On Sat, 17 Jun 2006 00:07:56 +0200 Nils Larsch <[EMAIL PROTECTED]> wrote:
> > was initially designed to allow binary only drivers. these days I think the > > belpic way is better - copy opensc, rename opensc to foobar and add your > > driver. open source everything except card-newdriver.c, but include the > > *.o file (and specify which linux/compiler/... you used). > > sorry but why is this approach better than supplying a binary driver ? > I think this is even worse as this will create unnecessary forks ... With or sithout fork, I believe that binary modules are very difficult to maintain. Look for example at the eID landscape where OpenSC is the ideal vehicle to support multiple eIDs in a single middleware for multiple platforms. Even if you complile only for a single platform, keeping up a binary module with the evolution of OpenSC is not guaranteed and (due to the difficulty to get test cards) is likely to remain untested. But even if you manage to get this working, things will surely break when you want to complile for different platforms, maybe some nice embedded device running linux on an embedded processor... So in my book, open source drivers are the only reasonable solution--at least for eIDs that should be used universally wherever one needs strong authentication.. This is also why I think that a regression test for all supported cards would be a nice thing to have.. best cheers -b -- Ing. Bud P. Bruegger, Ph.D. +39-0564-488577 (voice), -21139 (fax) Servizio Elaborazione Dati e-mail: [EMAIL PROTECTED] Comune di Grosseto jabber: [EMAIL PROTECTED] Via Ginori, 43 http://www.comune.grosseto.it/ 58100 Grosseto (Tuscany, Italy) http://www.comune.grosseto.it/interopEID/ _______________________________________________ opensc-devel mailing list opensc-devel@lists.opensc-project.org http://www.opensc-project.org/mailman/listinfo/opensc-devel
