> http://svn.debian.org/wsvn/pkg-scicomp/atlas/?rev=0&sc=0 > > the package builds in pbuilder, so all is fine. > > I'll see what is the best way to achieve the goal. One approach is to > generate quilt patches using the current packaging, but I fear it will > be easier to start from scratch, with the new upstream version. I think you are right. It seems pretty tricky to maintain.
> In > this case, I'll probably create a testing repo just for this and I'll > ask some DD to upload it to experimental, so that we can all try it > and improve it. And use the old repo to fix possible (RC) bugs with > the current packaging in unstable. > > And I am sure people will join in. In the Debian Scientific > Computation Team we use svn, so let's just use svn. If the whole team > migrates to git, we'll migrate to git, otherwise not, so that all the > packages are at one place. > > And I'll do the same with lapack and blas. I'll be out of town till > Sunday, but then I'll work on it. I can help for lapack and blas. I have been using them for Scilab. Sylvestre -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
