> Steve: > > I like the idea. I think that at present Caldera, and possibly SuSE, carry > LiS on their distribution medium. Caldera because of their Netware > software. SuSE I am not so sure about, but they do seem to mirror the LiS > ftp site. > > I wish that I could add some significant economic weight to your case, but > Gcom is a minor player in telco technology. We do have some significant > customers in the telco field but we mostly serve the transaction world that > uses SNA, X.25 and Bisync.
And those customers are probably creating billing systems, X.25 is still extensively used in billing systems. I notice that one of the three paths in the CGL roadmap is for "Management applications handle traditional service and billing operations, as well as network management". Our customers are continually being forced into buying new hardware, which requires a new Linux, which requires a new Streams and X.25 stack. >From my perspective the single biggest obstacle to deploying linux, especially in the >carrier world, is stability and backwards compatibility. Carriers are used to this >from the traditional unix marketplace (SCO et al on Intel, HP, Sun, Unisys, AT&T on >the big servers). The traditional unix marketplace unfortunately is hurting bad for >various reasons. Perhaps all that effort put into stability was ill-spent. But for carrier's these computers are mission critical, and carriers think in time frames of decades not months. The goals in CGL are laudable and do appear to address these issues as others. But the focus appears to be on technology (RAID, IPv6, IPSevV6, etc) rather than long term maintenace and support (application loading, live upgrades, diagnostics) all of which are priority 2 items in the Requirements specification. IMHO this is the wrong ordering of priorities. That said. Talk to the telcos and carriers about their billing and management systems, you'll find more low speed protocols there and more use of STREAMS. I believe some on this list have implemented SS7 in a STREAMS environment as well. The use of legacy protocols like X.25/B.X25, SNA, bisync will drive the requirement for LiS as many are currently STREAMS based (no matter how much effort has gone into making the core engine independent of the OS). And there really is no other non-proprietary stack/message passing mechanism as well suited for protocol development. (again IMHO). Ragnar Paulson The Software Group Limited _______________________________________________ Linux-streams mailing list [EMAIL PROTECTED] http://gsyc.escet.urjc.es/mailman/listinfo/linux-streams
