Posted by:
Adam Twardoch Thanks for the link to John Nack’s article. The writeup of his colleague Hart Shafer is also a good one:
http://blogs.adobe.com/hartshafer/2006/10/soundbooth_and.htmlOf course FontLab Studio 5 exists in Intel code, because we do have a Windows version, and that’s the version where the core development is being done. In the past years, we have developed our own streamlined process to port the code to PowerPC-based Mac. To switch to universal binary (and from Metroworks to XCode), we would probably have to abandon that process and do it differently.
As with Adobe Soundbooth -- if we were starting FontLab Studio for Mac right now, we would go a different path. Perhaps we would also make it Intel-only, perhaps not. However, in contrast to people who do audio or photo editing, type designers don’t switch their hardware so often. Font editors usually consume modest hardware resources so it would be unwise for us to abandon PowerPC support and only switch to Intel Mac. This means that we need to support PowerPC for quite some time, and this again means that in the next future, we will stick to our current way of building software -- simply because it runs very well under Rosetta on Intel Macs.
Adobe’s Hart Shafer writes:
"A different chip architecture means a different test bed, which in turn means a lot of additional testing time. Every aspect of the application has to be re-verified--all that Intel-specific code we would have re-written for PPC would have inevitably introduced new bugs and testers are rightfully distrustful of the idea of "oh, don't worry, the software will work the same on both chips." Engineers can write all the code in the world, but without a good QA team running through all of it you probably wouldn't want to buy what they produce..."
Just like Adobe (in case of Soundbooth), we think that under the current circumstances, it is wise to produce only one "series" of Mac code. They chose to only maintain the Intel Mac build, we will, for the time being, only maintain the PowerPC build -- and rely on Apple’s Rosetta to do the Intel translation on the fly.
This does not mean that we won’t test on both PowerPC Macs and Intel Macs, but at least we don’t have to develop on both. Because, as Shafer writes, developing universal Mac code is more than just checking the "Make universal" checkbox.
Altogether, I welcome Apple’s move from PowerPC to Intel, and I think in the long run, it will help both the hardware industry and the software makers. But right now, Apple made it easy for themselves by turning on a publicity machine aimed at end-users that suggested "It’s so easy for software vendors to make universal apps, you just tick the checkbox". That’s nice and I kind of understand why they did it, but it ended up in both Apple and the end-users putting a lot of pressure on software vendors. To some extent, I think it was a little unfair from Apple towards software vendors. The posts on the Adobe blogs kind of confirm that -- it seems that even a large company such as Adobe needs to "defend themselves" now. It is similar with Fontlab Ltd.
I know that it wasn’t your intention, LLtype, to have this kind of effect, and I’m glad that you understand our points. The really good news from our side is that, really, we can say that FontLab software seems to work really well on Intel Macs even though it’s not "universal".
Best,
Adam
