Reference Formats
CSV
A plain-text tabular format that represents rows of data as lines of comma-delimited fields, widely used for exporting and exchanging spreadsheet and database data.
Extension
.csv
Full name
Comma-Separated Values
Introduced
in informal use since 1970s-era mainframe data exchange; formally documented as RFC 4180 in October 2005
Vendor
No single vendor — a de facto format later documented by the IETF (RFC 4180, authored by Yakov Shafranovich)
What is inside the container
CSV has no container structure — it is plain text where each line represents a row and commas (or occasionally other delimiters like semicolons or tabs) separate fields, with quoting rules for values that themselves contain commas or line breaks. Unlike XLSX or ODS, CSV cannot hold multiple sheets, formulas, cell formatting, or embedded media — it is data only, which is precisely why it's a common lowest-common-denominator export format between otherwise incompatible systems.
How to open it
For archives
CSV is an excellent format for archiving pure tabular data in a DAM because it's plain text, application-independent, and documented by RFC 4180, but it should not be treated as an archival replacement for a full spreadsheet: it drops formulas, formatting, multiple sheets, and charts entirely. A DAM should store CSV as the data-interchange layer for tabular assets while keeping XLSX or ODS as the master when those richer features matter, and should record the delimiter and character encoding used, since neither is self-describing in CSV.
CSV predates any formal specification by decades — comma-delimited text files were used for exchanging tabular data between mainframe and early business systems going back to the 1970s, long before any standards body wrote down the rules everyone was already loosely following. That informality became a real interoperability problem as different tools handled quoting, embedded commas, and line endings inconsistently, which is what prompted Yakov Shafranovich to publish RFC 4180 in October 2005 — an informational IETF document that codified the de facto conventions and registered the text/csv MIME type, rather than inventing a new format.
CSV is one of the most common tabular formats a DAM encounters precisely because it’s the least opinionated: it’s the export format of choice when moving data between a spreadsheet, a database, and a script, since virtually every tool that touches tabular data can produce and consume it. For a DAM, that simplicity cuts both ways — full-text and column-level indexing is trivial since there’s no binary or XML layer to parse, but the format’s total lack of type information (a date, a number, and a string all look identical as raw text) and lack of a declared encoding mean a DAM’s ingestion pipeline has to make explicit decisions about delimiter, quoting, and character encoding rather than reading them from the file itself.
- Delimiter isn’t always a comma — semicolons are common where commas serve as the decimal separator
- No declared character encoding — a DAM has to detect or be told it explicitly
- No type information — dates, numbers, and text all look identical as raw characters
- No formulas, formatting, multiple sheets, or charts — data only
Frequently asked
Why do CSV files sometimes show scrambled characters or broken columns when opened in a DAM preview?
CSV has no self-describing character encoding or delimiter declaration, so if the previewer assumes the wrong encoding or delimiter, text and column boundaries can appear broken.
Can a CSV file store multiple sheets like XLSX or ODS?
No, CSV represents only a single flat table; multi-sheet workbooks need to be exported as separate CSV files, one per sheet.
Does CSV preserve formulas from the original spreadsheet?
No, CSV stores only the calculated values as plain text, not the underlying formulas — this is one of the key limitations to flag when CSV is used as an export from a formula-heavy spreadsheet.
Is CSV a good format for archiving spreadsheet data long-term in a DAM?
For pure tabular data, yes — it's plain text, application-independent, and documented by RFC 4180 — but it shouldn't replace a richer format like XLSX when formulas, formatting, or multiple sheets need preserving.
What delimiter does CSV always use — is it always a comma?
Despite the name, real-world CSV files often use semicolons, tabs, or other delimiters (semicolons are common in regions where commas are decimal separators), so a DAM's parser needs to detect or be told the actual delimiter rather than assuming comma.
Can a DAM validate that a CSV file is well-formed before ingesting it?
Yes, checking for consistent field counts per row, correct quote escaping, and a declared encoding are standard validation steps before treating a CSV as reliable structured data.
Why is CSV commonly used to export data from a DAM's asset metadata?
Because it's universally readable by spreadsheet software, databases, and scripts without any proprietary library, making it a practical lowest-common-denominator format for bulk metadata export and import.
Can CSV files carry malicious content that a DAM should scan for?
CSV can be exploited via formula injection if a cell value starting with =, +, -, or @ is opened in spreadsheet software that auto-executes it as a formula, so a DAM exporting user-supplied data to CSV should sanitize such leading characters.
Sources
- RFC 4180, 'Common Format and MIME Type for Comma-Separated Values (CSV) Files,' was published by Yakov Shafranovich in October 2005, formally documenting existing conventions rather than defining a new standard. checked 2026-08-07 — IETF RFC Editor - RFC 4180
- CSV has been used for data exchange between spreadsheet programs since long before it was formally documented, with the format never having a registered MIME type prior to RFC 4180. checked 2026-08-07 — Library of Congress - CSV, Comma Separated Values (RFC 4180)