NIELIT BCC common mistakes
NIELIT BCC · Understand the traps and how to avoid them.
Treating 1 KB as 1000 bytes and 1 MB as 1000 KB when converting file sizes in numerical questions, so answers come out off by roughly 2.4% to 4.9%.
Candidates carry over decimal SI prefixes from school mathematics or from storage-marketing labels (where a '500 GB' drive is decimal) and never learn that memory addressing and most exam questions use binary multiples: 1 KB = 1024 B, 1 MB = 1024 KB, 1 GB = 1024 MB.
What to do instead
Before the exam, memorise the ladder 2^10 = 1024 and practise five conversions each way (bytes to KB, KB to MB, MB to GB). In any question, first ask whether the context is RAM/cache/address space (binary) or disk marketing (decimal); when a question gives no context, assume binary. Write the multiplication factor on your rough sheet before calculating.
Confusing ROM, RAM, cache and virtual memory when asked which memory is volatile, which is non-volatile, or which is fastest.
All four are described as 'memory' and candidates memorise isolated facts ('RAM is volatile') without building a hierarchy, so under time pressure they swap cache and RAM or claim virtual memory is a physical chip.
What to do instead
Draw one hierarchy diagram from registers, cache (L1/L2/L3), RAM, then secondary storage, marking each as volatile or non-volatile and fastest to slowest. Add a note that virtual memory is a disk-backed extension managed by the OS, not a chip. Redraw it from memory the day before the exam.
Answering 'which device is input vs output' by naming the peripheral's purpose rather than the direction of data flow, so a touchscreen or a modem is classified wrongly.
Candidates learn short lists (keyboard, mouse = input; monitor, printer = output) and have no rule for hybrid devices, which appear in modern syllabi but not in older lists.
What to do instead
Adopt the rule: input means data flows into the CPU, output means data flows out, and a device doing both is input/output (touchscreen, modem, network card, headset). Classify ten mixed devices using the rule, not memory, and mark the hybrids explicitly on your revision sheet.
Mixing up the OSI layer number and the TCP/IP layer name for the same function, for example calling IP a transport-layer protocol or placing TCP at the network layer.
The two models are taught side by side, and candidates memorise layer names as a list without attaching protocols to layers, so they map by vague association ('IP is networking, networking is layer 3' is right, but the same reasoning fails for TCP and UDP).
What to do instead
Build one table with seven OSI rows, the matching TCP/IP layer beside each, and the protocols that live there: application (HTTP, FTP, SMTP, DNS), transport (TCP, UDP), internet/network (IP, ICMP, ARP), network access (Ethernet). Also note TCP is connection-oriented and reliable while UDP is connectionless, since questions test that contrast.
Choosing an IP address or subnet answer without checking whether the given address is private, loopback, or valid at all, so 192.168.1.10 is called public or 127.0.0.1 is called a network address.
Candidates learn the address ranges as trivia but never apply them as a filter, and they skip the quick sanity checks (each octet 0 to 255, private blocks 10.x, 172.16 to 172.31, 192.168.x).
What to do instead
Run a three-step check on every address question: octet range validity, private versus public using the three reserved blocks, and special cases (127.0.0.1 loopback, 169.254.x link-local, 255.255.255.255 broadcast). Practise on ten addresses until the check takes under twenty seconds.
Using the wrong spreadsheet cell reference type, so a formula copied across rows gives wrong results because a relative reference shifted when an absolute one was needed.
Candidates practise formulas only in the cell where they typed them and never copy them, so they never see relative references drift. The dollar-sign syntax (A1, $A$1, A$1, $A1) is memorised as symbols rather than as 'lock column, lock row'.
What to do instead
Learn the rule: each $ locks the part after it, so $A$1 locks both, A$1 locks the row, $A1 locks the column. Build a small multiplication table with one locked header row and one locked header column, copy it across and down, and verify the results by hand once.
Writing formulas that reference a cell by typing its value instead of its address, or omitting the leading = sign, so the sheet stores text rather than a calculation.
Candidates who are used to calculators treat the cell as a place to write the answer, and in office suites a formula without = is displayed literally, which looks correct on screen until the source data changes.
What to do instead
Adopt a fixed habit: every calculation starts with =, always points to cell addresses, and is verified by changing one input and watching the output update. Practise the core functions SUM, AVERAGE, COUNT, COUNTA, MAX, MIN, IF and the difference between COUNT (numbers only) and COUNTA (non-empty).
Treating email fields as interchangeable, so a whole team is placed in To, or a recipient who should only see a copy is placed in Cc rather than Bcc.
The difference between To, Cc and Bcc is a convention rather than a technical rule, and candidates who rarely send formal mail never learn that Bcc protects addresses from disclosure and that Cc is for visibility, not for action.
What to do instead
Memorise the purpose of each field: To for people expected to act, Cc for people who need to know, Bcc for large lists or privacy so recipients cannot see each other. Also learn that Reply goes to the sender while Reply All includes Cc recipients, and that forwarding can leak earlier thread content.
Judging a website's trustworthiness by its visual design or its position in search results rather than checking the URL, the padlock, and the domain owner.
Candidates equate professional appearance and top ranking with legitimacy, but sponsored results and copied designs are cheap, and the address bar is the only reliable signal of who actually operates the site.
What to do instead
Apply a fixed checklist before entering any personal data: read the domain from right to left (the part before the final dot is what matters), confirm https and the padlock, look for a verifiable organisation behind the site, and treat urgency, prize claims, and requests for OTPs or passwords as phishing signals regardless of design quality.
Confusing a digital signature with a scanned handwritten signature or with encryption, so questions about authentication, integrity and non-repudiation are answered by guessing.
The word 'signature' pulls candidates toward the paper analogy, and the underlying public/private key roles (sign with private key, verify with public key; encrypt with public key, decrypt with private key) are learned as two similar sentences that get swapped.
What to do instead
Fix the key roles with one sentence each: signing uses your private key and proves origin and integrity; encryption uses the recipient's public key and provides confidentiality. Then attach the property names authentication, integrity, non-repudiation to signing, and confidentiality to encryption, and test yourself on which key does what in each direction.