Back to the Future with RSE

If there's one thing that folks in the IBM i community love, it's a name change, right?

Well, here's one that we never thought we'd see - one that puts us back to where we were a few years ago. Nope, IBM i is not changing its name to iSeries or AS/400 (but that would have been a good April Fool's joke a few weeks ago, wouldn't it? Wish we had thought of that.)

We're sure you have all followed carefully the progression of packages and names for the Eclipse-based development toolset from WDSC (WebSphere Development Studio Client) to RDi (Rational Developer for i) and then to RDP, or RD Power as some prefer (Rational Developer for Power Systems). 

We're kidding of course. From the number of questions we get personally and see on forums, Internet lists, LinkedIn groups, etc., the litany of names has been a big source of confusion for many.

Guess what--we're going back to RDi again. Rational Developer for i. They've added "RPG and COBOL Tools" to the end of it, but still, it would seem that V9.0 (for those who have lost count, the current version is V8.5.1) will be called Rational Developer for i - or, we presume, RDi for short, as before.

We haven't seen any information yet on pricing, but we're hopeful that there won't be a significant increase beyond perhaps the (sadly) all-too-predictable 15% that seems inevitable with each turn of the crank.

As far as enhancements to function for IBM i RPG and COBOL developers, we haven't seen a lot. It appears that IBM Data Studio is now packaged with the new version vs being a free download that you could install yourself at 8.5.1.

Also listed are "Significant improvements to Remote Systems Explorer qualities of service." This sounds like lawyer-speak for bug fixes. We hope we're wrong but don't hold your breath. Then again, bug fixes would also be welcome.

There are two new packages, in addition to the basic RDi package:

1) Rational Developer for i RPG and COBOL + Modernization Tools, Java Edition - this would appear to replace the Rational Developer for Power Systems Software Power Tools for i.

2) RPG and COBOL + Modernization Tools, EGL Edition - this would appear to replace the oddly named RDi SOA toolset, which never changed to RDP when all the similar products did, for some reason.

From what we understand, if you have Rational Developer for Power Systems (what we've referred to as RDP) and have an active SW maintenance agreement, you will be entitled to a no charge license to the V9.0 RDi package. Someone reported on the WDSC-L list on Midrange.com that it will require a new Activation Kit to be used to install it, however.

Likewise, if you have a license to the RDP Power Tools, you're entitled to V9.0 RDi  + Modernization Tools, Java Edition, and if you have a license to RDi SOA, you're entitled to V9.0 RDi  + Modernization Tools, EGL Edition - both assuming active SW maintenance, of course.

So is all this good news, bad news or no news?

Until pricing is clear and, of course, we see how well the actual V9.0 code performs, it's hard to say. The new packages do seem to more accurately reflect what people really need. AIX/Linux developers have their own product and IBM i folks have RDi - as we did before. If you want/need more significant Java and/or EGL support, there are additional features for that.

Anything that simplifies the ordering process is a step in the right direction of course. But will the Rational reps stop trying to write up orders for RAD when IBM i shops call to get "Rational" to develop their RPG code? We can but hope! 

Back to the Future with RSE

If there’s one thing that folks in the IBM i community love, it’s a name change, right?

Well, here’s one that we never thought we’d see - one that puts us back to where we were a few years ago. Nope, IBM i is not changing its name to iSeries or AS/400 (but that would have been a good April Fool’s joke a few weeks ago, wouldn’t it? Wish we had thought of that.)

We’re sure you have all followed carefully the progression of packages and names for the Eclipse-based development toolset from WDSC (WebSphere Development Studio Client) to RDi (Rational Developer for i) and then to RDP, or RD Power as some prefer (Rational Developer for Power Systems Software).

We’re kidding of course. From the number of questions we get personally and see on forums, Internet lists, LinkedIn groups, etc., the litany of names has been a big source of confusion for many.

Guess what - we’re going back to RDi again - Rational Developer for i. They’ve added “RPG and COBOL Tools” to the end of it, but still, it would seem that V9.0 (for those who have lost count, the current version is V8.5.1) will be called Rational Developer for i - or, we presume, RDi for short, as before.

We haven’t seen any information yet on pricing - we’re hopeful that there won’t be a significant increase beyond perhaps the (sadly) all-too-predictable 15% or so that seems inevitable with each turn of the crank.

As far as enhancements to function for IBM i RPG and COBOL developers, we haven’t seen a lot. It appears that IBM Data Studio is now packaged with the new version versus being a free download that you could install yourself at 8.5.1.

Also listed are “Significant improvements to Remote Systems Explorer qualities of service.” This sounds like lawyer-speak for bug fixes. We hope we’re wrong but don’t hold your breath. Then again, bug fixes would also be welcome.

There are 2 new packages, in addition to the basic RDi package:

1) Rational Developer for i RPG and COBOL + Modernization Tools, Java Edition - this would appear to replace the Rational Developer for Power Systems Software Power Tools for i

2) RPG and COBOL + Modernization Tools, EGL Edition - this would appear to replace the oddly named RDi SOA toolset, which never changed to RDP when all the similar products did, for some reason.

From what we understand, if you have Rational Developer for Power Systems (what we’ve referred to as RDP) and have an active SW maintenance agreement, you will be entitled to a no charge license to the V9.0 RDi package. Someone reported on the WDSC-L list on Midrange.com that it will require a new Activation Kit to be used to install it, however.

Likewise, if you have a license to the RDP Power Tools, you’re entitled to V9.0 RDi + Modernization Tools, Java Edition, and if you have a license to RDi SOA, you’re entitled to V9.0 RDi + Modernization Tools, EGL Edition - both assuming active SW maintenance, of course.

So is all this good news, bad news or no news?

Until pricing is clear and, of course, we see how well the actual V9.0 code performs, it’s hard to say. The new packages do seem to more accurately reflect what people really need. AIX/Linux developers have their own product and IBM i folks have RDi - as we did before. If you want/need more significant Java and/or EGL support, there are additional features for that.

Anything that simplifies the ordering process is a step in the right direction of course. But will the Rational reps stop trying to write up orders for RAD when IBM i shops call to get “Rational” to develop their RPG code? We can but hope!

 

 

 

 

 

 

 

 

Lawson Software Users Flock to RSE (or at Least They Should!)

It's conference season again! For us, that means travel season--even more so than our normal travel schedule. Most of the conferences we attend are hosted by local user groups. We were in Wisconsin in March for one of our favorite events--the annual WMCPA conference, followed by our own RPG & DB2 Summit, which involved quite a lot of interaction with the MMSA user group in Dallas. One interesting aspect of both conferences was the increased number of younger developers in attendance. WMCPA always has a few because their president, Jim Buck, usually brings a group of his students along to assist and participate. But we were very pleasantly surprised to see the number of younger folks at Summit--we're used to Aaron Bartell being the youngster!

The following week (now) we're at the NEUGC event--a unique event where multiple local user groups in the area pool their resources to put on a successful and fun conference each year. Once again there seem to be a few younger folks around, but the really good news is that overall attendance is up. We should all applaud that. Next week, we head to Fort Wayne, Ind., for an all-day session focusing on RSE hosted by the local user group there. Then the day after returning from Fort Wayne, we head to Sweden for a developer conference there. The following week our own local user group, TUG, holds its annual TUG TEC event, which is followed by the grandaddy of user group conferences, COMMON in Anaheim. We'll be speaking at COMMON this year for the first time in a while; it will be nice to see some old friends again.

So what does all this have to do with Lawson? Today at one of Susan's RSE sessions at NEUGC, she had the pleasure of making one person really happy over the idea of using RSE to edit source code. In fact she was positively ecstatic. Many SEU users are happy when seeing the productivity features of RSE than leave SEU in the dust. But ecstatic? That is the reaction you get when you're a programmer who can't even use SEU to edit her RPG source code!

What would programmers be using instead? How about EDTF (Edit File)? Never heard of it? It's the green-screen editor for IFS stream files. You can get to it from WRKLNK. Imagine: no PDM or even SEU. Just WRKLNK and EDTF to edit your source code. Next to EDTF, even PDM and SEU look like the greatest programmer productivity tools ever invented. If you've not had the pleasure of using EDTF, perhaps you can remember eons ago using a truly rudimentary editor called EDLIN. EDTF is a tad more advanced that EDLIN but not a lot.

Why has this person been editing RPG source with as primitive a tool as that? Because her source code is stored in the IFS and it was the only editor that she knew of that could edit IFS stream files.

She's not alone. More and more shops are finding themselves in the position of having to edit source code in the IFS because Lawson's recent releases started storing and shipping all of their source in IFS files rather than source members in regular source physical files. Who knows why they would decide to make that rather strange decision? Perhaps someone out there among our readers can enlighten us. Perhaps even stranger is the fact that Lawson doesn't seem to have told their users that RSE is the way to go for editing the code.

Regardless of the reason, the effect has been a dramatic drop in programmer productivity among those RPGers who must make changes to the Lawson code because EDTF is the only obvious way to do it--at least if you're limiting yourself to green-screen tools.

Now the person in Susan's NEUGC session today has learned that RSE (packaged in RDP/RDi/WDSC) has no problem whatsoever working with source stored in IFS stream files, and hence her unbounded joy. RSE can seamlessly handle IFS files for editing, compiling, etc., in exactly the same way it works with source members for the rest of the IBM i community.

This individual wasn't the first Lawson programmer who couldn't wait to get her hands on RSE. One of our recent private classes was to an entire group of Lawson programmers who were equally ecstatic over the prospect of leaving EDTF in the dust.

Now, if we could only get the rest of the SEU users still out there to a have similar revelatory moments about how much more productive they could be ...

What We Are Thankful For

We live in Canada and consequently our Thanksgiving was in a few weeks ago. However, as in most years, we were working in the U.S. that day, so we tend to celebrate in November. So as we near the Thanksgiving celebration, our thoughts turn to things we are particularly thankful for.

We have many personal things we are thankful for. These include our health (including our wonderful Canadian healthcare system), our families (including four exceptional grandchildren), our home (which we sometimes wish we could spend more time in than we do), and many more things that we could go on for hours listing. We are truly blessed.

But in this post, we thought we'd concentrate on those things related to our work with IBM i that we're particularly thankful for.

First of all, we're grateful to still have work and a good income, despite the state of the economy. We know many who are not so fortunate. Many IBM i shops seem to have decided that an uncertain economy is a good time to exploit their investment in existing applications by moving them forward, enhancing their applications' value and the developers' productivity. That just happens to be our specialty, so it has worked out well for us.

We're grateful that the developers of our favorite programming language--RPG--have never even considered resting on their laurels. They could have simply cashed in on the maintenance revenue stream for applications running the businesses of thousands of companies worldwide. Instead, they have continually made the language more modern (/Free, data structure enhancements), more powerful (native XML processing, enhanced file handling), and especially recently, more easily expandable (via RPG Open Access or OA--more on OA in a moment)

We thank our lucky stars every day for RSE (the toolset currently packaged in RDP, earlier in RDi and WDSC). It's almost painful to recall our days of SEU development, with DSPFFD to get externally described field names and spooled compiler listings to find compile errors (shudder!).

And we're delighted that Rochester has given us a new friend to play with by embracing PHP on our favorite platform. If nothing else, PHP allows us to breathe a sigh of relief that we don't have to keep beating our heads against the Java brick wall. PHP gives us another option for a modern, browser-oriented language environment that lets us use OO to the extent we choose and offers an amazing wealth of open-source tools and applications. All with relatively simple integration to RPG applications and IBM i function. And best of all, it makes sense to those of us with RPG brains.

We said we'd come back to RPG OA. That's simply so that we can remind you that it was about a year ago that we made a plea for rational thought (pun intended) to prevail related to the Ts and Cs for OA. We have seen some progress. A trial version is available, at last. But we're still waiting for the true solution that we still hope will come--OA support to be included free-of-charge with either the OS or the RPG compiler. We have read rumors to the effect that we may not have much longer to wait for this. It would make a really nice present to find wrapped up under our Christmas tree!

Most of all, we're simply thankful to have had the opportunity to make a career working on the best platform on the planet for business applications.

Last, but by no means least, we're very thankful to all the readers of this blog. When the IBM Systems Magazine folks asked us to begin writing this some years ago, our first reaction was "but who would want to read what we have to say every week?"  Well, apparently, quite a few of you would. We're gratified and humbled at the number of you reading our weekly posts. Thank you all.

And a very happy Thanksgiving to all--whether in the USA or elsewhere--we all have a lot to be thankful for.

Jekyll and Hyde

This is one of those blog posts where we're not quite sure of exactly why we're writing it - in all probability, it's out of a sense of frustration/despair/je ne sais quoi we suspect but perhaps if we get on with it you'll understand where we're coming from.

As we said in our previous post, last week we were in Montreal (which is why we had to include the French reference above!) and some of the folks we met there reminded us of the Jekyll and Hyde aspects of our favorite system. On the one hand, our old systems can run unchanged from way back in the S/36 and S/38 days (Dr. Jekyll) and, on the other hand, it can continue to run our old systems ... (Mr. Hyde). Nowhere is this dichotomy more apparent than in the S/36 compatibility arena.

Written in the old, old versions of RPG and/or COBOL many of these applications continue working to this day thanks to the S/36 Execution Environment (EE). But is that a good thing?

It seems to us that what was initially a good idea has ended up condemning many IBM i users to being second-class citizens at a point in history when they most need help. After all, typically the longer an application is in service, the harder it is to maintain - but there are few tools available to assist in this task. Sure in the early days of the (then) AS/400, there were all sorts of tools available to help you to migrate your S/36 applications to native mode - and from there, of course, you could easily move to the new ILE languages when they became available. At the time, for those who subscribed unthinkingly to the "If it ain't broke don't fix it" mantra, the S/36EE was very tempting. Why convert when it would keep on running just the way it was? But that was 23+ years ago and now the vast majority of those programs should probably be considered "broken" or at least need some significant modernization because without it they are easy prey to a Windows "solution." Trouble is that too few users made the move to justify the continued evolution of the early tools and most appear to have vanished and no new ones emerging.

Case in point - we were teaching RDp in Montreal and a student asked if it would help with their existing S/36 RPG. Simple answer - No. It can only treat RPG II as if it were RPG/400 (RPG III) and that means that frustration lurks around every corner as any attempt to use the built-in Help, etc., is likely to lead to the use of op-codes and other options not actually supported by the S/36EE compilers. Not only that but the very tools that would potentially be of the greatest help, namely the Outline view and Application Diagrammer, are of limited functionality (Outline) and non-existent (Diagrammer). Not to mention that Service Entry Points for debugging are not available without writing an ILE wrapper. Then there''s the screen designer and ...

If a user does decide to seek help to move forward then Googling "migrate s/36 application" will bring up three references to Applied Logic's NCS followed by a forum question that dates back to the year 2000! Not exactly a lot of help out there. We suspect that Databorough's X... tools may be able to handle the task too as they can certainly document S/36 applications - but it isn't shouted out on their website and hence Google doesn't pick it up.

So what is the point of all this? Perhaps that we should reconsider whether what we have is really working - or merely continuing to run. Perhaps when IBM tells us that we should really be re-engineering our databases from DDS to DDL that we should resist the temptation to say "Why? It still works." Perhaps we should make a point of converting our RPG400 programs to RPGIV the next time we have to make even the smallest of changes. Perhaps we need to reconsider what we mean by "broken" lest even those of us not stuck in the S/36EE world find ourselves in the next "you can't get there from here" backwater.

 

 

More (or less) Rational Announcement

Last Wednesday, we attended a meeting hosted jointly by our local user group, TUG, and by IBM/Rational to review the latest announcements related to Power Systems and Rational. We blogged about the Rational announcement already, so if you didn't see it, you should probably review that first. After getting more details from this meeting, we're revisiting the topic.

First of all, many thanks to TUG and the folks at the IBM Toronto Lab (where the meeting was held) for the event. It was very informative. We're sure many such announcement meetings that have been held around the world recently focused first and foremost on the POWER7 hardware. But we were at the Toronto lab--home of the developers of the Rational tools featured in the announcement. So guess what took center stage for our announcement meeting--development tools, which suited us programmers just fine.

Having said that, we should say kudos to Chuck Wallace from the Rochester Executive Briefing Center for making the (relatively short) hardware announcement portion of the meeting far more interesting than we would have ever thought possible. After all, like many programmers, our response to "How many programmers does it take to change a light bulb?" is "None, it's a hardware problem."  But somehow Chuck managed to hold our attention and we missed out on our planned afternoon nap.

In our previous post on the subject, we covered the renaming and repackaging of the tools that we have only recently learned to call RDi. There were no surprises in the presentations, with one exception. We had mused in our post about what the "short name" for "IBM Rational Developer for IBM Power Systems: RPG and COBOL Development Tools for i" might be. We weren't too surprised to hear that it was most often referred to as RDP (although we think we heard RDPower once or twice).

But what baffled us completely was the way it was written on some of the chart. We've seen the same way other places since the announcement: RDp.  What does the "p" refer to? The thing that springs to our minds is System p. We've never seen anyone be tempted to spell Power Systems with a lower case p.  So what's up with RDp? 

Jon asked about whether he'd have to continue to run Windows on his Mac indefinitely in order to use the Rational tools. Hayden Lindsay (Rational VP) replied that they're willing to look at porting to Linux or Mac if they get enough requests for it. So send in those cards and letters!

During breaks we talked to some of the developers themselves, so we were able to get some of our detailed questions answered. We did ask why the lower case p, but we didn't find anyone who seemed to have an explanation for that. Fortunately, for the other more important questions, there were plenty of answers.

For example, we asked for clarification on the process existing RDi customers could use to upgrade to RDP (note, we're refusing to perpetuate the lower-case p silliness.) In our blog, we had said there would be "very little effort" required to upgrade. Unfortunately, we were a bit optimistic there. Even though the new package is V7.5 and the existing RDi package is also V7.5, it's not just something done via the "normal update process" as we said in our blog. You will need to order, download and install RDP. You can keep your RDi alongside if you want (not sure why you would), but it's a re-install step. Not nearly as "little effort" as we had hoped.

The "no charge" for existing RDi customers was correct, thank goodness. There was some concern expressed about the much higher price that appeared on some of the Web pages for the new package and the impact that would have on SW maintenance fees, even if there was no cost for the upgrade itself. Turns out, the confusion was caused by a mistake on the Web page, which had originally only listed the price for a "floating user" license ($1,510), which is new for RDP. The only type of user license available for RDi is/was an "authorized user" ($860). Even that price is higher than you may remember from the original RDi pricing. Turns out Rational has been increasing most of their software prices slightly each year--and not just the client-based development tools.

Another question we raised in the earlier post was whether the new print-file designer would be a Technology Preview, like the screen designer was for years after it was introduced. Good news: the print-file designer is fully supported in RDP and it's, of course, built on the same base as the screen designer, so it has a very similar look and feel to the screen designer. (Whew!)

One last question we posed to the development crew at the lab was related to the debug enhancement for attaching to "dynamic jobs." So, what, we wondered, is a "dynamic job" and how does this support help them?  It turns out there are some rare situations where using SEPs (Service Entry Points) wouldn't work for starting a debug session. The example given was related to programs started via messaging services, such as those using MQ. So some new support was added so the RDP debugger can connect to programs in those situations. This new support does require some extra code to be put into the programs. If you think you may need this support, better check the details on how to make that work.

So there you have our Rational Announcement, revisited. Not much here to mitigate any of the issues raised in the blog comments. Identity crises just seem to come with the territory for those of us who love i (System i, iSeries, AS/400 ….) Hey, at least it's still RPG. (Oops! Maybe we shouldn’t have said that out loud….)

The Latest Rational 'Announcement'?

New announcements are out today from Rational. And joy of joys--more repackaging and renaming. RDi is dead--long live "IBM Rational Developer for IBM Power Systems: RPG and COBOL Development Tools for i" Sort of rolls right off the tongue, doesn’t it?

Oh yes, and some nice new long-awaited features. More on that in a moment.

Actually the repackaging is nothing like what happened with the big shake up from WDSC to various flavors of RDi and the break up of the host compilers and tools. This time it almost seems a no-op, so far anyway. For existing RDi users, there is no charge--hurray--and very little effort required--to get this new product. RDi users will apparently simply use the normal update process get the new RDPSRCDi.

OK, we doubt anyone will really call it that. But what will we call it? This toolset has been plagued by an identity crisis second only to that of the platform it supports. A significant momentum exists right now for adoption of this modern toolset--and now we all have to pause and re-jig our brains again around what this thing is called. And in every discussion about it, we have to establish that we're talking about the thing that used to be called RDi, which used to be called WDSC.

We could think positively--maybe that's why Rational has given it a new name--so bloggers and other writers will talk about it. They say there's no such thing as bad publicity. Let's hope.

Rational Team Concert for i (RTCi) has also been renamed Rational Team Concert for Power Systems. However, no name change for Rational Business Developer (aka RBD which includes EGL), presumably because it is used on platforms beyond Power Systems. Rational Application Developer (aka RAD) has "for WebSphere" appended to it, but no mention of Power Systems. However, all four products--renamed or not--are listed in the announcement as part of the "Rational Developer for Power family of products." RDi SOA (which contains EGL and Web tools) does not appear to have been renamed, repackaged or included in the "Power family."

With regard to the RDi replacement product so far it sounds like just a new name. It seems that the base product is now "IBM Rational Developer for Power Systems" and there are features that plug into it. Today only one such feature is available--RPG and COBOL Development Tools for i. But the statement of direction in today's announcement calls for two future features for AIX development--C/C++ and COBOL.

So while for today, we can say that Rational Developer for Power Systems is just another name for RDi, when the statements of direction are fulfilled, we'll need to be more specific when we refer to RDPS--RDPower Systems, RDPower, that Rational thing?

OK, enough of the whining over yet another name change. What's in this new product/release/update?

A screen designer! Not a Technology Preview not-to-be-used-for-production toy--a fully supported graphical screen designer. But wait, there's more--a print file designer! Hopefully based on the same tool as the screen designer, unlike (shudder!) RLU. And hopefully not Technology Preview.

There are also long-awaited improvements in the compiler error feedback associated with embedded SQL and improved debug support for attaching to "dynamic jobs." Now we just have to figure out what dynamic jobs are.

If you want to read more about the Rational announcements yourself, here are a few links.

http://www.ibm.com/software/rational/announce/power/--Despite the word "announce" appearing in the URL, the word announce doesn’t exist on this page. It's a marketing page that talks in motherhood-and-apple-pie terms about the renamed product set.

While working through the announcement links we were excited to see a nice logo for an "IBM i can do it e-Kit" http://www-01.ibm.com/software/rational/smb/modernization/. We clicked the link to register for the e-Kit. The first disappointment was that on this page are links to the now defunct RDi and RTCi tools. But we persevered and clicked on the "Register for the e-Kit" link. After signing in, we were still none the wiser what we would get with the e-Kit, but we were intrigued. So we clicked on the "Get the e-Kit" link and were led to a page with links to case studies for EGL and HATS, white papers and podcasts on Rational Team Concert--the "for i" version, not the new "for Power Systems" version and some other links to things that we're pretty sure we could have found elsewhere on Rational's web pages before now. We never did find the "kit" just a series of links to existing pages. Very disappointing.

The actual announcement letter--the only place where we found the "what's different from yesterday" information: http://www-01.ibm.com/common/ssi/ShowDoc.jsp?docURL=/common/ssi/rep_ca/0/897/ENUS210-020/index.html&breadCrum=DET001PT022&url=buttonpressed=DET002PT005&specific_index=DET001PEF502&DET015PGL002=DET001PEF011&submit.x=7&submit.y=8&lang=en_US.

Happy reading!
 

Blog Posts Revisited

This week we thought we'd just follow up and provide some updates on a few of the posts we've made in the past few months.

"Make Yourself Happier - Give" 

Couple of updates here. For ourselves we have been back to the Habitat house a couple of times--including the day the roof that we had worked on was lifted into place. Check out Jon's photostream on Flickr  if you want to see what a flying roof looks like! In addition to continuing to work on this house we're now looking at some of the other opportunities to work on Habitat projects in Canada, the U.S. and elsewhere.

Meanwhile our friend Aaron Bartell has reached his first milestone. If you visit his donations page at http://www.firstgiving.com/aaronbartell you'll find a link to the hilarious video of him attempting (with the "assistance" of the smallest member of the Bartell family) to shave his legs. Even funnier than the video are the off-screen comments from his wife and (we suspect) sister, offering helpful advice on how to shave one's legs.

"Asking the Right Questions" 

This entry resulted in an interesting response from David that those of you who only read the intial post may have missed. He pointed out that Eric Raymond has an excellent discussion on how to ask questions. Although targeted at people asking questions in online forums, it is nonetheless an interesting and thought-provoking read. It certainly pointed out to us why we don't always get the right answers to our own problems!

"Japanese Imports and a New Role for COMMON?"

We were happy to see that Martin Fincham is leading the charge in Europe and look forward to seeing what his group comes up with. It was interesting to note that in one of the comments to his original post, a contributor noted that in Italy alone more than 19,000 of our favorite systems are in use in small businesses of all kinds. A pretty big opportunity and one that is obviously open to an attack by Microsoft-based "solutions".

We are saddened though that COMMON has remained completely mute on the subject. No public statement of support (something that we might also have expected to see from COMMON Europe)--not even a comment. Nothing--and nothing from the North American ISV/BP community either. That's sad.

"These are a Few of Our Favorite (Free) Things!"

Hans Boldt added his own personal choices to our list, and with one exception we agree with him. In fact Jon has added Open Office (OO) to his own personal list of favorites having been using it more and more over the last few months. He has encountered so many compatibility (not to mention reliability) issues with the Mac 2007 version of Office, that he had to either upgrade or find an alternative. So far he's very happy with OO, in part because some of the things that he hates most in Word and Excel are much easier to handle in their OO equivalents. With luck he thinks he can avoid enriching Bill Gates further by avoiding an upgrade.

Comments on Your Comments re: i Community Needs

What great comments some of our readers came up with in response to our iDevelop blog entry "What Does the i Community Need?". In reaction to many of the comments, we'd really have nothing much more to add than "Yeah--what he said!" However, a few specific comments sparked a few further ideas for us and we thought we'd take the opportunity to add our 2 cents worth to your 2 cents worth!

Several comments were made about the urgent need for marketing the strengths of IBM i, because. as Scott pointed out, without customers coming to the platform, it will die. We couldn't agree more with that sentiment. Sadly, we also agree with Nathan that IBM really seems to be firmly entrenched in a "pick a server, any server" kind of marketing direction rather than touting the benefits of IBM i for business. The idea of the IBM i faithful becoming its best marketing force isn't a new one.

Malcolm Haines was possibly the last person in IBM marketing who really understood the platform has sadly left IBM--probably in extreme frustration that no one else could see what seemed so incredibly obvious to him. Malcolm started the all too short-lived "iSeries, My Series" marketing campaign. The
campaign ended abruptly when the system was suddenly renamed System i within weeks of the campaign kickoff. <sigh>

Malcolm also envisioned iSociety, which in the beginning was supposed to provide support and leadership for the army of the iSeries (the proper term at the time) fans. His idea was that we, the faithful, could do a much better job of getting the right words out about the platform than IBM could because IBM was hamstrung by political correctness, corporate politics and company-wide homogenized "marketing messages."

Without Malcolm's on-going guidance, iSociety drifted. Now, controlled by COMMON (with that organization's embrace of all things Power), it's simply another COMMON web site. Its opening page declares "Welcome to COMMON’s Power Networking Center… where the COMMON Community connects!" In other words it has nothing to do with its intended purpose of promoting IBM i to the uninitiated. We think that perhaps we should just start over--perhaps with the YiPs Web site as a better place to put that kind of support.

Speaking of Yips--that leads us to another of your comments--what a great idea about making a system available as a sandbox for young (or even not-so-young) developers to get a feel
for the system (thanks to Chris and Aaron for those comments). This is a problem that we have been complaining to IBM about for a long time now. Even when you manage to pique someone's interest in the platform, where can you send them to A) Learn more and B) Play. By contrast, if someone suggests that you should try using .Net, you simply download the tools from Microsoft and away you go. Want to try Linux? Sure. Just boot from this handy freely downloadable CD of Ubuntu or ... and you are off and running. If you want to do the same thing with IBM i, it is virtually impossible--there are some Rational tools you can play with but you can't really explore the underlying system or do any real programming. IBM's virtual loaner program should have been a step in the right direction, but you must be a software vendor to use it. And why oh why does it require that you use a VPN connection with the associated problems? Even if you are prepared to pay for access there are very few timeshare sites left--and almost none with up-to-date software. Sure you can buy a used system relatively cheaply, but try to bring the software up-to-date and see what the final bill looks like. This is an area IBM must address if it really is serious about the platform.

Since we're talking about tools etc. Jack commented "Having recently returned to a System i developer role, here's what on my wish list: A low cost IDE. Very excited to be working with Eclipse-based IDEs (WDSC v7 and , ZendStudio ...). Not surprised to learn that IBM has once again decided to tinker with developer licensing at a premium price." In previous iDevelop posts, we've written about our problems with the way that IBM changed the pricing model for development tools. We don't think anyone would argue that there was justification for change of some kind, but why penalize your forward-thinking users?

We have customers who have moved most of their developers to WDSC but in the current economic climate are having real trouble justifying the cost of moving to RDi. IBM (or Rational to be more precise) has made matters worse by apparently deciding that 7.5 is a new version and not a simple free upgrade for 7.1 users. That has just added to that pain.

"as400 guy" went beyond the development tools and addressed the application front with the comment: "What the '400 camp needs is customers, and lots of em. But there is only one way to get them, and thats to have the apps that customers want to buy. Why? Because companies do not buy systems, they buy apps." We agree wholeheartedly--and so when we saw the news the other day that IBM's new new small and medium business packaged systems--the Smart Cubes were now to be available in North America we sighed "at last",  but when we located the actual details (not an easy task sadly), they were disappointing. ("Now Open" explains a bit about the Smart Cube concept.) Where are the IBM i based accounting or warehousing solutions? Where are the retail, local authority, medical office, etc. packages that would take advantage of the i's strenghts and offer small businesses a real solution in these difficult times?

One last comment. At least there are some hopeful signs of progress in the academic world. For ourselves we'll be headed to Rochester this summer to join the members of the Academic Initiative at their annual Summer School. We had to miss last year's event due to a prior engagement but are looking forward to meeting with the professors again this summer. In a similar vein, we hope to be working soon with students from Seneca College in Toronto. These students fought for permission to do their class project, over the next two semesters, based on developing an application using RPG, DB2 and IBM i as the foundation. They had to do significant arm-twisting with the professors to be able to do this because the course is on object-oriented applications. Clearly these students understand the power of RPG and i,
despite being bombarded with other languages and platforms. They felt strongly enough about it to fight to do it the "right way".

Beware of IBM Rational Bearing Gifts (or at Least Fix Packs)

As most of you know, we are huge fans of IBM's Remote Systems Explorer (RSE) for our IBM i development work. Neither of us can imagine going back to SEU for our RPG coding needs. But today Jon is feeling a little less love--at least for RDi. We decided to write about his latest experience in the hope that it will save our readers some grief.

The problems began when Jon decided to install the RDi 7.1.0.3 fix pack, which was announced by IBM in early January. He was suffering a couple of the problems identified in the Release Notes and so decided to go ahead with the update. Normally we would wait a little longer, but the only reported problems appeared to relate to RDi SOA, and we hadn't had any problems with updates for some time. Besides, Installation Manager allows us to roll back changes if they cause problems. Big mistake.

After applying the update, all went well for a day or two, but then his workbench started to hang with no way out except to Ctrl-Alt_Del and kill the application. This appeared to happen after his machine had been suspended (or some other communication break occurred) while source members were open. When this behavior continued even after reloading the workbench and rebooting XP, we decided to roll back the change and revert to the previous fix level.

Problem. The Installation Manager insisted that the roll back information was unavailable. Strange--we were certain that we had set the option to keep roll back data. We checked the preferences and sure enough the option to save roll back data was selected, and the dialog showed over 800 MB of saved data! So why couldn't we roll back? Hmmm, time to check the WDSC list at Midrange.com and see if anyone else had experienced these problems. It quickly became obvious that EGL users (RDi SOA) had experienced some difficulties, but nobody was complaining about basic RDi. So we posted a message and asked if anyone else was seeing the same problems. The only reply received (to date) was from the IBM Lab's Eric Simpson who is on the RDi team (thanks Eric, even if we didn't like the answer). Eric didn't have any comment on the bugs per-se but it was the rest of his news that we didn't want to hear!

For reasons that we hope IBM might choose to explain one of these days, RDi apparently does NOT support roll back. So once you apply a fix pack, the only way you have to get rid of it is to completely reinstall the product and then re-apply the known good fixes. Not a good option and a waste of an hour or so that frankly we just didn't have. So it's back to WDSC for the time being until we have time to spare.

All of this raises some questions. RDi is the first version of the IBM i development tools that we are have to directly pay for. In a "for fee" product is it so unreasonable to expect that any rogue fix packs should be able to be rolled back without a reload? Eric noted in his response that RDi had never supported roll back, but we don't recall any warning to that effect during the fix installation. Since we had selected the option to save the roll back data we think it was a reasonable expectation on our part that we could roll back as we had with WDSC. We'll certainly be asking IBM to comment on this and post any responses we get.

In the meantime--for all you RDi users--a warning: Rather than applying fixes when they first become available, it might be a good idea to wait a while and see if others have problems before applying them yourself. Of course if all of us wait ...