From mboxrd@z Thu Jan 1 00:00:00 1970 From: Ralf Baechle DL5RB Subject: Re: callsign limit Date: Thu, 1 Jun 2006 14:25:00 +0100 Message-ID: <20060601132500.GA10812@linux-mips.org> References: <200605310927.56706.vk3heg@iinet.net.au> <20060531093716.GA21395@cloud.net.au> <620c90570605310403o1e4b94bcgf4ab922ef26983e4@mail.gmail.com> <20060531111919.GA22980@cloud.net.au> <20060601014047.GB27504@linux-mips.org> <447ED7B2.3070701@sktc.net> Mime-Version: 1.0 Return-path: Content-Disposition: inline In-Reply-To: <447ED7B2.3070701@sktc.net> Sender: linux-hams-owner@vger.kernel.org List-Id: Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit To: "David D. Hagood" Cc: linux-hams@vger.kernel.org On Thu, Jun 01, 2006 at 07:04:02AM -0500, David D. Hagood wrote: > Ralf Baechle DL5RB wrote: > >One of my crazier ideas to deal with the issue of the longer callsigns > >was using a hash value instead. The probability of colissions is low but > >how to do the reverse mapping from hash to callsign ... > > Since a callsign is always from the set [A-Z0-9] (36 chars, 5.2 bits of > data per character for a naive encoding) could you not simple encode the > callsign of an otherwise too-long call as 6 bits per character packed > into the octets, with a specific leading character to indicate > compression (e.g. '?') - yielding about 8 chars/call. That is a modest proposal which gives you a little bit of extra headroom. But the WRC03 rules set no limit for callsigns length. And, let's face it, the current format of AX.25 MAC addresses sucks anyway, 7 bytes with various other junk thrown in just because space was available make processing more complicated than needed. 73 de DL5RB op Ralf -- Loc. JN47BS / CQ 14 / ITU 28 / DOK A21