[Got confirmation from William to quote him in public - please everybody rise your opinion here. Thank you.]
----- Forwarded message from Andreas Tille <[email protected]> ----- Hi William, On Thu, Jul 26, 2018 at 05:50:52PM +0100, William McCaffery wrote: > I've been working with Tony on Bio-linux 9, So Tony did not told you that I try to force people to open discussion on our mailing list? ;-) Honestly I care a lot for having a public record and keep the team informed about all issues except really private things and I have quite some record of shamelessly ignoring the netiquette by simply answering private mails in public. Since you are new and I'm not sure whether you want to see your e-mail address exposed on a publicly archived list this is just a warning and the question whether you give permission to bounce my mail to [email protected] and we keep on discussing there. Please be aware that I will do this without question in the future for very good reasons if no private content is in your mail. > and he mentioned that he had > planned to base the new package on med-bio. To this end, he suggested I > asked you about the process of doing this. Tony mentioned that I would have > to start by submitting a bug report? Well, you can do a bug report but we can also discuss this on the mailing list where typically more people pay attention to a topic that deserves a wider audience. I can assure you that I very open for integrating Bio-Linux and will definitely support your attempt. We just need to find out what might be the best solution for you. > I also was unsure of how to structure the meta-package. I have read the > blend documentation, but I am still confused by the structure of the deb-med > package source. Good so far. I suggest you register on Salsa for a login and I will add you to the Debian Med team to grant you commit permissions to the tasks. > Am I right in saying that we would only need to change > debian/control? Not really. debian/control is auto-generated. You might need to create a new (several new?) tasks biolinux (biolinux-featureA, biolinux-featureB, ...) and than run `make dist` which will recreate d/control and adds those task(s) to debian/control. > I also wanted to consult with you on the best way to fit in > a bio-linux-desktop meta package, since the package would have an overlap of > some packages from other meta-packages already within debian-med. Overlaps are no problem at all. Tasks are no exclusive categories but rather sets of packages that are needed for certain tasks. So don't mind about this. > In > addition to this I am not completely sure how Tony wanted the new package to > be based on the med-bio package. He said that you discussed this prior; is > this what he meant? We've been exchanging emails but I'm still a little > uncertain. Well, this depends. If med-bio (and may be med-bio-dev are fine for you you can simply Depend from these and be done with a zero maintenance effort. If we add new packages these will be mentioned there (and I have also developed tools to verify that everything is really included so you are safe that no new package will be missing). So lets discuss this all on the mailing list not wasting input of others (including Tony's) which have a valuable opinion (specifically since I'm not a user and thus I'm probably bad in guessing what users really need. Kind regards Andreas. -- http://fam-tille.de ----- End forwarded message ----- -- http://fam-tille.de
