Professor says agree on a standard!

Why? Because volume means cheaper? What does that mean? Well riddle me this Batman, if the world is supposed to go wireless make it easy on my local Best Buy sales rep in being able to identify a solution for me when I ask him to find me a coverage for my roaming laptop. Don’t give him a map with colors on it that will cause him to give me poor information.

Wimax is the first to take off, with deployments ranging all over the 802.1G spectrum. It has a prominent provider of CPE based gear in Zyxel. It already has developed countries singing its praise. All accept the United States. The US does not have one competitor in agreement or where to go next. Granted LTE utilizes most of the current infrastructure as 3G networks go, however WIMAX should be the future for 4G. Not because it was the first to hit the market, but because it just plain made sense.

What ever is chosen, we should know that it will exist over a predominantly IP infrastructure. 3GPP standards hasn’t even agreed on a standard, but Wimax can overly a GSM wireless infrastructure and is ready to do so now. What is the hold up? I’m tired of ordering T1’s only to be raked over the coals of its expensive and ridiculous installation costs. I want to order a connection and have it provisioned just as easy as installing an access point in the 4th floor cafeteria.

I know that is a little naive and I may be a little biased, but I have participated in a Wimax deployment and it was a work of art. Granted Wimax is not without its problems, but I would happily sacrifice growing pains then be a part of the non scalable legacy copper provisioning that has been happening over the last quarter century.

As we stand now, AT&T and Verizon will participate in the When-will-we-have-a -standard LTE. While Sprint and whoever Sprint buys next well be participating in WIMAX. What does this mean? Expensive CPE equipment that does little to fix the mobility problem the US has today. One scenario is that the rest of the world follows Korea and goes Wimax and some of us will be stuck using American only phones. The other part of the equation would be higher CPE costs because of the limited distribution of the technology and the cost that competitors imposed to stay competitive. Meaning Sprint. 802.1G was supposed to be the answer for 4G why not all consolidate our resources to agree on one standard. Count chips after that.

Way to easy. If you want a comparison of the two technologies please go HERE

This is more or less a very objective discussion on technology in general. Looking back in the early stages of IT, forums were the way to bring in new ideas and open discussions on technology strategies. I attempt to participate in a variety of discussion in order to stay afloat in my profession in order to have knowledge as to what the future holds as far as technology is concerned. It is very difficult to follow at this stage. And the need to focus on active participants from the consumer industry is needed.

Technology companies are the only companies that still utilize R&D budgets today. There are no more companies with desire to pay groups to provide solutions for business needs as far as technology goes now. This has hurt the playing field in many ways. Costs have been reduced. Solutions picked on basis of convenience and technology is driven by those who are profit driven. And have you seen requirements for CTO’s nowadays? The requirements are no longer for elite technical professionals, they are for business driven executives with a little bit of tech on the side. Where is the career path going for technical folks? 

This is where participation comes into play. Consumers should run the world, not technology companies. The flow should focus on business needs. R&D groups should be formed in hopes to provide custom based needs for corporations. Technology needs that are out of scope and out of budget should be outsourced for a vendor supplied solution. 

Supporting Example:

Enterprise verses traditional Telecommunications carriers. 

Having eyes in both areas provides me with more insight than most people have. For starters, network monitoring solutions. Most vendors that claim that their solution is the best are fooling only those who are too lazy to invest time to see what they actually want. Any solution should begin with need and followed by how we can achieve this. Most commonly in today’s workplace it is followed by how can we “obtain” this? And you will go on a bender putting two or three solutions together and achieve nothing but higher operational costs. “All” management tools can be created in house with local tools. That’s right, all, in fact if companies sat down and provided resources for excellent R&D they could save allot of money on operational costs and the purchase of commercial solutions. 

In most telecommunication companies, in house development of management tools has always been a necessity. New equipment arrives onsite and vendors participate with the consumer in developing the best API’s or MIB management interfaces so that the company is provided everything they need to manage. 

Now this is just one example and one area where this model can be helped. However, all areas can be helped by this and this is why forums were developed in the first place. To assist the consumer base, and not to assist the commercial market in developing new solutions.  Companies settle for solutions. Enterprise Security and VOIP have been bottle up into solutions and provided to consumers who take on the operational support costs and not fully understand if this solution is what they are actually looking for or does the name solve their last assessment. 

If I had the time I would put together a forum. As you can tell from the site, the professor gets busy. However, to fix the open community I would make time. I would like to take all small business into a forum and assist in providing solutions that may or may not include those that necessarily cost money. Develop security solutions that do not require a huge support contract in order to maintain. Help take open source VOIP to provide telephony at cost verses one size fits all solutions that are overkill. 

And next provide forums that comment on technology and participate on a consumer level and not a vendor level. 

For example:

In parallel to what is being developed in the mobile ip community, there is a protocol being written to provide mobility using IPv4 and can be used in conjunction with VOIP and mobility platforms. This protocol is called HIP. (Host Identity Protocol). The HIP protocol works in conjunction with dns and public keys to keep host identity intact for hosts. Consumer feedback would enable development for enterprise solutions as well as carrier solutions. You can all see how it applies to carrier based solutions, but in the enterprise, how cool would it be to have users that are quite mobile and still have access to resources provided by their own company. This is just one application, but it would be secure (certificate based) can survive without the need of a vpn tunnel.  

You could carve out a bunch of need for this technology.  In order to do so, we need to change how we participate. Lets not wait until it becomes relevant and then put in a feature request. Let’s mold it relevancy at its infantry. 

I have been involved in VOIP on and off for 10 years. So I have been eager to see it develop into an everyday phrase. There is not a phone call being made during the day that does not involve VOIP to some degree. Even if you still have that Verizon line at home and haven’t given in to Vonage, Comcast, or Time Warner, you have to know that even your TDM Verizon connection traverse a backbone of H.323 and SIP trunks to make it to the DS0 that is connected to your Grandma’s rotary dialed telephone.

Professor’s Conspiracy Theory

Q1: If the government can force you to go digital television, why won’t they force you to go VOIP?

Q2: If I purchased VOIP service from Comcast or Verizon, then why haven’t they converted my traditional lines with the same lines I use to connect my computers? (Cat5)

Q3: I hear about VOIP, I even have Voip service, where are all the bells and whistles?

Like you I have always asked these questions. When you order VOIP service from Vonage, they provide you with an ATA to convert your analog to digital. They also provide 802.11 and regular ethernet VOIP phones for you to connect to their service. They are not a big enough outfit to provide CAT5 cabling to you as well, but for the most part they provide the cheapest offering because they only have to invest in their core and infrastructure. You depend on your Internet Service Provider.

When you purchase VOIP service from Comcast, they have already done the math. They connect your Telephony Demarcation directly to the MTA so that all of your phones will not be changed and all of your existing wiring will be used, rerun or maintained.

This isn’t because they are making your life simpler. Your life would be simpler if you can purchase an 802.11 (wireless) cordless phones and have video conferencing from an IP handset in your bathroom. Softphones for remote travel options. In other words, IP flexibility. However, think of all of the money Comcast will lose if they no longer had to send a tech to your house because your know using 802.11 wireless router from Best Buy, which is plug and play, and not have to send a tech to search your wires and charge you $90 per hour. Dem boys are Union!!

Comcast has the technology to provided these services. They also have the technology to provide ip presence and other IP related features that exist today. But they don’t.

Okay, Asterisk: How to live free!

What exists in the market today are two different models to support subscribers. An IP PBX and a Class 5 server. Both are similar in terms of provisioning application based services, but are different in terms of scalability. You will not configure a PBX to handle 10k CAPs. (Call Attemps Per second).

Asterisk is an IP PBX. Asterisk can provide voicemail, text to voice, voice to text, trunks, extentions, anything a TDM pbx can provide. Asterisk is more or less a core system. You will have to provide a front end. Other than that I would put Asterisk against any of the PBX’s that I have experience on. Such as the Nortel CS2k/ CS2100 and the Cisco Call Manager.

For the purpose of this lesson, I will tell you how I use Asterisk and how it provides a very cheap and flexible alternative for my home living. I have built an Asterisk server for the sake of providing a cheaper service. Asterisk is free, however the time it took to compile and get working was less than 8 hours. There are many white papers that would assist you out on the web.

Cost: Minimal. I used an old computer with a P4 processor. I purchased refurb for $150. I did not purchase a Digium card used to connect to TDM trunks. So I am not using a T1 or DS0 to connect. This is a purely sip connection. So total price of hardware would be $150

Service: I chose to go with a very lightweight service provider. The quality is not perfect, but it isn’t bad either. I am tinkering around that 50ms range which becomes noticeable. But I am pay $14 a month for the service and a DID.

So with $14 a month I am connected to a Sip service provider and I am paying .0012 per call globally. With comcast I pay $19 a month for all US calls. So if I am just comparing Nationally based calls then I make a very minimal profit due to quality concerns.

Applications:

Voicemail is additional with phone service. It is included with Asterisk. Nice part is that with Asterisk you can do what you want with your Wave file. What I have done is configured my Asterisk server to email my wave file to me when I get a call. What I am working on is providing voice to text emailing. There are services out there that I can send my WAV file and they email it back transcribed, but that thats the fun out of it. So bottom line is free.

SMS: Now Comcast supports SMS to Voice services for free. Right now it is a demo, but there is a Perl module that you can use to send SMS to a public SMSC that will forward your messages, but that will take text to voice and I am not a big fan of text to voice. So until that changes I will not be doing this.

Mobility: I can connect to my IP PBX via a soft client. I use many of them and they are all free. One that is readily accessible is Xlite. Comcast has yet to support softclients so this is an added bonus.

Now the way I have my service configured is that everyone in my home is a different extension. So I only have one DID, but for an additional 10 bucks I can get more DID’s. There are other services cheaper, but the total cost of the sip trunk goes up. I don’t need that. I rather have an operator ask you want extension you would like.

Manageablility:

Now you do have web access for Comcast, just like I have web access to my server and any subscriber has access to alter his/her account. I like mine better because it is free. But you can go on the server and look at your voicemail in .WAV files.

Asterisk is a very cool tool that is free and very customizable. I like using pure IP, but you can purchase a card that will have you connect via TDM if you wanted to do so. Only problem I had is RTP proxy. You must have a firewall that does RTP proxy or your voice will never make it in. I use Ubuntu as a firewall so I compiled it very easy. And it is activated in my ipchains rules when every my firewall is restarted.

For a medium sized organization looking to go VOIP, this would be a perfect solution. Even for a large size corporation if you scale it correctly. However, Asterisk is not for the technically weak. So you must be willing to get your hands dirty and your solutions hat on hand. Enjoy.

Ten years ago, majority of the information systems staff were operators. People who managed a system around the clock using commands that were supplied to them in hopes to keep the network up and running 24/7. This included making backups, verifying disk space, and also making sure they ran certain commands that needed to be run at off peak hours without fully understanding what the purpose was. It wasn’t their job to find out, it was their job to perform what was asked of them. Similar to my grandmother when she calls Dell support to find out what is wrong with her internet connectivity.

There are thousands of professionals out there who take what is provided to them to perform certain functions day in and day out. These could be network professionals or System Administrators. It is not solely their fault, this is what has become of the field due to vendors who advertise “IT for dummies” solutions. If you are one of those people who purchased a CCNA book, got CCNA certified, and wondered why your salary hasn’t broken the bank yet, you are one of these people.

I have never seen IT that way and I refuse to look at it that way . For every problem, there is a solution. Just because no one has provided a vendor solution, it doesn’t mean that it doesn’t exist.

Example:

I was working a Carrier VOIP project. The softswitch I was using supported SIP, so it was a SIP registrar. Every night at a certain time all SIP connections were going down and causing endpoints to re-register. This problem had been occurring for a long time and without the there were finger pointing without much proof.

I always travel with a linux (Ubuntu) laptop. So what I did was configure a Sip registrar (Asterisk) on my laptop and croned a nmap job as well to verify open ports to the endpoints. I used another script to register to the laptop so that a session was established to another registrar other than the Class 5 server. That night at 2am the same problem occurred. The client lost connectivity and nmap caught the ports that were closed.

The problem easily pointed to the firewall that was setup to detect and act upon attacks and there seemed to be a very mild DOS using port 5060 that cause the router to reject traffic to that port for a period of time. Simple automated approach.

I just used this example because tools aren’t purchased they are improvised. You can ask all of those idiots who purchased a flute years ago and are still trying to find everyday use for them.

People go out and purchase network management tools and utilities everyday  all they are is a bunch of scripts running snmp get and set packaged in pretty java/html pages. When the same can be done on a regular linux platform with cron and referenced in man pages. Why wait for the vendor to upgrade the software to manage switches now, when you can walk the MIBS of the device and update your own scripting.

This is what is wrong with IT. If the paid “network engineers” and “system administrators” cannot develop tools of their own then what is their value add? You can pay the vendor to train a monkey to be an operator.

Three months ago I wrote a perl script for a client that enabled him to automate a packet trace to isolate a problem that was occurring off hours. Every night from 2 to 3 am the script took the trace and then ftp’d it to my ftp server so that I could review and take a look at what was occurring. Now the client is looking into 3rd party vendors that could do the same for them because they fell in love with the concept. I did not have the heart to tell them that it was only 5 mins of work, but this reinforces my view.

Technology is ever changing. Things you purchase today have been available for years. Just because Cisco or Microsoft offers it does not mean they invented it. There are always cheaper, more flexible and cost effective alternatives. Be an engineer, not an operator!

Searching for a security candidate provides you with individuals that mostly know security from an application standpoint. Knowing how to configure a firewall makes you less of a security engineer as it is to know how to spell security. If you do not know how to decipher an ethereal trace or to expose typical threats on a a user’s pc, without the use of tools and scripts, then what makes you a security expert?

To know how to configure a router is far easier than to know the in’s and out’s of routing protocols. Why would you consider yourself and expert if you know BGP, but cannot articulate when and why to use RIP? Isn’t being an expert knowing when to use a certain technology and tool? I blame that on Cisco, but only a little bit. I mostly blame the industry. Again to properly hone your skill set and evolve with technology, but know the basics is what is missing in networking.

If you qualify yourself as a VOIP expert, then I take my hat off to you. Voice over IP is a hybrid of Telecom and IP. Knowing Voice over IP is very demanding because if you do not know the ins and out of TDM technology, you will not know how it is supposed to work in the IP world. Being able to bypass tolls have more to do with how to design a Voice network and less to do with how to get Voice to properly work in a IP environment. The routing is more complex and is disguised in complex and confusing Dial plans, and the rules you configure are enough to allocate a whole server to perform. Yet, there are certified experts in the market who can configure a call manager and run a Voice probe.

An expert is a term that goes hand in hand with perfection. No one can ever obtain this title, but only the good ones pursue because the chase is worth it.

For me this became the very first commercial push for VOIP that I could remember. The enterprise was ready, but VOIP wasn’t. H.323 was built for voice and video, and found it’s niche being used to trunk particular gateways together. A VOIP gateway is used to mediate between TDM and standard IP. However, for signaling between the phones themselves and the callmanager, Celsius created SCCP, otherwise known as “Skinny”.

This was the first of many errors vendors made when developing products for VOIP. Cisco wasn’t the only vendor with a proprietary protocol used for delivering services and call setup in VOIP, all of the major vendors utilized proprietary protocols. Remember, this was in 1997 and even though companies bought into the idea of VOIP and cost savings, none of them had any models to build from.

Today, it is a different story. Carriers worldwide have converted their core infrastructure as a necessity to provide the service demands at a reduced cost. “Every” major telecommunications provider today provide phone service off a predominately SIP core. Meaning a call coming in from San Jose, CA destined for Boston, MA will be routed over a sip trunk over an IP backbone to terminate to a VOIP Gateway in Boston for call termination. A a cost savings mostly because of the compression ratio of proprietary Gateways on the market. (Some that offer up to 10:1 compression). Not only is sip being used for IP trunking, sip is also being used to provide phone service for Vonage and many of the most popular instant messaging clients worldwide.

VOIP offerings differ as you move out of the US. I have been a part of many projects in which IP phones are being used in the Residential and Enterprise domains in some cases making the need for tdm connectivity obsolete. Witnessing a housewife make a phone call via an 802.11 phone makes my heart warm. Still in the US we deal with corporations still rolling out proprietary units without looking at the maturity of the technology.

Each year more and more VOIP phones are arriving on the market. Cisco has devoted resources to develop their sip offering. They have already released SIP trunking capability in their product line. With the maturity of SIP and their competition growing, Cisco has to realize that their primary selling point, “their phones”, will be challenged by lower cost and equally reliable IP PBX solutions that offer Open Source interoperability and freedom to use any phone with equally as rich feature sets.

Sooner or later, people will realize that there is no true benefit by going with a specific VOIP vendor. Telephony will still be as autonomous as it was in the past, the only reliance is the infrastructure. And when that is realized, let’s see what Cisco comes up with next.

Note: I pick on Cisco, but Cisco is not the only vendor that tried to fight Open Source interoperability. However, it is the most successful vendor in promoting it’s VOIP porfolio as an enhanced feature set of it’s Network equipment offering.

I travel quite a bit and I am one to tell you that if a movie came out in the states, I will have to wait 6 months to see that same movie in Europe. Due to piracy, this time table has moved up dramatically. I now have to wait sometimes less than a month. This will go a long way in promoting cable television and on demand services. When in Kuwait this past January, I was practically watching a close to real time Apprentice Celebrity edition. Not bad considering that prime time television is the entire day.

The need for IPTV is not seen as a necessity in the US as it is in other countries, mostly third world, who already exploit the benefits Voice over IP and Video on demand services, but it should be. The US economy have long relied on technology convergence. While fiber is being rolled out in most countries at a record pace, the US is still focused on finding a technical strategy.

Dubai is one of the countries that are leading the pack. Along with Tandberg, a company known for video conferencing making a strong push in IPTV with proprietary measures to transmit MPEG-4 to a nation and not just to an organization.

Operator- An individual that performs routine tasks and maintenance. The duties can be related to backups, running routine scripts (thank God for cron, network health monitoring, and/or handling some System Administrator tasks.

System Administrator- Glorified Operator. The tasks differ because the System Administrator is responsible for providing tasks and duties to the Operator. User management is the ultimate responsibility.

Network Technician - Responsible for maintenance of the information infrastructure. Responsibility includes network drops and troubleshoots network related issues.

Network Engineer - Now this is a multipurpose title today and could have more uses than comet, but it shouldn’t be that way. A Network Engineer used to be a title that represented a person that integrated network equipment into an infrastructure. To be more specific, when I was a technician I used to have to be on-call when ever a network engineer was bringing a new site up online. Troubleshooting and Integration are two different operations within the network world.

When I was a network technician I was allowed to troubleshoot routing issues, however if i needed to make changes I escalated to get approval. As an engineer you make configuration changes and introducing another router or switch to the environment really takes someone with experience that knows what they are doing.

Network Architect- The role of a network architect is basically keep an overall picture of the infrastructure and design solutions to best provide to the overall mission of an organization.

Solutions Architect- The role of a Solutions Archetect is the same as a Network Architect, but not as narrow. A typical Solutions Architect may work for a professional services outfit and be responsible for researching and designing solutions that are not limited to routers and switches.

I just rattled off most of the titles that have been around for the last 10 years. All of them are summed up today as a network engineer and that is what makes the field frustrating. I remember looking at a resume of an individual who was a Senior Network engineer. I was really impressed with his resume so I brought him in for an interview. When doing interviews I allow the person I am interviewing to control the topics we discussed. This interview exposed the fact that this individual never configured any routing protocols, and could not give me one any sized project that he lead.

With that little of experience you shouldn’t have the right to put network engineer on your resume but people do. In fact, it should be embarrassing to use the title Senior Network Engineer if you cannot perform with mastery of routing, switching, and advanced networking tasks. If you have been a network engineer for 20 years and you still cannot perform these tasks then I am sorry, you are still a Network Engineer.

Specialization fits in the same category. Do not call yourself a Sans Engineer if all you know how to do is manage a Sans switch. Please do not call yourself a VOIP engineer if all you know how to do is configure a call manager platform. You must know the technology and become a reference of sorts on the protocols and inter-working of your craft.

Which reminds me, I once interviewed a person who had a countless number of the word “Expert” on his resume. His title he claimed was “Solutions Architect of the America’s”. See why I had to bring him in for an interview? To make a long story short, he barely knew any of the information he represented on his resume. He was basically a project manager of all the said “designs” that were listed on his resume. I found humor in interviewing him, but sadness in what this field has actually become. Someone will hire this guy with dreams of grandeur, and be thoroughly disappointed and maybe bitter. It will reflect on the rest of the network community.

I misrepresented the term hierarchy when in fact all of the titles and duties that I presented above are autonomous of each other. Needless to say that a good technician would not neccessarily be a good engineer. Most organizations interchange these duties when in fact all you are doing is creating guys that are proficient in the art of cut and pasting when it comes to newly introduced configurations. And a good Network Engineer may not be suited in the art of troubleshooting, however they may been very good at integration.

I once hired an engineer that knew networking inside and out. I never had to proof his configurations. He knew all advance routing protocols and was very proficient in all other technologies he was expected to know. He was a very poor troubleshooter. If it was a BGP issue he would most likely find the issue, however if it was a reported issue that the root cause was not known, he could not deal with eliminating other variables that may or may not be network infrastructure related.

I could use myself as an example, if you want a patch cable made you may get one faster if you ordered one online then wait for me to make one. If you ask for a 10 foot cable, you may get one 8′ feet long by the time I was done making it. I can not express enough the importance of a good network technician. I have had the privilege of working side by side with a lot of them during my career.

Soon we will be required to submit to lie detector tests and have full background investigations. All because individuals cannot properly classify themselves. You heard it first from the Professor.