REPLY: 2:280/464 6076abca   
   MSGID: 1:153/7001 6076dd3e   
   CHRS: UTF-8 4   
   -={ 2021-04-14 12:17:02.619723556+00:00 }=-   
      
   Hey Wilfred!   
      
    WvV> So your algorithm based on unixtime has a bug!   
      
   That was just a demo which probably is accurate for time based serialno's   
   generated by abandonware which will likely never get fixed.   
      
    WvV> It should take into account it can be called multiple times   
    WvV> within the same second.   
      
   Sure.  How about this?   
      
     6076da69-1689a348   
     6076da69-16bf0244   
     6076da69-16f7701c   
     6076da69-172abd44   
     6076da69-175c7788   
     6076da69-179027b0   
     6076da69-17c263b4   
     6076da69-17f51008   
     6076da69-18287584   
     6076da69-185be1cc   
      
   Note that all of the above were created in the same second and there are zero   
   collisions.  Also note that the following 8 hex characters are in sequence and   
   are indeed unique.  The above routine has a shelf life of just over 2 billion   
   years since the leading seconds hex field is not a fixed field and will expand   
   by one hex character when it is required.  The second part after the ascii   
   dash is from strftime()'s %N (nanoseconds) which only requires 8 hex   
   characters for all time BUT in my future proposal, if it gets that far, will   
   allow for picoseconds when they become available by adding 2 hex digits which   
   will make it a fixed 10 hex digit field.   
      
    WvV> If you do that correctly, it's superior to the random one...   
      
   See above but just for fun I will do it side by side just like before.   
      
     6076e0bd-22d50e70 n3yM8uFm   
     6076e0bd-246302d4 sL4Irxdg   
     6076e0bd-25d0d1f0 8fq4vU7e   
     6076e0bd-276236a4 Aez4QZog   
     6076e0bd-28f6291c jbQiqKYc   
     6076e0bd-2a842704 gElrH3UT   
     6076e0bd-2bf24d98 8qmyLLQR   
     6076e0bd-2d8103dc Pbeaul1Z   
     6076e0bd-2f0436d0 IHsKacHr   
     6076e0bd-30a01b60 0An6jRkx   
      
   My best guesstimation is that the random one is superior given that it doesn't   
   require a rewrite of current FTN standards.  However I do plan to change that   
   and then for sure the unixtime based one will be superior and unique for all   
   time, nevermind 3 lousy years.   
      
   I doubt there is anything better currently in use to either of the above   
   routines and most definetly nothing even comes close to the unixtime based   
   one.  It is a work of pure genius.  :-)   
      
   Hm.  Interesting tagline got tossed into the works considering the content of   
   this reply.  It looks to me the Anglo-Saxons might have had working knowledge   
   of the Laws of Thermodynamics long before it's time.   
      
   Life is good,   
   Maurice   
      
   ... Eal þæt þu her sceawast hit is sceaduwa gelic, æll hit gewitað.   
       All that you see here is like a shadow; it will all vanish.   
   --- GNU bash, version 5.1.4(1)-release (x86_64-motorshed-linux-gnu)   
    * Origin: Little Mikey's Brain - Ladysmith BC, Canada (1:153/7001)   
   SEEN-BY: 1/123 18/200 90/1 105/81 106/127 114/705 120/340 123/120   
   SEEN-BY: 123/131 124/5016 129/305 153/105 135 250 757 802 7001 154/10   
   SEEN-BY: 203/0 220/70 221/1 6 360 226/17 30 227/114 229/101 424 426   
   SEEN-BY: 229/452 664 700 1016 1017 230/0 240/5832 249/206 317 400   
   SEEN-BY: 250/5 8 261/38 267/800 280/464 5003 282/1038 288/100 292/8125   
   SEEN-BY: 298/25 301/1 305/3 310/31 317/3 322/757 335/364 340/1000   
   SEEN-BY: 342/17 200 396/45 423/81 120 460/58 633/280 712/848 770/1   
   SEEN-BY: 770/100 330 340 772/210 220 230 4500/1   
   PATH: 153/7001 757 221/6 1 280/464 770/1 317/3 229/426