Timeline for MD5 Checksums For Different Versions Of Mathematica
Current License: CC BY-SA 3.0
19 events
| when toggle format | what | by | license | comment | |
|---|---|---|---|---|---|
| Oct 27, 2016 at 18:37 | answer | added | ktm | timeline score: 4 | |
| Oct 27, 2016 at 18:31 | comment | added | ktm | It might also be useful to link to one of Wolfram's articles on checking MD5 check sums: support.wolfram.com/kb/25578. | |
| Oct 27, 2016 at 17:56 | history | edited | William | CC BY-SA 3.0 |
added 148 characters in body
|
| Oct 12, 2016 at 5:05 | history | tweeted | twitter.com/StackMma/status/786070286663839744 | ||
| Oct 12, 2016 at 2:34 | comment | added | William | @user6014 might that be you? :) In all seriousness that is a good idea. | |
| Oct 12, 2016 at 1:48 | comment | added | ktm | A longtime user of Mathematica would be able to go to user.wolfram.com and attempt to download all the versions available to them to see all the revent MD5 checksums... | |
| Oct 11, 2016 at 23:39 | history | edited | Karsten7 |
edited tags
|
|
| Oct 11, 2016 at 18:40 | history | edited | William | CC BY-SA 3.0 |
added 148 characters in body
|
| Oct 11, 2016 at 18:36 | answer | added | William | timeline score: 3 | |
| Oct 11, 2016 at 16:56 | history | edited | William | CC BY-SA 3.0 |
deleted 9 characters in body
|
| Oct 11, 2016 at 16:55 | comment | added | Jacob Akkerboom | @EmilioPisanty yes, I should have been more specific, I should have instead said " Unfortunately providing unsigned checksums on the same site as the downloadable application ...". Anyway the second link was only meant to provide a second source that you should not have the wrong expectation of such (specific) checksums. | |
| Oct 11, 2016 at 15:14 | comment | added | Emilio Pisanty | That said, I'm not sure how strong MD5 is to chosen-prefix collision searches, particularly where the 'prefix' is the full MM .exe installer, possibly with some malware attached. If the ebay seller can modify the installer and then add junk that will make the hash match that from the original file, then that is the main worry. | |
| Oct 11, 2016 at 15:12 | comment | added | Emilio Pisanty | @Jacob I don't understand how your second link applies. The point there is that an MD5 hash should be considered as compromised as the file (security-wise) if obtained from the same source. The point here is to create a trusted repository (or as trusted as this community) of hashes, which can be protected with https in transit. | |
| Oct 11, 2016 at 7:53 | comment | added | Jacob Akkerboom | Another thing to note is that changing a file like KeyEventTranslations.tr will change the checksum. | |
| Oct 11, 2016 at 7:52 | comment | added | Jacob Akkerboom | So my point is that if we are going to do this, we should post other cryptographic hashes, possibly alongside the md5 ones. | |
| Oct 11, 2016 at 7:51 | comment | added | Jacob Akkerboom | Note that md5 checksums are not very reliable anymore. From its wiki page: "Although MD5 was initially designed to be used as a cryptographic hash function, it has been found to suffer from extensive vulnerabilities. It can still be used as a checksum to verify data integrity, but only against unintentional corruption". Unfortunately providing unsigned checksums on a site leads people to believe that checking such a checksum is the main measure against intentional data corruption, but again it is not. | |
| Oct 11, 2016 at 4:09 | history | edited | William | CC BY-SA 3.0 |
edited tags; edited title
|
| Oct 11, 2016 at 3:59 | history | migrated | from mathematica.meta.stackexchange.com (revisions) | ||
| Oct 11, 2016 at 3:53 | history | asked | William | CC BY-SA 3.0 |