Applescript Controversy

So most of you have heard the news that Sal Saghoian has been let go at Apple. The best roundup of the news is at Michael Tsai’s blog. It’s worth noting that not only was he fired but his entire team was disbanded. At first, like many, I was extremely concerned. As I’ve thought more about it I’ve arrived at a more nuanced view.

First off despite the hue and cry automation isn’t going anywhere. Siri pretty well presupposes automation. Almost certainly these events are due to conflict between the Applescript team and the Siri and perhaps even Swift teams. My guess is that most of the responsibilities get rolled into those teams.

The second point is more important. For all of you who remember Applescript during its heyday in the 90’s, what’s the most significant AppleScript related technology of the last 15 years? I can only think of Automator. Yet here’s the thing. I don’t know anyone who uses Automator. It’s too limited for pros. It’s still too confusing for novices. More importantly after the first year it largely got ignored. New apps rarely make Automator Actions and most users I talk to don’t even know it exists. So the biggest success was still a failure.

Now look at everything done since. We had the abysmal Scripting Bridge that came out when the far superior Appscript was available. Further trying to use Scripting Bridge with Swift is an exercise in frustration. We had JavaScript for Automation (JXA) which again had huge limits and never was a success. (It was almost impossible to do GUI scripting with JXA – at least I could never make it work) Again quickly forgotten. There were a few other minor releases but the one thing they all had in common was they didn’t work well and quickly were forgotten.

During this time there certainly were things Apple could have released. Applescript debuggers (available from third parties), help doing GUI Scripting for those apps (most of them) not supporting Applescript, better Swift tools for scripting, making it easier to add scripting to applications, etc. 

By and large automation the past 15 years has been dismal. The interesting technologies outside of Apple were ignored in preference to proprietary solutions that worked far worse. So anyone saying this is bad needs to grapple with what has been good.

Now what this all means isn’t yet clear. The needs and focus of the Siri team certainly aren’t the same as the needs and focus of pro users. However at least the Siri team is producing products that do useful things. There are of course big problems with Siri. Arguably at least as big of problems as the Applescript language itself has. But at least there in the Siri team we have excitement, innovation and more importantly VP focus going on.

Edit: Jason Snell, who I respect a lot, took me to task for comments like the above. Now Twitter isn’t a place where it’s really possible to have discussions. It’s always frustrating given its constraints. He makes some good comments though. His point is that Sal didn’t like what his team released anymore than anyone else but he wasn’t given the resources by Apple to do a good job.

While there’s something to that, one might say that Apple then should have removed his team earlier. However this doesn’t explain in the least his team writing Scripting Bridge with all its numerous problems rather than just adopting the far superior AppScript. Likewise it doesn’t explain why they released Javascript that has the same problems Scripting Bridge had plus far more. That is, I just don’t buy the idea that all the problems were Apple. While the comments are no longer available, at my old blog Hamish Sanderson noted the problems and how he’d submitted them to Apple.

Edit2: Jason Snell put up a good response to this post over at Six Colors. I should also note an important tweet he put up: 

It’s tough to see someone you know being blamed for things you know he was actively opposed to

I’ve tried to keep separate complaints from the person myself, but perhaps I could have done a better job. From everything I can see Sal’s heart was undeniably in the right place and he was quite beloved in the community. Likewise in a fundamental way the blame rests with Apple and senior Apple VPs. They just haven’t given enough support to automation. That said I don’t think that excuses everything that was released. 

By all measures Sal kept automation alive when few at Apple cared. I’d certainly give Sal the benefit of the doubt that he wanted very different products and was primarily limited by resources. In particular it’s been surprising how many Apple products do have very good Applescript support. When iWork was rewritten by Apple, I was quite upset that it initially didn’t really support scripting. While I have no idea how much of this was due to Sal and how much due to complaints by users, the fact is scripting was added back to Pages and Numbers. I use scripts for them regularly.

Leave a Reply

Your email address will not be published. Required fields are marked *