Where do I find info on the 5250 data stream?
Jeff Johnson
5250 data stream seems like pulling teeth. Any suggestions? Oh, and I'm
looking for something free, not a RedBook or the like. And preferably
available via the Internet.
Dave McKenzie
A good starting point is Mike Madore's site at:
http://www.blarg.net/~mmadore/5250.html
He's writing one for Linux and lists a lot of reference info (not to
mention his source code :-)
--Dave
Mark Pottorff
that just where you expected to find it :) The pub number is
SC30-3533-04 and you are going to spend a lot of time there, so I would
suggest you order the hardcopy. It is also available on the web at
http://publib.boulder.ibm.com:80/cgi-bin/bookmgr/BOOKS/CO2E2001/CCONTENTS
The front-end to that was http://www.as400.ibm.com/infocenter and then
go to the Online Library and in to the V4R4 books.
I an always curious, why is it that everyone feels they need to create
their own Telnet emulator? I realize that keyboard mapping is poor etc.
But there are a small pile of emulators of various types already
available on the market. Why go to all the effort of creating your own?
And then debug pop-up windows, and then debug read-under-format, and
then debug dynamic screen manager and UDDS?
Jeff Johnson
Mark Pottorff <ma...@totale400.com> wrote in message
news:373EED36...@totale400.com...
> I an always curious, why is it that everyone feels they need to create
> their own Telnet emulator? I realize that keyboard mapping is poor etc.
> But there are a small pile of emulators of various types already
> available on the market. Why go to all the effort of creating your own?
Time I got. Money I don't. If there were a freeware TN5250 emulator out
there, I'd be on it.
There's also the learning experience.
Alex Brainman
Mark Pottorff <ma...@totale400.com> wrote in message
news:373EED36...@totale400.com...
>
> I an always curious, why is it that everyone feels they need to create
> their own Telnet emulator? I realize that keyboard mapping is poor etc.
> But there are a small pile of emulators of various types already
> available on the market. Why go to all the effort of creating your own?
> then debug dynamic screen manager and UDDS?
>
Because, for some, it sounds ridiculous to pay around $100+ a pop
(for mid-size company with 500 users = 500 * $100+ = $50,000+)
for something that should be 1/100 of that price if not free, and get
a monster of 10MB or bigger (while should be ~500-700KB at most)
that impossible to manage and give troubles every time you change
something on your system.
Alex
Thomas
Possibly. But the data stream itself is only part of what an emulator
must contend with. There are a few things an emulator must take care of
beyond just interpreting the data stream.
However, the bigger problem is that it will most likely simply cost more
than $50K to develop. A decent programmer working alone is going to take
quite a chunk of time to develop an emulator that you're essentially
going to "bet your business" on (if it's going to be used by every
user). Unless it's a seriously underpaid programmer (who might find a
better deal two weeks after the emulator is distributed to users), I
think $50K is a pretty low estimation of what this'll cost to develop.
And if it's a team of programmers, the cost will simply accumulate
faster.
With newer products such as the Client Access Express client, issues
such as the client footprint get less troublesome. And you're likely to
max out on the number of licenses you actually pay for somewhere under
200, much less than 500 that you're quoting. And as long as you have a
software subscription, what's the real cost anyway? Further, what
happens to your home-grown emulator when a PTF comes along and breaks it
or a new release updates the 5250 specification? Getting support from
IBM (or whatever the vendor is) can be pretty handy... especially after
that underpaid programmer goes elsewhere and no one remains who knows
5250 inside and out.
Tom Liotta
In article <7hsu8k$2ap$1...@eplet.mira.net.au>,
--
Tom Liotta
AS/400 systems programmer
--== Sent via Deja.com http://www.deja.com/ ==--
---Share what you know. Learn what you don't.---
Brian
> Mark Pottorff <ma...@totale400.com> wrote in message
> news:373EED36...@totale400.com...
> >
> > I an always curious, why is it that everyone feels they need to create
> > their own Telnet emulator? I realize that keyboard mapping is poor etc.
> > But there are a small pile of emulators of various types already
> > available on the market. Why go to all the effort of creating your own?
> > And then debug pop-up windows, and then debug read-under-format, and
> > then debug dynamic screen manager and UDDS?
> >
>
> Because, for some, it sounds ridiculous to pay around $100+ a pop
> (for mid-size company with 500 users = 500 * $100+ = $50,000+)
> for something that should be 1/100 of that price if not free, and get
> a monster of 10MB or bigger (while should be ~500-700KB at most)
> that impossible to manage and give troubles every time you change
> something on your system.
>
>
> Alex
At NLynx, the price is closer to about $20 bucks a pop for something like
500 users. BTW, I am the author of the Mac tn5250 emulator and it fits on
a floppy (installer and everything is 880K bytes). That includes 128 bit
encryption, scriptability and probably the best keyboard remapper in the
industry.
Thomas
If money is the problem, two immediate (and cheap) solutions come to
mind. Forget TN5250 and use TN3270 instead -- QWS3270 is an example that
can be used with few problems, just execute SETKBDMAP in your initial
program to get the key action you want if you don't like the default
AS/400 handling of 3270 keys. Or use the workstation gateway and just go
in with your browser.
Tom Liotta
In article <34h03.335$ku3....@news15.ispnews.com>,
"Jeff Johnson" <pawp...@com.geocities> wrote:
>
> Mark Pottorff <ma...@totale400.com> wrote in message
> news:373EED36...@totale400.com...
> > I an always curious, why is it that everyone feels they need to
create
> > their own Telnet emulator? I realize that keyboard mapping is poor
etc.
> > But there are a small pile of emulators of various types already
> > available on the market. Why go to all the effort of creating your
own?
>
> Time I got. Money I don't. If there were a freeware TN5250 emulator
out
> there, I'd be on it.
>
> There's also the learning experience.
>
>
--
Jeff Johnson
Thomas <tho...@inorbit.com> wrote in message
news:7i0158$5vs$1...@nnrp1.deja.com...
> If money is the problem, two immediate (and cheap) solutions come to
> mind. Forget TN5250 and use TN3270 instead -- QWS3270 is an example that
> can be used with few problems, just execute SETKBDMAP in your initial
> program to get the key action you want if you don't like the default
> AS/400 handling of 3270 keys.
The AS/400 is only a small part of my job function nowadays, so I'm afraid
I'm a bit out of touch. Is QWS3270 a PC program? If so, where is it? (I'm on
V3R7.)
> Or use the workstation gateway and just go in with your browser.
Huh? What's the "workstation gateway"? Is it included in the OS software or
is it a separate item? Does it require special hardware?
Thomas
QWS3270 is a PC TN3270. Go to Yahoo and type 'qws3270 download' for
sites. Note that there is QWS3270 Plus which runs $44/copy. Make sure
you mind the difference.
Workstation gateway is one of the TCP/IP servers. It's started via
STRTCPSVR *WSG and accessed by:
http://as400hostnameoraddress:5061/wsg
Assuming you allow HTTP/port 5061 though your firewall, access should be
simple. If you allow TELNET to your AS/400, HTTP is just as likely. It
is NOT a substitute for a real emulator because of the limitations
imposed by running in a browser window. But it's certainly workable in a
pinch.
Tom Liotta
In article <_z_03.312$cF3....@news13.ispnews.com>,
--
ssen...@my-dejanews.com
> Time I got. Money I don't. If there were a freeware TN5250 emulator
out
> there, I'd be on it.
>
> There's also the learning experience.
>
Guys,
Check out http://www.mochasoft.dk It is as close to TN5250 freeware as
you can get. They have it for most client platforms and they also have
a java applet version that is served from the AS/400 http server. So,
the linux guy who started this whole discussion is also covered. He
just needs a java enabled browser. We use it; it works great; learn from
our experience.
Kind regards,
Steve Senatori
Phillip
you can find Information over 5250 Telnet (Protocol) in RFC1205, too.
L.Friedrich
Jeff Johnson schrieb in Nachricht
<34h03.335$ku3....@news15.ispnews.com>...
>
>Mark Pottorff <ma...@totale400.com> wrote in message
>news:373EED36...@totale400.com...
>> I an always curious, why is it that everyone feels they need to create
>> their own Telnet emulator? I realize that keyboard mapping is poor etc.
>> But there are a small pile of emulators of various types already
>> available on the market. Why go to all the effort of creating your own?
>
Alex Brainman
> At NLynx, the price is closer to about $20 bucks a pop for something like
> 500 users.
>
Good for them. I just wonder why I haven't heard about them
before but Client Access.
> BTW, I am the author of the Mac tn5250 emulator and it fits on
> a floppy (installer and everything is 880K bytes). That includes 128 bit
> encryption, scriptability and probably the best keyboard remapper in the
> industry.
BTW, does your '128 bit encryption' comply to a standard
that as/400 hosts support?
Alex
Alex Brainman
>
> ... There are a few things an emulator must take care of
> beyond just interpreting the data stream.
>
> However, the bigger problem is that it will most likely simply cost more
> than $50K to develop.
>
Going by the same token you must agree that it would cost more
then $50K to develop a reasonable web browser, but I doubt there
are many people willing to pay one cent for it this days. From my point
of view it has nothing to do with the cost.
>
> With newer products such as the Client Access Express client, issues
> such as the client footprint get less troublesome.
>
You mean instead of Pentium 130 you need Pentium 400 now?
>
> ... And as long as you have a
> software subscription, what's the real cost anyway?
>
I wonder why IT expenditure goes up in my department.
We used to get away with 1 person admin/installer/fixer/help_desk_support
to up to 3 now doing the same thing with help of 2 servers full of
software management software (it is not from software subscription,
but is sold by the same company that'll sale the subscription to you)
just to keep track with PTFs, patches, service packs,
that merely don't bring any functionality
but fix old problems and introduce new ones.
> Further, what
> happens to your home-grown emulator when a PTF comes along and breaks it
> or a new release updates the 5250 specification?
>
As I said before, generally, PTF fixes product bugs, and doesn't affect
protocol itself; but when it does, it introduces new things without
affecting
(thank you IBM for that) old programs. BTW, 'home-grown' emulator can
accommodate new features much faster because all you have to do is
'change the code' and 'post the changes on the web' that can take
from 2 hours up to 2 weeks for major change - I doubt, big companies
where development team is 50+ can act that fast.
>
> IBM (or whatever the vendor is) can be pretty handy... especially after
> that underpaid programmer goes elsewhere and no one remains who knows
> 5250 inside and out.
>
Yes, if you need someone you can call and complain about your problem,
then IBM is good; but I wonder why there are so many questions about
CA asked here in the newsgroup and been answered by people that has
nothing to do with development and support of the product? I must admit,
some people feel better if they pay big money for support, just as insurance
...???...&5~`%3Z..., not a mean to guarantee a solution to their future
problem.
And some products (take telnet.exe that comes free with 95 or NT),
don't require much support if they are made simple enough and with the
thought
to minimise the maint in mind.
Sorry, if I've been harsh somewhere.
Alex
Paul Nicolay
The subject is getting a bit off-topic, but anyway;
1. The new Client Access Express has indeed a much smaller footprint
regarding active modules and system resources. This should express itself
related to CPU usage as well.
2. Software subscription can indeed be hard if you planned your budgets way
in advance. However, you're still free to follow a slower migration path.
Just think about how Microsoft forces you much harder to keep up (just try
to buy a Windows 95 and you'll notice).
3. IT expenditure... hmm, just think that the AS/400 programs written 10
years ago still run without a single change to them, and remember that even
the microprocessor has changed in the mean while. The expenditure is the
result of Microsoft Windows... and the single fact that it all needs to be
graphical. Basically you're still doing the same, so I wouldn't think about
what that stupid GUI has cost you !!!!
4. Fixing. A home grown emulator is something different than CA Express.
Remeber that their software is used worldwide, and has to be tested for all
those possible configurations. Related to that, a team of 50 people might
be smaller than you yourself writing your (very) limited emulator only.
5. IBM. We had indeed better times with IBM, but they had to cut down costs
as well (under presure from companies like M$ who don't even care about
support). Have you ever tried in getting support from Microsoft ? Ever
compared the PTF mechanism from IBM with M$ ones (if it isn't in SP3/4 bad
luck) ?
A final word about CA... it isn't perfect but as far as I'm concerned it
works pretty well if you take some things into account (the most important
is to use it only on native TCP/IP... but yeah, some people don't even
believe this).
Kind regards,
Paul
-------------------
Alex Brainman wrote in message <7ifiak$2d6$1...@eplet.mira.net.au>...
Alex
The contents of this message express only the sender's opinion.
This message does not necessarily reflect the policy or views of
my employer, Merck & Co., Inc. All responsibility for the statements
made in this Usenet posting resides solely and completely with the
sender.
Thomas
> Going by the same token you must agree that it would cost more
> then $50K to develop a reasonable web browser, but I doubt there
> are many people willing to pay one cent for it this days. From my
point
> of view it has nothing to do with the cost.
>
Hey, YOU were the one who brought up the price. And if you think price
isn't related to cost, it's no wonder you're so far off base.
> >
> > With newer products such as the Client Access Express client, issues
> > such as the client footprint get less troublesome.
> >
>
> You mean instead of Pentium 130 you need Pentium 400 now?
>
The FASTEST PCs I run Client Access Express on are 16Mhz. The slowest is
currently a P-100 and it runs perfectly well.
> >
> > ... And as long as you have a
> > software subscription, what's the real cost anyway?
> >
>
> I wonder why IT expenditure goes up in my department.
> We used to get away with 1 person
admin/installer/fixer/help_desk_support
> to up to 3 now doing the same thing with help of 2 servers full of
> software management software (it is not from software subscription,
> but is sold by the same company that'll sale the subscription to you)
> just to keep track with PTFs, patches, service packs,
> that merely don't bring any functionality
> but fix old problems and introduce new ones.
>
Huh? Is this relevant at all? Of course the numbers of PTFs are going
up. For one simple reason why, just compare the number of OS/400
optional parts in Version 4 to those in Version 1. Today there are FTP
servers, TELNET, Java, etc.; not to mention the huge increase in
complexity of the hardware itself. Do you hink it's trivial supporting
multi-processor systems running about every type of server that exists?
Think the IFS is simple? Think it's simple supporting a system that
scales to the range of AS/400s? Every version expands the capabilities,
and it's a matter of practical fact the the number of fixes will
increase.
Now, although we have a decent help-desk support center where I
currently work with a multi-national network including four AS/400s in
our local data center alone, all AS/400-related PTFs, patches and
service packs, etc., are handled by a single person. And it doesn't take
more than a quarter of my time. If you require three people, well, all I
can say is you have my sympathy.
> > Further, what
> > happens to your home-grown emulator when a PTF comes along and
breaks it
> > or a new release updates the 5250 specification?
> >
>
> As I said before, generally, PTF fixes product bugs, and doesn't
affect
> protocol itself; but when it does, it introduces new things without
> affecting
> (thank you IBM for that) old programs. BTW, 'home-grown' emulator can
> accommodate new features much faster because all you have to do is
> 'change the code' and 'post the changes on the web' that can take
> from 2 hours up to 2 weeks for major change - I doubt, big companies
> where development team is 50+ can act that fast.
>
I cannot disagree that a "'home-grown' emulator can accommodate new
features much faster". I would have to emphasize "_CAN_ accommodate"
however. When the local 5250 expert goes elsewhere, I sincerely doubt
whether it "_WOULD_ accommodate". There's a significant chance that it
would in fact "_NEVER_ accommodate" once the author was gone.
> >
> > IBM (or whatever the vendor is) can be pretty handy... especially
after
> > that underpaid programmer goes elsewhere and no one remains who
knows
> > 5250 inside and out.
> >
>
> Yes, if you need someone you can call and complain about your problem,
> then IBM is good; but I wonder why there are so many questions about
> CA asked here in the newsgroup and been answered by people that has
> nothing to do with development and support of the product? I must
admit,
> some people feel better if they pay big money for support, just as
insurance
> ...???...&5~`%3Z..., not a mean to guarantee a solution to their
future
> problem.
> And some products (take telnet.exe that comes free with 95 or NT),
> don't require much support if they are made simple enough and with the
> thought
> to minimise the maint in mind.
>
True, telnet.exe can be had for free. And guess what? The AS/400
'accommodates' it!
A TN5250 emulator isn't quite in the same class. Pure telnet has many
more potential users. This is like comparing TN5250 to "free" browsers.
Sure it'd be nice if TN5250 could be had for a hundredth of the going
price, but the potential market is much less than 1% of the market for
browsers. And just who would supply it? I doubt if IBM could reasonably
supply such. That would put 3rd-party developers at a significant
disadvantage. So who? Browsers are supplied mostly as "loss leaders" or
gateways to more profitable products such as Netscapes server products
or to enhance something like MS's Win95. What vendor (other than IBM)
gets anything from supplying free TN5250?
However, I can't answer why most people post questions here except to
give three possible reasons. Keep in mind that these refer to questions
normally answered through IBM support, not questions about how to
program day-of-week in CL or *WSCST object creation, etc.
First, some people possibly don't take as much advantage as they could
of the support they have. Second, some don't have a support contract.
And third, some questions cannot be answered fast enough through
support. E.g., I've asked two questions in this newsgroup in the past
year that should've been answered by IBM first. Both were pretty
obscure. Neither received a single response from the newsgroup. Why did
I ask here? Well, just in case someone knew the answers.
As far as CA questions go, I'm not sure at all. I've never had a
CA-related support question that wasn't easily fixed through a support
call except for one about CM/400 under OS/2 accessing data queues for
imaging workstations. And that question was nothing more than "When will
CM/400 support data queues?" And I've worked with CA and PC Support
since the first month it was available on S/38s. It looks to me as if a
significant number of CA questions are tied to very particular
combinations of hardware and/or other software running on the same PC.
At least, only a small percentage seem set up the way I set PCs up for
CA.
> Sorry, if I've been harsh somewhere.
>
> Alex
>
>
--
Tom Liotta
AS/400 systems programmer
Sent via Deja.com http://www.deja.com/