Hello folks, some bad news from placements. The top 25 from CSE, the top 15 from ECE and the top 20 from IT departmenst went to the Coimbatore Institute of Technology (CIT) to attend offcampus Yahoo! placements. The pre-placement talk was good and then we had the AT. The AT spelt doom for all 60 from TCE. UNIX questions were truly difficult and most of my friends didn’t even try those questions. The greatest disappoinment to all of us was that inspite of neing the top 60 students of the college, not even one cleared the Aptitude test and went to the second round.
Some of the Yahoo! placement AT questions include shell scripting and PERL. For example, we were asked what would be the output of the script “kill $$” followed by “echo “hello world” “. There were lots of questions on UNIX commands and few from Operating Systems as well.
We had the AT from 3:15 to 4:45 and then were let free. The results were expected to be announced by 9. But they were announced only after 11. Three of our students (Sridhar, Arun and Balaji) were waiting for the results. Finally when it was announced, we were all totally disappointed. The fact that none of us could get through even the AT was really troubling and hurt us all a lot.But then, after a day now, on detailed analysis, we were able to find a few solutions. The solutions that I was able to come up with are,
- Complete change in syllabus atleast for CSE and IT .
- Change in exam pattern and methodology.
- UNIX classes for interested students after college hours.
A change in syllabus includes the fact that the syllabus must be the same as what the industry wants and not what all other colleges/universities follow. Possible suggestions include introducing GNU/Linux or UNIX in the earlier semesters. Advanced UNIX programming as an elective for those interested in operating systems( this could be an option for Advanced Databases) Another major issue is the role of SIGs. The SIG for FOSS must have sub groups for OS and networks as FOSS is extremely wide and has everything in the world of computers.
Change in exam pattern is the most important. Instead of testing the memory of the student by asking questions like explain/describe/write notes on, we can have questions like analyse/design/*. The questions of theory papers must also be thought provoking and interesting. For example, a question on file systems can be like “Design a file systems with support for the following ——” instead of asking explain file systems and “Describe a file systems with the following properties——“. I would also recommend giving mini projects for students every semester.
Activities of GLUGOT can be made more useful and more practical oriented instead of discussing about Emacs and SVN, there could be discussions on issues like memory management design or something like the design of emacs itself.