Orate

Who builds Orate, and where the numbers come from

Every article on this blog carries a byline, and this is the page behind it. It is here because the useful thing to know about a comparison is who measured it and on what — not that a company exists.

About Gajendra Singh Rathore

Two native apps, not a wrapper

Orate is two separate codebases built to one product: Swift 6 and SwiftUI on iOS, Kotlin and Compose on Android. There is no shared module and no cross-platform framework between them. That is a more expensive way to build an app and it is the reason the camera behaviour on each platform is what that platform actually does rather than what a bridge exposes — which is most of what these articles are about.

Both apps ship the same fifteen languages, translated in full: English, Hindi, Bengali, Marathi, Tamil, Telugu, Spanish, Portuguese (Brazil), Indonesian, Thai, Arabic, Hebrew, Czech, Danish and German. They run on iOS 18.0 or later and Android 8.0 (API 26) or later. The app is published by Planify.

Where the numbers in these articles come from

The figures quoted across the blog are first-party and were measured rather than collected, and this is the detail that makes them checkable:

  • Crash recovery was measured on a OnePlus KB2001 running Android API 34, killed with am force-stop — a SIGKILL, so no lifecycle callback runs and nothing gets a chance to tidy up. A roughly ten-second take left 20.9 MB on disk and came back on relaunch as a playable 8.0-second take. What is lost is the fragment in flight, and nothing more.
  • Fragment interval is two seconds in Orate. Apple’s AVCaptureMovieFileOutput defaults to ten; CameraX’s default muxer writes its index only on stop, which is why an unmodified Android recorder loses the whole take rather than the last few seconds.
  • Reading pace is set per writing system, not per language: 150 words a minute for Latin scripts, 130 for Devanagari, Bengali and Arabic, 115 for Tamil and Telugu, and 250 characters a minute for CJK and Thai. The unit changes with the script, which is the whole argument against a single words-per-minute constant.

What is not claimed here

The comparison pages are written by someone who makes one of the products being compared, which is stated at the top of each of them rather than buried. Three rules follow from that, and they are the reason those pages are worth reading at all:

  • Competitor figures are dated snapshots, and unverified cells stay blank. Where a claim about another app could not be checked, the comparison says so instead of filling the gap with something plausible.
  • There is no invented social proof. The structured data on this site carries no aggregateRating, because there are no ratings yet, and a made-up one is how a site stops being cited.
  • No prices are quoted on this site. Every figure on the paywall is computed from what the App Store or Google Play returns, so a number typed here would be a fourth place for a price to go stale.

Corrections

If a measurement here is wrong, it is worth more to me to know than to be right — write to contact@planifyapps.com and say which figure and what you got instead. Articles carry the date they were published, and a corrected one gets its date changed rather than edited quietly.

Elsewhere

The Android app is on Google Play. The iPhone app is built and not yet published; it will be linked here when it is. Orate posts on Instagram, YouTube and X. There is no newsletter and no Discord. Support goes to the support page, which is the same address read by the same person.

Gajendra is gsrathoreniks on GitHub and on X.