Congratulations to Marco Arment on shipping the new Overcast. The new version is a complete rewrite and redesign that runs faster and uses Apple’s latest technologies.
By far, my favorite applications are the ones developed by solo or small teams that are obsessive about design and performance. Marco fits that mold with Overcast. Looking through the new app, you can see that he was intentional about every button and screen.
Last week’s CrowdStrike outage is pretty concerning. It’s shocking how a mistake by one software company can shut down so many global industries.
While I understand that antivirus software is necessary, its existence always represents a vulnerability. It has to embed itself deep in the system to do its work. Any mistake in that software, whether the result of negligence or malice, can cause the exact damage the software is engineered to avoid. Fortunately, Apple uses an Endpoint Security framework and therefore avoided the CrowdStrike fiasco.
I’m a little late with this one, but I wanted to congratulate my friend Brett Terpstra on the release of Conductor. This new command line utility is an add-on for Brett’s successful Marked app. As the name implies, this utility acts as a train conductor for Marked 2, allowing you to set up and run custom processors to make Marked 2 even more powerful and automation-friendly.
Brett goes deeper into Conductor in this post. He even added some functionality a few days ago.
MacWhisper has been updated to version 8 with some new features, including a video player. Multiple apps use the Whisper model to perform transcription. I bought a license for MacWhisper early, and I’ve been using it a lot ever since.
One example: We use a Notion database to manage all the MacSparky content (this blog, the MacSparky Labs and Field Guides, etc.). With the addition of Notion AI, we’ve found value in keeping text transcripts of released content in the database. This allows us to ask questions like, “When is the last time I covered MacWhisper?”
MacWhisper 8 adds new features:
A new inline video player has been added that allows transcribing video files. The video player can be popped out into its own window. Subtitles display directly on the video, and translations appear as separate subtitles, too. This will make the above Notion workflow even easier
You can now choose different Whisper engines like WhisperKit for your transcriptions. WhisperKit offers distilled models for faster transcription speed, and transcriptions stream in real-time. WhisperKit can be enabled in Settings → Advanced.
There are a bunch of other improvements keeping MacWhisper at the top of my list for transcribing audio on my Mac.
I will be curious to see if Apple incorporates the Whisper technology into the Mac operating system at WWDC. It seems like it should be built into the operating system. Moreover, if they incorporated it onto the chip, it could really scream. But it’s too early to tell exactly what Apple’s vision is for incorporating AI into macOS, and this may be a bridge too far. In the meantime, I’m very happy to have MacWhisper around.
The folks at Bare Bones Software just announced a big upgrade to their ubiquitous text editor, BBEdit. It’s got too many features to mention, but here are a few that jumped at me:
You can read more about the new features and get upgrade pricing if you have an older version of the app.
I’ve seen BBEdit grow for the last few decades and it’s nice to see it’s still going strong in 2024.
One of the bigger questions around the looming release of the Vision Pro is software. Specifically, will there be any, and will it be any good? We don’t know yet. We’ve seen some offerings from Apple and some limited offerings from third parties, but now that we have a shipping date on the hardware, announcements are starting to roll out. The Omni Group announced OmniPlan will be on the new hardware. I suspect they’ll be announcing more. This is where the companies that adopted SwiftUI will get their payoff.
But it’s too early to tell whether a healthy software stack will be available to us on Day One. The device could be focused on enterprise-style software, given its cost. But I could equally see a lot of the better developers getting apps on it despite its small initial adoption so that they can have their flag planted. We’ll see.
MacWhisper, the app that uses an AI Large Language Model to create transcripts, just got its third major update. Improvements include:
I can’t help but feel that we are at an inflection point with voice to text and it is getting a lot easier to use and implement. MacWhisper is part of this wave.