Nomad: Report based on the experience of the WSF 2005 in Porto Alegre

Introduction: Basic knowledge about the Technical Interpretation Systems

The basic issues that have to be tackled with the Technical Interpretation System are three:

1. The public must be able to hear the interpreters.
2. The interpreters must be able to hear the speakers.
3. The interpreters must be able to hear each other, on demand. This is the case that the interpreter does not speak the language of the speaker, so he interprets another interpreter (relay interpretation) [1].

Picture 1: The initial design for the interpretation system. We can see the FM transmitters and the consoles (mixers) in each booth. Those consoles would be connected linearly. We can also see the use of Targ for archiving and Internet streaming.

We should note that the 2nd and 3rd point are not separate issues, technically: the speakers can be considered as an additional interpreter’s booth, that we have to transfer its output to the other booths.

The initial plan

The expectations for the interpretation system in POA where high. There had been previous experience in Quito, Mumbai and Paris, and the International Committee of the WSF had taken the decision, early enough, that a “Nomad system” was to be used.

The initial design (see picture 1) was that:

1. for the communication to the public, FM transmitters and receivers would be used
2. for the relay interpretation, a special sound console for the interpreters (mixer, picture 2) was to be developed,
3. Targ, a software developed by the Nomad team to handle the relay interpretation, was to be used only as an additional feature, to archive and stream to the Internet the speeches and the interpretations, and not for the relay.

We should note that the Nomad team had proposed the use of computers running the Targ software as the tool for the interpreters, instead of using the consoles, but this solution was not selected. We do not explicitly know the reasons, but we suppose those have to do

1. with the big number of computers that this solution required,
2. with skepticism, regarding the reliability of this solution, due on the one hand to issues of maturity of the Targ software and on the other to the reliability of the computers, given moreover the fact that the electrical network in Brazil is not absolutely stable,
3. with the fact that the technical team in Brazil had knowledge about using sound systems, not computers.

For the implementation of the initial design, the team that was named Nomad Brazil was formed, consisting mainly of people with experience on electronics and local community radios (radio activists). Some persons from “Nomad International” did two invited visits to Porto Alegre from May 2004 onwards, to discuss about the system, while one person came to POA two months before the Forum to overview the procedure. The whole procedure was under the responsibility of the organization committee, of the Communications section.

Picture 2: The “Itairtron”, interpretation console, of Nomad Brazil

Nomad Brazil was responsible for the design of the console. The construction of the about 130 consoles that where required, was given to a local cooperative, under supervision from Nomad Brazil. Regarding the FM transmitters, there was the problem that the normal FM band, 88-108 Mhz, is too congested in POA. The solution that was selected was to tune the transmitters to transmit to lower frequency, 82-88 Mhz, that is the frequency of channel 6 of TV, which was free in POA. That had as consequence that special radio receivers had to be constructed (more correctly: standard radio receivers had to be modified) to be able to tune to those frequencies. The construction was assigned to some companies (cooperatives).

The revelation of the disaster

The date for delivery of the equipment had been set to January the 20th (the Forum begun on the 27th). The radio receivers had already been delivered, in two models, one analog and one with a digital tuner. Problems that emerged were that on the one hand the digital tuners could not tune under 85 Mhz (to the frequencies 82-85 Mhz), which made them practically useless, while several of the analog ones shifted (gradually lost tuning) and one had to tune them periodically. Several more of the analog receivers were not correctly calibrated, and tuned to different frequency than the one indicated.

Given that the delivery of the rest of the equipment (consoles and transmitters) was delayed, on Saturday the 22th a meeting was held, among the person in charge of the Communications section, Nomad Brazil and Nomad International, in which it was discovered that only about 10 out of the 130 required transmitters were ready, and 5-6 consoles! The circuits for both were constructed and were in the cooperative, but they were not assembled. In essence, four days before the beginning, they realized that there was no equipment. The situation was critical and required special measures
to be taken.

The proposal of Nomad Brazil was for the Forum to provide more volunteers that could help in assembling the transmitters, so that the cooperative could devote itself completely to assembling the consoles. My own proposal was to abandon the construction of the consoles immediately and do the relay using radios, and to give absolute priority to constructing the transmitters in the cooperative, due to that we should have the equipment early enough to have time to do the installation. The
Nomad International team did not have a common position (one of the Indians favored more the second position) and finally a kind of solidarity and comradely trust in Nomad Brazil functioned, and the first solution was selected. Moreover, two members of N.I. were to go to the factory the next day, to do an autopsy on the situation.

The plan was a complete failure: naturally, there was difficulty in immediately forming a group of technical volunteers to assemble the transmitters, while the construction of the consoles in the factory was having enormous problems and was not advancing, although it worked on a 24 hour basis. It seems that the designs for the consoles had arrived late, and the staff, excluding two persons, did not have the exact knowledge of how to do the soldering. The next day, the N.I. report for the situation in the factory was that there was no probability that the consoles would be manufactured within the next days, while there was a problem with the delivery of the transmitters as well. On January the 24th a new meeting was held, without the basic person in charge of the
Nomad Brazil team this time, and with the person in charge of the general organization of the Forum, where the fact that we should go for a “plan B”, an emergency plan, was acknowledged. That plan said that we would do the relay using radios, while the construction of the consoles was abandoned (in essence it is the second proposal that I mentioned before, but delayed for two days). There was one more big technical problem to be tackled: how we would connect the microphones to the transmitters, because there was no microphone input provided (pre-amplification of the mic was required before the signal would reach the transmitter). There were two possible solutions, the one using computers (we would use the sound cards that we already had, four per computer, as preamplifiers) and the second using sound equipment (mixers or pre-amplifiers). We decided to go for both in parallel. The reason was that on the one hand it was not sure that the computer solution would work, on the other hand there was no sound equipment available in POA to hire (everything was already rent for the needs of the Forum). Finally, it proved that the computer solution could not work (the software was not completely ready to cope with the situation, and furthermore there was not enough time neither for the installation, nor for the training of the special coordinators that would be required). But sound equipment (mixers) were found, which came by airplane from São Paulo. With these facts, the emergency system that was finally used in POA was as follows:

1. For the communication with the public, FM transmitters were used, the assembling of which was outsourced to a company in POA, as well as the analog radio receivers.
2. For the interpreters to hear the speakers in each room, one sound mixer was used, with input from the Public Address system (PA, the system with the amplifier and the loud phones), and 3-4 outputs directly to the headphones of the interpreters. That meant that the interpreters did not have volume control, because the control was on the one mixer, which was put in one of the cabins. This also meant that in some cases only one of the two interpreters in a booth could have headphones, because there were not enough outputs on the mixer.
3. For the relay, the interpreters would use radio receivers, just like the public, and they should tune to the station of the booth they wanted to hear, and use the radio’s headphones.

The delivery of a first part of the equipment was on the evening of the eve, and the installation of the first rooms took place at night, and continued on the first day of the Forum, January the 27th. There had been almost no training of the interpreters to that devised system, neither of the volunteers that supported the sound system (many of whom did not have almost any technical knowledge), resulting to the system “crashing” continuously, a fact that demanded the presence of somebody from the Nomad team to handle it. The whole situation created enough annoyance to the people that had come for the Forum and expected to have interpretation, as well as enough anxiety and much fatigue to the interpreters, who really did heroic efforts to save whatever they could, using consecutive interpretation or whispering, wherever the system was not set up or was down. Another fact that cost heavily to the procedure of installing and operating the system was that while the Forum had started and it was on the second day, and many rooms where still with no interpretation equipment set up, there emerged an issue regarding the payment of the workers in the cooperative: the Nomad International perceived as their duty to secure the payment of the workers before they proceeded to do the installation and the operation of the system. Luckily, after almost a full day of
debate, the decision was to make the efforts to secure the payment in parallel with installing and operating the system.

What went wrong

The Porto Alegre case gives many and useful conclusions to use for the next attempts. The positive in the whole procedure was that there was decision to apply alternative techniques, and there was action to that direction. The negative was of course the failure of the whole attempt. Below, we give the reasons, according to our opinion, for that failure: The basic reason was that the ones responsible in the organizing committee for managing the whole procedure did not have the technical knowledge required to cope with that duty. That fact inevitably led to assigning de facto the whole responsibility of supervising to the people who did the implementation, that is Nomad
Brazil primarily, and “Nomad International” secondly. Nomad Brazil was carried away by the vision of a technological success, which could probably have even commercial significance: the building of the interpretation console. That construction became self-purpose, and the general aim of providing a reliable system of interpretation for the realization of the Forum became a second priority. Nomad International, independently of their wishes or the side reasons [2], was proved not to have the capacity to follow the procedure of implementation of the system. Moreover, the planning and carrying out of the training of the volunteer supporters, as well as the planing and carrying out of the installation of the system on due time, proved once more a much more demanding task than the Nomad team could perceive. At this point, we should also note a political differentiation of the team from the main political agreements of the Forum, which, combined with the tactics of the group for political use of the technology, had results that from the viewpoint of the Forum could be considered problematic (i.e. “strikes” on critical moments). Furthermore, it is important that in spite of the creation of a team named “Nomad” in Brazil, the reality is that, during the whole process, but even more after the results, there was and is no communication, common political agreement or common sense of identity between the Brazilian part and the hard core of Nomad International (even Brazilians that take part in the electronic mailing list of Nomad International do not have contact with Nomad Brazil).

Summarizing, we could say that the tragic fact was that the whole procedure of designing and implementing the system depended on void, until the dead-end in which it was driven was proved.

Thanasis C., Greek Social Forum