Recommendation letters and institute-sealed copies of academic records have always some sort of problems for the applicant, either it be for further education or for the job. The problems in general compound when an applicant has different degrees from different countries, has employed in another country and have had studied in the institution where only one set of original academic records are issued. Usually we get two-three weeks time for detailed application package for the job (where there are several stages of appointment process as in World Bank, UN, ADB; applications --->preliminary sorting ---> detailed application package ---> selection for interviews ---> selection) and we need to complete within the this limited time period. Somebody argues that the best combination of referees for PhD studies would be master thesis supervisor, undergraduate project supervisor and the longest-duration employer if any (we can drop one of the last two when the university requires only two referees), for job application after PhD degree would be PhD dissertation supervisor, current employer if any and master thesis supervisor (we can drop one of the last two when only two referees are required) and for master course would not matter much who the referees are given the recommenders give good write-ups. However, the combination of referees also depends on the type of job. There is always a problem for a recommendation letter from the current employer. When you are performing well at your current employment, the boss hesitates to recommend you because you are planning to leave him/her and if you are not doing well, the probability of getting a good recommendation from the boss is very low except few who really are nice. As khbhandari experienced, the recommendation letter from non-English spoken professors are multi folds. First, they do not want to recommend to English-spoken professor due to their poor English ability and also a long slot of time they require to write it up. They really hate multiple recommendations. In fact, the English competency in a recommendation letter does not matter much. Surprisingly, they recommend very well in their native language, if translated from translation professionals. Some professors even ask applicant to write a recommendation letter on behalf of him or her. This is the most disturbing thing we have to do sometimes. Third, when he or she wants the applicant stay at his/her lab, they try not to recommend. However, if they recommend, they usually do a good job because you are the one who has satisfied him or her in academic achievements. Fourth, if you are currently employed with the help of a professor and you want to change the job, it is almost impossible to get the recommendation letter from him or her. The best way is to convenience him or her supported by a set of logical and professional reasons. Some professors even suggest that the job is not suitable for you or you do not qualify for the job, if he/she does not want to recommend or does not have time to recommend. In this case, it is better to change referee. Some professors are so nice that they are always there at any time to recommend you! We must be lucky enough to get such advisors. The institute-sealed copies of academic records is also a major problem. When the institute does not issue multiple copies, we have to certify them by the concerned issuing authority which usually requires physical presence. Some institute require applicant's physical presence or a representative. Some accept credit cards for the processing fee and some does not. Based on my personal experience, we have to have a list of at least 7 or 8 referees, pooled when required, for two or more job applications within a year's time. Oh! it's too long