Hello all, Giorgio left this comment on the Confluence post and I thought it was helpful to put it into this thread.
Giorgio: My idea for having and growing the community it to have something to present at the next Embedded Linux Conference in Lyon, the call for paper ends beginning of July. This will give us the right visibility for saying "We are back!". As always depends on us. Some thoughts: 1. For the server i don't know what you have since in the repo we have just an excellent set of python servers. We can follows two paths or just one, port all to Modern C++ (programming in modern c++ is not so scary) or move the ideas to go grpc, depends on what you have. 2. My vision for the server architecture is a federated system of different microservices both for the RPA and DTA. Here we have a lot of challenges on different things for example load balancing (look at envoy proxy), fault tolerance, kubernetes support. Brick over brick. The first thing is to a docker-enabled D-TA server. Then will come the RPS. 3. The current repos needs to be reorganized. For the c++ milagro library after the bug on config64.py will be resolved (filed today), i have a CMakeLists.txt and a conan recipe. Conan https://conan.io/ is a decentralized package manager for C++ (for java developers is like gradle). So the library will be present in the JFrog repo. 4. Further ideas: a. Modern Cloud enviroments support HSMs and an opportunity to evaluate is the to integrate pkcs11 support in the servers for storing and generating secrets. (https://docs.aws.amazon.com/cloudhsm/latest/userguide/pkcs11-library.html ) b. The rpa/rps clients might be as well cars so we could explore how to use intel/arm in the tpm module for storing sensitive information in a board. I imagine a RPS client that from a car does authentication before dowloading navigation datas and stores in our servers sensitive information because they trust us. c. For the javascript demo would be nice to hava a react/login component to be integrated in asp.net core application. Just plug and play. Estimated time a couple of days. On 1 Jun 2019, at 17:46, Brian Spector wrote: > Hello all, > > I have spent quite a number of days on Apache Milagro over the last > couple of weeks, and I’m hopeful that we now have the tools at our > disposal to succeed. > > On that note, the committers have accomplished the following in May: > > 1. Updates and bug fixes to the core Milagro crypto library. > 2. 3 new committers - Giorgio Zoppi, Chris Morris, and Alessandro > Budroni and 2 contributors - John McCane Whitney and Howard Kitto. > 3. 1 new PPMC - Giorgio Zoppi because he’s such a rock star > 4. Setup a Milagro Confluence site where all contributors and PPMC > members can create content at will. > 5. Got myself added as admin to Milagro JIRA, and have added some > contributors, all committers and PPMC members so that we can > collaborate > at will and efficiently. On that note, I have also setup the Kanban > board and cleaned out the dead/redundant/duplicate issues. > https://issues.apache.org/jira/projects/MILAGRO/issues/MILAGRO-18?filter=allopenissues > 6. Got back in touch with Patrick Hilt, which was nice, and Patrick > also > want to remain involved on the PPMC and contribute where he can to > Milagro development. > 7. Had Qredo sign a CCLA, and have ICLA’s on file for all > contributors > working at Qredo. > > Shout out to Justin Mclean who has been exceptionally helpful. > > To Do: > > 1. File Podling report for June, due in a few days. > 2. Move existing website on SVN to new githubpubsub ‘asf-site’ > branch publishing system. Will file JIRA request with Infra. > 3. Travis - CI for milagro-crypto-c and milagro-javascript repos. Will > file JIRA request with Infra. > 4. Sadly, we will need a new mentor as Nick Kew has been unresponsive, > which is hardly surprising given the disappointing progress made > before > this month. Hopefully we can get him re-engaged at some point. I will > send a solicitation to IPMC to see if there are any kind souls willing > to help. > > GOING FORWARD: > > What I need from all committers and engaged contributors is some > feedback and discussion on direction for Milagro. I have put these > thoughts and proposals into a blog post on our Confluence page. > > REQUEST: Please put your comments into the reply on this thread, not > in > the comments section on Confluence. Tracking comments in both areas > will > slow down consensus building. > > Please review the post here: > https://cwiki.apache.org/confluence/display/MILAGRO/2019/06/01/Updating+Milagro > > Thanks > Brian
