Everything Else

Between Observations of Radio Silence

4 May 2009 — As human beings, exploration is part of our nature. Our ability to adapt to new surroundings is what has helped us survive. We’ve spent most of our existence exploring our planet, yet only in the most recent span of time have we begun to expand our search beyond its confines.

Where shall we go? We’ve set foot on the Moon, and some of us may live long enough to see one of us set foot on Mars. But we already know we can only visit these places; we’ve found no suitable place to live amongst the stars. If we ever do find such a place, what kind of place will it be? And how will we have to adapt to life there?

What little we know about space flight is informed by our theories of relativity, tempered by our present technological limitations. We know that anyone taking a really long space flight — if he or she is lucky enough ever to return — will find that everyone they ever knew would be long gone.

Thus a long space flight is probably not a trip that ends at home, but at a new home.

The process of getting Helix from where it was in 1996 to where it is today bears some striking similarities to long space flight, remembering again how little we really know about it. We have expanded our orbit somewhat, but the orbit has still been focused on home, which, for Helix, is a place called “Classic.”

This home — Classic Helix — has served us well. We have maintained it as a place to see what Helix is.

Soon, however, it will be where Helix was.

Constant attention to our plan has brought us this important day

One thing cannot be emphasized strongly enough: No plan can ever succeed without the ability to adjust to changing conditions.

Since undertaking the conversion of Helix from Classic to macOS in the waning hours of 2002, we’ve had to adjust quite a bit. We set out under the assumption that the answer to each of these three central questions was yes:

  1. Is Helix worth preserving?
  2. Are there enough people left using Helix to make it worthwhile?
  3. Can we make it good enough again to attract new users so Helix does not die as the current users retire and development is no longer sustainable?

While there were a lot of obstacles in our path, our plan encountered four unforeseeable twists in the road. Each one forced us to reevaluate our direction. Would it still be possible to achieve our objectives when…

  1. … the new G4 & G5 Macs were no longer able to boot in Classic?
  2. … the new Intel Macs were no longer able to run Classic at all?
  3. … macOS 10.5 was released and Classic was gone altogether?
  4. … Classic-capable machines became scarce and the ability to keep them running became a losing battle?

When Apple introduced Intel Macs, we were forced to confront a number of issues: switching from our familiar development environment (CodeWarrior) to the rather unproven (and somewhat buggy) Xcode, making Helix Universal while maintaining a Classic Client for customers needing to keep at least one old machine on their network because some hardware connected to it had no macOS equivalent, and many other hurdles both large and small.

Each was a project, without which, Helix simply could not remain viable. We faced a tremendous amount of work for us just to catch up to the current state-of-the-Mac.

Helix 6.1.3 Release Announcement

And through all of this, for more than a decade now, Helix has had to “observe radio silence,” a military term for avoiding the spotlight, at all costs. But now, today, at last, we are lifting radio silence, if only briefly.

Clearly, we’ve been very busy, and not very communicative of late. Nearly four months have passed since our last update, so there’s a lot to talk about. We sincerely hope you’ll return here to read the rest of this edition of The Latest Word after reading this next sentence: Helix 6.1.3 is available now.

In the weeks following our failure to deliver these tools as intended on New Years’ Eve, we were treated to many terrifying glimpses into the abyss. What we thought would take a matter of days to resolve stretched again into very emotionally charged weeks and months.

All of which makes it the more joyous for us to be able to state the following “small print Helix boilerplate” text: “If you already own a Helix 6.1 license, this is a free upgrade. If your licenses are for Helix 6.0.x, 6.1.3 products are available either as a paid upgrade or as a free PowerPC-only version, to fulfill the promise we made to you. If you are using Helix 5.3 or earlier, Helix 6.1.3 is a paid upgrade.”

Helix Server performance on Intel machines is getting rave reviews from users. Even those who still have predominantly older machines and continue to run Classic Clients report that some functions perform significantly faster when connected to an Intel Mac running the Universal Helix Server.

“Those who fail to plan, plan to fail,” it is often said. Thanks both to our adherence to and frequent reevaluation of our plans, there are now six Helix products, Helix Server, Helix Client, Helix Engine, Helix Utility, Update Collection and Helix Maintenance Manager, all running in macOS. The User Mode side of RADE is already in macOS; you can see it today in the macOS native Helix Engine.

All that remains untouched by macOS is RADE Design Mode, the essence of the icon-driven development interface that is the jewel of the Helix crown.

Universal Helix is here at last

Our last few releases have divided Helix into three distinct ‘flavors’: Classic, macOS for PowerPC, and macOS for Intel. That changes today. We are now offering “Universal” Helix products, dropping the “for Intel” version we shipped in Helix 6.1.1 and 6.1.2. This means you can now install one version of Helix Client, Server, or Engine on all of your macOS machines. For the foreseeable future, Helix Client Classic will remain available for those Macs that must run OS 9, as well as to access functions as yet not implemented in the macOS Preview Release products.

And so, today we are announcing that the PowerPC-only version — along with other older versions of Helix — have been taken off the price list and out of the web store and are no longer available for purchase. We are entering a new era, putting our full confidence in Helix 6.1.

But there’s a wrinkle: those who bought Helix Server 6.0 were promised a free update that would support a macOS Client. In order to keep that promise, we will — as long as it remains practical — continue to produce a distinct “for PowerPC” only version as a separate product. So if you are a Helix 6.0 user and you don’t want to upgrade, we won’t force the issue. Just be sure download the ”for PowerPC” version of the installers and use your Helix 6.0 keys to enable them. (Yes, you can use 6.0 keys to enable the PowerPC-only 6.1 products, and that is a bit confusing, so the installation dialog includes a paragraph that states whether it accepts 6.0 or 6.1 keys.) Helix Client (and Engine) for PowerPC can run on your Intel Macs using Apple’s Rosetta translators; you just won’t be able to take advantage of the speed and stability benefits that come with Intel-native Helix code.

We will continue to support older Helix products, but that support will be lukewarm at best. So many things have been fixed and improved between 6.1.2 and 6.1.3 alone that it is very likely that problems you’re experiencing in older versions can be eliminated by upgrading to 6.1.3. That’s a bold statement, but for most of you this is true without qualification.

Hundreds of issues have been identified, catalogued and processed since we released Helix 6.1.2, some of them reaching back to the Helix Express days. The vast majority of issues noted in the 6.1.2 Preview Release have been resolved. The only features still incomplete are Power Query and advanced Document Management. Basic Document Management is now functional in Helix 6.1.3. We could have pressed on and finished those features now, but the improvements over Helix 6.1.2 are so vast, and the outcry from our user base has been so clear that we chose to draw the line here and put this much-improved version into your hands today.

Do not interpret any of this to mean that Power Query and Document Management are in any way insignificant; they are quite the opposite. But according to the survey we took when we started this journey, they are also among the least understood and thus are among the least used features of Helix. This is due in almost equal parts to three factors:

  1. People are slow to integrate new functionality into their applications.
  2. The usage of these functions was never communicated clearly enough to enable the average user to take advantage of them.
  3. The way they functioned was not intuitive or, in some cases, were outright buggy.

We’ve held these back until the next release because we wanted to take the time do them right. As a result, Power Query will be much more intuitive and natural than before. Document Management will support drag and drop. Things will work the way a Mac user expects them to work.

Document Management — the use of the binary large object data type (aka: BLOB) — was once something that put Helix years ahead of its field. Today, it’s a trick that many other applications perform, some better, some worse. We plan to revisit Helix’s Document Management tools and assure that they become world class once again. macOS RADE will give power Document Management users the tools to take advantage of the improvements.

When Power Query and the remaining functional pieces of Document Management are ready, the Preview Release phase for our macOS Client and Engine products will be over and we will turn our attention to the final phase: RADE Design Mode.

A few ways we can help you help us

As always, we need your support: upgrading gives us the funds we need to continue on to the next phase.

If you need additional reasons to make the jump to Helix 6.1.3, consider these:

  1. Your computers. That wonderful Titanium PowerBook G4 you’ve lugged around the world three times is going to die some day. Maybe soon. That office full of colorful iMacs? They are getting old and can’t visit some of the websites your users need access to. G3, G4 and G5 towers in blue, gray and metal are dying, heading for landfills or recycling centers. And when that happens and you go shopping for a new machine, all you will find are Intel Macs. If you step up now, begin the transition by upgrading to Helix 6.1.3, you won’t face a crisis when that new Mac lands on your desk (or in your lap).
  2. In light of the hardware issues, we have worked out a deal so that we can sell you both your Helix upgrade and a new Intel Mac in a single package.* On top of that, you’ll save 10% over purchasing both separately. At your request we will pre-install and configure Helix for you before shipping it. We will even come to your location and help you set it up, if you wish.** Click here for additional information.
  3. We’re making it easier for you to pay for Helix Server. The price of a Helix Server upgrade can be daunting, especially during these economically difficult times. So if you need to break your purchase up into two or more payments over time, give Gil a call — he’ll work it out with you.***

* This offer valid in US only. QSA ToolWorks reserves the right to limit the number of machines purchased at a discount.

* This offer is not available through our web store. You must call QSA ToolWorks for details.

** On site installation by QSA ToolWorks staff or authorized representatives. Additional fees apply.

*** This offer does not apply to single-user products.

We’ve also made it possible for you to purchase — at a substantial discount — macOS 10.5 (Leopard) and the new Mac Box Set (macOS 10.5, iWork ’09, & iLife ’09) from our web store. Log in and click the Purchase New link for pricing.

Transitioning to macOS takes understanding, patience and work

When we began the transition to macOS, we warned that it could be a bumpy ride. Helix users are used to upgrades that amount to little more than dropping your collection onto Update Collection, running through Helix Utility and going back to work. All along, we’ve said that this would not be a simple upgrade.

When we released Helix 6.1.2 last July, the Preview Release pages contained a list of the changes that you could expect, along with tips for reworking your collections for macOS.

The past nine months have been spent teaching baby Helix User Mode how to walk and talk like a citizen of macOS should. In trying to accurately gauge our progress, we sometimes found ourselves frustrated by “reference frame fatigue.”

Classic Helix is our baseline — our natural frame of reference — but it is the past, not the present or the future. Helix now runs in macOS. Ordinary Helix tools like form queries, quick queries, check boxes, radio buttons, popups and combo boxes now all update the look and feel of Helix collections in a consistent way.

As each part of Helix became functional in macOS, two apples-to-apples questions always had to be asked. Open the same exact collection on the same exact machine and do the same exact thing. Compare this in Classic and macOS, then ask: Is the result the same? Is the performance faster or slower?

Perceptions of progress are generated within a frame of reference, in this case, doing the same things on the same machines with the Classic and macOS products.

It is a given that Classic Helix produces the ‘right’ answers, so making sure the result is the same in macOS has been our sole focus so far. If the result is wrong, how fast (or slow) you get there just doesn’t matter.

If the result is correct and performance is better in macOS, that’s a bonus. But when we reached the point where most functions were generating the correct results, our focus shifted more to how Helix functioned and performed in macOS.

We wanted to be able to say categorically that a particular aspect of Helix was this much faster or slower than in Classic Helix, but the comparison was problematic. On an Intel Mac, there is no Classic, so the apples-to-apples comparison mentioned above can’t be done. Comparing Helix on a G5 in Classic and macOS is one thing, but we want our macOS tests run on an Intel Mac, and Classic cannot run there. The assumption is that everything should be faster on an Intel Mac, but that assumption is not quite valid or fair because direct comparison is no longer possible.

Mythbusters to the rescue

In dealing with this conundrum these past several months, we repeatedly bumped into a misperception: that macOS is intrinsically faster than OS 9. But as anybody who has worked with any large application (such as Photoshop) can tell you, it simply isn’t true.

The primary benefit of macOS is not speed, it is stability. Remember how often you had to restart your whole computer in OS 9 and earlier? All it took was one misbehaving program to bring the whole system crashing down around a “Type 2” error.

When is the last time your macOS Mac required you to completely restart because you had lost control of the computer? No, the promise of macOS was not speed, it was stability. It wasn’t until the advent of Intel Macs (during the macOS 10.4 era) that it was possible to start talking about speed gains again.

Some Helix functions run much faster (and better) in macOS than they ever have before. Sadly, however, there are things in Helix that function more slowly in macOS than in Classic. Occasionally the difference is remarkable. For now we simply have to note these issues as ‘Known Problems’ and move on. When primary development is complete and we’re ready to focus on performance, we’ll address them with a fresh perspective.

[What we’re trying to say here, perhaps not so successfully, is that while you may be disappointed to find some things slower in macOS, rest assured that we are disappointed too, but that it will get better.]

Based on feedback from that release and from working with some businesses that have stepped up to the role of beta testers, we have worked not only to restore functions that were missing in 6.1.2, but also to try to lessen the number of places where changes are required. Some still exist, but they are much less onerous than they were when our first Preview Releases appeared.

The release of Helix 6.1.3 will make it much easier to switch from Classic to macOS for good, whether you’re already in version 6.1 or still using an older version.

While 6.1.3 is strong, solid, stable and dependable, please bear in mind that the macOS Engine and Client are still classified as Preview Releases. There are still two significant things (Power Query and advanced Document Management) that remain to be completed before we are willing call these products finished. And there are still areas where they may crash, but those should be significantly fewer than in the previous Preview Releases.

If you rely on one of these features, you don’t have to continue to sit back and wait. Just use that older Mac to drop back to Classic Helix temporarily, make the change you need, then take it back to macOS and carry on. So often we tell people how easy this is to do, yet it still seems to confuse people. This short video illustrates the technique by showing how you can set a Power Query macOS.

The critical variable is your collection

Factoring in the collection itself practically invalidates all assumptions. No two Helix applications are alike.

That said, there is still a simple rule of thumb that you can apply: the faster the machine, the better the performance. Especially in Client/Server situations. Intel Macs are faster than the PowerPC Macs they are replacing, and Helix runs significantly faster on Intel Macs.

Without having hard numbers to give you, what’s faster? Printing — and it is laser sharp once again! Entry views, even very complex ones. Popup menus! Even dynamic popup menus with thousands of items in them. (Yes! Popups are no longer limited to just 253 items as they are in Classic.)

And what’s slower? Long lists. Forms with nested subforms.

But you may still encounter places where views don’t look the way you expect, or performance is less than desired. If this happens to you, please keep in mind:

  1. Some design change is inevitable. macOS is just different than OS 9 and its predecessors. If you designed your collection back when all we had was the 9" black and white screen of the Mac Plus/SE era, and you have a lot of 9 point type in rectangles butted up tight against each other, it’s time to spread out and take advantage of the screen real estate that everybody has now.
  2. We have not yet devoted any resources to improving performance. We have a lot of ideas about how performance can be enhanced, but we still have functionality to complete, so we’re not going there just yet.

Most of the performance issues we’ve observed are found in particular designs. You may choose to ‘ride out the storm’ and put up with sluggish performance in one or two places for a while, but Helix gives you so many ways to arrive at the same destination, that the best long-term solution might be to get into your collections and rethink the places where your design doesn’t perform so well.

For example, instead of requiring your users to open a 50 page list and scroll through it to find the information they need, consider making a searchable subform list instead. That will not only make Helix run faster, it will make your users more productive than ever. Much has changed since most of your collections were first created. You owe it to yourself and your investment in Helix to take advantage of how well Helix works with new technologies.

But really, none of this should deter you from getting into 6.1.3, because if you still have a Classic-capable Mac available, the functionality you need is there today via Helix Client Classic & Helix RADE Classic. As the video on the right illustrates, all the functionality is still there.

All of us can stand to learn a thing or two, from each other

Still trying to decide what to do? You’ve read that there may be some significant work involved in making the transition. If you’re like most Helix users, once your collection was doing what you needed it to do, you started using it, leaving visits to Design Mode for the occasional tweaks.

If you’re like some Helix users, you may not have even designed your collections. And you may no longer know where to find the person who did. Or worse, you may know where they are but no longer speak to them!

And if you’re like a rapidly growing cross-section of the Helix user base, sometime between now and the time that RADE appears in macOS, you will lose your ability to get into Design Mode at all because your Classic-capable Macs are dropping like flies.

Fear not! There are still plenty of Classic-capable machines in the Helix universe. And many of them are in very good shape and in the possession of people who know a great deal about how to use Helix. People who use them professionally.

There has long been a dichotomy in the Helix universe: do it yourself or hire a professional. These two ends of the user spectrum have almost always been practically at odds with each other.

Users reel at the idea of having to pay someone to do something they could do themselves. This is especially true among users who have written their own systems for business.

Developers reel at the lost productivity they see when called in to work on a collection designed by somebody whose true talents are in running their business, not in writing software. Sometimes they will scoff at the work done by these ‘amateurs.’

Is there no détente possible? Is there no way for both to work together? Could there possibly be a better time than now for them to meet somewhere in the middle and elevate the state of the art, while baby RADE for macOS is still gestating?

One of the things we’re going to attempt to do, once we finish Power Query and advanced Document Management and finally have a way to deploy all Helix applications in macOS, is to ramp up the Helix Professional Developer program. It has been running in a very informal way for the past several years, but now we believe we need good developers more than ever.

Without giving away any specifics, let’s simply say this: between now and the time we ship RADE for macOS, we are going to make sure none of you will be unable to make structural modifications to your collections. From the simplest problem to the most apparently complex. If you built it yourself but can no longer get inside, we’ll help you. If someone else built it and they’ve shuffled off this mortal coil, we’ll help you. If the person who created it can no longer get inside, we’ll help him or her. If it is just that you do not have access to a Classic-capable machine anymore, we’ll even help you there.

“No Helix Collection Gets Left Behind.”

In the end, there’s this…

Why are you still using Helix? Do you hope there will still be a Helix in the future? Would you like to run RADE in macOS? Maybe want to teach it to your grandchildren?

Here’s why we’re still using Helix: We’re tool makers who enable people to control how their data becomes information they can use.

The process has been long and continues to get more interesting. The only thing most of this trip has not yet been is fun. But now the interesting parts are also going to start to become fun, because soon comes RADE Design Mode, the re-entry phase of our flight.

We’re not re-entering where we started. We’re going to our new home: macOS. When the Client and Engine Preview Release phase ends, we will once again be under radio silence, which also happens to be the term for what astronauts experience for a brief time during re-entry. For us, it may not be so brief, but rest assured that when it begins, we will be hard at work moving RADE Design Mode to macOS. We can’t say for sure when the first RADE Preview version will appear, but once it does, subsequent updates should begin to appear quickly, bringing us safely home in an exciting new world.

Find PreviousFind Next