GLG News by Samuel Greenholtz

Implications: 1.      Level 3 is faced with an expensive, work-intensive, but doable undertaking of dealing with SONET boxes that came from a couple of dozen different suppliers. 2.      One way would be almost akin to building a dam for a river – only doing it numerous times around the network. 3.      Level 3 will need to have a totally clear vision of how it wants to end up at the conclusion of the process.

Analysis:  In constructing a dam, one puts kind of an alternate river around where the dam is going to be built.   The dam is completed – close off the alternate – then the river goes back to the original path.  In the same way, Level 3 could take each network –build something parallel to it – dismantle it – and then bring it back.  Essentially, it is a stepped program to migrate customers to an intermediate network until the carrier can get the complete integrated network working.  So, it does entail two moves for every subscriber.  Three to four hour network maintenance windows would need to be scheduled in the middle of the night on a customer-by-customer basis (with their approval).  Of course, just because it is at night, it does not mean the subscriber wants the network to go down.  In fact, the networks are often in use 24 hours a day.   

A more typical approach minimizes downtime.  Carriers do an overlay on a new pair of fibers for all of the traffic – build all of the circuits on a new set of boxes – have it all ready to go and then essentially at the end points, go off and flip them over – just move client connections from one box to the other box.  In this way, the service provider gets more customers done and is going to affect fewer people for a shorter period of time.  

Implications: 1.      Run-of-the-mill ISPs might be very attracted to operating a municipal fiber network, especially one with open access. 2.      However, they lack the necessary experience and support equipment. 3.      Also, with not much skin in the game, the level of their commitment is going to be questionable.

Analysis:  Look at a successful ISP that has been in business for a while.  The company started out doing dial-up and then it shifted over to unbundled copper loops from the phone company -- and it is selling to both businesses and residences.  It likes the idea of the city’s fiber optic infrastructure and it says that it will come in and run it.  The management has certainly done marketing before.  But is it prepared to conduct door-to-door sales efforts?  Because that is what it is going to take, period.  

Let’s say the ISP believes it can engage in such rigorous marketing activities.  It seems like a pretty sweet deal for the provider.  Potentially, it gets an exclusive for triple play (while also competing against, let’s say, Comcast and Qwest), but there is no upfront capital necessary for the ISP.   

Then one starts drilling further down.  What is the company going to use for a switch? Perhaps the response is that it is talking to Avail about an IPTV headend, which is about $200,000 to $400,000.  Of course, the provider really knows nothing about video really or any of the issues.  The ISP is asked again – what are you using for a switch?  The answer might be “We’ve got this Linux server in the backroom.”End of discussion.

Implications: 1.      There are signs at British Telecom that it has been frustrated with Huawei regarding optical transport systems. 2.      It is not out of the question that the Chinese giant could be com pletely knocked out (at least in this particular space) at BT. 3.      Both Ciena and Infinera are in the running to displace Huawei.

Analysis:  Apparently, British Telecom has become disenchanted with Huawei in terms of supporting its transmission infrastructure products.  There are rumors that the products are quite as advertised. 

A pullback at BT would obviously be a major blow to Huawei because the supplier (with no incumbency) became a preferred vendor for the carrier’s 21CN project for transport and other gear.  The win has been considered a key stepping-stone to other potential service providers outside of Asia. However, you can be sure that Huawei will not fade into the sunset quietly. 

As with other PTTs, Infinera is putting on a full-court press in getting into BT.  Earlier in the year, there was some speculation that the vendor was going to get a deal with the telco.  It is likely that Infinera is going to try to undercut Ciena on price.   

The situation for Ciena at BT is exceptionally good right now with a contract that remains solid  The carrier has been very pleased with how the products from its incumbent vendor work including the 4200 and the CoreDirector.  Evidently, there have not been any major problems with Ciena’s equipment – and any trouble was serviced in a timely fashion.       

Implications: 1.      The new Sirius XM is now the second biggest radio company in the US with revenues of $2.2 billion.   2.      It is only second to Clear Channel. 3.      The odds are better that Sirius XM becomes the largest radio company in the States than whether 4G goes beyond “functional aspects.”

Analysis:  The author of the source article writes in the middle of June that XM and Sirius “have managed to attract only 17 million subscribers” – (actually the combined number is now close to 19 million subs).  To put it in perspective, the largest CATV provider, Comcast, has fewer than 25 million subscribers.   

Because of government inertia, these two companies have been in a state of limbo for about 18 months.  Now a larger combined audience for its most popular shows will mean even greater advertising revenue (not all of its stations are ad-free).  

One of the firm’s biggest strengths is its CEO, who has an outstanding track record in the radio business.  Mel Karmazin got the merger completed against great odds.  (For example, the two satellite TV companies, DirecTV and DISH were not allowed to combine.)  And he did so without making one draconian concession.  Now Karmazin is highly confident that the company will eventually become profitable – and perhaps he should be now taken even more seriously (no pun intended).  

It is a risky game to predict the demise of any type of well-established radio technology.  AM has been around for over 80 years.  There is a constant audience for exclusive talent, particularly live shows and events, in a closed network situation such as Sirius XM

Implications: There is a high likelihood that Qwest will expand its optical equipment budget for its interexchange network.   2.      Qwest has evaluated at least three suppliers for 40G technology: Ciena, Infinera, and Nortel. 3.      In addition to other factors, Ciena’s seven-year plus relationship with Qwest, gives it the upper hand.

Analysis:  Infinera has been pushing hard at Qwest.  There had been conjecture in the past that the carrier might have had its system in its network – apparently untrue.  It appears that Qwest does not feel totally comfortable with going with the relatively new manufacturer.  Price is evidently not an issue for the vendor.  In fact, in order to make penetration into tier-1 service providers, including its recent win at Deutsche Telekom, Infinera has evidently been very aggressive on price.  

Qwest likes Nortel’s 40G solution.  However, it seems that its prices were considered to be unacceptable to the service provider.  

As a result, Qwest is likely to go with Ciena.  At times in the past, the carrier’s purchases have represented a high proportion of the vendor’s revenues.       

Implications: 1.      As the source article points out, carriers across the board are stressing managed services -- their version of ‘convenience.’”   2.      So, any positive impressions by equipment suppliers in that space are likely to influence decisions on equipment selections in the public local and long distance sides of the RBOCs. 3.      For optical vendors in particular, the timing in getting into enterprise networks that are run by the RBOCs is good right now.

Analysis:  Before writing out a fat check to Telcordia for OSMINE approval, manufacturers will often test the waters first by trying to get into the managed service businesses of the RBOCs.  Historically, Nortel was perhaps the most proactive of any supplier in encouraging its enterprise customers to pressure the telcos to use its equipment.  The problem for Nortel on the optical side is that its 5200 is old and very expensive, opening up the door to new solutions.  

It appeared that Cisco Systems early on with the 15454 got traction in the managed  services operations of the RBOCs before being deployed in the public networks.  Of course, once a supplier breaks through, it does not necessarily mean it is a permanent situation.  Cisco appeared to play pricing games and was not too responsive on requests for upgrades – and thus, the supplier lost its grip relatively quickly in that market segment with the 15454.  And like Nortel’s product, the 15454 is getting long in the tooth and its previous disruptive cost advantage that Cisco was able to offer in the past has disappeared – again, providing opportunities for replacement.  

There are lots of other examples.  ADVA’s success in the enterprise allowed it to get a shot by the RBOCs.  Unfortunately, the supplier lacked the organizational infrastructure to deal with the complexities of these big carriers.  Ciena’s 4200 has been an eye opener in terms of competing so aggressively on cost that one or more of the RBOCs has actually lost business to the vendor.  In addition, taking this path will eventually be the way that Chinese vendors, such as Huawei and ZTE get into these Bell companies in a big way.  

In addition to OSMINE often not being required for managed services, there are not the long RFP cycles that one sees at the local and long-distance divisions.     

Implications: 1.      There appear to be some clues that the first thing that Infinera will do to play in the metro market a little better is to develop a 4 x PIC. 2.      The vendor should be able to slowly work its way into the space. 3.      It is becoming more known that Infinera’s involvement with Level 3 goes much deeper than what appears to be on the surface.

Analysis:  Infinera will likely need to do more on the aggregation side to get into the metro space.  It will have to make the transition to grooming lower rates and things of that nature.   

A 4 x PIC would certainly give Infinera a cheaper offering, especially as compared to its 100G transponder.  But a 4 x transponder is likely to still be considered expensive in the metro market.  Therefore, Infinera will need to target the bigger guys and getting into additional Tier 1 service providers, beyond Deutsche Telekom and the ex-MCI, will be even more critical.  Establishing a solid metro niche business might be the most likely outcome.  On the manufacturing side, achieving the necessary yields will take on a greater importance.  

There seem to be signs that Level 3 may have a metro prototype from Infinera in its lab.  It is also becoming common knowledge that as part of the Level 3 deal, the supplier acquired some of the intellectual property of the Corvis technology in the ex-Broadwing network.  Evidently, Infinera is maintaining that equipment for the carrier.    

Implications: 1.      Several years ago, the difference in the terms, 2 1/2G and 3G, were not clear. 2.      Certain technologies/releases that were recognized by standards bodies as being third generation conflicted with the perspectives of the trade press and others that it was more 2G+. 3.      Just as what happened with 3G, suppliers are going to stay as far away as possible from a completely 4G standards-based solution in order to maintain competitive advantages.

Analysis: There were several examples in the past in which there were debates as to whether to say solutions were between second and third generations – or just the latter.  For instance, it was not certain whether EDGE should be considered to be fully 3G.  Also, despite IS-2000 being recognized as a 3G technology by the ITU, it appeared that the press still muddied the waters by referring to it as 2 1/2G.  

Of course, carriers are going to prefer that the promise of 4G will be the ability to use multiple vendors in an end-to-end path to manage costs and encourage technology development.  They will want the manufacturers to adopt a totally standards-based implementation of the next generation architecture.  However, this kind of dream will never happen.  It is inevitable that when the industry starts with a multi-vendor approach, the end result will be rammed down to the lowest common denominator of all of the suppliers, including Ericsson, Alcatel-Lucent, etc., -- and they would become noncompetitive quite quickly.  So, the major obstacle to the multi-vendor approach is all of the detail that manufacturers will concentrate on for differentiation – all of the network and configuration management issues – all of the actual little widgets of functionality that kind of look irresistible when you pick the thing off the shelf. 

Implications: 1.      During the telecom bust, some folks in the industry actually would hardly have predicted that Nortel would wind up investing a fortune in 40G technology for the long-haul market.   2.      In developing an entirely new system, the 6500, Nortel seemed to sacrifice a lot of its existing customer base. 3.      Therefore, it opened the door to other players.

Analysis:  Evidently, Nortel’s new 10G system was not terribly successful.  The zero-dispersion-compensation story did not play well.  At the time the product was being built, dispersion compensation went from being very expensive to almost free.  But the biggest problem was that Nortel did not make the 6500 work on its existing line systems.  And it pretty much abandoned its older product.  

Nortel’s competitors were happy to play on the fact it was forcing its customer base to go to a different solution.  Ciena appears to have been wise in staying with its CoreStream line and just construct a new multi-reach generation that addressed both regional and long haul to leverage its existing base.  The supplier apparently believed with the technology that was out there it could do enough to its existing box to continue to differentiate it against brand new systems.  

While the whole idea of re-attacking the high-end optical space is a good one, with Nortel appearing to have a lost a good deal of its 10G business, it has made it harder for the company to introduce its new 40G technology.  While the vendor may tout about 14 customers for its 40G solution, it really demonstrates that it still has a lot further to go in trying to get back that customer base.

Implications: 1.      The Long-Haul (LH) hardware market is much tighter than in the metro or in the access markets in terms of margin. 2.      It is an interesting dynamic in the industry because at an initial glance, one notices that there are fewer equipment suppliers for interexchange purposes. 3.      However, they are all desperate to keep the business – and there are fewer opportunities in the LH.

Analysis:  Of course, the margins on long-haul service providers are the hardest to generate of any telecom sector and so there is a greater amount of pressure on the manufacturers to keep their prices low. Another big part of the problem is that LH is not the same everywhere.  In the U.S., it is a 2,500-3,000-kilometer kind of system.  With perhaps the exception of China, India, and other places in Asia, LH is probably 2,000 kilometers or less in the rest of the world.  Therefore, the vendors that produce systems to go a couple thousand kilometers or more really have a very limited number of carriers that see value in it – and unfortunately, there are still at least seven or eight notable players including Ciena, Infinera, Alcatel-Lucent, Fujitsu, and Nortel.  Anytime one of those deals comes up that uses the specific technology that gives them an advantage to go more than 2,000 kilometers, there is still a willingness to be more aggressive on price to get that business.  

In fact, it could be argued that for the past decade, if not longer, there were probably too many vendors in the LH space.  When the bottom dropped out during the bubble burst, expectations by some people that suppliers were going to bite the dust left and right never came to fruition.  With the inevitability of further LH carrier consolidation, as much of the wholesale service market continues to decline, the number of opportunities will decrease further. 

Implications: 1.      It seemed that in the past, the safest strategy for a fiber optic vendor with a new technology was to go up to the subsystem level, especially because the timing of market takeoff is always so uncertain. 2.      There was supposedly the best of all worlds – far up enough to avoid getting killed on cost – but not having to deal directly with the hard sales/support processes that involve service providers. 3.      However, it appears that Opnext was able to get StrataLight at a reasonable price because of the apparent generic weaknesses with the latter’s subsystem approach.

Analysis:  Despite having potentially compelling technologies, optical newcomers will have to think twice about going with a subsystem strategy.  First of all, it is at the subsystem or even in some cases at the lower module level, where system vendors generate most of their margins.  So, as it turns out, the idea of buying it through a third party is not that appealing.  Most vendors will sell in commons cheap to get footprint and will make their money on the transponders.  So, unless there is a desperate need, it is hard for subsystem vendors to hold onto business for the long term.   

The models of suppliers such as StrataLight (before the purchase by Opnext) and Mintera  are apparently less likely to be successful in the long run.  The threat to such vendors is that the system equipment vendors will utilize them, in this case, as long as 40G was not cost-effective.  But once it proved in on cost, and running in volume, these full systems suppliers would just develop their own solutions – leaving the subsystem players out in the cold.   

The other problem in going with a subsystem business is that there is not on-going, direct contact with the end-customers.  One of the most valuable aspects of an optical product manufacturer is the ability to leverage an existing installed base of service providers to build other applications.  The only possibility is if the OEM leverages the subsystem supplier to develop new relationships with the buyers – which is not always going to happen.   

The subsystem manufacturer could try to directly get in the door of the carriers themselves and engage in a push-pull method of getting those telcos to put pressure on their systems integrators to adopt the new technology.  StrataLight was successful with this approach.  However, the ultimate goal for a subsystem vendor is to get bought out by a systems vendor – or in the case of Opnext – a larger component company.  It appears doubtful that StrataLight could have remained an independent firm indefinitely.

Implications: 1.      It is always a good thing that there are firm believers in technology.   2.      However, the idea that 100G deployments will be realized in a couple of years is just implausible – and that it will almost certainly drag out for a much longer period of time.   3.      A reasonable, optimistic projection for the technology is at least six to eight years.

Analysis:  It would not be shocking if it took a very long time before 100G took off in a significant manner.  Standardization has to catch up, and that is the least of the issues.  Just as with 40G, the focus in the industry is on polarizaton muxing.  The use of 10G would be spectrally inefficient in the case of 100G.  So, everyone is looking at four 25G streams, which, of course, lacks the performance of 10G.  In seeing what is available off the shelf along with ASICs that a manufacturer will have to develop itself, the cost is just way up there.   

Even the next-generation 100G solutions that are being talked about for 2010 will likely be very expensive.  They will be more than 10x 10G and more than 2 1/2x 40G.  And it is doubtful they will have the 2,000-kilometer reach – probably the minimum requirement to overlay existing networks. 
 
There are at least a couple of vendors saying that they can get as much as 12db gain on FEC to try to get 100G up to 2,000 kilometers.  But these techniques will be hard to implement – and the problem in the past has been the amount of overhead that would be required.   Late last year, it was announced that Verizon experimented with Alcatel-Lucent’s   100G gear. 

However it was for a short distance (just over 500 kilometers).  The whole project had to be astronomically expensive (although the RBOC probably did not pay for it).  But the most incredible statement in the press release was that “[t]he move from 40 Gbp... to 100 Gbps will be exponentially quicker than the move from 10 Gbps to 40 Gbps.”

Implications: 1.      We are talking about a telecom incumbent that has been around for over seven decades.   2.      Some industry analysts apparently look upon what amounts to be a very small adjustment in expectations at ADC -- as a significant development. 3.      With just two large telco powerhouses in the US, it should be anticipated that subtle, unexpected shifts could occur – such as with FiOS sales.

Analysis:  The apparent trend with FiOS at Verizon is that during the warmer months in the second and third quarters, there is going to be a marked increase in installations – in particular, in the backbone, where ADC really benefits.  In the colder first and fourth quarter months, the stress is on selling and connecting FiOS customers.  In the first quarter of this year, Verizon gave away free 20-inch HDTV sets, if the subscriber signed on for a two-year commitment.  As a result, sales were especially huge during those three months.  In the second quarter, there was nothing significant offered as an incentive – but the results for FiOS additions still compared favorably to performances announced in the past.  

The situation is that the score for FiOS additions sticks out like a sore thumb – because it is all about growth.  In comparison, the net increases in business lines will not be as noticeable.  Verizon’s sales performance in the first quarter essentially blew everything out of proportion – and the RBOC could probably foresee that the latest announced numbers were bound to result in a negative reaction in the industry.  ADC appears to be indicating that the RBOC will now put somewhat more of an emphasis on selling FiOS services – with more of a tilt toward gear interfacing with the CPE – to make sure that the third quarter is hopefully even better.  In particular, with a major advertising campaign just kicking off for NYC, Verizon feels even more pressured to show a good outcome there as well.

Implications: 1.      Despite assertions to the contrary heard in the industry, the 40G market has still not taken off. 2.      Despite implementing 40G with some sort of 10G streams right now, with 10-gig optics continuing to drop, the cost of these systems is still above 4 X 10G.   3.      The only reason that the space has gotten to its current position remains that Cisco Systems has been pushing it.

Analysis:  Cisco Systems has successfully used 40G as a differentiator.  It is still the only factor driving 40G volume.   Cisco is telling the world that 40G interfaces perform better on routers.  However, a router is a logical device.  Cisco could just treat groups of four 10G ports as a logical port – and base all of its routing algorithms on groups of four.  

Essentially all of the vendors are implementing 40G with some kind of combination of 10G streams right now – they are polarization muxing them into a 40G stream.  The expectation is that they will ride the economies of scale as those applications are rolled out – resulting in cost reduction.  But the polarization mux givers up a little bit of the performance characteristics of the 10G.  Also, there is not a good migration path from 10G to 40G – with all of the new implementations, they cannot sit next to 10G waves without disrupting/interfering with them.  Therefore, guard bands between the 10G and 40G are necessary – in other words, providing large empty spaces in the spectrum to protect the 10-gig waves.  Of course, that just winds up wasting wavelengths.  Moreover, there continues to be the problem of not getting a full 40G worth of capacity in using four 10Gs.   

Implications: 1.      Getting Rob Adams was a big coup for Ekinops.   2.      As a high-level veteran of Ciena, Adams gives the newcomer a lot of credibility.   3.      Ekinops absolutely believes it can compete against Ciena successfully.

Analysis:  Ekinops makes an astounding assertion in its presentations combining its “T-chip” along with optics to create different transport systems.  The supplier tells carriers, from a transport perspective, to forget everything they have heard about getting to Ethernet quickly because it is more cost-effective.  It claims that it can deliver all protocols at an Ethernet cost point.  And Ekinops states that its chip technology can enable delivery of any combinations of the following while staying at that same low cost: 1) client and line protocols (including SONET/SDH, Fibre Channel, ESCON, video, etc), 2) lengths (from access to ultra long haul), 3) varying amounts of wavelengths options, 4) flexibility on rates (2.5G/10G/40G, etc.), 5) various capabilities (ADM, FEC, etc.) – while claiming interoperability with existing systems in the network.  

Ciena made great headway with its programmable port technology.  Ekinops argues it has taken this innovation to the next step.  In downloading new firmware along with the programmability of its T-chip, it can give the carriers the flexibility and speed to upgrade a customer’s card in the field.   Ekinops says that its solution is more cost-effective because it is a single chip on a board (instead of having eight to nine chips) – providing a much smaller form factor and lower power consumption.   

Ekinops contends that Ciena could not reach this capability without a total re-architecting   of its systems.  Undoubtedly, Ekinops is hoping that a supplier, such as Ciena, will ultimately buy it out to get to this next-generation capability, assuming the former makes a significant amount of penetration into the market.

Implications: 1.      As far as the two large RBOCs are concerned,  regarding the TA 5000, Adtran has ironed out all of the inevitable bugs that arise with any new product. 2.      With both companies representing a combined 35 percent of its revenue last quarter, it is important that the supplier has a new solution, which could be used by these telcos in a very big way for several years. 3.      There is an excellent chance that British Telecom will be buying Adtran’s new box.

Analysis:  The product evaluators at Verizon have been rumored as giving the thumbs up to this product -  TA 5000.  With it being determined that the solution can do everything as advertised, it is now possibly in the hands of the sourcing folks at Verizon.  The RBOC appears to be in the process of making a  deal for the solution.  

There are indications that AT&T is doing some trials with the TA 5000 and that the equipment is progressing well there.  In being further ahead than Verizon with the gear, it would not be surprising that within the next six months, deployment of the solution will start to happen at the biggest RBOC.  

Conservatively speaking, the window of opportunity for the TA 5000 is at least a billion dollars worldwide over the lifetime of the product.  It seems that one of the most important functions of the gear at the incumbent telcos is that it facilitates changing their mature, established networks to Ethernet.  

British Telecom, which has lately been inclined to buy the same kinds of products as the RBOCs, is undoubtedly taking a hard look at the TA 5000. Adtran fits the mold of other mid-tier suppliers that BT has been focusing on lately – including Ciena and Tellabs.  

Implications: 1.      Infinera decided to keep its chip as a proprietary ASIC. 2.      So, it has been the only manufacturer of the technology. 3.      What could Infinera have done differently to improve its competitive position in the market?  

Analysis:  In going down the path of totally owning its PIC technology, Infinera arguably opened itself up to being a bottleneck for its customers in terms of the supply chain.  It also provided a gate for its buyers to get new technologies.  There is no question that it has taken Infinera a long time to develop native 40G capability.  And as a public company, the odds are lower that it is going to spin new versions of its PIC very often because it would be very expensive.   

Infinera could have considered licensing its PICs to other manufacturers so that it would get some help in furthering the technology, in particular, up the yield ramp.  Other vendors would have made it easier to improve the process and also would have provided multiple sources for it on the systems side of its business.  

The path that a startup chip vendor, Ekinops, has taken in going all the way up the food chain -- is providing the value in the firmware.  The supplier wanted to make sure that it could use off-the-shelf hardware packages.  The idea was to avoid not getting tied to a technology that costs tens of millions of dollars to spend every time it wanted to introduce a new flavor.  In addition, Ekinops argues that it achieves economies of scale because it takes advantage of the volume benefits of using hardware from larger manufacturers.  

In fairness, it should be pointed out that Ekinops is unable to upgrade line-side cards in the field, such as Ethernet, from 10G to 40G.   

Implications: 1.      An important priority for AT&T is to manage expectations very carefully. 2.      The carrier has to be concerned about being too optimistic about equipment CAPEX during an economic slowdown – because it could be interpreted negatively – that despite the negative pressures on the company, it is still being forced to spend. 3.      However, if AT&T is overly pessimistic on CAPEX outlook, it could be seen as the carrier overly suffering from the downturn in the market.

Analysis:  In prognosticating expenditures at “a slightly lower level,” it avoids the industry leaping to either extreme as mentioned in the key implications.  Barring a major downturn in the economy, it is hardly out of the question that there will be a net increase in AT&T’s CAPEX the rest of this year and perhaps even further out.  The apparent hope at AT&T is that by exceeding anticipated expenditure levels, the service provider can use that to demonstrate that it is executing effectively in the market.  

AT&T is making it clear that it is not going to decrease spending on the growth sides of its business.  One can anticipate even more expenditure in the mobility area -- and U-verse, at least for now, seems to be untouchable.  In terms of the older technologies, AT&T appears to be only buying just enough to adequately maintain those portions of the network – while at the same time looking to transition the network to IP and other newer solutions.  It can be reasonably expected that actual IP service orders will be rolled out promptly.  But if there is heavy demand for IP expected in the next year or so in a particular office – it would not be shocking if those kinds of expenditures could be delayed.  

Implications: 1.      It is no secret that network integration at AT&T is going slower than expected.   2.      Apparently, there is still not full reconciliation in choosing which of the processes (of the three major carriers being combined) will ultimately be utilized. 3.      Telcordia’s OSMINE, the traditional OSS of the Bell companies, has not been as helpful as initially believed.

Analysis:  There are still way too many merger related matters that are slowing down AT&T’s overall development.  The sheer number of things that have to be brought to bear in terms of future direction can seem mind boggling.  The carrier could obviously be moving forward a lot faster if it could get rid of the high level of inertia that has been created by this tremendous combination of assets.   

How long will it take to get rid of the lethargy at AT&T – to end the picture of network planners and engineers struggling their way through to achieve a desirable synthesis of all three corporations?  There is hope at the service provider that it can actually begin to attain some cost savings within six months to a year.  

But even in terms of integrating the wireless and wireline networks, there still appears to be quite a lot of work.  A primary goal has been to get circuits off third parties and onto its own facilities, now that AT&T Wireless is fully owned by the single entity.  The pessimistic view is that there is about a 10 to 25 percent probability that the financial benefits of all of the hard effort will not be realized for two to four years.  

Although the commonality of the OSMINE has helped in the BellSouth situation, it has not resulted in it being any less of a tough sled in integrating the legacy AT&T network.  Moreover, the company still has as many as four different OSS groups that still need to be combined.   

Implications: 1. It is no accident that Chinese telecom manufacturers have placed a lot of focus on Libya and Nigeria in terms of infrastructure investment. 2. The two countries together have about 75 billion barrels of oil reserves.  3. In general, China has invested for a long time into Africa in order to obtain resources of all kinds including copper, cobalt, gold, aluminum and platinum – to provide new markets to keep its high level of economic growth going -- as well as to potentially alleviate its population problem.

Analysis:  The “Beijing-based consultant” is only giving part of the story regarding telecom equipment deals in China.  It is part of a much greater investment by the country in Africa’s infrastructure including the building of roads, electricity distribution networks airports etc. to facilitate getting badly needed resources out for its own use.  In the last ten years, China’s usage of oil has increased 35-fold and Africa is supplying a third of its needs.  China and India are projected to make up 40% of the growth in demand for oil by 2030.  

Despite the source article’s assertion to the contrary, unfortunately, there are often no “questions about product quality and specs” because of the large amount of autocratic control by governments in several countries on the continent.  In addition, in such situations, it is disingenuous to state that Huawei and ZTE have to “dispel any misgivings and concerns” because of “familiar names like Ericsson and Alcatel.”  

China does not make any distinctions in terms of the types of governments it is  willing to deal with in Africa.  While there can be positive results in these countries, it can also lead to an exacerbation of “political turmoil” – or a lot worse such as the tragedies that are taking place in Sudan and Zimbabwe.