Re: TN5250 0.16.2 released! -- LINUX5250

On Tue, 28 Aug 2001, tonyl wrote:
>
> If the product is supposed to emulate a 5250 terminal, I would suggest that
> the emulation should cover the whole terminal and not just the display. Maybe
> it already does, maybe it's just too poorly documented to get up and running
> exactly how people want - as the mailing list archives illustrate? (I wonder
> how many just go away and drop the product because of this issue)

The documentation of how to make the keyboard mapping work is lacking,
but, as has already been discussed here...   people are working on it.
But...  you CAN make it work, it's not THAT hard...  This isn't even
purely a tn5250 issue.   Try telnetting from a FreeBSD machine into a
Linux machine sometime!  Some keys work, some don't!  Try using the telnet
program called "CRT" under Windows to log into a Linux or FreeBSD
machine.   Some of the keys aren't right.  You have to go through the same
steps to fix the issues as you do with tn5250.

We are working on ways BOTH to document the existing fixes better AND to
find more user friendly ways to handle it.  But this is a complicated
issue, and it's being worked on by people who do so in their spare time,
without pay.  It's not going to happen overnight!


>
> The .kmp files you're so against are very simple text files.  You know, those
> things unix lovers are always raving about :-) A conversion tool can not do
> tn5250 any harm. Ahh, now I see your point. You feel tn5250 would have to
> read a whole extra file at run time...?

So write a conversion tool.   It wouldn't do me any good, and I don't have
Client Access to do any testing with...   so I'll leave it to you to write
the tool.   If James doesn't want to use your tool, he doesn't have to...
thats the beauty of a conversion tool, as opposed to building it into
tn5250.

>
> >From a personal point of view, I just wonder from what authority you speak
> when you try and tell me how I (or others) should be using a keyboard?

I'm pretty sure he was just expressing his opinion, not trying to be
authoritative.

> Why
> can't I use whizzing cursors and lasoo texts? I don't like emacs, but I don't
> go round telling everyone to use vi. I would guess that field-exit is one of
> the most important keys on the 5250 keyboard, very much so to data-entry
> staff where every single key depression affects their throughput.

Why can't you?  Because the code hasn't been written to do that, yet.

If you stand to save a lot of money by saving keystrokes and thus having
better productivity, perhaps it'd make sense for you (or someone more
appropriate at your company) to contribute code to the project that does
these things?

You see, thats how open-source works.  People who want/need things done
will write code and contribute it.   Others benefit from these
contributions, and then make whatever contributions THEY can...  and
things continue to improve for everyone...

>
> Souldn't you be more concerned that the doze version doesn't work, or maybe
> you feel tn5250 belongs to *nix? (guess that means there's no mac version
> too) It never occured to me until recently (thank's to a couple of post in
> this list actually) how much it must cost as/400 sites just to talk to their
> boxes. It's probably more obscene that qshell! I think it'll be a long long
> time before doze gets replaced in the office, but that's another issue.

It's a matter of priority.   Since there are upwards of 3 dozen different
5250 emulators for Windows, it seems like unix/linux should be the
priority...

If Windows is a priority to you, then make it work under Windows and
contribute those changes to the project, as I explained above.  Otherwise
you'll have to wait until it makes it to the top of someone elses project
list.

>
> Maybe you've already completed that wonderful howto and all the relevant info
> is already available, and I'm just being a prat?
>

Is "prat" actually a word?  What does it mean?   Is that slang from your
area of the world?