In This Edition:
Elegance, Simplicity, Complexity and Reality
31 December 2008 — Three years ago this month, we put our first OS X Helix product in your hands: Helix Server for OS X 6.0. While it addressed the very pressing need to allow Helix Client/Server users to replace aging Servers with new Macs that can not boot in Classic Mac OS, two facts about it left some users extremely confused:
Those two issues were easily explained (to those who contacted us to inquire) but beyond that, like any new Helix release, it left most users anxious for more.
Immediately following that release, we set about putting an OS X face on Helix, in the forms of Helix Engine and Helix Client, for PowerPC. But within a month Apple announced the transition to Intel Macs and we had to retool to avoid being behind the times yet again. Another year of delay brought us to a year ago this very night, when we put the first “edible” fruits of that labor on the table and introduced you to the Helix “Preview Release” concept.
Elegant simplicity masks recalcitrant complexity
We get a lot of feedback, running the gamut from encouragement to near damnation, but regardless of the tone of the correspondence, one thing they all seem to have in common is high praise for the elegance and apparent simplicity of Helix.
Many users appear to assume that those characteristics also extend to Helix’s underlying structure. Sadly, the “dark secret” is that the other side of the Helix looking glass is anything but simple; making Helix is a heck of a lot harder than using Helix. All that elegance and simplicity you experience as an end user masks an almost mind-numbing degree of complexity underneath. (For instance, no other program allows you to create complex queries and guarantees that you can’t make a syntax error.) But like glass, the code that does this ‘magic’ is fragile and tends to break when modified, particularly with the type of large-scale remodeling we’ve been doing to bring Helix to OS X and Intel. Further complicating the process is that what breaks doesn’t necessarily have anything directly to do with what’s been added or changed. It doesn’t break in the same place that it’s stressed.
In the process of moving from Helix 6.0 to a solid release of Helix 6.1, this has happened to countless functions we take for granted. And many of these, once reassembled, fall apart again when new progress is made in other parts of the program. To describe this process as frustrating barely scratches the surface of how difficult it is. But for those of you out there who are wondering why this takes so long, that’s the principal reason. Very few steps forward come without at least one step backward.
The excitement that lies ahead will have to wait just a tiny bit longer
When we set out on this road to OS X Helix, the only things we knew for sure was that if it was done properly, it was going to be a very long and difficult process. We made a commitment to do it that way, doing the difficult and often excruciatingly dull work first, saving the more exciting work for last.
That’s where we stand right now. Aiming to ring out 2008 by releasing the fourth (and hopefully last) Preview Release of Helix Engine and Helix Client, we found ourselves tonight just a few days short of realizing that goal. While disappointing in light of the tremendous progress we madeincluding solving hundreds of problems in the last six monthswe are close enough that we can go to bed tonight with a feeling of accomplishment at what we were able to do in 2008 on a shoestring budget. But we can’t turn the page on the year without a couple of last minute announcements for the Helix community.
The first is to our intrepid beta testing group: we had planned tonight to give you what would have been the final candidate 6.1.3 betas of Engine and Client for OS X. But we didn’t really expect you to abandon your New Year’s Eve plans to grind out another test cycle — too many marriages might suffer — so releasing it tonight was more symbolic than anything. But that became a moot point as we discovered a couple of rather obvious bugs that force us to push back the release until Friday. But if you’re in the beta program, don’t make any serious plans for the first weekend of 2009 — other than planning to test Helix!
And whether you are you are a beta tester or not: as you raises your glasses tonight to toast the year to come, please add a wish that when these guys do get down to work, they don’t find anything significant enough to keep us from moving putting out the next Preview Releases of Helix Engine and Client, which will be very close to finished products.
One way or another, 2009 will be the year of RADE
The second announcement is for the weary among you, for those of you have kept the flame burning through this long, dark night. The news for you is that we although you haven’t seen anything for six months now, we have made serious progress, and the Preview Release process for Helix Engine and Client will soon be drawing to a close. In the months to come, that process will at last be applied to the final piece of the Helix puzzle: RADE.
Helix 6.1.3 will be here soon. The Classic components — RADE and the Classic Client — are ready to go right now, as is Helix Server. But the OS X Client and Engine need another week or two before we are comfortable allowing them out into public view. As soon as we are confident in them, we’ll release the whole package and you will have OS X Helix products that print beautifully, perform document management tasks correctly and provide a dramatically improved overall experience for those who are running the earlier Preview Releases of Helix Client and Engine in OS X. The only “missing feature” left is Power Queries — they are the last piece of the OS X User Mode interface to be done before we are (finally!) able to focus all of our attention on the OS X-native Helix RADE.
So please hang on just a little longer. The really fun stuff is coming soon. Happy New Year and thanks again for all your support and dedication.