carlkesselman
commented
6 years ago Not quite sure what you are asking here. What pipeline? We should talk. You can probably leverage a bunch of the code I have already implemented for interacting with the data management system.
There is no specification as the design changes frequently given bill’s whims.
Carl
Sent from my iPhone
On Aug 17, 2018, at 7:06 PM, aandreev notifications@github.com<mailto:notifications@github.com> wrote:
I currently have no design doc for data storage system. That makes it hard to ask proper questions, let alone design pipeline. We ideally want to be able to write single script that runs whole pipeline given few input parameters, that ultimately generates figures for paper. Right now there are many steps already, and I don't think anyone but Bill knows them and their order. For paper we'll also need some description of the pipeline.
Thus, spec and design doc will increase bus factor, as well as server for better design of code, and ultimately fill paper's methods section
— You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHubhttps://github.com/informatics-isi-edu/synspy/issues/67, or mute the threadhttps://github.com/notifications/unsubscribe-auth/ADbjXhxT60ndqnBxhZ_PxO7uFyfaPuOgks5uR3azgaJpZM4WCYuD.
karlcz
I currently have no design doc for data storage system. That makes it hard to ask proper questions, let alone design pipeline. We ideally want to be able to write single script that runs whole pipeline given few input parameters, that ultimately generates figures for paper. Right now there are many steps already, and I don't think anyone but Bill knows them and their order. For paper we'll also need some description of the pipeline.
Thus, spec and design doc will increase bus factor, as well as server for better design of code, and ultimately fill paper's methods section