![]() |
This is a running job jar of items in support
of the toolcraft
MD5 Checking Project.
Status |
Date |
Description |
2004-06-14 | I just found a source for a CRC-64 checksum calculator, which is different than MD5sum but might be useful anyhow. <http://lists.boost.org/MailArchives/boost/msg27062.php>. | |
2004-04-02 | The Unix/GNU/Linux implementations of MD5sum may use file names and locations (paths) that don't work in an interoperable way. Any MD5check tool needs to be portable or be able to support reconciliation of toolcraft and platform conflicts. This needs to be reflected in the requirements and the process of coming to a conclusion on this. And it is all to be confirmable. | |
2004-04-02 | One interesting problem is how do we verify something that produces a one-way cryptographic result. This is a nice place for some performance architecture. In addition to boundary conditions, what we have to go on is existing implementations and our ability to duplicate the results. Then, even if a common implementation is "incorrect," we have interoperability to deal with. There are also boundary conditions to deal with. That is, a 0-bit message, a 1-bit message, etc., what the padding rules are, what the transitions for multiple blocks are, etc. Then all of the work to crack MD5 hashes (that is, find an useful collision of a real hash). | |
2004-04-01 | Consolidate all open-source material on cryptography, hashing and such on compagno, as a resource for further work here. | |
2004-04-01 | Notice that hashing is not the same as signing, though the hash is something to verify, and that signing is not complete without also having a reliable timestamp. So I will have to know about timestamp services before this is all completed. There is a discussion on how these all go together to satisfy the various requirements of a good signature. Find that and incorporate it somewhere, around the authentication of software, data, and other filed materials. | |
2004-04-01 | Make a start on requirements for confirming MD5 implementations and also understanding the format of MD5sum files (*.md5) that carry hashes for several files in a directory, or package. | |
2004-04-01 | I have collected substantial material on MD5 under open-source crypto work, and also material on related topics. These need to be mined from centro's open source, the crypto folder I started here, and the notes and references I have under IETF RFCs and the MSC-CC course discussions on security. Also, the homework I did in Java is relevant to some of this. | |
2004-03-19 | Find the history of the hash and the Microsoft use of it. (apparently in challenge-response authentication) | |
2004-03-19 | Look on SourceForge and in various libraries for appropriate tools. | |
2004-03-19 | MD5 hashes are provided in HTTP responses under appropriate conditions - look at that too. | |
2004-03-19 | There is use of MD5 in IETF specifications, dig further there. | |
2004-03-19 | There is discussion and implementation of MD5 in Scheier's Applied Cryptography -- review and cite it. | |
done | 2004-03-19 | Add job jar and diary to P040201 so that notes on MD5 can be tabulated, at least. |
2004-02-28 | Create initial toolcraft project on MD5 checksums and the files that include MD5 sums. It is an example of spiraling of a kind, because one thing to MD5 check is the MD5 tool. So, there should be a project sheet for the MD5 checksum, and some steps to the project. Get a starter there. The first item is to gather existing materials and the email discussions. This is going to be a gathering of collaborative material that Bill Anderson and I can claim jointly, perhaps referring to it from any articles we produce. There are going to be holes that need to be filled in. That needs to be spiraled also. | |
![]()
|
created 2004-03-19-12:12 -0800 (pst) by orcmid |