System iNetwork

Basically true, but it would take a lengthy response to deal with all the issues raised, too lengthy to deal with here.
.
. First, color, mouse, menuing, and windowing used in 5250 "green" screens make for productive work environments. I have never seen someone complain that a green screen wasn't productive.
.
. They complain of not being able to get to data, of not being able to work with it as needed, or of it not doing what they need now after a business rule change, but not about how it looks. And web pages don't change any of that.
.
. On the other hand, I have seen package after package upgrades purchased for the AS/400 with GUI screenscrapers, and every time the people that actually work with the screens were not interested in putting a mouse between them and getting real work done. We confuse glitz with substance. They are not nearly as confused.
.
. Having said all that, the AS/400 is dying fast because IBM decided that their Websphere browser server was our new interface. Why? Because they could. No one else would put up with their nonsense. We have no choice.
.
. Look anywhere else and ask yourself why no other computer pretends to have a web server as their interface, why no else is described as not having a GUI interface, why no else is dying. They have a choice. We don't, except to join them.
.
. IBM has to provide a thin rich GUI client as a native interface for the AS/400 for us to survive, and they have to do it now.
.
. EXFMT needs to talk to it with an XML stream of data that can be converted to any display protocol, HTML or any of numrous more complex display interfaces on the way.
.
. A standard way once again for AS/400 programmers to program, delivered transparently through EXFMT, used by the AS/400, based on Eclipse. Now.
.
. If IBM wants the AS/400 to survive.
.
rd

I apologize for the double response. I got a "CGI error" (both times), but the posts took.
.
rd

I am not so sure that just making software look better(i.e green screen to GUI)automatically makes the software better for the business. I have had more than one data entry person tell me where I could shove "slow my point and click" and wished they had their old green screen back!
It is our job as technical people to "educate" the non-technical management. I have seen people "go GUI", spend a lot of money and turn a stable IT department a three ring circus! Don't do something to just hop on the band wagon; do something because it makes sense for the business.

AS far as what sells when we walk down a grocery store aisle. Everybody at one time or another has bought something in a creative colorful package- only to find out it was a piece of trash.

I agree there are many viable alternatives for green screens and we should pursuit those avenues-in an orderly, disciplined and financially responsible manner!!!

Never lose sight of the fact that the computer is a business machine-just like any other machine in the company. Sometimes we forget the obvious: A computer is there to help
The company to generate revenue, reduce cost or both. The IT budget is not there so that bored technical people have the latest "toys" to play with all day!

I absolutely agree that if the interface doesn't change and change natively, OS400 if not the AS400 will die, sooner or later.

I think we also need to stop talking about GUI and mice, verses character based displays and deal with the problems. They are...

low screen resolutions
small colour palete
no automatic refresh
No bit level display control.

IBM needs to think about how they can make what they have better progresively, not try to ape slow interfaces that feel disconnected from the machine. Can you amagine how quickly our mature and intellegent develpment community would jump on DDS control of all of the above and create world beating applications.

Ibm simply has to defend the Iseries to survive. If the Iseries goes down, next topic will be those legacy "green screens" of IBM's mainframe systems.

And then the circle is closed
IBM lost the minicomputer (desktop) battle from Microsoft Windows,
IBM may lose the midrange battle from Microsoft Server 2003, Unix & Linux
IBM may lose the mainframe battle..... from clustered "wintel" machines with hundreds of pentium processors running ... linux.

Unthinkable?? Could anybody 5 years ago imagine the dying state of the ISeries in 2005?

Ibm has to defend their Midrange and they know it!

Are you all High or just MS Advocates?

The iSeries like UNIX and Linux and is stable, and is just what it takes to run a companies IT needs.

You don't need a fancy, retarted GUI to run a decent server. If you think Windows is the answer then please pull out your patch kit and start patching your Windows Servers, as for me I will be serving my customers rather than patching and worring all day about my crappy Windows Servers.

Yes I said it, Windows Servers are crappy. I think that MS has gotten it all wrong, and the problems can be seen every week when you get another security patch for those wonderfull servers.

You don't need a GUI, do you? and if you do then maybe the iSeries, and being an Admin are not for you and you may need to find a differnet line of work...try Circus Clown.

As I read through all the post an have to say I feel like you all don't get it and probably never will. Not all of you, but the ones off on a limb you need help. Not to say I don't, but at least I am not going to post something about how the iSeries needs a GUI...

We all know that OpsNav is good, but not great, but IBM was showing some great products to manage all thier server offerings, which I think is clever and will work for most IT shops.

Being web based give it the ability to be able to run on any platform, and is resricted to one OS.

Is the iSeries going to die....Keep on dreaming, you all have been saying that for over 20 years and look where it has gotten you? The iSeries is in a slump, and for that matter so are oil prices, and the economy too, so stop all the mush you are all trying to spread and get you chins up.

The iSeries is alive and well, and doing fine. IBM is going to have to watch what happens next and get a marketing group that is more agresive then they are now.

No one was referring to Windows.

rd

It would be nice the problem were so simple as presentation. Whether it's old school green screen or "windowing", I think is only a small portion of the iSeries future question. From my tiny POV, it appears that the high end service/contracting preference for the iSeries has put the iBox into a rarified atmosphere where fewer and fewer systems are sold. You can have all the contracts you want on every little piece of hardware or software you stick on the machine but if you only sell a few hundred machines a year you're going to be in trouble. Now I hope that's not what iSeries sales are like but it sure looks like it's going south! If we treat it strictly like the Kenworth/Freightliner of servers, then we can't ask for GUIs etc. because that's sort of like asking for a Porsche to do the heavy lifting. It can do it but not very well and vice versa. If the target market is SMB for now (market flava of the month/year?), then there needs to be a dedicated SMB iSeries platform that can do GUI, is sellable to young lions who prefer GUI, and is not solely dedicated to manage financial systems. It probably should not even be presented as an iSeries! However, under the covers, it should operate as an SMB iSeries and share the hardware and OS. Use any SMB OS upgrades as a path for migration into the larger platforms and for evaluating/testing LB platform upgrades and additions. That way, there is at least a positive potential for selling many smaller systems, keeping up some kind of software/hardware future technology that is not green screen dependant and yet relating SMB boxes to LB boxes. Plus, please do something unusual and outside the envelope for iSeries! It's nice that Countrywide et. al can run enormous businesses on huge multiprocessor systems reliably but take some iSeries machines out of that comfort zone. Otherwise, no one knows what else it can do.

I work for the largest vendor of iSeries systems to Outpatient Medical Practices in the US. I can tell you that the green screen totally kills new system sales to customers not already on the iSeries platform.

You can’t rationalize to new customers that green screens are more efficient even though it is very true. Usually the executives making the buying decisions are people who aren’t that familiar with data entry. They just look at the green screen as antiquated. You have to at least show that you've got a gui. Often times the customer doesn't end up using it but it gets you in the door.

Ibm simply lacks leadership in the IT information technology marketspace.

Remember the RPG/SEU/DDS classes we all followed decade(s) ago? For many 40+, 50+ guys, this knowledge has become a cult of its own, and they refuse anything else. Truth is these classes are no more and soon those guys will all be retired. Classes for SAP, ORACLE, .NET and JAVA on the other hand are today in high demand.

The result is that IBM is perceived in the eyes of decisionmakers as a follower, or uglier, a copier, of someone elses technology instead of being a leader as they were 10 to 15 years ago.

Holy cow, I can't believe what I'm reading. IBM has done everything they can to wean those ahem veterans like me off of RPG/SEU/DDS and done everything the can to replace it with Java/HTML. IBM is blameless in that.

I agree with Charlie above. The green screen is a powerfully productive interface that should be available without IBM's stupid green screen tax to punish people for not buying into Websphere but apps for the AS/400 should be displayed in a windowed GUI environment, something with good looks and performance based on Eclipse.

525o has all the fundamentals, it just needs more aesthetic options like fonts, colors, images, a GUI version of 5250 programmed centrally on the AS/400 as we always have, backwards compatible and future extensible.

XML from EXFMT will do it. But even that won't get the SMB business. It's too powerful, the apps too pricey. IBM doesn't have a clue.

rd

I am just thinking I should way in my two cents here...

I have been in a number of iSeries (AS/400) shops over the years... and one thing that I see a lot of is a segregation of the iSeries platform from the other platforms in a company... and usually, the people that don't work on the iSeries no very little or nothing about the iSeries platform, and those working on the iSeries know little to nothing about Windows/Unix/Linux servers and domains and routing and switching and stuff like that...

Now who can you blame for that lack of knowledge??? Maybe a little blame belongs to IBM, because, in my opinion, their training is a little pricey... since the colleges and universitys aren't teaching classes on the iSeries platform, IT managers are faced with a decision, do I go with a platform that is taught in every university in the country, where I can hire someone straight out of college that can (more or less) hit the ground running with my chosen computer platform, or do I go with a platform where I have a hard time finding quality people to work on it, and anyone I hire out of college, I have to send to at least a couple of expensive training classes (which aren't usually in convinient locations and not held very regularly).

Now, the blame isn't entirely on IBM here either... those of us that have worked on the platform for a while, have to get our feet out of the cement and venture out and learn new things... learn SQL, learn something about network authentication, read a non-iseries security book... no one is ever too old to learn something new... we need to broaden our knowledge into other areas so that we can learn how our platform can work and play well (seamlessly) with the other platforms in the company.

we also have to take the initiative to change the image of this platform within our companies or in the eyes of our customers... in some cases that is going to involve putting a GUI interface on a legacy application... not necessarily for an ease of use imporvement, but for an image purpose... it's like dressing up for an interview... if you put on a new suit and a new tie and a pair of polished shoes you will make a much better impression than if you wore your 20 year old suit, your shirt with the colar that has a wing span comparible to a small prop plane and a pair of shoes that were instyle during the Nixon administration... its all about appearance for those not familiar with the platform... and we have to make a good first impression... then once we get their attention, we can dazzle them with the capabilities of the platform...

we, as users of the platform, need to stop pointing the finger just at IBM when it comes to the poor image of the server... we need to get excited about this platform, and demonstrate its capabilities to the non-iSeries world... and educate ourselves so that we can speak intelligently to non-iSeries people about the platform and its capabilities..

I agree whole heartedly with Ralph. I think a lot of the problem with the Iseries platform is that a lot of it's programmers do not have a strong technical foundation. Many of them have a business background and learned to program in RPG on the job. A lot of them started out as a computer operator or in accounting and switched over to being a programmer, so they don't have the apptitude to sit down and learn a new programming language/platform. Many programmers I know that have a strong IT education can start out programming in Visual Basic and make the switch to C++, Java, etc. by just sitting down with a manual for a few days. Or they can make the jump from Unix to Linux, etc. This would be kind of a stretch for most RPG programmers.

I'm not going to get terribly involved in this discussion, because I strongly disagree with just about everything being said here. I will, though, make a couple of dissenting comments.

First, the browser is the fastest growing user interface option today. If it weren't, you wouldn't see so many books and resources dedicated to web design. To say otherwise is simply ignoring common sense.

Second, specifically to Mr_natural, if you're saying that learning VB in school somehow makes you a better programmer than learning order entry in RPG on the job, then I think you have an entirely unsupportable viewpoint.

Feel free to write an MRP generation or a cost rollup from scratch, or debug a subfile program, and then tell me how easy it is. Personally, I believe that ANY competent programmer can, if they choose, pick up ANY language -- it's just a matter of syntax. It's a lot harder to teach the core concept which differentiates programmers from coders, which is the ability to take user requirements and turn them into specifications (and eventually into working applications).

I'll take a solid RPG programmer over a coder (in any language) any day.

Joe

A lot of end-users in my shop have been very happy with the green-screen front ends I put on my apps. Happy because they are fast, don't slow down the PC, and I have squeezed every last feature I could out of the DDS to make them user friendly - popup windows overlaying the main screen so they can see what they were doing, rather than going to as whole new screen, context-senstive help, etc.

But the end-users aren't the ones making the decisions as to technology used. IT upper management are. As I write this, plans are underway to move a lot of apps to a 3 tier web architecture. It may be slower turnaround and a big expensive project, but using Java/Web tech lets the app be migrated to whatever platform they want. I think I-series is probably a good platform to run this from, maybe the best for high volume data serving, but to the top brass, it's a money pit that you get mired in the more you use it. Especially with the orice of disk. All the other boxes archive to the SAN, but not the I-series! All the other developers have free deployment and source magagement tools, but not the I-series!

Some of this will be fixed I hear, and some of this is due to the fact that you can't easily program freeware at home for an I-series and there's not that many of us to boot, but the primary objection I hear is money spent on hardware, combined with arcane technology.

The I-series as part of a web environment is being talked about a little here, but only speculatively.

IMO, the best bet for us would be to simply license OS/400 and let it run on anything, like Unix. Soltis already says this is possible. But, IBM sells iron primarily, not software...

Stability? If users are used to Unix levels of stability, they won't notice I-Series not crashing. Unless these things are measured and compared (not in my shop) this will fly over people's heads.

Would a GUI help? I don't think so. Unix has no native GUI. There's X-windows, but I-series has something similar.

No, I see Java becoming the lingua franca for programmers, and an accross-the-board decoupling of programming from the platform. Which is a good thing, for those who can make the switch - much bigger market. Native coding on any platform will probably be for optimizing IO (RPG native methods for example) and coordinating system events.

That's what it looks like from here.

Part of the problem with disk prices, I think, is that people are using the iSeries disk incorrectly. They're storing images and other large, static data on the box when that should be archived to a cheaper storage device. The iSeries disk should be for your high-volume, volatile, mission-critical data... NOT for storing web page images and scanned documents.

Joe

Attractive and robust GUI interfaces are a key to attracting new users and new types of applications to the iSeries, and HTML is currently the best GUI for most server applications. I can't quite agree that green-screens are killing the platform, though. The 5250 interface is a critical asset in my mind. It doesn't deserve bad press.

UI standards supported by Internet Explorer and Firefox will likely be the most successful for server based applications. I'd also suggest keeping an eye on Mozilla's XML User Interface Language.

IBM is NOT in a good position to invent or promote UI standards. The people who control Windows, Linux, Mac OS, and Palm OS are in better positions. HTML clients are the de-facto standard for client-server applications, and growing. People are better served by applications that adopt a GUI which is native to the client OS. I wouldn't expect IBM to come up with a better browser than Internet Explorer or Firefox.

Don't fall into the trap of deploying server-centric applications to a desktop-centric OS, just because it's GUI.

Nathan.

This article is dead on. I am glad to see an OS/400 related web site has finally figured out and published an article about the source of the OS/400 problem. GUIs are more productive than text based user interfaces. IT IS NOT UP FOR DEBATE, IT IS A FACT. OS/400 needs a native GUI. Instead of worrying about the next goofy BIF for RPG, Rochester really needs to figure out a native GUI solution. Rochester needs to see the “Big Picture”/think outside of the box. With at least 50% of the AS400 install base basically purchased and moving to an Oracle GUI solution (J.D. Edwards purchase), the only way OS/400 can compete is with a GUI. After all, it is the 21st century. A computer and its operating system solution has to sell itself. JUST A MARKETING CAMPAING WILL NOT CUT IT. Anyone fighting a change from green screen to GUI is just reluctant to change. In today’s business environment, "reluctant to change" is not an option.

"The O/S has to sell itself" is a statement I couldn't agree more with, especially in the IT sector where a lot depends on trust and mouth to mouth advertising. I mean if you push a new brand of vodka by a massive ad campaign, even if it tastes awefull, you could succeed, but not with IT.

Since someone mentioned disk drives here, I have a story here that gives the Iseries a serious blow in terms of reliability. Few years back, I had several new disk drives failing in a short period. When I pulled the drives, I noticed the label said model DSYS-xxxx 36GB drive 10K RPM, costs about $2,000.00 each. Accidently I had a PC intel server I needed some extra disks for, so I ordered some 36GB disk from IBM. The label said DSYS-xxxx 36GB 10K RPM, costs about $200 each. I compared with the Iseries disk and the labels were exactly the same (!!!!), except for a line in small print saying "fru part xxxxx ". When I asked the IBM service engineer, he said, no, these Iseries 36GB disks have slightly different microcode, but regretfully, this microcode is so incredibily buggy, that it may cause the drive to fail within weeks. Soon after, IBM sold his Storage division to Hitachi. Too much bad press, I guess.

Now picture yourself an IT manager who needs to replace his 5 year old depreciated Iseries with a new system. He knows he has to pay a more than ridiculous price for the 5250 tax, a more than ridiculous price for failing IBM disks ....

If it wasn't for that pile of dreaded 5250 RPG programs that runs his business, what would he choose?

I actually have a similar story to Ugerts...

several years back (when I was still new to the iSeries platform... I was trying to setup a 5394 remote workstation controller. We were having problems getting the CSU to communicate with the WSC. After lengthy troubleshooting we determined that it was a bad cable between the CSU and the WSC. I told the IBM service rep "No problem I have another serial cable right here"... he told me that we couldn't use that because IBM flips the transmit and receive pins in their cables (presumably so that they can carge $200 for a $10 serial cable)...

This kind of crap has to stop if IBM wants to succeed with this platform... in the world of tight IT budgets.... charging insane prices for proprietary equipment won't go over with IT Management.

I was a senior developer in a key position at IBM Rochester. The only thing I will say in Rochester's defense is that the IBM Corporation never understood the AS/400.

The rest of the story is that the Rochester lab is not what it seems to be. Rochester is a town in the middle of nowhere that has strived for decades to be completely self-sufficient - this is where the term Fort Knox, referring to the IBM program that gave life to the AS/400, came from. It is the most smug town I have ever lived/worked in. That attitude is brought right into the lab through the front door, even though it's also not the most innovative gem in IBM's crown by any stretch of the imagination.

Then there's the leadership at Rochester. How can anyone expect great things from directors that have more skeletons in their closet than some politicians? Sexual harassment suits, nepotism, deals made outside of work that have nothing to do with "work" but obligate one to favors, and a complete lack of technical vision manifest in people who literally could not get or hold down a job anywhere else in this industry. You can read "sour grapes" into this, but I've stuck to the facts and the truth. And BTW, Frank Soltis is nothing more than a "face" as most long-time, lab visiting, partners already know. Frank is no more the father of the AS/400 or iSeries than Elvis is to Michael Jackson.

I listened to years of Microsoft propaganda in that lab that was completely untrue. I've since been inside of Microsoft, as a guest, and it *IS* the IBM of the past - the one I hired into. Noone had any more admiration for IBM, and the AS/400 in particular, than Bill Gates and he has done everything he could to make Microsoft into a much improved version of IBM. One cannot know OS/400 and Windows Server (and where it's heading) without seeing the obvious path of imitation. Watch for a modern, improved version of single-level store in the next 3 years or so in Windows Server. Can you imagine seeing thousands of happy developers excited to be at work and succeeding every day? This is what IBM development once was and will never be again.

Finally, the whole key to achieving successful in business can be traced to effective marketing. Marketing can make mediocre products and services shine, while mediocre marketing makes great products dull or just invisible. Make the analogy to Windows and the AS/400 and you get the point. We spent countless hours seeking ways to improve the iSeries and to cheapen it to "keep up" with the competition. Did it help to charge $2,000 for $200 disk drives? Absolutely not, most of us personally thought actions like this were stupid. I even wondered if loading the disks with microcode to enable their use on iSeries, and preventing the use of $200 disk drives practically impossible, was legal.

So, is it any one the i5/iSeries is where it is today? With a development lab left to it's own devices and a two-bit marketing organization, I recognized by 1997 that the iSeries was dead. Today the workforce there is demoralized and fearful.

Move-on if you can.

It's Time To Improve Our Image could apply to a native interface or marketing, or both. But is marketing really an answer for the AS/400? I say no, not without a native GUI interface, and even then, only to software vendors to prodice apps that take advantage of OS/400 and its new interface.

IBM's message for the AS/400 since at least 1998 has been simplification, with some kind of vision of the AS/400 as a disk server for multiple operating systems, OS/400 seemingly an afterthought in the list of Linux, AIX, and Windows.

It used to be a mantra of one backup command, or something similarly inane, but has evolved to trying to convince people to put all their Linux, AIX, and Windows servers on the AS/400 for this On Demand stuff, fancy terminology for putting extra CPU's, disk, and memory on a box and charging for its use as needed.

Does anyone really believe that IBM isn't going to charge for the cost of the hardware one way or the other? No, at least outside of IBM marketing.

There's also a hint of getting rid of admins implied although I have no idea how administering four OS'es on one box magically eliminates the need to know four OS'es, as if any of these OS groups would do this anyway. I guess that's why IBM is going the CxO route now, and proud of it. They are driving trickle down demand from the CxO's, they proclaim. Holy cow, how lame can it get?

Rather than simple, the native interface for the AS/400 actually has to be more powerful than lowest common denominator web pages. Even XWindows wouldn't do now, although we have allegedly had it for years.

It would take a real architecture, from OS inherent development and deployment to native integration with the rest of OS/400, something as good as 5250 is and was on the AS/400, to put a solution set out there that isn't a me too golly gee I can serve web pages solution. Firefox XUL or some similarly robust open source client could be that solution.

It has to be in the realm of a GUI 5250, somewhere between HTML and Windows, something more solid than AJAX, a thin rich client if you will, backwards compatible and future extensible, and we'd be on our way.

The image would be improved, and it wouldn't be IBM's idea of an improved image, a guy spinning around on a merry go round saying gee the AS/400 is simple.

No, whoever dreamed up that mess is.

rd

It seems sad that perhaps too many people's futures are overly dependent on what IBM and Microsoft do, or don't do for that matter. The answer which may seem obvious to some is to become more self reliant, as individuals, as organizations, and as IT departments within organizations.

On a more positive note, it might be helpful to know that there are ISVs that are growing, primarily on a strategy of deploying new applications with graphical user interfaces, which are attracting new users to the iSeries platform. Our revenue nearly doubled last year from the previous year. Most of our new applications are written in RPG, and some in Java, using HTML & JavaScript for the user interface.

Folks,

In June 2003 I was made redundant from my position as Senior/Analyst Programmer at a financial organisation in West London, England.

For months afterwards I strived in vain to get a job in RPG programming. After months of failure I changed tack and launched a new career in technical software testing. I now work for JPMorgan Chase & Co, am better paid, better valued and happier than I ever was in my last 2 fearful years on the AS/400.

There is life after the AS/400. Believe me. Do not hitch your wagon to the falling star of IBM.

One thing holding folks back from moving to GUI on iSeries is response time. In my shop I've been pushing the team to get experience with Navigator and Management Central, and consider doing more of their work with them. It is, after all, a powerful and convenient tool.

Everyone on the team has had bad experiences with the tools from previous environments, and they perform poorly in ours too. We don't know where the bottleneck is, but even simple operations can take forever to accomplish. These are well-configured desktops, running on a healthy LAN, and no one here seems to know why it always runs so slow! Any ideas out there?

In response to the last poster.

Whatever the cause is of the slow response times, you can tweak and configure all you like, os/400 is simply a dead horse. It will never get away from its traditional habitat which is "green screen seu, rpg and batch programming" and it will be buried with it.
Look at the reponse times of Websphere or Java on AS/400, for crying out loud, it's a living joke!!! Yes, maybe you can get it up to speed if you buy an I5 with horsepower of 500,000 CPW costing 500,000$, but you can get better response times from an ordinary intel pc server running dual pentium 4's 3GHZ, 2GB main storage serving 1000 simulteanous clients at a purchase cost of 5,000$. Just 100 times less pricy and on top, better performance.

The sounding motto for AS/400 (or as its name changes each month now, Iseries, I5):

!!!!!!!!!!!!!!!!!!!!!!!!!!!
!!! MOVE ON IF YOU CAN !!!!
!!!!!!!!!!!!!!!!!!!!!!!!!!!

For those who are saying that PCs are taking over AS/400. I first programmed on an Apple in 1980 using Basic. So PCs are not the buzz word of today, they have been around for 25 years. I was a Mainframe Cobol programmer those days and was told the days of mainframe are over. Last year I made a conversion from COBOL/390 TO RPGIV to discover the mainframe and Cobol are still alive and kicking. No, the conversion was not a shift from 390 to AS/400. It was to expand the software market to AS/400, not to get away from 390.

25 years after the prediction of its death, mainframes and Cobol are still not dead. It is gonna be a while before AS/400 and RPG dies.

Yes, we need to change our image as a freaked out overaged green screen programmer. My 65 year old former lady boss is still programming in true ILE and going for GUI now. Shame on me who is under 50!

For certain applications, a GUI is fine. But in other parts of industry, where RF is used heavily for instance - such as distribution centeres - GUI is the worst thing. You have handheld devices with small screens and no room for a mouse on the keypad. It all sounds good from a presentation point of view, but our business side refuses to consider an alternative to a "green screen" in this and in customer service as it would dramtically reduce productivity.

Rather than remove the capability, the software provider needs to consider the sector they are marketing to and determine if GUI or green fits the business model of the customer and code to those needs rather than try to appease everyone.

I get too many vendors trying to market bells and whistles when the functionality to perform our core mission is missing. This includes IBM as they are trying to shove x-series down the throat by offering them in the new servers. Quit wasting my IO space with unwanted features.

If IBM would be customer friendly instead of a confused morass of marketing iniatives, the iSeries could be marketed to meet the needs of the customer rather than entangling all of us in a mass of buzz words.

What the heck was I talking about when I wrote the comments above? Sorry for that little bit of...ummm....crap!

As I sit in and learn more about IT and the changing face of the datacenter I think the GUI is the way to the user hearts and that will in turn help IBM. If you make applications that are vital to the business and are easy to navigate you will in turn sell more systems.

That is all. Thanks Scott for a great point of view and again I must have been the one that was high that day. Sorry. I am still shaking my head.

Scott is right! Our sales suffer because of our Green Screens.

Anyone who said that Green Screens are better for Data input are also right.

BUT, the bottom line is that the people buying (Company Boards, CEO's etc) have no idea what happens in their IT departments and whether data input into Green Screen is better, they just see either a Green Screen (means OLD) or a pretty screen GUI, Windows or whatever (means NEW).

WebSphere and related products are not a solution, they are slow to develop and use and expensive. Most Scrappers are too expensive to maintain. CGIDEV2 is great, but only for new development.

There are several Vendors of Migration software but few Vendors that give us Pretty screens inexpesively with low development and maintenance costs and allow us to stay on the iSeries.

I have spent over a year looking at various solutions and most go part of the way but leave a lot of development which equals cost.

IBM could have done a better job if as previously suggested they had gone to HTML and XML and CGI rather than Java and WebSphere. By now PDM should have given us all these things. Why are they so focused on Eclipse and Java and WebSphere. Now with the i5 they have no excuses, we should be able to build graphical interfaces from within the i5 and not have to go to PC based products.

I would suggest that it is not just Green Screens that have been killing the iSeries, but IBM as well.

Hopefully in this new look at the iSeries we can get the upgrades and changes we should have had at least 5 years ago without having to swallow WebSphere, Java and Eclipse!

IBM is too late to turn this machine around. They had their chance in 1995-99.
Languages that made then the e-jump to GUI and OO have a chance to survive.
All that IBM added to RPG (ile) are some bells and whistles, not even worth to spent 1 word of comment on.

Again, for those on the AS/400 not yet near the retirement age, one moto:

MOVE ON IF YOU CAN !!!!!!!!!!!!!!!!!!!!!!!!!!

The sad part is that this scenario was predicted already 5 years ago. But IBM choose the easy money from "services" instead of "hardware". Fi., I've met tens of IBM consultants labeled "SAP" or "ORACLE" who didn't even know what an AS/400 was, never met any IBM AS/400 consultant. This strategy of neglect led to today's status, and frankly, the way you see IBM turn and turn to please AS/400 ISV's and customers today, compared to the arogant IBM they once were, should make us all laugh. What a joke they've become.

Move on to WHAT??????

There is no better machine than the i5.

As for you not meeting any IBM AS/400 Consultants, perhaps you are in a different circle and so would not. I have met many very talented IBM AS/400 consultants. I have also seen the code and file structure created by SAP. You can keep it!

Times are changing. Companies are no longer happy to throw millions into software or hardware. I have seen and worked with "PC" teams. C++, Java, VB. They are all more expensive than RPG programmers and generally (in my opinion) do not produce products to the same standard in the same time.

If I were you U.Geerts, I would change my little slogan to

MOVE BACK TO COMMON SENSE IF YOU KNOW HOW!!!!!!!!!!!!!!!

OK, I've read all the previous comments here. It bothers me that we have become so dependent upon IBM to give us something. We are just as divided as IBM about the platform we love to hate!! More marketing, better developers, different interfaces. I think we are all looking at this all wrong. IBM is all over the map with it's direction for the iSeries. We don't know what we want because our only investment is in what we know (or even better what we feel comfortable with).

I challenge any developer that has been on the iSeries for over 3 years to ask themselves, "What have you done in your shop to change the view of the iSeries? Have you suggested the iSeries to do something that a PC or Server could?" If you haven't then you are part of the problem too. Things will change at IBM when we start doing, instead of complaining. We control the ultimate direction of the iSeries not IBM. Look at the open source movement, Linux would not be what it is today if it wasn't for developers that have vision and do something instead of complain. We say we love this platform but treat it like a red headed step child, "because IBM didn't market it right or they didn't do this or that". Stop the crying and step up to the plate and become part of the solution instead of the problem.

IBM has issues, I'll be the first to agree with that. But don't make IBM's issue our issue. Developers are the driving force that can sell things to upper management. Yes, they love the look and feel of GUI. They hate the cost that iSeries puts on interactive. My challenge to you is become a leader instead of a follower.

What do you love about the iSeries? Can you remember? If you can't then you are just as dead as the platform you've been complaining about. So let's stop whining and be REAL developers and define solutions.

Finally, this is not competition between us and them. This is defining solutions for our customers. We have the tools, the question becomes are we willing to use them?

Well this is a sight to see. As a person that has done everything on the AS/400(Client Server distributed application with VB 6.0 .NET, CL, RPG,II,III,IV, COBOL, DDS, SQL).

I can't tell you how sick and tired I am of the MS hate mongering. Neither IBM nor MS pay me a dime to do development. It is my choice to witch platform I develop on. I have done the impossible with AS/400 and had the people on this machine and the communities at large vilify me and my efforts because I use MS tools to do it. I don't want this machine to die off, but the way things are in he community it is certainly headed that way.

I have a great architecture for building advanced application for the AS/400 that would leave the RPG and COBOL people very much involved. It’s has been tested and used.
But does anyone in this arena want to hear it. No!!!

The Web is only viable to a certain point. If you are going to build Windows applications then build Windows applications. Stop trying to force green screen mentality on the GUI. I have a distributed application that can be launched from a browser or just go to the web site and run it as a Windows application. It’s using .NET on the front end and an iSeries on the back. The AS400 is a wonderful machine but you can't get blood from a turnip. Stop hating MS for competing, no matter how slimy it may be, and start re-architecting your applications.

Over the past 10 years I have figured out how to do just that. With UML OO and true distributed system design techniques, I can build just about any AS400 business application you want. I used to work under a frequent commenter on this site “Joe Pluta�, if this is the same guy that used to work for SSA in Chicago.

I am willing to teach the techniques as well, but I won’t deal with the hate. It's about architecture and design now folks not the code. You guy's have all the business knowledge in the world to re-build the AS400 market and give your users what they want and not what you want.

If I’m wrong then let me know.

James C. Posey

He is. I was there too, a programmer in the cave. He was across the hall, I didn't meet him. That was before Covey threw the consultants out and hired kids out of college to rewrite BPCS in C++. The rest is history.

I took a quick look upthread to see what you might be responding to and I see I have done it before. In May of last year I posted, "No one was referring to Windows."

rd

Maybe you did RD. Sorry about the re-hash, but others implied the comment even thogh they didn't say it directly.

Anyway, it's good to meed a fellow brother of the technical downfall of good product. To be honest, that is where a lot of things came together in that place.

C++ huh, that had to hurt.

James

Why the heck does my name keep being used in vain?

Hey James, wassup? By the way, I have an architecture that works great using the web as a front end to RPG and COBOL business logic. It can just as easily talk to thick clients.

One version of the architecture can even be used to directly replace a display file. I published a book about it back in 2000.

Joe

It's all in vain Joe, James. Many people, including myself, developed prototypes back in 2000 to interface a Web browser directly with Cobol/Rpg programs, even WITHOUT(!) the use of an HTTP server or CGI. I did this with direct socket communications to the pc client and, probably the first in industry, the exchanged html pages were not stored in an IFS or other hierarchical file system but stored in an sql indexed file as variable-length ascii data strings. The strings were simply captured from Ms Frontpage designer. This system was blazing fast and very suitable for a direct interactive dialogue between the browser's user and the Cobol/Rpg program with the same functionality as a 5250 session. Even the exchange of rich datatypes like gif images or avi audio was no problem, but they had to be inserted in each html page.

So nobody denies it is technically impossible to realize this on an as/400 or iSeries.

But the reality today is that iSeries is going under fast, faster and now accelerating to light speed. What do you make of the 18% sales drop in 2005 Q4 verse Q3? With all the hoopla of the i5 introduction, revenue 2005 was only crancked up a mere 1% higher than the disaster year 2004. Is this a turnaround? And those who set their mind on buying an i5 in 2005 have done so, so who will be hungry enough to buy in 2006? Because of power5+ architecture verse power5? Will that '+' or the next v5r4 release be a sufficient incentive for customers? Nobody has a crystal ball, but a lot of us have a gutt feeling this thing will be dead and gone in no time.

BTW, heard about Coca-Cola? The TOP 500 company was a long time advocate of iSeries. They began investigating SAP around 2002, but I was told by an insider in august 2005 it was only a preliminary investigation which would take 10 years at least and it would never replace the home grown, rock solid, RPG ERP system on as/400. I contacted him again (januari 2006), out of work, Coca-Cola closed down their as/400 operations and made the transition to SAP end of 2005.
Amen.

PS: as a side note, SAP is running on iSeries, but os/400, except for powering up or down the system, is not used anymore.

You want to know one reason why the iSeries is in decline? There are others, but here is one. IBM has the wretched habit of renaming common concepts with their own new words, creating confusion.

For example, after 10 years of heavy AS/400 programming and support, I finally learned what an 'activation group' actually is. The linux equivalent is a child shell. For example, if you are using the bash shell and you type 'bash' at the command line, you open a child shell that is independent of the parent shell. Shell variables from the parent are not defined in the child unless they have been exported using the 'export' command. Variables created in the child are independent of the parent.

I had this explained clearly in a couple of minutes and immediately noticed that this is the equivalent of an activation group. This also explained why I never seemed to have any problems keeping all my work in the same activation group, and only had problems when a new one was created.

in IBM speak, you only get a confusing and incomprehensible explanation and the feeling that you are somewhat incompetent of you don't use activation groups religiously.

Bye Bye AS/400. Glad to see you go.

Long live RPG! Long live Java Websphere! Long live .NET front ends! Its all useful in its place.

I find the discussion from the ex-SSA people contains elements that need to be considered by any any organization that has a large investment in iSeries server applications. There is no single architecture that will effectively address the business needs of these organizations. For some applications that involve heads-down data entry the 5250 UI may be the best solution. If it is not broken don't fix it. For other applications such a business intelligence, a fat client (VB.NET, J2EE Java Application, Visual RPG etc) that uses the iSeries as a database server may be the best solution. For anything that is widely distributed, thin is good. Thin could be 5250, J2EE Websphere, or .NET with ADO access to the iSeries. I have spent the past few years focusing on Java Websphere and found that the marriage of J2EE and the 400 is not always a happy combination. Lately, we have been moving our Websphere applications off of the iSeries on to Windows 2003 Servers. We have found that unless you can carve out alot of memory into dedicated LPARs for Websphere, either the Websphere suffers from paging requirements of other apps or the rest of the 400 apps suffer because Websphere grabs all the memory it can. We are also finding that Websphere on Windows Server 2003 and RPG (both ILE and OPM) on the iSeries provide higher performance and better maintainability than pure J2EE Java on any platform. Lately, as our applications have been scaling, we have found that using JDBC with Java on Websphere for complex iSeries database operations is resulting in a growing problem with stale connections. We have recently begun writing server side code in RPG and accessing from Websphere with PCML. All of a sudden our recent hires who did not come up through the RPG ranks are excited about this new back-end technology that we have for procedural logic - RPG. We are going through an RPG renassance. If we find bottlenecks in our Websphere apps, lately one of our first considerations has become "should we rewrite the logic for the offending area in Java?" There is no reason that a good iSeries developer should only consider one tool for a job. All of the languages and tools discussed in this thread have both positive and negative attributes depending on the context that they are used in.

Hey Joe! Sorry to use your name in vain. It's only in the highest respect you know.

I knew you would create something like that. Many of the concepts I used in the early stages of my OO indoctrination came from our SOM project with SSA.

Love to share some stories with ya. Just hit me up on my email adress jamesposey@modelsysllc.com.

I work in both worlds (Java, .NET) but most I'm into Architecture and Desgn with UML and RUP. Keeps me out of the language wars as much as possible.

I still like the AS/400 and what it does. I don't sell hardware so I can't dictate on that level.

James

Honesty in 2006!

Honestly this thread needs to die.
Honestly the iSeries will be placated by any number of other systems.
Honestly any number of other systems will be placated by the iSeries.
Honestly iSeries is a great system for thousands of businesses.
Honestly ditto for windows, unix, linux, mainframe, etc.

Honestly why the heck does Ugeerts keep bashing the iSeries?
Honestly Ugeerts from reading your threads you've done it all, so why do keep coming down from on high to lecture everyone?(I mean the .html,.wmv,.avi,.gif,.jbg,.wellgeejusteverything, in an ascii variable records accessed by SQL in 2000!!!!! And socket communications - whoa you go girl!) Wow you have done it all!

Honestly do you think we care?
Honestly, we understand that you see no future in the iSeries.
Honestly are you going to keep saying the same thing throughout 2006 too?
Honestly for all your knowledge you claim to have, you sure are one repetitive, boring individual.

Honestly just who the heck pays you to keep up your mis-information and iSeries bashing?

And so honestly just why do you keep buzzing around here?

We get your point, the same one, over and over, now go pick on (manufacturer inserted here).

PS Directly from a job listing a the Coca-cola site:
Install, test or implement upgrades to existing systems, software
tools or infrastructure (e.g., Unix, Mainframe, NT, AS400, Visual
Basic, Windows, DB2, CICS) in order to support information system
infrastructure.

First you say Coke dumped the /400 for SAP, then you say SAP still runs on the /400. You're one confusing guy. A company like CC will have many uses for a variety of technology and it is un-likely that they were ever satisfied being a one-horse annie. So more to the point, CC may use the /400 only for SAP but then again it may have never been the only horse in their stable to begin with. You're stretching yet again. I mean do you think they use PC's?!