A consumer view of residential Internet service in Fiji

Intro
Let's be honest, Fiji's ISPs have room for improvement. The problem with residential broadband in Fiji is value for money. Value in this case means a combination of 1.) actual, as opposed to promised, speed, 2.) reliability of service, and 3.) monthly data volume. Why is this value lacking?

Wholesale
Last week there was a session on the Southern Cross Cables Network (SCCN) presented by George Samisoni of FINTEL (video of the session is available, thanks to Franck from PICISOC). The three most noteworthy points of the presentation for me were:

1. FINTEL splits the revenue roughly 50/50 with local ISPs for data transfered. Some undisclosed portion of FINTEL's share then goes to SCCN.
   
2. The SCCN is nowhere near its capacity - only money/pricing and the infrastructure between FINTEL and your door is preventing users in Fiji from having faster access to the Internet.
   
3. In the new deregulated environment, FINTEL will remain the sole agent for access to the SCCN. This means that FINTEL maintains a practical monopoly on wholesale Internet access until a competitive sea-floor cable is run to Fiji or satellite connections become more reliable and affordable.

What does this mean for Fiji's residential Internet users? It means that competition will only benefit residential consumers through any economic efficiencies that can be made and reliability that can be gained through infrastructure between FINTEL and your doorstep. In other words, there will be no changes in the near future on the wholesale side.

Retail
The following table lays out the current options for residential Internet consumers.

Residential ISP pricing as of May 9, 2008
based on ISP websites supplemented by phone calls to sales reps.

ISP	Advertised speed	Price (rounded off)	Usage cap	Setup costs (rounded off)	Notes
Connect	256/128 down/up kbps	$40	3 Gb/mo.	$100	need a land line
	512/256 down/up kbps	$90	3 Gb/mo.	$100	need a land line
Kidanet	128 kbps	$50	4 Gb/mo.	$300	install visit required
	256 kbps	$100	5 Gb/mo.	$400	install visit required
Unwired	256/128 down/up kbps	$50	none	$100 24 mo. $200 no contract	burglar bars may cause interference
	512/256 down/up kbps	$100	none	$100 24 mo. $200 no contract	burglar bars may cause interference

There are a number of points of interest on this table:

• Internet companies lacking commitment to their own website: Kidanet's website is woefully out of date. They are certainly turning potential customers away due to the high startup pricing quoted on their website ($599 for the modem and $99 installation). If you call or go in person, they will quote you the lower pricing that I have provided in the table above and tell you that there is now a data cap. Connect and Unwired both have some minor inconsistencies on their sites as well.
  
  
  
  Kidanet's broadband pricing on their website as of May 9, 2008 - no cap and a very expensive modem
  
  
• All you can eat?: Kidanet's nationwide launch was based on no monthly download limits. Now, only a few months later, they have quietly clamped on limits starting at 4 Gb per month. In response to Kidanet's launch last year, Unwired dropped the monthly download cap that had been in place since at least mid 2006 (when I signed up). Now Unwired's website claims that they have always had no cap. Other than Connect's special After Dark plan, which allows access only on evenings and weekends, Unwired is the only ISP currently offering no download limits. Can they keep it up? The reluctance of ISPs to offer residential customers no limit Internet access is an indication of significant pricing pressure on data volume procured through FINTEL.
  
  
  Unwired's dubious claim of "always have been" from their website on May 9, 2008
  
  
  
• Student plans: All three ISPs offer reduced rates for student users. If you are a student, consider signing up for a rate reduction of up to 20%.
  
• Lock in: Both Connect and Unwired have service packages requiring no long term commitment other than the purchase of necessary hardware - although packages with special pricing do require contracts with a term of one year or longer. In contrast, Kidanet requires all users make a two year commitment (the website says three) and the penalty for early cancellation is a painful six months of fees. Increasing customers' switching costs is not an uncommon practice for many service vendors, but it is not a method for forging strong customer relationships.
  
• Misleading advertising: All three ISPs advertise attractive data communication speeds, but my informal survey of residential customers of all three vendors reveals that no one sees the advertised speeds for any significant period of time, and others claim that they do not see the advertised speeds at all. In fact all customers that I spoke to claim to have experienced significant episodes of down time. None of the ISPs offer a meaningful service level agreement to residential customers and terms of service small print always reveals that the advertised speeds are nothing more than maximums. It would be invaluable to have a third party carry out customer satisfaction and actual throughput surveys across all major vendors. Based on my informal research, I assume that Kidanet does not offer less for $50 per month as it might appear on paper; rather they are merely more honest than the competition in advertising their residential transfer rates.

Conclusion
I am not confident that deregulation will substantially improve the lot of Internet consumers in Fiji in terms of data transfer rate or cost per packet. Without competition at the upstream or wholesale stage, the opportunity to drive consumer pricing down is severely limited. And, at least for the short term, competition in Fiji is a facade. The ATH Group owns controlling stakes - directly or indirectly - in all of FINTEL, Connect, Kidanet, and Vodafone Fiji. Unwired is the only meaningful competition to the ATH juggernaut. However, increased competitiveness in the future, should it come, is likely to drive an increase in quality of service to residential consumers as vendors will seek non-price related means of differentiation. I look forward to a day when Fijian ISPs' primary method of maintaining customers is through creating an outstanding customer experience rather than relying on switching costs.



Photo by: publicenergy
Diagram by: activeside

Blocking anti-military blogs may harm military

There have been numerous reports in both the Fijian and international press over the past few of days about the interim government working with FINTEL to censor troublesome blogs. I think that this would be a mistake. In fact, blocking these blogs may actually harm the interim government.

Here are four reasons why Fiji's interim government should reconsider asking FINTEL to block access to anonymous anti-military blogs:

1. Clearly, some of the remarks in Fiji's anonymous political blogs regarding members of the interim government are libelous. Fiji's Defamation Act and supporting Common Law allows for an injured party to ask the court to instruct an Internet Service Provider to turn over records relating to a customer who has published defamatory remarks. No additional legislation or special measures are required. However, some anonymous bloggers cover their tracks and will not be caught by this method.

2. The perception of government tampering in FINTEL, Fiji's pivotal Internet Service Provider through which all Internet traffic flows, would dampen high-tech investment in Fiji. Off-shore service centres rely on the perception of data communications security in order to assure their customers that the confidentiality of their data will not be compromised.

3. Information warfare theory shows that censoring blogs may cost the interim government more than it gains. Power is the ability of a combatant to take an action and have it result in a desired effect. In information warfare, where the weapons are ideas and the battlefields are people's minds, cultures, and ways of life, the power of a combatant is limited by the combatant's perceived legitimacy. Every action that can be perceived as taking the interim government one step further from a return to democracy is going harm the government's perceived legitimacy in the eyes of major international stakeholders - the EU comes to mind here - and in the eyes of at least some proportion of the population of Fiji. This reduced perception of legitimacy then constrains the interim government's future efforts to exercise their power. In an information war, which is certainly what Fiji is currently experiencing, combatants must carefully weigh the consequences of any action before executing it as one's own actions can be more damaging than an enemy attack.

Figure 1.

Any action that weakens Fiji's interim government's perceived legitimacy weakens its power to act.
[diagram from K A Taipale, 2006.]

4. Most importantly, blocking blogspot.com or a hand full of offending blogspot sites will not be effective. Anonymous political bloggers and their readers will simply shift their activities to other available Internet resources, such as:

• A new blog on blogspot.com (unless they just block the whole blogspot domain)
• Any other free blog site such as Wordpress, Livejournal, or countless others
• Any social networking site such as Hi5, Facebook, MySpace, or others
• Any free website provider such as Google Pages, Geocities, or others
• Any instant messaging service from the venerable IRC to jabber, yim, gtalk, or others
• Any group discussion technology from the venerable USENET to Google Groups, Yahoo Groups, or others
• Any p2p technology to circulate files or to chat such as Skype or FWD
• Or users could do any of the above, including reading any blocked blogs, by using open proxy servers, an anonymizing service, or The Onion Router
  

In the final analysis, little or nothing would be gained by the interim government by blocking access to Fiji's anonymous political blogs - Fiji's anti-military bloggers and their readers will carry on regardless - but there certainly would be a price.

Photo by: FelipeArte