Tamil Nadu HSE Second Year Computer Applications common mistakes
Tamil Nadu HSE Second Year Computer Applications · Understand the traps and how to avoid them.
When asked to convert a decimal fraction such as 0.625 or 0.3 into binary, the candidate converts only the integer part and either drops the fractional part or writes it unchanged after the binary point (e.g. answering 0.101 for 0.625 by guessing digit patterns instead of multiplying).
Textbook worked examples in this unit are almost all whole numbers, so candidates memorise the divide-by-2 table and never rehearse the multiply-by-2 procedure for the fractional side. Under exam pressure they pattern-match the visible digits rather than perform the repeated multiplication, and they do not notice that a terminating decimal like 0.3 produces a non-terminating binary expansion.
What to do instead
Drill the two-column method explicitly: integer part by repeated division by 2 reading remainders upward, fractional part by repeated multiplication by 2 reading the integer digits downward, stopping when the fraction becomes zero or the required precision is reached. Practise a set that includes 0.625 (terminates), 0.1 (repeats) and 13.75 (both parts), and always state the radix point position in the final answer.
The candidate writes 2's complement subtraction as 'flip all bits of the subtrahend and add 1' but then forgets to discard the final carry out of the most significant bit, or reports the result as negative because a carry appeared, giving answers like 1110 for 5 - 3 in a 4-bit register.
The rule is learned as a three-step ritual without the underlying idea that the carry out is the sign that the true result fits in the register width. Candidates also confuse the carry out with an overflow condition, which in 2's complement depends on the signs of the operands and the result, not on the carry alone.
What to do instead
Work every subtraction as A + (2's complement of B) and then state three things explicitly: the fixed register width, whether a carry left the MSB, and whether the sign bit of the stored result is 0 or 1. Verify each answer by converting back to decimal. Add two overflow-check questions using same-sign operands so the distinction between carry out and overflow becomes concrete.
In questions on memory hierarchy and cache, the candidate lists registers, cache, RAM and secondary storage in the wrong order, or claims that cache is slower than main memory because it is 'further from the CPU', and mixes up SRAM and DRAM characteristics.
The topic is studied as a vocabulary list rather than as a speed-cost-capacity trade-off, so ordering is recalled by feel. Candidates also carry over the everyday meaning of 'secondary' as 'extra' rather than 'lower in the hierarchy', and never link volatility to the SRAM/DRAM distinction.
What to do instead
Build one table with columns for typical access time, relative cost per bit, typical capacity, volatility and whether the CPU addresses it directly, and fill it for register, cache, main memory and secondary storage. Then answer past-style questions by reading values off that table. Memorise the single anchor sentence: faster means smaller and costlier per bit, and SRAM is fast, expensive and used in cache while DRAM is denser and needs refreshing.
The candidate defines a booting or operating-system question in one word ('starting') or lists only GUI functions, omitting the distinction between cold and warm booting, the role of the bootstrap loader in ROM, and the categories of system software versus application software.
Booting and OS functions appear early in the syllabus and are skimmed as easy marks, so the answer is written from general knowledge rather than from the defined sequence. Candidates also blur the boundary between an operating system and utility programs, which costs marks in classification questions.
What to do instead
Write a fixed six-line answer template for booting: power on, POST self-test, BIOS/bootstrap loader from ROM, loading of the OS kernel into RAM, initialisation of devices and services, transfer of control to the user. For classification questions, keep a running two-column list of system software (OS, language translators, utilities, device drivers) versus application software (word processor, spreadsheet, browser) and rehearse ten examples per column.
While answering a word-processing question, the candidate describes a mail merge but cannot name the two documents involved, or says the main document contains the variable data; in spreadsheet questions the candidate writes =A1+B1*C1 when the question asks for the total of the row, ignoring operator precedence and the need for brackets or SUM.
These practical features are learned by clicking through menus rather than by naming the components, so the terminology is fuzzy. In spreadsheets, arithmetic precedence is applied the way it reads left to right, and the candidate does not check whether the intended grouping matches what the formula actually computes.
What to do instead
For mail merge, memorise the pair: main document holds the fixed text and merge fields, data source holds the records, and the merged output is a third document. For formulas, always rewrite the intended calculation in words first, then bracket it, and verify precedence with a small numeric example (A1=2, B1=3, C1=4) so that =A1+B1*C1 and =(A1+B1)*C1 are seen to differ.
When asked to trace a loop, the candidate executes the body one extra or one fewer time, or reports the loop variable instead of the accumulated value, e.g. giving 10 for a sum of even numbers below 10 that should be 20, or counting the final increment as an iteration.
Tracing is done mentally in a single pass rather than in a table, so the update statement and the test condition are conflated. Candidates also confuse the loop counter with the accumulator, especially when both are named i and s in the same short program.
What to do instead
Trace with a mandatory four-column table: iteration number, value of the loop variable at the test, values of any accumulators after the body, and the test result that decides continuation. Write the condition's truth value as yes or no in every row, and finish by stating the final printed variable by name. Do this for a for loop, a while loop and a do-while loop on the same numbers so the difference in when the test runs becomes visible.
The candidate writes a function that modifies its parameter and expects the caller's variable to change, or declares an array of size n and then loops from 1 to n inclusive, reading or writing one element past the end of the array.
Pass-by-value and pass-by-reference are taught as definitions but not as observable behaviour, so the candidate assumes the language 'knows' which variable was intended. Off-by-one array indexing comes from habitually counting from 1 in mathematics while the languages in this syllabus index from 0.
What to do instead
Write one short program twice, once passing an integer by value and once by reference, print the caller's variable after each call, and keep that pair as a reference note. For arrays, adopt the fixed pattern of declaring size n, looping i from 0 while i < n, and stating in the answer that valid indices are 0 to n-1. Add a deliberate out-of-bounds example and note what is printed versus what is undefined.
In SQL questions, the candidate writes SELECT * FROM student WHERE marks = MAX(marks), or puts an aggregate such as COUNT or AVG directly in the WHERE clause, and omits GROUP BY when a per-group aggregate is asked for.
The candidate translates the English sentence word by word into SQL, treating MAX as an ordinary function usable anywhere. The evaluation order of a query, in which WHERE filters rows before grouping and HAVING filters groups afterwards, is not internalised, so the placement of conditions is guessed.
What to do instead
Memorise the clause order FROM, WHERE, GROUP BY, HAVING, ORDER BY and the rule that aggregate functions may appear in the select list and in HAVING but not in WHERE. For 'highest marks' questions, practise the subquery form WHERE marks = (SELECT MAX(marks) FROM student) and also the ORDER BY marks DESC with a row limit, and write the evaluation order in the margin before composing any query.
The candidate confuses a hub with a switch, or a router with a modem, and states that an IP address identifies a computer permanently while a MAC address changes with the network; in HTML answers the candidate writes tags without closing them or uses deprecated presentational tags such as <font> and <center> when the question asks for CSS.
Networking devices and addressing are learned as isolated definitions without the layer or the purpose that distinguishes them, and HTML is typed from memory of old tutorials rather than from the current separation of structure and style. The candidate also does not re-read the question to see whether CSS was demanded.
What to do instead
Anchor each device to one sentence about what it forwards and to which layer: hub broadcasts to all ports, switch forwards by MAC within a LAN, router forwards by IP between networks, modem converts between digital and the transmission medium. For addressing, state that MAC is the hardware address fixed by the manufacturer and IP is the logical address assigned per network. For HTML, keep a checklist of required closing tags and of presentational tags replaced by CSS properties (font to font-family, center to text-align, bgcolor to background-color).
In the emerging-technology and cyber-safety section, the candidate describes cloud computing only as 'storing files online', cannot match IaaS, PaaS and SaaS to examples, and in safety questions recommends forwarding a suspicious link to verify it or advises sharing an OTP with bank staff.
These topics are recent additions to the syllabus and are revised last, so answers come from advertising language rather than from the service and deployment models. Cyber-safety items are answered from everyday instinct, and the candidate does not distinguish authentication factors from information that must never be disclosed.
What to do instead
Learn one example per cloud model: IaaS as rented virtual machines, PaaS as a managed application platform, SaaS as a ready-to-use application such as webmail, plus the public, private and hybrid deployment terms. For safety, memorise a short refusal list (never share OTP, password or PIN; never forward a suspicious link; verify through the official channel) and a short action list (strong unique passwords, two-factor authentication, report and block, keep software updated). Rehearse by classifying ten scenarios as safe or unsafe and justifying each in one line.