diff options
| author | jvoisin | 2019-05-11 00:19:17 +0200 |
|---|---|---|
| committer | jvoisin | 2019-05-11 00:19:17 +0200 |
| commit | dd9ead4ebeb04a82e8d1e7c72be319520aacd4d1 (patch) | |
| tree | fdc11e744e3f42a90d5598b70381e5edc4f7496b /doc/comparison_to_others.md | |
| parent | d0ab2c3023fd3be4e2eb1f76a8d8b273a099257b (diff) | |
Document how mat2 compares to other software
Diffstat (limited to '')
| -rw-r--r-- | doc/comparison_to_others.md | 51 |
1 files changed, 51 insertions, 0 deletions
diff --git a/doc/comparison_to_others.md b/doc/comparison_to_others.md new file mode 100644 index 0000000..8939009 --- /dev/null +++ b/doc/comparison_to_others.md | |||
| @@ -0,0 +1,51 @@ | |||
| 1 | # Exiftool | ||
| 2 | |||
| 3 | mat2 is in fact using exiftool to extract metadata from files, | ||
| 4 | but not to remove them. The previous iteration of mat2, MAT, | ||
| 5 | was using exiftool to remove metadata, which lead to several cases where | ||
| 6 | they weren't correctly removed, if at all. | ||
| 7 | For example, [Exiftool's documentation](https://www.sno.phy.queensu.ca/~phil/exiftool/TagNames/PDF.html) | ||
| 8 | states the following with regard to PDF: | ||
| 9 | |||
| 10 | > All metadata edits are reversible. While this would normally be considered an | ||
| 11 | advantage, it is a potential security problem because old information is never | ||
| 12 | actually deleted from the file. | ||
| 13 | |||
| 14 | To remove metadata, mat2 usually re-render the file completely, eliminating | ||
| 15 | all possible original metadata. See the `implementation_notes.md` file for | ||
| 16 | details. | ||
| 17 | |||
| 18 | |||
| 19 | # jpegoptim, optipng, … | ||
| 20 | |||
| 21 | While designed to reduce as much as possible the size of pictures, | ||
| 22 | those software can be used to remove metadata. They usually have very good | ||
| 23 | support for a single picture format, and can be used in place of mat2 for them. | ||
| 24 | |||
| 25 | |||
| 26 | # PDF Redact Tools | ||
| 27 | |||
| 28 | [PDF Redact Tools](https://github.com/firstlookmedia/pdf-redact-tools) is | ||
| 29 | a software developed by the people from [First Look | ||
| 30 | Media](https://firstlook.media/), the entity behind, amongst other things, | ||
| 31 | [The Intercept](https://theintercept.com/). | ||
| 32 | |||
| 33 | The tool uses roughly the same approach than mat2 to deal with PDF, | ||
| 34 | which is unfortunately the only fileformat that it does support. | ||
| 35 | It's interesting to note that it has counter-measures against | ||
| 36 | [yellow dots](https://en.wikipedia.org/wiki/Machine_Identification_Code), | ||
| 37 | a capacity that mat2 [doesn't possess yet](https://0xacab.org/jvoisin/mat2/issues/43). | ||
| 38 | |||
| 39 | |||
| 40 | # Exiv2 | ||
| 41 | |||
| 42 | [Exiv2](https://www.exiv2.org/) was considered for mat2, | ||
| 43 | but it currently [misses a lot of metadata](https://0xacab.org/jvoisin/mat2/issues/85) | ||
| 44 | |||
| 45 | |||
| 46 | # Others non open source software/online service | ||
| 47 | |||
| 48 | There are a lot of closed-source software and online service claiming to remove | ||
| 49 | metadata from your files, but since there is no way to actually verify that | ||
| 50 | they're effectively removing them, let alone adding unique markers, they | ||
| 51 | shouldn't be used. | ||
