After the completion of the 1st step of the project, companies listed below were selected to participate in the 2nd step of the project. These are Bodicomputer, Interactive, Orgiltech, Engineersoft, Infocon, JEMR, Datacom and BSB Co.Ltd. The companies had a meeting and decided to create a working group under MIDAS. Mr. Natsagdorj, who is the Vice-Chairman of the MIDAS Council, a member of National ICT Committee and the General Director of the Bodicomputer Co.Ltd was elected as a leader of the group. Engineersoft Co.Ltd expressed not to join the group due to the nature of their business and their unavailability.

The MIDAS secretariat was also tapped as the temporary secretariat of the project in order to handle data entry job registration, dissemination, checking, receiving from the companies, sending outputs back to USA, financial issues related to the job and maintain a connection with the participating parties. E-job@ict.mn e-mailing list is serving e-job the participants for the sharing of information.

An FTP was also established for downloading job and uploading results using the Bodicomputer Co.Ltd ISP.

The working group was operating on testing of the data entry job. Especially, they were spending more time for developing a technology, which can be suitable and profitable for the companies.
At first, sample data entry job materials were distributed to the companies and shared information about the technologies, which they will use for the job. During the meeting they also discussed advantages and disadvantages of the technology according to their experiences. There were several different technologies presented. However, the main technologies that the group opted were the following:

- Separate entry by two typists and comparing the outputs by third person
- Entry using OCR and checking

Mr. Ganbold has developed software for the 1st technology - for the automatic comparison and correction of the documents which is done word for word.

The disadvantage of the first technology was its high cost due to double entry and comparison feature. For the second case, the companies use different softwares for the OCR. Correctness of the output from OCR was 80-85 percent. Thus, it was cheaper and quicker way for the data entry.

Two options were sent to Jeremy for comments and discussion.

After the discussion the group has developed another 2 options of the technology, which are mentioned below.

The first option is: the first typist will type directly then the second typist will check while typing again. The second option is: two people will use separate OCR for the data entry and compare the two outputs. Then will use HTML tagging software in order to place tags.

Mr. Ganbold has developed the following new software:
- SW1 to compares two text files
- SW2 for tag generation and tag validation
- SW3 compares typed text with existing text.

Achievement:
- Established working group and procedures of cooperation
- Technologies have been developed, selected and adapted
- Necessary software have been developed
- The software will be useful not only for this project but also it will be useful for other purposes such as web site design, document comparison and etc.

Problems:
- Communications between Jeremy and the working group were sometimes confusing because of language barriers
- Even though we have received instructions for entering tags during the testing, a lot of questions still arose related to the tags.

Future perspective:
- Improve and advance the experiences of the companies involved in the data entry job, in order to be experts or specialist in this job
- Involve other interested parties for this job
- Create separate a permanent secretariat for the management of this job.
- The working group will establish strategies to independantly obtain international assignments.