Documents menu
Date: Mon, 12 Jun 1995 12:01:18 EST
From: "Pimienta Daniel " <pimienta!>
Subject: More info on Caribbean Networking

On Caribbean Networking

By Daniel Pimienta <pimienta!>, 12 June 1995

Dear colleagues,

FUNREDES is completing the information you have taken the initiative to publish in your gopher about networking in the Caribbean. Please, feel free to update your information base with this note (or with selected parts of it).


First of all, let us congratulate you for the useful and comprehensive work you are realizing in collecting data on networks in Developing Countries. Only recently, thanks to the first Internet full connection in the Dominican Republic (see note 1), we could get to visit and browse and enjoy it.

We also want to thank you for presenting, in your gopher, information about our organization (FUNREDES) and REDID (the Dominican research network). We sincerely appreciate this contribution in defense of pluralism.

The conditions are (at last!) present for the creation of REDID and FUNREDES.ORG internet nodes and subsequent gophers and webs (see note 2). It will take an estimated 4 months to get them in operation. We will then inform you and we could proceed to organize the required linkages.

As for Haiti (REHRED), I am afraid it will take an additional 3 months to get there (see note 3). Meanwhile, we will try to provide you with more accurate and up to date information.


In our last trip to , we discovered, with a tremendous interest (we have been so eager to get them for years!), the CUNET traffic figures.

The report of CUNET is misleading when it presents the .DO figures. It should be specified that these figures are limited to PUCMM.EDU.DO (the CUNET component of Dominican Rep.) and that they do not include REDID.ORG/EDU.DO.

As you may know, the .DO domain, still under management in Puerto Rico, is shared by the two networks. With the imminent creation of, at least, two more Internet commercial providers (in fierce competition) and the migration of REDID to the Internet, the managing of the .DO by a Dominican entity is soon becoming a priority. This is a touchy issue, requiring fair concertizing between the involved entities. FUNREDES is taking actions to initiate and smooth the process.

As for the correspondent traffic and user figures for REDID.DO, they are shown in note 4 and 5.

You will notice that REDID ALONE represents as much traffic as the ENTIRE CUNET (including the CUNET .DO component)

To enhance the Caribbean perspective it will be worth having more data, in particular the corresponding Jamaican and Cuban figures. For Cuba, I hope the CENIAI Director can show the figures. For Jamaica, I hope the people of University of West Indies could provide corresponding figures.


In the persistent lack of user directories (or total user numbers) from CUNET, these figures are the closest way we got for approximation of the numbers of users in CUNET (see note 6). For those interested in methodology evaluation matters, they offer the most reliable indication, 3 years after the parallel start of REDID and CUNET. REDID was designed with the "REDALC methodology" (see note 7) which focuses in building the user base and institutions and relies in partnership with telecom operators (the Codetel company have been helpfully supporting the entire international and national telecommunication costs of REDID, since may 1992).


The point is hardly to compare projects in terms of methodology vs results. The question about what project is doing better is secondary. The real issue is, independently of how and with what project, TO GET MORE PEOPLE USING THE NETWORK IN THE CARIBBEAN RESEARCH COMMUNITY.

The installed base of CUNET is a lot better than nothing, but let's be fair: it is not enough. The traffic figures alone speak for themselves (and networking is much more than traffic).

The critical point is that, as an undesirable side effect of the excellent marketing of CUNET, funding agencies tend to consider that the RESEARCH NETWORKS PROBLEM IS SOLVED IN THE CARIBBEAN.

Unfortunately, THIS IS JUST NOT TRUE, although it does lead to DRAWING AWAY the badly needed SUPPORT, that otherwise would have been available for around here.


This issue calls for prompt action. It begins by recognizing the objective real situation. The rest should follows naturally, namely the assignment of the required resources, with participation of the key players in this field. The Caribbean is not another northern district (indeed it is not quite Latin America either), it has proved to hold a delicate mixture of values demanding a tactful approach.

It's still a long way to Caribbean integration, even in the cyber-Caribbean. . .

FUNREDES have repeatedly insisted that a "user connectivity" component was still missing to facilitate the research and academic users growth and institutionalization (as it is tried, for instance in Cuba and the Dominican Republic). By "user connectivity" (instead of "physical connectivity") we mean a comprehensive set of actions, from decision makers sensitization, user training, user group forming, to the production and organization of national and regional information base. The reinforcing of the level of institution and the technology transfer being the key criteria of success, besides the number of real and satisfied users.

FUNREDES have been unsuccessful so far to convey this message to those interested in the Caribbean and susceptible to support actions. As a matter of fact:

  1. REDID received no further support after 1993, and its operation and growth have been, as a consequence, seriously affected in the period mid-1994 to mid-1995, with a total traffic decrease. With a total investment of say 75,000 US$, REDID would have turned today more than 70% of the targeted researchers into real users (instead of 25%). This is less than the cost of standing a regional meeting for making decision about what to do to solve the very problem of the Dominican network :-(.
  2. FUNREDES, although a consistently active and productive field player in the Caribbean, since 1988 (see note 8), has never been called to participate in the consultive or decision regional meetings on Caribbean networks, and, furthermore, is totally absent from the list of funded NGOs in that region and in that field. It is ironical and sad to confess than FUNREDES, a non profit field oriented organization, have more success getting support from private companies than from the "funding" agencies...


If the grass-root people are going to disappear, because of the lack of support, who will implement the nice strategies discussed in the regional meetings where several international functionaries gathered from all over the world to decide the future of other people who are not even involved in the process?

How about a 10% "field tax" collect on these meetings to go to the people who are doing the job, back in the field? That will probably be enough to maintain the field organizations alive without preventing these meetings to happen. . .

One can expect that the simple facing of these figures will give some matter for reflection to the funding agencies about the productivity of their investments and about the real importance and productivity of those who are struggling in the field, with reduced support, to keep grass-roots efforts going on.

How about a fair and constructive vision of Caribbean Networking?

Best regards from the Dominican Republic,

Daniel Pimienta and colleagues from Foundation Networks and Development (FUNREDES)

PO BOX 2972- Santo Domingo
Dominican Republic
Santo Domingo

tel: + 1 809 535 2422
fax: 535 6646



May 4th, the private company AACR announced the start of its service and its willingness to cooperate for totally free access to REDID members. This was reported in KISKEYA and SALSA lists. The two other telecom operators CODETEL and TRICOM are about to market their respective services (and, hopefully, complementary or alternative offers to REDID).




These results are obtained from partial data collected from both the codemail node and the UPR2 node. We are in the process to complete the data and cross-check the results. Meanwhile, one can use the average figures since there are strong and relatively stable patterns.

Stable observed patterns (period from 10/94 to 5/95):

Which makes the 30x2. = 60 Mbytes monthly figure we used.


REDID publishes on a regular basis its user listing. After the difficult period of 1994/95, in terms of user support and installation (as a consequence from the absence of external support), REDID is in the process to reinforce its operation (with the cooperation of FUNREDES) and to update the listing. There are now some 120 end-users from 60 member institutions (not all of them having "real users"). The growth factor which have been relatively stable, between 1 and 2 new institutions per month, should climb now to an estimated 3 to 6 with the improvment of the operation and the strong Internet media attention in the country.

The number of user per institutions varies from one unit to a maximum of 26 and 16, respectively for UNDP and PUCMM in Santiago (among the rare institutions having been capable to organize the REDID access via LAN). The NGOs are the most active users of REDID. Although connections are organized in several universities, the systematic in-networking of the Dominican universities is still a pending item in the national agenda (this issue escapes from REDID responsibility). Besides REDID and CUNET/PUCMM.DO, several BBS initiatives (as in INTEC) are trying to fill the gap, still open, for the students connectivity.


If one uses, as a projection factor, the ratio "data volume/user" of REDID (during the period 10/94 til 4/95) and backup it with some traffic assumptions, it is possible to guess some user figures for CUNET. Obviously, this is just an extrapolation, provided for the sake of comparison, and should be treated then as such. I wish that Cunet node administrators will correct the extrapolated figures and publish the real ones (together with the associated directories, so that to facilitate the in-Caribbean communication).

Traffic assumptions:

This comes from the simple scheme:

The variance is obviously very high, and then the extrapolation process is uncertain where the user figures is very low, as in most Cunet nodes (one single active user may generate as much as 20 times more Bytes than a slow one!).

It is worth noting that:

For the CUNET node, the average of the published data has been used.


                  ISO-CODE AVG MBytes  PROJ.  POPULATION  
   COUNTRY           COUNTRY  Per Month   USERS  (MILLION)
   Antigua and Bermuda AG     1.06         2       .07                  
   Bahamas             BS     4.15         8       .25  
   Barbados            BB     8.59        17       .26       
   Belize              BZ    13.74        28       .19       
   Dominican Rep CUNET DO     3.62         7      7.00     
***Dominican Rep REDID DO    60.00       120      7.00       
   Grenada             GR     0.63         1       .10   
   Guyana              GY     0.10         1       .80 
   St. Lucia           LC    11.72        24       .14          
   Vincent and Gren.   VC     0.79         2       .11
   Suriname            SR     2.29         5       .42
   Trinidad/Tobago     TT    17.14        34      1.30      
TOTAL CUNET                  66.92       129      10.64
TOTAL REDID                  60.00       120       7.00

Note: If one takes into account the size of the population, some CUNET nodes (such as St Lucia and Belize) hold notable figures in terms of market penetration. It will be worth studying what factors have produced those interesting results.