Ulrich Schroeter wrote to Roy Witt about MakeNL v3.4.1 Release:
          No pseudo-French spellings, please!

 AL>>> That decision is up to the ZCs.  I'm all for it.  Since it is a
 AL>>> top-down option, it needs to be enabled at the ZC level first,
 AL>>> ...

 RW>> This is where my RC's system breaks down. He tried using MNL for
 RW>> OS/2 and it didn't work.

 US> When did he tried it the last time? with which revision?
 US> probably with an older revision < 3.4.x

I'm not sure anymore. Maybe a year ago.

 US> Yes, there was an annoying bug with an OS/2 version, that produces a
 US> runtime error under OS/2 but not on other revisions. This bug has
 US> been fixed in v3.4 and the ticket is marked closed-fixed
 US> https://sourceforge.net/p/makenl/bugs/17/

There's an old saying that goes something like this; once bitten, twice
shy.

Meaning that once you've tried something and it didn't work as expected,
in fact crashed your system, the more reluctant you are to try it again.

 RW>>  I enabled the allow unpublished and his MNL/OS/2
 RW>> couldn't handle it. He's still reluctant to use it. One other
 RW>> complaint he has is that MNL won't submit a netseg with no changes
 RW>> in it. NLMake, which is what he uses, won't except a file that has a
 RW>> date over 8 weeks old. So an NC must make a change to the netseg,
 RW>> forcing one to enter a number in an IP only node to make that
 RW>> change, until he gets a working OS/2 version.

 US> You can forward these infos also with the FORcesubmit 1  parameter
 US> setting and probably all problems are gone  -;)

Yes, I know this now, Uli...back then it was like muddy waters to me.

Thanks for your input, Uli.


        R\%/itt


--- GoldED+/W32 1.1.5-31012
--- D'Bridge 3.92
 * Origin: South Texas Hub - Gulf Coast Backbone (1:387/22)