Exhibit 2

August 17, 2011

VIA EMAIL:

Mr. Shane Coppola
Special Committee
Global Traffic Network, Inc.
880 Third Avenue, 6th Floor
New York, New York 10022

Mr. Gary Benson
Special Committee
Global Traffic Network, Inc.
880 Third Avenue, 6th Floor
New York, New York 10022

Mr. Stuart Romenesko
Special Committee
Global Traffic Network, Inc.
880 Third Avenue, 6th Floor
New York, New York 10022


Dear Messrs. Coppola, Benson, and Romenesko:

We are writing this letter to you in your capacity as the Special Committee of Global Traffic Network Inc. (GNET) and our capacity as significant shareholders of the company. Our fund currently holds 1,575,000 shares of GNET, representing 8.2% of the fully diluted outstanding shares. Based on our extensive analysis of GNET's past and projected financials, public statements made by management, and the multiples of comparable transactions, we believe that the $14.00 proposed acquisition price by GTCR grossly undervalues the company. In a world where proven, open-ended growth stories command massive valuations (in public and private markets), we are being asked to sell such a company at a sizable discount to fair value. As we intend to illustrate in this letter, in a base case scenario, using pedestrian assumptions, we believe that GNET is worth at least $16.10 per share today. Further, in negotiating the tender offer with GTCR, the Special Committee has seemingly ignored the potential asset value of Global Alert Network (GAN), giving it to GTCR for a pittance.

As you may know, we are a San Francisco-based investment adviser whose clients/funds invest primarily in companies with market capitalizations under $3 billion. We currently have assets under management of approximately $320 million, and our historical success has been a function of our ability to identify mis-valued companies. Based on GTCR's rapacious $14.00 bid, it would appear that they have a comparable skill-set. This letter should come as no surprise to the Committee. In fact, the conference call on August 3rd announcing the tender was accompanied by a Q&A session marked by disbelief and anger among shareholders (selective comments on the call below):

"Well, that's very unhelpful"
"I'm just a little disappointed frankly"
"This just looks like an inside job to me"
"You have some investors who don't seem very happy about the price"
"We are being taken advantage of here"

Before discussing why we believe the current GTCR proposal fails to compensate shareholders appropriately, it is worth noting that we have a deep admiration for CEO Bill Yde and the business he has built. Over the past year, we have developed tremendous respect for the GNET franchise, the long-term opportunities, and the strong cash flow dynamics of the company. We believe that Mr. Yde has developed a disruptive platform to help broadcasters and television networks shed unnecessary fixed costs, while also creating unique value for advertisers. In the three markets that GNET has entered, the economic payback has been exceptional, while also building a unique channel for national advertisers to reach highly targeted audiences. Based upon our research, we believe that GNET has the opportunity to expand its operations into a handful of new markets, while concomitantly generating an operating profit from the four existing business segments.

Ironically, we share the Special Committee's assessment that the inherent nature of the GNET business model makes the private domain a more appropriate setting for growth. In our view, the best strategic decision for GNET owners would be an immediate entry into a handful of new markets, such as Germany, Brazil, and India. However, under the quarterly microscope of the public eye, this would hit short-term profitability given the losses new markets create, and a potential near-term stock price decline as well. As such, management has stated publicly that they would wait to enter new markets until Canada became profitable - a feat that finally occurred with the recently disclosed FYQ4'11 results. GNET's unique ability to fund its de novo growth with $41 million of cash and the Australian "cash cow," would allow private ownership to enter new markets with a clear ROI expectation devoid of concern for any short-term stock price impact. We believe GNET could thrive under private ownership as expansion would undoubtedly occur in new geographies. But given the choice between the ignominious tender offer of $14.00 and keeping GNET public, we believe the latter is more compelling based on expectations for a higher exit in the future (even if this does result in the stock declining short-term).

GNET Australia represents the template for future markets. Amazingly, GNET has been able to build a sizable business in Australia despite the country's relatively diminutive size when compared to Canada and new market opportunities. Australia is your most mature market. In fiscal 2011, GNET reported revenue from Australia of $69.7 million with $20.4 million of EBITDA (29% margin). There are several points that may have been overlooked by the Special Committee when negotiating with GTCR:

>    Australia is a virtual monopoly for GNET, with 84 radio stations and 14 television stations (just for traffic reports).

>    In Australia, GNET dominates 19 markets, including ALL of the top 10.

>    According to AC Nielsen, the weekly reach in Australia is 9 million radio listeners in the top 5 markets, equating to 100% of commercial listeners. GNET also reaches 7.1 million TV viewers, representing 50% of the commercial markets.

>    According to GNET's own marketing material, GNET controls the single largest media network in Australia.

>    This monopoly has been accomplished with a mere $2 million of investment before turning a profit, demonstrating the capital efficiencies of the model -"We were all in for Australia for $2 million before we turned a profit." [1]

>    Australia should continue to produce $20 million-plus of EBITDA with virtually zero incremental capital expenditures.

>    Monopolies should command high multiples

>    The contribution margins in this business are outstanding, as illustrated by the chart below:

(millions) USD
Australia
FISCAL YEAR (ends June)	2006	2007	2008	2009	2010	2011	2012*
Revenue	19	28.4	44.3	42.7	56.5	69.7	78.1
growth YoY	-	49%	56%	-4%	32%	23%	12%
EBITDA	1.3	6.7	13.8	12.7	16.3	20.4	23.3
growth YoY	-	415%	106%	-8%	28%	25%	14%
contribution margin	-	57%	45%	69%	26%	31%	35%
*2012 is projected by management in 14D-9

GNET filings *** GRAPH ABOVE DOES NOT INCLUDE Q4'11

Based on the $14.00 tender ($41 million of cash at 6/30/11 and 19 million shares out), GTCR is proposing to buy the entire company for only 9.66x Australia's monopolistic FY'12 EBITDA annuity stream while getting Canada, the UK, all future markets, and Global Alert Network (mobile) for free. And we believe, based on public commentary, past investments in other areas are on the cusp of returning outsized profitability to shareholders.

As you are aware, Canada was the second country GNET entered and has been steadily growing. With CPMs around $90 (still 40% below those of Australia), there is an enormous runway for Canadian growth and margin improvement. As of two months ago, GNET Canada had 94 radio stations and 4 television stations. According to BBM Canada, GNET reached 12.8 million radio listeners, representing 47% of commercial listeners, while reaching 2.6 million TV viewers. This trajectory is very similar to Australia, and Mr. Yde recently confirmed the promising financial expectations.

At the Noble conference on May 16 and 17, 2011, Mr. Yde expressed his enthusiasm about Canada, stating his expectations for an "enormous amount of growth you are going to see over the next couple years in Canada." He further commented, "You are going to see Canada take off and be substantially more profitable than what you have seen in Australia." We believe these statements to be credible given the accelerating trends and Mr. Yde's proven ability to execute. This inflection point in Canada was reinforced by Mr. Yde on the August 3rd conference call:

Jeff Osher > : Hey Bill, do you still believe and I know you said publicly for several quarters if not years now that you thought Canada can be very comparable to Australia and the trajectory definitely has supported that thesis over some period of time?

William L. Yde> : Yes, nothing's changed in my belief about Canada

(millions) USD
Canada
FISCAL YEAR (ends June)	2006	2007	2008	2009	2010	2011	2012*
Revenue	0.5	3.3	6.7	6.9	9.5	15	19.7
growth YoY	-	560%	103%	3%	38%	58%	31%
EBITDA	-2.4	-3.6	-2.4	-2.3	-3.0	0.1	3.5
growth YoY	-	50%	-33%	-4%	30%	-103%	3400%
contribution margin	-	-43%	35%	50%	-27%	56%	72%
*2012 is projected by management in 14D-9

GNET filings *** GRAPH ABOVE DOES NOT INCLUDE Q4'11


Mr. Yde has argued in his public remarks that Canada is following the same trajectory as Australia. His thesis has been supported by the financial progress (seen above), as well as the projections made by management in the tender documents.

Another asset that appears grossly undervalued in the tender offer is Global Alert Networks, or GAN. We understand that GNET bought the GAN IP from Telcordia for approximately $150,000 several years ago and that GNET has spent about $3 million developing GAN, on top of the $8 - $10 million that had already been spent by its previous owner. [2] Based on public commentary, and investor excitement over location-based services, we believe that GAN has been materially mis-valued by Moelis (in their fairness opinion) and by the Special Committee in its negotiations.

Mr. Yde recently described the immense opportunity for GAN at a conference, explaining, "We invented the process that is the best way to deliver traffic reports to people in the world." Over the last few years, GAN has been a corporate expense with zero payback. This profile changed when the application was introduced in February and commercial marketing began in March. GAN is a free app that consumers can download that provides hands-free audio alerts for traffic and weather information. In just a few months since its commercial launch, GNET already has nearly 300,000 subscribers with virtually no marketing. [3] What is more amazing is this traction has been accomplished with no iOS (Apple/iTunes) version available yet.

The model is elegantly simple: provide hands-free alerts to subscribers in real-time (unlike GPS devices or Google traffic which relies on delayed satellite data) and monetize the alerts with advertising. There are minimal delivery costs and exceptional margins. As new users are added, new ad inventory is created, which attracts even more advertisers, hence driving up potential CPMs. In fact our profitability analysis was recently corroborated by Mr. Yde himself, "the important thing about this business [GAN] is that it is going to be massively more profitable than our core business." [4]

How profitable can GAN be? Well, we can simply take the company's public communications, all of which have been verified as reasonable by analysis in the online advertising space. Mr. Yde said at the Noble conference that they expect to have one million downloads of GAN by the end of the year. To be conservative, we have haircut this figure by 20% in our analysis (even though you are nearly 1/3 of the way there after just three months and no iOS). Further, management has publicly stated that GAN is delivering two, or three, alerts per subscriber per day [5]:

Jeff Osher > : Okay. And based on the most recent public information you guys provided, I believe it was two or three alerts per day?

William L. Yde> : That is what we're expecting per subscriber, yes.

Mr. Yde said in May at an industry conference that GAN will be delivering about one billion alerts per year once they get to one million subscribers. We have haircut the high-end by one-third, assuming only two alerts per day. Further, the company has stated consistently that they believe GAN can generate approximately $0.05 per alert. This would be very much inline with CPMs for other online advertising companies that have highly targeted audiences. Nonetheless, we have cut our revenue/alert expectations by 20% and only assume $0.04 per alert (it is worth noting that we believe the alert opportunity is massive and could extend to public service announcements around Amber alerts, hurricanes, or national emergencies). Finally, we believe that the EBITDA margins for GAN could be fantastic. Based on our understanding of the operating model, GAN's cost structure is largely fixed, with the only variable component being commissions for ad sales. As such, we believe Mr. Yde's public comments about GAN being "massively more profitable than our core business" are quite reasonable. Specifically, management has stated at various investor events (most recently in May) that they believe GAN can generate a 50%+ EBITDA margin with a critical mass of subscribers. As you can see below, we have taken a major haircut to that assumption, and have only assumed an EBITDA margin slightly greater than the core Australia business. Even with those conservative assumptions, we can't reconcile the fundamental drivers of GAN with the projections and assumptions found in the tender. The tender offer appears to value GAN at next to nothing ($4.4 million according to Moelis to be precise). If our sensitivity analysis is directionally correct, then GNET has agreed to sell the GAN asset for a tiny fraction of its potential value.

GAN - company assumptions
downloads	1,000,000
alerts/day	2.5
revenue/alert	0.05
GAN Revenue	45,625,000
GAN EBITDA margin	50%
GAN EBITDA	22,812,500
Harvest Small Cap Assumptions
downloads	800,000
alerts/day	2
revenue/alert	0.04
GAN Revenue	23,360,000
GAN EBITDA margin	35%
GAN EBITDA	8,176,000

Based on management's publicly stated objectives, GAN could be nearing these figures within a few years. In March, GAN delivered 65,000 alerts, just one month after its commercial launch. In April, GAN delivered 700,000 alerts, 10x the number of alerts from March. Given the progress, we were confused to see such muted projections in the tender documents. Something doesn't reconcile and we are quite certain that our assumptions are accurate/conservative based on management's public commentary and the metrics to-date. Should our haircut assumptions prove to be remotely correct, then the nature of this financial model, and open-ended opportunity, would warrant a multiple well in excess of the 9.7x that GTCR has bid for the projected EBITDA over the next twelve months (with nothing for GAN). Should GAN deliver anything close to $8.2 million of EBITDA, it could be worth in excess of $6.00 per share alone, and management seems to agree with our assessment of the potential, "We have been working about eight years on this project to get this going. We have got some very interesting patents around it. And we think in the audio, automatic alerting system we have a great advantage over virtually anybody that is out there." [6]

Before discussing how we value GNET - which would hopefully be comparable to how Moelis & Company is presenting GNET in the go-shop window - we feel obligated to highlight several concerns with the sale process:

1.   The timing of the proposed acquisition and the refusal to disclose Q4'11 (June) results on the deal conference call raises eyebrows. On the August 3rd call, several callers inquired about when Q4'11 results would be released. Further, the prepared remarks cited a 14x multiple based on trailing twelve month results (through Q3'11, even though Q4'11 had concluded over a month earlier), "The transaction with GTR represents over 14 times multiple of our adjusted operating income for the last 12 month period ended March 31, 2011." [7] While the numbers are what they are, we believe that, at a minimum, management's unwillingness to even disclose estimates of the Company's blowout Q4'11 results on the August 3rd call reflects poor judgment. Had the most recent "last 12 month period" been used (through June 30th), the acquisition multiple would have fallen to 11.5x from the 14x cited on the call. [8]

There has still been no press release disclosing Q4'11 results. But in the tender materials, we learned that GNET had record results in Q4'11 and would have crushed analyst expectations. While it is not even worth speculating how much the stock may have risen based on Q4'11 results in the absence of the tender, it is an objective observation that Q4 results were significantly better than consensus expectations:

	Q4'11 Actual	Q4'11 Estimate *	% Surprise
Revenue (millions)	33.5	28.3	18.37%
EPS	$0.20	$0.06	233.33%
EBITDA (millions)	5.70	3.21	77.57%

* Bloomberg consensus estimates from 7/28/11

Given the magnitude of the Q4'11 outperformance, it is not unreasonable to assume the stock may have been close to the tender price on its own merits. After Q1'11 and Q2'11, which were beats of a far smaller magnitude, the stock increased over the proceeding two days by 12% and 18% respectively. Again, we will never know how the stock would have reacted had Q4'11 results been released prior to the tender offer. But the chronological decision to seemingly obfuscate the excellent results from Q4'11 raises questions about the Committee's rationale for such an unusual decision.

2.   Moelis & Company has a stellar reputation and we have zero reason to believe they have not acted 100% in the best interest of GNET. However, we do think it is within our rights as shareholders to question elements of their analysis:

a)   Why is Moelis & Company using a "comp" list of deteriorating radio broadcasters for comparisons? The median revenue growth from 2010 - 2012 for the two comp groups Moelis & Company used is a torpid 10.5% and 3.5%, averaging out to just 7%. This compares to GNET's fiscal 2010 revenue growth of 54.7%, fiscal 2011 growth of 26%, and projected growth of double digits for the next five years. Further, the 2012 projected median revenue growth for the comp group is a meager 4.5%, with EBITDA growth of 8.8%. This compares to projected revenue growth for GNET that is nearly 150% greater than the comps and expected EBITDA growth of 41% (vs. 8.8% for the comps) for the next twelve months. Worse still, the comp group's growth for next year may not even be indicative of real secular growth due to the one-time benefits from the 2012 elections and Olympics. The point is we believe the comp group that was used to justify the median multiples may have been poorly selected.

b)   Moelis & Company used several different methodologies for valuing GNET. In two of their methodologies, arguably using numbers that were materially understated for GAN, they arrived at fair value ranges that had a high end materially above the GTCR offer. Had they used the same multiple that GNET's CFO highlighted on the conference call (14x) on fiscal 2012 financial projections, the fair value price for GNET would have been $20.00 (on $24.8 million of EBITDA and $41 million of net cash - a figure that conservatively excludes the free cash flow that will be generated in FY'12). Page 16 of the fairness presentation runs a simple analysis of the implied GNET tender price based on the median share price premium for all M&A and LBO only deals. As discussed above, the comp group appears to have less desirable financials than GNET. Nevertheless, had GNET simply received the median premium that these less attractive deals received, the implied offer price would have been between $15.58 - $16.10 (all M&A) and $14.42 - $15.24 (LBO only). This anomaly is only amplified when one considers GNET's more favorable growth characteristics.

c)   Examining the six selected M&A deals that were used as comparisons, the median EV/EBITDA multiple was 9.7x - and it is worth noting GNET appears to be growing faster than all of them. Based on Australia's projected forward twelve months EBITDA ($23.3 million), GTCR is effectively paying 9.7x for Australia and getting everything else for free!

d)   Why is the total enterprise value assigned to GAN only $4.4 million? As discussed above, we believe that GAN has immense potential. It is difficult for us to envision a platform with GAN's existing install base, cash flow dynamics at scale, and momentum selling for just $4.4 million in a private transaction. This equates to roughly $4.40 per subscriber if the company is able to achieve its goals for subscriber growth in the next year. Using the valuation Moelis & Company assigned to GAN, the payback on GAN would be just 72 days at a 50% EBITDA margin (2.5 alerts per day at $0.05 per day over 72 days = $9 of revenue/sub, and $4.5 profit/sub with a 50% margin).

e)   Moelis & Company's DCF assumes virtually no traction for GAN and a ramp for Canada that looks nothing like the ramp in Australia. Why is this?

3.   How were new markets valued? As recently as two months ago, management was strongly hinting in public comments that the entry into new markets was imminent. As stated above, as large shareholders, we would encourage the company to grow as quickly as their infrastructure would support. The costs of entering new markets are immaterial when compared to the fantastic cash flow characteristics once critical mass is established. We believe that investors may have rewarded GNET for an entry into Brazil, India, or another market of similar size despite the upfront costs. Some of the quotes that we were able to find in public presentations and conference calls since the middle of May that supported our view of imminent new geographies are (as your Special Committee knows, Mr. Yde is not a promoter, he is instead very humble and has a reputation for under-promising and over-delivering - making these remarks even more interesting):

-- "[Our] continued growth opportunities [are] bigger than what we have had over the last five years"

-- "We have been exploring India, we have been exploring South America, and other European countries, like Germany and France. There is virtually no competition in these markets so it is not difficult to get in. It is our plan to look into those very shortly."

-- "We've been exploring a lot of countries. [Our] goal was to get Canada to where it was a contributor... even though we have made good EBITDA growth over the last five years, we were losing money in Canada, we were losing money in Great Britain, and we were losing money developing mobile. What our goal was is to get all of those as EBITDA contributors. I think we are nearly there now." (our emphasis added)

-- There are "190 countries that we are still not in yet."

4.   We can't emphasize enough that Moelis & Company has earned their reputation as an excellent boutique investment bank. Further, we believe that they will cast a wide and thorough net over would-be acquirers. And with the tens-of-billions of private equity dollars that have been raised over the last five years desperately needing to be put to work, we are hopeful a higher bid will materialize.

Despite our high opinion of Moelis, we want to make sure the Board and Special Committee understand the nature of Moelis & Company's past relationships with GTCR. The reality is there are always potential conflicts that could bias any fairness opinion or go-shop activity. We hope the Committee was able to gain comfort around the following disclosures given the entirety of all other data points (emphasis below is ours):

"In the past two years, Moelis has provided investment banking and other services to affiliates of GTCR. GTCR and its affiliates have paid fees to Moelis of approximately twice the amount of the fees being paid to Moelis in connection with the Transaction. In addition, since Moelis' engagement by Global, Moelis was engaged by an affiliate of GTCR to provide financial advisory services in connection with a potential acquisition of a public company, for which it will be entitled to receive customary compensation."

5.   Coming back to GAN, we are unable to reconcile management's projections, or Moelis & Company's valuation assumption ($4.4 million of enterprise value as discussed above). On the August 3rd call, GNET's CFO responded to two different callers stating that total GAN costs for fiscal 2011 would be somewhere around $1 million. This is consistent with past statements. Therefore, we continue to be utterly confused about how GAN will not be a material contributor to operating profits going forward - something that seems to have been neglected in the financial projections. As highlighted earlier, we believe the business will have extremely powerful economics in the very near-term. Mr. Yde seemed to agree on the third quarter earnings call, "I don't think it's going to take long for it to be an EBITDA positive thing because our cost structure is so low." So based on the basic drivers that have been disclosed, how can the GAN projection of a $1 million loss for fiscal 2012 be accurate if revenues are projected to jump to $1.5 million (a figure that seems very low) if the cost structure is only "around" $1 million? The math means the implied costs jump from $1 million in fiscal 2011 to $2.5 million in fiscal 2012. This is completely inconsistent with our understanding of the model. Aside from marketing payouts that have been publicly disclosed to be 10%, there should be minimal incremental costs. This is a very serious concern of ours given our view of the potentially massive opportunity and subscriber economics for GAN.

We are confident that with the Special Committee's oversight, Moelis & Company will run a comprehensive "go-shop" process through the September 13th deadline. With that said, we are hopeful they are highlighting a business that has gone from $32 million of revenues to $117 million of revenues in just four years. During that same period, EBITDA has expanded from break-even to $20 million, representing a contribution margin of 23.5%. What makes this performance even more impressive is that it was accomplished with three of the four segments losing money (the UK turned its first profit in 2010). Adding to our skepticism about the sale process is the fact the company has failed to highlight that all four businesses should positively contribute to EBITDA in fiscal 2012 for the first time ever. The momentum is accelerating for GNET at a time when investors (both public and private) have a paucity of sound growth stories at reasonable prices.

Additionally, we are hopeful that the free cash flow dynamics will be highlighted - something that was surprisingly absent from most of the Fairness Presentation (EBITDA was highlighted extensively, but given the minimal capital requirements of the GNET business, uFCF is arguably more important). In fiscal year 2012, GNET should generate $17.7 million of free cash flow based on their projections. We have demonstrated the financial projections seem to be highly conservative, an opinion that is only reinforced by the 14D-9 disclosure that revenues in July (the first month of fiscal 2012) grew 30.4% year-over-year - a rate that is highly inconsistent with the implied deceleration in the projections. Should the $14.00 tender price be accepted by a majority of GNET shareholders, GTCR will be buying an accelerating free cash flow stream for just 12.7x, or an 8% free cash flow yield. The "plundering" of shareholder value becomes even more apparent when looking at the out-fiscal year projections, which would represent a free cash flow yield closer to 10%. We tip our cap to GTCR for their negotiating acumen... where else in the world is this type of theft legal, yet alone a sign of adroit analysis?

In short, we believe that GNET should be worth considerably more than $14.00 today, with very little "hope" embedded in our assumptions. While picking EBITDA multiples by segment is challenging, the exercise does demonstrate the apparent shortfall of GTCR's bid.

Given the monopolistic positioning of GNET Australia, its reliable free cash flow and low double-digit growth profile, we believe that an 8.5x multiple on EBITDA is appropriate, and perhaps even conservative given the nature of the monopoly. This yields an enterprise value of $173.4 million. Based upon Canada's accelerating profile, high contribution margins, and similar trajectory to Australia, we are modeling for EBITDA of $11 million in fiscal year 2014. We then discount that back at a 15% rate and use a 9x multiple based on the early stage of growth and established Australian precedent. This yields a discounted enterprise value of $71.5 million today. We only value the U.K. at 6x EBITDA, despite the optionality of this segment and enormous market size. As you can see below, based on undemanding assumptions for the three core businesses, we arrive at a base case enterprise value well above the $14.00 tender BEFORE assigning a control premium, any value to GAN, or new markets.

	EBITDA (million)	Multiple	Enterprise Value (millions)
Australia	20.4	8.5x	173.4
Canada	14	9x	71.5*
UK	3.5	6x	21
Total EV			265.9
Cash 6/30/11			41
Total Cap			306.9
Shares Out			19.06
Stock Price - Base Case			$16.10
* Canada EBITDA is FY'14 discounted back at 15% to today for EV calculation

But as we demonstrated above, GAN is a clear wildcard. However, going back to our base assumption for GAN that significantly haircuts management's own public assumptions, we arrive at $8.2 million of EBITDA for fiscal 2013. If this is remotely accurate, then this business will command a multiple inline with other location-based advertising companies, and hyper-growth technology darlings. What is that? 12x EBITDA? 15x EBITDA? We don't know with precision. But if we assume a multiple similar to several of the public social advertisers (approximately 17x), we get a potential value for GAN on our estimates of around $136 million, or about $7 per share. Additionally, we have built no valuation assumptions into our model for new markets.

The bottom line is that we believe the base case fair value for GNET is $16.10 today assuming none of the new initiatives are successful (but continuing to include their costs). This represents a 15% premium to GTCR's $14.00 tender. Given the open-ended nature of GNET, the Australian "cash cow", the strong balance sheet, and the optionality around GAN and new markets, we are loath to accept any bid below our modest, base case fair value price of $16.10 per share.

Please consider our points carefully. As shareholders, we rely on the Board, and by extension the Special Committee, to ensure all steps possible will be taken to protect our rights. In the interim, we will continue to consider the $14.00 bid from all angles, which is consistent with our own fiduciary responsibility. While the process to-date has fallen short of expectations, we are confident that with the Special Committee's oversight, Moelis & Company will communicate the unique aspects of the GNET story to the myriad of potential buyers. We are hopeful that our analysis will help stop GTCR from their shrewd attempt to steal GNET from its existing shareholders.

Respectfully Submitted,

Jeff Osher
Managing Partner


CC:
William Yde - Chairman, CEO
William Mower - Director
Dale Arfman - Co-Founder, Director
Scott Cody - CFO, Treasurer

[1] Bill Yde - Noble Financial Conference May, 2011

[2] Noble Conference disclosure - Bill Yde

[3] 14D-9 filing

[4] Noble Conference as well as FYQ3'11 call

[5] Noble Conference and August 3rd conference call

[6] May 16th Industry presentation

[7] CFO Scott Cody on the August 3rd call

[8] Based on $19.7 million of EBITDA from the Fairness Presentation