On Tue, 2006-06-20 at 12:12 +1200, Jim Pye wrote: > Matthew et al. > One topic that seems to be missing from the future exams is email.
To me, it's more than just e-mail. I put this under the domain (in my 8 domain model of services**): "Collaboration and Mail Services" What an ISP uses for e-mail is not what an enterprise WAN would. An enterprise is going to be interested in shared folders, scheduling and other server-side stores and resolution. **NOTE: That list is here: http://en.linux.wikia.com/wiki/ELResource#Domains I will be filling in the subdomains by concept (technology) and practice (actual services). Eventually I'll write some tasks (and hopefully enlist the help of others so we'll have a real resource, and not just another set of "standalone" HOWTOs that don't help much). > I do not probably have the skills required to become the lead on this > topic I don't think any of us do. E.g., I haven't deployed all of the new NFSv4 RPC services myself, but am used to doing Kerberos-authenticated and authorized NFSv3. I'd say 99.99% of people who have messed with NFS outside of an enterprise environment have only used the old, old "sys" model (i.e., trust IP, trust ID). > but to stretch the knowledge of the candidates I see the following > topics: > MTAs > Sendmail > Postfix > ?? > Protocols > SMTP > POP > IMAP > ESMTP > Secure POP, IMAP E.g., you can get into a whole can of worms with authentication and access control on IMAPS. That's why we really need the "auth/dir/name" exam -- because we don't need to recover those concepts, only how a IMAP service might use them. ;-> > Security > Certificates > Encryption Again, more "auth/dir/name" foundation. Let's not recover these things for Apache, Mail, etc... exams on their own. ;-> > Virus scanner (ClamAV*) > Spam (spamassasin*) If we're just talking e-mail, that might do. And it would fit fine in an ISP-geared examination. But at some point, we should look towards enterprise collaboration. And that means WebDAV, emerging CAP (Calendar Access Protocol), etc... That's why I split these two domains: - Collaboration and Mail Services (enterprise WAN/Intranet services) - Internet and Web Services (most others, like ISP/ASP-oriented) Both of which already built on (like everything else): - Directory, Authentication and Name Services But don't recover anything in: - Network File Services > Email Clients > Reading the discussions on Samba and requiring client configuration > knowledge. This topic could also have the issue of including or not > including configuration of email "client" software. Again, I'd argue a "Desktop" exam instead. We can quickly get lost with all the choices. Remember, open standards-based services. From there, people can choose solutions -- especially since there are always dozens of clients for any 1 open standard service. E.g., in the Collaboration and Mail Services exam, we cover iCal, so many different Linux (and even more non-Linux) clients can be addressed. If CAP takes off, then that could be added as well. > * These 2 because they are OSS. However the exam may not include > specific scanners but how the MTAs allow integration of these type of > scanners. The MTA is going to be a tough one. If we choose Postfix and Sendmail, a lot of their configuration (even if Postfix legacy compatibility for Sendmail) do overlap. But there are others like QMail (yes, I know, not likely) and Exim (now this is used more than people realize), etc... How far we do go on "inclusion"? I don't know if that can be so addressed so easily. If you haven't noticed, I'm trying to build a domain that gives enterprise administrators and architects an Exchange replacement. Remember, just like ADS, Exchange is just a set of services. In fact, Exchange is a rather poor solution and incomplete (unlike ADS) when it comes to collaboration -- it relies on a _lot_ of _client_-side resolution with Outlook. Exchange is _not_ a true scheduling/collaboration back-end, just a store with intelligent client in Outlook. -- Bryan J. Smith Professional, technical annoyance mailto:[EMAIL PROTECTED] http://thebs413.blogspot.com ---------------------------------------------------------- The existence of Linux has far more to do with the breakup of AT&T's monopoly than anything Microsoft has ever done. _______________________________________________ lpi-examdev mailing list [email protected] http://list.lpi.org/mailman/listinfo/lpi-examdev
