephone number!

Login could be accomplished using a keypad PIN or for bonus points an IR link with some crypto keys in my palmpilot. (Double bonus points for being able to use my palmpilot telephone list as speeddials!)

Lazy question

[kaphka]

I should just try this out myself, but I'll let you guys do the work instead...

When I last tried voice over IP (about four years ago,) the biggest problem that I could see was the latency. The one or two second delay completely destroyed my precise comic timing, which is the only thing preventing people from seeing me as the evil, hearless bastard that I am.

Is it any better today? The latency, I mean?

Re:Lazy question

[Anonymous Coward]

I currently develop a (better) IP phone for 3Com. We also have a several orders of magnitude larger market share with our NBX phones which can also talk IP.

You are correct that latency is important. I wish more people would cite a latency (measured in units of time) when I ask about a DSL/cable-modem speed instead of a throughput (measured in bits per time). With TCP and congested conditions, bandwidth is allocated by latency anyway. Also, for the TCP ramp-up (where most web pages are loaded) latency is a big governor.

In Internet Telephony (not counting that unusable crap over dialup modems) everyone basically uses G711 (PCMU) with 20ms packetization, so the inherent round-trip packetization delay is only 40ms. My typical ping times across the Internet backbone are less than 80ms making a total of 120ms + a little jitter buffer which is as good as a cell phone. For perspective, sound travels about a foot per millisecond anyway.

A big problem is trying to make the phones autonomous. So many people implement a master slave protocol like MGCP/SGCP which allows for dumb phones. Strangely when you use this approach, you usually want the phones totally dumb, and the protocol specialized to the specific plastics, so everyone uses a proprietary one. The phones don't even know what their buttons mean. The have to tell the server. Yuck.

My phone uses SIP an IETF standard. The phones can make direct phone calls without any server assistance. SIP has features for PGP, and the next phones will support strong encryption, like 3DES and blowfish. They currently already use MD5 digest authentication.

Internet telephony will happen. The voice infrastructure grows at 7% per year, but the Internet at 300 to 1000% depending on how you measure it. As bandwidth becomes free (finally a commodity which really is too cheap to meter), people will naturally migrate to Internet telephony for the services.

Now, if only 3Com could build DSL & cable modems fast enough to satisfy demand. :-) -- Rick Dean

Re:Lazy question

[tomorrows_joe]

It all depends on the network environment you're having to go through. If you have some of the various QoS methods implemented, then the delay is pretty much unnoticeable.

I personally don't know how far the carrier IP network (esp. in the US) has progressed in this matter, but I do know that QoS is at least starting to be implemented industry-wide.

One thing I noticed, though, is that people here seem to be expecting that you can just hook up these phones and have instant clean and delay-free VoIP from your home to the home of your friend on the other side of the country. While this may be the case (like I said, not familar with the US market), most "IP Phones" utilize the IP Type of Service field. This is one of the QoS methods: "best-effort" traffic is given an IP TOS of 0, while VoIP traffic is given a TOS of 5 or higher. The router then throws TOS 5 traffic into the low latency queue, while TOS 0 traffic is just treated as normal internet packets. Now, if you are paying for for the privilege of sending traffic marked as you see fit to your ISP, you would see no delays perhaps. But it doesn't take much effort to relabel all of a "best-effort subscriber level" user's packets as TOS 0 (well, the settings are simple).

This doesn't necessarily mean non-business private individuals can't use an IP Phone, but you may experience the delay traditionally associated with VoIP.

But as seen on the IDEO IP Phone information page, this is mainly aimed at businesses, who have full controll over their own internal networks and can pay for the nice extra services (Low Latency Traffic, VPNs, etc), to their upstream providers when connecting remote offices.

Incidentally #1, with a full QoS setup, VoIP works surprisingly well....even with thrashing a router with best effort traffic, a voice call goes though and despite the heavy load most people would hardly tell the difference between VoIP and a normal TDM line.

Incidentally #2, I believe this is just another step in an overall trend of making all networks packet-based, even the traditional phone network (so don't worry too much about the phone company getting jealous). There have been lots of recent developments in SS7 gateways to connect VoIP type networks with traditional PSTN.

(sorry, too lazy to make multiple posts, so I just crammed everything I wanted to say into one reply, heh heh)

Re:Lazy question

[StanSmith]

Very funny comment about the comic timing :)

Now, about the phones...
We just took a looong look at these at work, as we were opening a new headquarters and had to decide between buying a lot of regular (nortel) phones and doing the whole PBX thing or going the cool Cisco route.

We all really wanted to go the Cisco route, but it just wasn't quite ready for business use. The handsets were actually the big shortcoming, but overall the tech just wasn't there yet.

The decision, made by the guy who started my company, whose telecom knowledge is remarkable, was that the Ciscos would be ready by late summer. So one generation past what you see now, if you're a business intending to buy a lot of them and rely on them heavily. Or if you're just a normal guy who doesn't want to buy a new phone again for a while ;)

Martin

Re:bypass phone companies completely?

[AppyPappy]

Yeah. We said the same thing about CB radios in the 70's. Nice try but no cigar.

Re:ABC of IPphones

[mosch]

Just curious, a pet peeve of mine, why is it that so many people consider a binary protocol to be harder to debug than a text protocol? Really, is it that much easier to look for text than a binary string? I'm not familiar with the protocols in question, but I'd wager a guess that any differenece in ease of implementation is based mostly on sanity of protocol design, not whether there are text or binary fields.
----------------------------

And [sniff sniff] what's that smell?

[Tony Shepps]

It's the Public Utilities Commissions in 50 US states, sharpening their knives and getting ready to keep their power by finding new ways to apply new regulations to the net.
--

Need for phone book protocol.

[etwilegar]

Now that we have the hardware, we need to have a email like protocol to handle the registration and traking people.

First the ip of the phone needs to be registered with the identifier of that phone. Friendly name would be like email. Something like somebody@phone.com or whatever the domain of the server. The mac of the phone would be it's identifier. The phone would have setup options that would allow it's owner to enter the server's it is to use to register itself, much like an email client would. Phone keypads would be replaced with small keyboards or touch screens.

Their might be a problem with security. If someone can make their device have the same mac as you, then they could take over your phone. I imagine we would have to have a password to keep people from changing your ip on the server side.
If everyone runs their own server and registers their own domain, than there would be no risk.

Time for new last names. Come up with a domain name, and identify you family with it. When you get married come up with a new one. :)

Only seven digits?

[TheGreek]

No area code? So only 10,000,000 people can have one? Nuts. :(

Cisco IP phones are available now, but...

[Nate237]

this particular phone isn't available yet. The VIP 30 is their flagship phone, and its pretty much as basic looking phone. This is technology that Cisco aquired when they bought Selsius. I'm in the process of deploying Cisco's IP phone systems in two cities now. Eventually we will link them both together and have toll-bypass via our Internet connection. They are pretty cool, but it is not a total PBX killer just yet. There are a few more features needed, but they are probably not far away. As for latency, I haven't noticed any, and the sound quality is as good as regular phone systems. We have had some issues with drop-outs of very quiet calls, but I'm pretty sure its just something we need to tweak on our h323 gateway. For more info on the core server that runs all this, see http://www.selsius.com/univ ercd/cc/td/doc/pcat/7830.htm

3Com NBX100

[COLUG]

The 3Com IP phone system has been out for some time.

Mutant Ethernet

[Detritus]

What happened to the modified version of Ethernet that allocated 10 mbps to data and 6 mbps to voice/video? I think it was being pushed by National Semiconductor. It looked like a clean way to run voice over Ethernet without the delay and loss problems of standard Ethernet.

Seven digits?

[Mike Schiraldi]

I didn't see anything in there about seven digits .. what happens when they use up the 10,000,000 "phone numbers"?
--

ABC of IPphones

[Ion Berkley]

Seems like enough people are interested that its worth my time writing a little helper on IP phones here. Standards: There are 3 standards in the play at the moment that are IP Phone candidates. H.323 which is the incumbent mature protocol in IP video conferencing and also the underlying protocol of MS netmeeting is the one that these phones and others will be addressing first. H.323 is somewhat ungainly based on a binary protocol thats a bitch to debug and firewall unfreindly. MGCP which is the basis for forthcoming IP telephony in the digital cable TV system, and a much easier protocol to understand being text based. And lastly SIP which is the new darling of the IP telephony world and already supported in beta versions of Cisco IOS. The actual cisco phones do NOT talk any of these standards, but insted talk something nicknamed the "skinny stack" which cisco inherited from Selisius. Other Cisco equipemnt in the network translates to the'standard protocols'. Expect to see many more IP phones from other companies in the next few months, cisco is not the only show in town. Interoperability: Until someone has a brainwave the user interface to make calls on all these phones is going to be a classic phone number which will be tarnslated into IP andthen ethernet by protocol engines transparant to the user. IP Phones interoperate with the legacy phone network through 'Gateways', just as present TollBypass cheap IP long distance works right now. You can expect that LDAP will be integrated in to buisness style deployments of these phones pretty quickly so throw that phone list away! Bandwidth: An IP phone will use anything from around 80kbit/S (G.711 audio) to maybe 24kbit/S (G.729/G.723), but in actual fact the killer problem is latency and jitter, so expect to need to use new protocols to make them work well. Cisco and others will use a standard called 802.3p which basicly adds a priority into the ethernet framing info so that switchs can expedite the delivery and routers can tranlsatethis into other protocols for delivery over the WAN (Diff-Serv is the ideal here). Enough I'm Boring myself! ION

Area codes?

[Spyky]

I imagine that these phones work fairly similarly to a standard PBX (digital or analog) phone system found in many offices. That is, it's going to require a device that manages multiple physical "trunk lines" that go out to the real world and the phone companies. This device also generally handles voice mail, receptionist routing, etc. The difference between this phone system and the PBX phones systems is that its traffic is routed over standard network Cat5, using the same hubs, switches, etc. as the rest of the network.

What does this mean for the adopters of the system? Well PBX systems are generally supplied by phone companies and contractors and are, in a word, fantastically expensive for what they really are. I think that CISCO/IDEO can probably provide this system for the same cost or even less then traditional PBX systems! Any additional cost of the devices (if any) will be offset by the cheaper installation, because it shares the same network cables as the computer network. Running extra sets of cables in offices (or expanding current runs) is extremely expensive, as evidenced by several estimates my company recieved when it expanded its office space and took over a suite next door. Also as far as administration goes, these phones, designed by the talented people at IDEO are probably far more user friendly then the gastly telco-designed PBX systems. Being able to move the phone physically (to a bigger office when you get promoted) just makes management of the system easier.

However, it (unless my thoughts on how this system works are totally wrong) will still require telcos to contract blocks of phone numbers and trunk lines, which allow communication with the outside world. The telcos aren't going anywhere, until we start being able to route IP voice traffic *outside* of internal networks. Note: this is going to require a more reliable internet service. Users are going to scream a whole lot louder when then can't use their phones than when they can't access their email (and they scream pretty loud about email, trust me)

As for those who mentioned that fax and modems are still going to require traditional telephone lines. This is true, except this doesn't really change anything. Most current PBX systems are digital internal to the site, and many use RJ-45 connectors instead of old RJ-11s. Fax machines and modems *already* require a seperate phone line under most current systems. However, modem support is mostly a legacy (but sometimes necessary) application, not every person's office is going to need a modem, every office requires a phone. Expect fax machines to move towards a similar IP-based internal system. Some current products allow the routing of incoming faxes to a "fax server" to be sorted and sent via email to the users email. The problem arises when handwritten cover pages are used (OCR can't figure out who the fax should be sent to). This technology is really interesting, but definitely has a ways to go. Personally I eschew the user of fax machines, but they are sometimes required when original documents with signatures need to be sent immediately.

At any rate, this system is definitely a god-send for us techs who deal with both computer and phone networks, and its likely to save the company money in the long run.

Spyky

You can't avoid the phone company.

[Jason Pollock]

You physically can't. You can't even do it if you forget about the physical link. Even if you are using IP for long-distance, someone still has to pay exchange fees to get it to a regular phone. The FCC has already declared that IP voice traffic changes are to be treated/tariffed as inter-LATA (long-distance) calls. In other words, your ISP, or the other ISP will have to pay $$ to connect the call to a regular phone, and you had better bet that they will pass that charge back to you.

Also, if an appreciable portion of the traffic starts to be carried over pure IP, the FCC will force ISPs and equivalent carriers to charge for the service, even going so far as to tariff all data calls as inter-LATA (billable).

It may seem incredibly strange, but prices are regulated up as much as down in the telecoms market. This prevents the big companies, like ATT, MCI, AOL/TimeWarner from undercutting veryone else overnight and putting everyone out of work. This would also prevent anyone else from entering the market.

Yes, yes, voice is data just like everything else. But, would you really trust your 911 calls to your cable modem and the Internet in general?

And before anyone starts ranting about routing around the regulations and changing ports/etc. How would you find the person at the other end? You want to pick up the phone and dial. You can't do that without looking up the other person and initiating a connection. The connection initiation can be easily detected, simply because the software has to meet a standard in order to be usefull. Most high speed network providers already run hidden http proxies, do you think that they couldn't run H.323 gatekeepers if required by the FCC?

Jason Pollock

You guys are stupid...

[Anonymous Coward]

Wake up... you can't just plug this thing in on any Ethernet network and start making phone calls. These are the telephones themselves. They require a gateway to the PSTN if you want to call anyone other than users you've got set up internally. You need software to configure the terminals and handle the calls. You cannot go out and get a pair of these for "evaluation" and expect them to work. Most everyone, including the author of the original "article", are confusing this system with the Aplio or other systems like it where you may use your existing internet connection to call another user with a like device. This is NOT anything like that. It is simply an IP-based PBX system, and just as proprietary and expensive as anything by Lucent, Nortel, etc. Unless you run telecom in a big office and need a new PBX, put your dick back in your pants. This is NOT a consumer device. Do your homework next time.

Bypass the numbering plan?

[XNormal]

If you want people with a regular phone to be able to call you you must have a number assigned according to the phone company's numbering plan. This means that even if everyone gets this kind of phone and we bypass the circuit switching phone network for virtually all calls we are still dependent on the phone company for assigning numbers. Sorta like our dependence on the InterNIC in the past, except that you can't assign your own hostnames under your domain - you must register each hostname seperately.

----

Cisco's IP Phone are a nightmware

[draggy]

I've had an horrible experience with Cisco's Selsius phones. The first strike is the call manager. Runs on NT and that is VERY far from the 5 9 (99.999%) availability that other PBX offers.

In a business environment, people can live without access to email for 5-10 minutes. But not to their phone system. Any system that bases their Call Manager and H.323 gateway around a NT solution is doomed. Cisco couldn't offer the reliability we were looking for

I don't really know why this IP Phone post is news, we've switched our own PBX to Selsius and then quickly took it out because of the unreliable call manager and the initial Selsius phone models were pretty bad.

We currently use 3Com's NBX100 product. It's worth checking out. It's been working great for us for the past year. Offers Layer2 and 3 telephony (why waste IP space when you're on a LAN).

Its Layer3 telephony has an "IP on the Fly" characteristic that assigns IP addresses on the fly to the phone only when it requires Layer3 addressing. This way, you don't waste an IP per phone, set up a minimal pool of addresses and they get assigned on the fly when required.

The actual "PBX" is a VxWorks powered box running an AMD Elan (x86 with integrated IO). It's got cards for analog lines, T1 lines, H.323 gateway and some more. It's offers CoS, ToS, Vlan tagging and all the things you'd expect from an IP telephony system.

There are also analog adapters, so you can plug any analog phone (we use it with cordless phones) to an end unit, which then let's you use your analog phone as an IP phone.

We've found the solution to be much more reliable than Cisco's, where you need to dedicate yet another NT box for your call manager and where the reliability just isn't there. The functionality of the Call Manager, though better now, has lacked trivial standard telephone options for a long time and just didn't cut it.

Another nicety of the NBX100 is that you can program/configure your phone through the NBX's web interface, check your voicemail through its integrated IMAP server. It's also TAPI compliant, so the Windows users can tie in their address book software with the phone system.

--
Let's not all suck at the same time please

VoIP for LANs

[bladel]

My company has been involved in VoIP for nearly 5 years, and while it is extremely sensitive to latency, the response-time and bandwidth for a single full-duplex call has decreased about 10-fold in recent years.

Maybe it's not ready to replace the PSTN yet, but would be ideal for a LAN PBX.

AFAIK, however, there are still some interoperability issues to be ironed out, pertaining to call transfering, multi-party conferencing and other call management features. Still, this is way cool, and brings us one step closer to packet-switched network.

Alaogue Phones R.I.P (we hope)

[JamesSharman]

I know slashdot is a usually a forum for slaging things of, making some kind of insightful or informative comment but I simply thought this was a really good idea.

The only issue here is will they try and patent the very principle and if not will systems from different vendors be interoperable. What we need here is a nice clean open standard so we can finally put the old phone system to rest along side the dodo, the dinosaurs and our freedom of speech.

Selsius?

[Anonymous Coward]

What was wrong with the Selsius Cisco IP Phones?

These ones are snazzier, sure, but the Selsius worked find and sounded great...

-D.

MCI bought UUNet

[J.J.]

MCI Worldcom (which is now beginning to drop the MCI part of their name and refer to themselves increasingly as just Worldcom) bought UUNet recently. They're currently in the process of completing the details of the merger within the companies. (employee pay, pension plans, chain of command, etc) For the most part, each is acting independently as of now.

While the bulk of MCI's revenue still comes from long distance, someone at the top there has some vision, and is positioning the company to do well as the Internet shifts the way they do business. They've been on a business purchasing spree over the last two years. A merger with Sprint PCS is currently getting gov't approval. The Worldcom acquisition, the UUNet acquisition, as well as many smaller companies.

One of the most interesting (read: risky and different) things I've seen them do is purchase many small MMDS companies in hopes of offering wireless Internet access at 310kbps for US$40/month. They're the first large company to attempt it on a broad scale. I believe their timeline is to begin to roll out the access on a large scale sometime early next year. They're currently in test trials in three cities in the South. (Memphis was the largest)

So yes - the phone companies do own a large percentage of the backbones in one form or another. But I don't think the phone companies can get by by simply raising access prices. AT&T will probably try, because they're a behemoth of a company that doesn't do anything well, but companies like MCI that can see the trends will, hopefully, mow them over.

J.J.

bypass phone companies completely?

[j1mmy]

Doesn't MCI own a fairly large chunk of the internet backbone here in the states? Kind of stuck with the phone companies either way.

Already done

[Wesley Felter]

The standard is called H.323; you can get an Open Source, patent-free implementation from the OpenH323 project.

Re:Bypass the phone company my tuckus.

[G Neric]

keep a consistent phone number (maybe)

yep, "is the phone number yours to keep" is the right question. In light of netJerk solutions claiming it owns our domains, you know what Cisco should introduce? A little box with with an RJ45: plug it in, and your domain is online... and yours to keep till you sell the box :)

BTW, toches is spelled toches.

Bypass the phone company my tuckus.

[MAXOMENOS]

I love this device. I do. But lest we forget, the people who own the T1 lines, and most of the backbone of the Internet, are the phone companies. It's cool in that you can avoid long distance rates and keep a consistent phone number (maybe). But we can count on LD companies raising their rates for backbone access as voice over IP becomes more popular.

The Second Amendment Sisters

Re:Time and $$ saved in wiring..other random thoug

[Detritus]

What about reliability? Where I work, we have a big Rolm PBX. It wasn't cheap but it has been very reliable. It has a hot backup CPU in case the primary CPU faults. The techs can do maintenance without taking it down. The LAN is much less reliable, probably due to the large number of components that are single points of failure. It was designed to be fast and cheap, not redundant and reliable.