www.justiceforchandra.com Forum Index www.justiceforchandra.com
Justice for Chandra Levy and missing women
 
 FAQFAQ   SearchSearch   MemberlistMemberlist   UsergroupsUsergroups   RegisterRegister 
 ProfileProfile   Log in to check your private messagesLog in to check your private messages   Log inLog in 

rdwrites iseries blog
Goto page Previous  1, 2, 3 ... 9, 10, 11, 12  Next
 
Post new topic   Reply to topic    www.justiceforchandra.com Forum Index -> www.rdwrites.com forum (part of www.justiceforchandra.com)
View previous topic :: View next topic  
Author Message
rd



Joined: 13 Sep 2002
Posts: 9234
Location: Jacksonville, FL

PostPosted: Tue Apr 15, 2014 2:20 am    Post subject: Reply with quote

I was doing some other experimentation with CGIDEV2 to come up to speed (as usual found the answer in a Scott Klement post) and saw a thread which gave me another thing to check out. First though I ran across this link from Paul Tuohy which is very helpful: http://search400.techtarget.com/tip/Getting-started-with-CGIDEV2-Part-3

Now, as far as using CGIDEV2, I'll post my little test code. Point here of this test is that minimal html snippet of img src has IFS file name set before showing the page. I did this as a parm where I just paste the parm into a file name (32 char max) variable in my CGIDEV2 template file. So it displays whatever image file name is typed in the url parm ( e.g. http://IBMiurl/cgiserver/helloimg?img=TN_123456789.JPG ) Also I am running the cgi server on one port and using apache on port 80 to show image file from document root folder, in other words, all defaults and minimal file path.

HELLOIMG RPG CGIDEV2 program
*****************************************************************
* CGIDEV2 Testing
*
* Program HELLOIMG
*****************************************************************
h dftactgrp(*no) actgrp('RDWRITES')
h bnddir('CGIDEV2':'QC2LE':'QUSAPIBD')
h optimize(*none) decedit(*jobrun) truncnbr(*no)
h debug option(*srcstmt:*nodebugio)

*****************************************************************
* Prototype Definitions
*****************************************************************
/copy cgidev2/qrpglesrc,prototypeb

*****************************************************************
* *Entry Declaration
*****************************************************************
d HELLOIMG pr
d HELLOIMG pi

*****************************************************************
* Data Structure Declaration
*****************************************************************
d qusec ds
d qusbprv like(int32) inz(%size(qusec)) Bytes Provided
d qusbavl like(int32) inz(0) Bytes Available
d qusei 7a Exception Id
d 1a Reserved
d msgdata 500a

* indicators for getHtmlIfsMult subprocedure
d IfsMultIndicators...
d ds
d NoErrors like(boolean)
d NameTooLong like(boolean)
d NotAccessible like(boolean)
d NoFilesUsable like(boolean)
d DupSections like(boolean)
d FileIsEmpty like(boolean)

*****************************************************************
* Stand Alone Fields Declaration
*****************************************************************
* like prototypes
d boolean s n
d int32 s 10i 0

d CGITESThtml c '/rdwifs/html/'
d nbrVars s like(int32)

d htmlMsgStr s 1000a varying

d savedQueryString...
d s 32767a varying

d imgparm s 32a inz(*blanks)

* constants
d newline c const(X'15')
d htmlHeader c 'Content-type: text/html'
d htmlErrMsg c '<html><body> -
d <p>An error occurred. -
d Programmer has been notified. -
d Please try again later. -
d </p></body></html>'

*****************************************************************
* Mainline
*****************************************************************
/free
exsr initProgram;

// read externally defined output html files
IfsMultIndicators = getHtmlIfsMult('' +
CGITESThtml + 'helloimg.html' + ''
:'<!-- $':'$ -->');

// output HTML message if fail to load HTML templates and exit
if NoErrors = *off;
exsr failLoadHtml;
exsr exitProgram;
endif;

// get input
nbrVars = zhbGetInput(savedQueryString:qusec);

if nbrVars > 0;
imgparm = zhbGetVar('img');
endif;

if imgparm = *blanks;
updHtmlVar('imgfile':'no img parm');
else;
updHtmlVar('imgfile':imgparm);
endif;

// complete send of buffered output to the browser
wrtSection('showimg *fini');

exsr exitProgram;

//**************************************************************
// Subroutine: failLoadHtml
//**************************************************************
begsr failLoadHtml;

htmlMsgStr = htmlHeader + newline + newline + htmlErrMsg;
wrtNoSection(%addr(htmlMsgStr)+2:
%len(htmlMsgStr));
wrtSection('*fini');

endsr;

//**************************************************************
// Subroutine: exitProgram
//**************************************************************
begsr exitProgram;

*inlr = *on;
return;

endsr;

//**************************************************************
// Subroutine: initProgram
//**************************************************************
begsr initProgram;

// clear messages
clrMsgs();

// clear the HTML buffer
if getHtmlBytesBuffered > 0;
clrHtmlBuffer();
endif;

endsr;
/end-free


HELLOIMG.HTML

<showimg>
Content-type: text/html

<html>
<body>
<p>/%imgfile%/</p>
<img src="http://IBMiURL//%imgfile%/"/>
</body>
</html>
Back to top
View user's profile Send private message Send e-mail Visit poster's website
rd



Joined: 13 Sep 2002
Posts: 9234
Location: Jacksonville, FL

PostPosted: Sat May 17, 2014 11:34 am    Post subject: Reply with quote

Responding to a question on thoughts on object oriented RPG:

It wouldn't be compatible with what we do. If someone determined that OO is right for what they want to do and include RPG as part of it I would suggest C++ and write the RPG in ILE modules.

This concerns an at home personal project, but I am currently rewriting a sizeable app I developed in Java to RPG. The code tends to translate very closely with class variables held in data structures, with most data structures dimensioned to hold each new() instance. The instances are well regulated in this app (number known beforehand) but if I didn't know how many instances there could be I would use alloc memory to hold the the data structures rather than dimensions.

RPG procedural is much more powerful for business logic and in this case even a non business server app.
Back to top
View user's profile Send private message Send e-mail Visit poster's website
rd



Joined: 13 Sep 2002
Posts: 9234
Location: Jacksonville, FL

PostPosted: Wed Oct 08, 2014 12:33 am    Post subject: Reply with quote

Saw this article in Federal Computer Week today:

http://fcw.com/blogs/the-spec/2014/10/darpas-unhackable-software.aspx
DARPA does math


It goes on to give a challenge for "developing programs that are mathematically proven to be highly secure".

The SPEC column author elaborates: "DARPA is looking for research that offers mathematical proof that a program is extraordinarily difficult to hack."

Instead, it's the environment, stupid.

Much more important than a program syntax is the environment it is running in. And if the government and corporations wanted to be secure, they would be running their programs on the IBM i. But they don't care as much about security as they do staying comfortable with what they know, Windows and Linux.

Now Linux never had the security rap that Windows had but these last two exploits, SSH Heartbleed and Bash have been monumental universal failures in security due to their access to root the system.

All these exploits are easily patchable, but the fact of the matter is is that government and corporation servers are being cracked every day, and the cracks are getting bigger all the time.

Thus this Federal government research quest for a mathematically proven safe programming.

Here's how you make a program exponentially difficult to crack.

Run it on IBM i.

Have an operating system that is object based, not file based. Many exploits trick operating systems into running an executable file that was loaded as a data file.

Have an operating system that doesn't run on x86 opcodes. That is one of the key routes to running many exploits in Windows and Linux.

Have an operating system that runs on hardware through a translation layer. Even if IBM i did run on x86, the translation layer wouldn't allow the exploits.

Have an operating system where common utilities are reworked to IBM's standards and carefully shipped into production. No ready for updates and constant patches slipstreamed into production with security surprises.

Support all open source leading edge technology, but in an environment hostile to crackers.

Support internationalization, but also translate from ASCII to EBCDIC internally and back to ASCII externally. Try to exploit that.

Have an operating system with robust process memory access protection, as in you won't be scanning memory for credit card info on this system.

Fully support the Unix POSIX standard, and run everything UNIX does, if and only if recompiled for a memory safe environment hostile to crackers.

Have an operating system that supports the full gamut of visual interfaces, and standard client interfaces for programming development with Java Eclipse, and web serving with everything from Apache to Websphere Java Enterprise to node.js server Javascript. All engineered to the highest standards of security.

Run this secure programming as compiled, typed code in a number of integrated languages. RPG, C, C++, Java. Why jeapordize the security of your servers because your programmers find it inconvenient to specify exactly what the programs will do and get clean compiles of it? Many exploits are enabled by manipulating dynamic scripts. You have to know what the code is before you can prove anything about it.

Have an operating system that doesn't allow this compiled, tested secure programming to be patched like a data file in Windows or Linux. What good is proving something if it is changed at will by the cracker?

For serving web pages with those secure programs, many on the IBM i have (and I am developing) web apps that assign non-deterministic session id's. And similar for each web page. And restrict actions to what is allowed from last web page.

We on the IBM i have a lot of experience in controlling what happens in sessions. Secure processing in secure programs in a secure system, the IBM i.

Mathematically, it doesn't get any better.

rd
Back to top
View user's profile Send private message Send e-mail Visit poster's website
rd



Joined: 13 Sep 2002
Posts: 9234
Location: Jacksonville, FL

PostPosted: Mon Oct 13, 2014 8:04 pm    Post subject: Reply with quote

Re: someone disappointed that IBM i shops rarely use DB2 SQL/PL -

Some are into platform independence very heavily, but in the first place if it's so important to you that you think we are missing out on this platform independence, why would there need to be special mention about particulars on the IBM i? For that matter take it on down a level to the SQL. The dirty secret the cross platform independents have is that SQL and all the trappings like this are very database dependent, so much so that about any product you care to examine will list what databases it supports, and often DB/2 of any flavor is not one of them.

I watched a major company (SSA, of BPCS fame) go down in flames trying to replace their complex business logic in RPG with multiple flavors of SQL. It wasn't pretty. I have no idea what the bankruptcy sale buyers did with the code but I'm pretty sure they don't say "gee, this is in SQL, runs on any database". Nope, you won't ever see that for anything complex.

Secondly, I doubt very many of you hard core advocates run a large company with your ideals. In fact, I think you're much more concerned with your place in life as a programmer being able to go to a non-IBM i shop with your skills than any experience at running a large complex programming environment that millions and billions of currency count on to survive.

So while ideas are good, and maybe a proven solution especially provided in open source might make a point and be used and adopted, I've seen some very large operations go down in flames because suits believed the talk and pulled the plug.

Conservative is another word to describe IBM i people. We're not startups, and failure isn't an option. I've yet to see a recommended architecture make a similar company leaps and bounds better and more cost effective to run. When some of you idealists get to that point, I have no doubt suits will be eager to buy in.

Until then, they have a solid architecture that works, and IBM i programmers that make it work.

Sometimes with 30,000+ lines of a code programs from a company that was leading the world with them until they tried to replace them with SQL.
Back to top
View user's profile Send private message Send e-mail Visit poster's website
rd



Joined: 13 Sep 2002
Posts: 9234
Location: Jacksonville, FL

PostPosted: Tue Oct 14, 2014 1:14 pm    Post subject: Reply with quote

To add to the causes of the CASE/SQL rewrite fiasco, we RPG consultants were shown the door. The rewrite was done by C programmers hired from college. All the talk about the virtues were talked, and they did the walk, and they went bankrupt.

One leading ERP company doesn't prove anything, but my point is the disparaged 30,000+ RPG programs ran the world's largest manufacturing intensive corporations, and the promised land platform independent CASE language programs with SQL for an attempted three supported databases bankrupted the company.

That which is disparaged works, and we have proof of that. That which is promised is just very expensive software, and no proof it ever replaced an RPG or COBOL shop more cost effectively.

I'm not against options, we can use SQL on the IBM i and we do use SQL on the IBM i, but there's no demonstration it's a better option than RPG.

I'll end my piece on that because people have their preferences and I'm not in the business of changing people's minds. My view is that I compete against software every day. The day other software ourperforms and is more cost effective than mine is the day a business has a better option. But I've heard the talk for a few decades now, and the companies running on the IBM i are the winners.
Back to top
View user's profile Send private message Send e-mail Visit poster's website
rd



Joined: 13 Sep 2002
Posts: 9234
Location: Jacksonville, FL

PostPosted: Tue Oct 21, 2014 5:08 am    Post subject: Reply with quote

The IBM promise of 20 per share in 5 years was indefensible. It was like They have no fricking clue what's going to happen in 5 years, or next year, or even next quarter. All they did was cannabalize the company and buy back shares to try to get the share price up. I saw the other day they are heavily in debt. To buy back shares when you are are heavily in debt shows you what a bunch of desperate people those CEO's were/are.

Now to the only thing I care about in this, the IBM i. Which is an operating system running on IBM hardware (correct that - as of today IBM designed hardware).

We knew all this was coming. Nobody but a complete idiot would believe any of that crap about made to order earnings per share, it was just a matter of how deep in debt IBM would go to buy back stock to artificially make believe they were accomplishing something. Next thing on the agenda will be a mea culpa for marketing not explaining the wonderfulness of the Cloud well enough. Problem solved.

We knew they were selling the foundries too. They don't need no scumbag workers holding down their profits. They're holding out for Watson to become sentient and sack the whole lot from Ginny on down and earnings per share become infinite.

Pending IBM's financial nirvana, they have an operating system on a tape and hardware some sucker foundry is making for them, and they charge from tens of thousands to millions of dollars which is almost pure return on investment at this point.

Which is second best to Watson becoming sentient.

IBM i will be around as long as someone can install and sell it. Watson is working on it.
Back to top
View user's profile Send private message Send e-mail Visit poster's website
rd



Joined: 13 Sep 2002
Posts: 9234
Location: Jacksonville, FL

PostPosted: Tue Oct 21, 2014 8:51 pm    Post subject: Reply with quote

While I enjoy and learn from what many software writers have to offer up, I disagree fundamentally with industry wisdom that large code is bad and small is easier and superior. There are tradeoffs, and all of it will work, but large in general is easier to deal with, by far.

I know very few agree with me but I welcome them to their distributed code world. I haven't seen any of these theories produce more cost effective code to develop, run, and maintain, function for function, as of yet, and quite a few years have passed. In other words an ERP in another language hasn't proved to be more cost effective to develop, run, and maintain than the ERP's in RPG. And those RPG ERP programs tend to the large size.

My contention is that a change that is easy in a small module is just as easy in a large program, and a change that is hard in a large program is much harder in distributed modules.

Modules are better for shared use, obviously, and we have that at many levels in RPG. But they also have more widespread effects from changes. Strictly honoring and adhering to each interface, on both sides, is an absolute requirement, but the more the interface is shared, the harder it is to have a universal contract to honor.

The modularity level I find that works the best is at a program call level. That's the way we've always done it in RPG and it still is most cost effective approach.

But there's no free lunch. Lazy people who want quick solutions without understanding what they're doing are just lazy people who aren't very good programmers. Maybe they can navigate myriad object calls better than following flow in a large program, but I doubt it. Certainly haven't seen it.
Back to top
View user's profile Send private message Send e-mail Visit poster's website
rd



Joined: 13 Sep 2002
Posts: 9234
Location: Jacksonville, FL

PostPosted: Thu Oct 23, 2014 6:26 pm    Post subject: Reply with quote

Clarifying based on some responses:

1990 was 25 years ago. Why are we talking about what RPG programs looked like then? Yes, many still in use if they are still performing the function, and I still make occasional changes to them. If I see dupe code I make subroutines out of them. If I see a Goto loop I replace with a DOU loop. If I have new functionality I write a program and call it. I don't see the big deal. Big programs are actually easier to follow than distributed code.

By big I do not mean duplicated code (or duped with minor changes), and I don't mean mainline where logic isn't grouped into routines, but I do mean big as in if the logic is particular to this program then it goes in this program, not spread over triggers and service programs and stored procedures because it must be easier to develop and maintain if the code is scattered around so the program isn't "big".

Of course if the logic is truly shared then placing in a separate program or subprocedures in a module, copy file, service program, or stored procedure (or all of the above as function dictates) is of course called for. As far as I know that is done now.
Back to top
View user's profile Send private message Send e-mail Visit poster's website
rd



Joined: 13 Sep 2002
Posts: 9234
Location: Jacksonville, FL

PostPosted: Wed Nov 05, 2014 9:43 pm    Post subject: Reply with quote

Responding to a question about setting up a web service on the IBM i:

Re: IFS file authority. With WRKLNK '/www/myapps/conf/httpd.conf' option 9 will show the access authority. They are standard Unix/Linux authority flags. The user you sign on with the Admin web app needs *RWX authority to use and change the file. A broader user(class) such as *PUBLIC needs *RX to read it and use it. You probably have the *PUBLIC setting.

Your path is different than the default configuration on 5.4 which is at '/www/apachedft/conf/httpd.conf' It might have changed in a later release or you are dealinig with a copy at a location you specified for your test server with Admin. Just so you're aware. The settings of the location are with the server setup file member.

There are a number of solutions, but in the API arena there are years of work in wrapping the HTTP API's in the CGIDEV2 service program, open source and free. It was done at IBM back in the day and is maintained by an ex-IBM'er. In my download link is a HELLOPARMS sample CGIDEV2 program. There is a community supporting it. We are many, many years from Brad Stone's book, which was helpful back then but not the level you want to be working at.

The Apache HTTP server is essentially used by all solutions, although there are some additional Java stuff with Websphere I think. As far as API calls go, they are in the CGIDEV2 source code. This is a good solution and the route to go if you want to deal with the API call approach.

You also stated you tried to start the Websphere SOAP web service. You would only have Websphere as your solution or another one such as CGIDEV2 (based) solution, not to mention a huge choice of additional product solutions. The Websphere SOAP web service is a reasonable approach. There are reasons to use others but those are preferences based on your development goals. If you have goals that require some more detailed control, you could get into that later. Your Websphere SOAP service will still be running.

As you know I wrote a program that uses HTTPAPI, so I think a lot of it, but with your SOAP server and client requirements with Tango (you need SOAP server when they initiate requests) the HTTPAPI software is not applicable to your needs, Henrik mentioned earlier it is client software.

I think the CGIDEV2 based free and open source prodct solution from Henrik is an extremely powerful way to go with everything you will need and might want to do already working. IBM has solutions but they are very stuffy with gobs of Java involved. It's some people's cup of tea, not mine. But choice is good.
Back to top
View user's profile Send private message Send e-mail Visit poster's website
rd



Joined: 13 Sep 2002
Posts: 9234
Location: Jacksonville, FL

PostPosted: Thu Dec 18, 2014 8:23 am    Post subject: Reply with quote

There has been ongoing software enhancements to OS/400 through the years just like Linux and Windows and every other OS. The AS/400 operating system of OS/400 then iseries i5/OS then Power Systems IBM i are just renames.

Obviously API's evolve amd new features are added that are more powerful than prior releases but the AS/400 is a business workhorse then and now with latest IBM marketing rename.

The fact is that we run a wider range of technology workloads than other systems. Some businesses do very well with this box, under all the IBM renames of it.

If some want to portray this as some new SQL and web server box, then people have been and will opt for Windows and Linux to do only that. It is our wide range of capabilities we've had all along that make this a business workhorse.

The people that understand this know what it is regardless of what IBM renames it. The people that don't aren't interested in it just for SQL and web serving no matter how much some rename enthusiasts would wish.

I spend my time writing code, and having more solutions this box runs and competes with very well is the only thing that will ever count for usage. I will continue to do my part on that. Software that outcompetes.

Let's pretend what Trevor say is true. Let's pretend IBM just came out with a new system that only runs SQL, Java, and an Apache web server. The OS is called IBM i ad it runs on IBM's Power System hardware. Pricing and licensing are as they are.

Let's say they hire Trevor and his trusted sidekicks as the sales and marketing team.

You know how many sales Trevor and his sidekicks will have?

Zero.

Zero however Trevor and his trusty sidekicks spins it to businesses.

There's your reality, gentlemen.

Now the rest of us have real businesses to run that rely on that trusted workhorse where perceptions are earned every day.
Back to top
View user's profile Send private message Send e-mail Visit poster's website
rd



Joined: 13 Sep 2002
Posts: 9234
Location: Jacksonville, FL

PostPosted: Sat Jan 10, 2015 7:40 pm    Post subject: Reply with quote

I have made more than my share of comments on IBM marketing, I will never forget the merry-go-round, enough that I am glad there is no discernible budget for IBM i marketing. Advertising is expensive and then lack of sales makes the product a money loser.

As it is, just ripping off some DVD's for an OS install costs a few dollars. The rest is profit. There is no higher margin than IBM OS sales at this point in the lifecycle. It is pure ROI on the billions invested by IBM in various software initiatives that make up IBM i.

I am taking a very brief break from programming to write this. It is my opinion that software that takes advantage of the IBM i infrastructure adds to reasons for buying the system. I am trying to do my part in that.

Others are a few ERP variations that are IBM i specific, which really still drive the sales, as ERP's always did.

Management and security along with legendary reliability are the other main driving forces for sales, and with the continued server breaches with hundreds of millions of credit card info stolen (many more than once of course, and re-issues stolen, etc.) along with massive confidential data breaches one would think executives would be looking long and hard at the security inherent in IBM OS'es, but the continued breaches show they aren't up to the challenge. They think pointing fingers at AV software not detecting and stopping the intrusion is where their responsibility stops.

Having said that, most software is now cross-platform (other than Windows software pending a more robust Mono alternative and IBM big box proprietary language implementations RPG and COBOL software). If software can run on multiple OS'es that IBM sells and supports, why should they push one of their children over another? It's the customer's choice.

IBM does make a lot of specialty software available only on AIX, so while not pushing AIX specifically they push their software which requires AIX. My memory is foggy on the exact memory addressing issues that must be addressed so to speak to make a successful recompile for PASE, but unless there is bad coding techniques there shouldn't be a reason that it isn't recompilable for IBM i. They seem not to want to do that for some reason. That is really where their "marketing" pedal doesn't hit the metal.

I can't see IBM not willing to press off install DVD's for tens to hundreds of thousand dollar sales, but I can see customers not willing to pony up IBM's licensing fees if they don't perceive the inherent value in the OS and software that utilizes the infrastructure.

That was easier to see in the past, harder to see with vanilla cross-platform software, as in, you mean I could just run this on Linux and not pay IBM anything?

So as always, software will drive the sales.
Back to top
View user's profile Send private message Send e-mail Visit poster's website
rd



Joined: 13 Sep 2002
Posts: 9234
Location: Jacksonville, FL

PostPosted: Sat Jan 10, 2015 10:21 pm    Post subject: Reply with quote

I diverge with the SQL only software people. It's their choice, but I will continue to use record level IO for transactions. That's my choice.

As for software of any type, obviously at this point it must be web software. I believe IBM i can run web software better than other platforms, I haven't seen a downside to running it on IBM i yet.

Having said the above, I do not believe all inhouse 5250 sessions replaced with web pages is productive in any way, but new systems can't be limited to 5250. Actually should be (and the way I develop my personal systems) interface independent, essentially client/server model, interface could be web, 5250, socket to client, etc.

As for a strategy of programming on another system, I am not interested in it. When I was unable to find another job in the 2004 timeframe, I taught myself Java (wrote a substantial program in it) and planned on going that route. Before I was ready to seek Java jobs, I did get another RPG job which I still have, but the Java experience paid off. I use it in my current job.

I would add Javascript to that now as even more important skill knowledge. But I have an iseries 520 now that I didn't have in 2004 (hah! my iseries 520 is a 2004 system and the price I've seen for my system new was very, very expensive).

And a business one is in transcends technology stacks anyway. I would have a hard time imagining anything running multi-billion dollar transactional businesses like the IBM i and predecessors do, but most of us oldtimers have done incredible things with low powered machines, so I'm sure we'll get the job done one or the other with the copious amounts of hardware available today.
Back to top
View user's profile Send private message Send e-mail Visit poster's website
rd



Joined: 13 Sep 2002
Posts: 9234
Location: Jacksonville, FL

PostPosted: Thu Feb 05, 2015 8:46 pm    Post subject: Reply with quote

"No matter how much code of mine or yours is still running since the days of the System 38 recast in later Systems..."

I don't know about you, but I've been writing new code every day since, and it was running in production at the time. My thoughts are the opposite of some here, the older it is, the less chance it's still running. But there's plenty still running, and plenty more added every day.

New requirements, new challenges, new opportunities. RPG on the IBM i has always been more than ready for them.
Back to top
View user's profile Send private message Send e-mail Visit poster's website
rd



Joined: 13 Sep 2002
Posts: 9234
Location: Jacksonville, FL

PostPosted: Fri Feb 20, 2015 12:19 am    Post subject: Reply with quote

Here's a hilarious tid bit from ITJungle.

CSC blames RPG programmer shortage for their revenue miss. They have an IBM i insurance app among other things. Here we go with the RPG programmer shortage again.

I saw people continuing to look for RPG programmer jobs, and I didn't see CSC posting about wanting RPG programmers for the insurance industry. If they did, I didn't see where anyone mentioned anything about it.

Meanwhile, while CSC alleges to not be able to find RPG programmers, I saw several excellent RPG programmers hired in same time frame. Who to believe, my eyes or an ex-IBMer president of CSC looking to blame their bad performance on something. Ha! Yeah, blame it on lack of those stupendous RPG programmers who would have pulled your fat out of the fire.

Remedy? Working with partners who have people with the programming skills they need. Well, coming from an exIBM'er, that's likely code for H1B's / outsourcing overseas.

And raised the salaries offered for the positions they can't fill. If anyone sees such as offering from CSC or an insurance industry job in DC metro area, then I'll believe it.
Back to top
View user's profile Send private message Send e-mail Visit poster's website
rd



Joined: 13 Sep 2002
Posts: 9234
Location: Jacksonville, FL

PostPosted: Thu Feb 26, 2015 8:05 pm    Post subject: Reply with quote

"RPG runs on only ONE BOX, 1 server, on 1 hardware box."

Technically it runs on Windows as well with the ASNA product and their DB2 clone database. You can develop one code base and deploy on IBM i and Windows.

That is one of the better ways to run cross platform in my opinion, with Java a distant second. Having said that, RPG only needs to run on one box, the only server I care about deploying on. IBM i will also run the cross platform stuff from Windows and Linux, but I am not concerned about RPG running on Windows or Linux. It takes advantage of the features of the server I want serving my code.

Now I know it's been said oh, just in case yada yada but not interested in just in case, I'm not going to water down to "just in case". RPG on IBM i serves business best, and I and the entire IBM i RPG ecosystem will develop for best case, not just in case.
Back to top
View user's profile Send private message Send e-mail Visit poster's website
Display posts from previous:   
Post new topic   Reply to topic    www.justiceforchandra.com Forum Index -> www.rdwrites.com forum (part of www.justiceforchandra.com) All times are GMT - 4 Hours
Goto page Previous  1, 2, 3 ... 9, 10, 11, 12  Next
Page 10 of 12

 
Jump to:  
You cannot post new topics in this forum
You cannot reply to topics in this forum
You cannot edit your posts in this forum
You cannot delete your posts in this forum
You cannot vote in polls in this forum


Powered by phpBB © 2001, 2005 phpBB Group