Ninety seconds, maybe two minutes, repeated daily. It's not a lot on any single day, which is exactly why nobody fixes it — the cost is invisible per instance and only obvious in aggregate.
The more interesting cost is attention rather than time. You're making a dozen small spatial decisions before you've started thinking about woA diferença é o gatilho. Uma regra é acionada sozinha quando uma janela é aberta; um script é acionado quando você o executa. Na prática, essa lacuna é menor do que parece - vincule o script a uma tecla de atalho ou botão de hardware e seu layout será pressionado com um toque, o que para a maioria das pessoas é funcionalmente equivalente a automático. expresseA verdadeira desvantagem é a fragilidade. O script depende da inicialização dos aplicativos dentro dos atrasos permitidos e é interrompido silenciosamente quando um deles está lento. Construir uma versão que pesquise a janela mais frontal, em vez de dormir por um período fixo, resolve a maior parte do problema, ao custo de uma noite.s this frDuas notas para quem segue qualquer caminho. O Pro lê suas preferências em com.knollsoft.Hookshot em vez de com.knollsoft.Rectangle, portanto, os comandos de terminal copiados dos guias precisam trocar o domínio. E qualquer que seja o caminho que você escolher, primeiro estabilize o arranjo manualmente - automatizar um layout que você ainda não definiu significa codificar uma decisão que você ainda está tomando.r screen aSe a atualização faz sentidois and whyOs layouts por aplicativo são a justificativa mais clara do Pro, mais do que fixar ou tamanhos personalizados, porque o que eles substituem é um ritual diário, e não uma tarefa ocasional.llowing aO teste é repetição e estabilidade juntas. Se você criar o mesmo arranjo todas as manhãs a partir dos mesmos aplicativos no mesmo hardware, o recurso elimina um custo recorrente e é reembolsado rapidamente. Se sua configuração variar de acordo com a tarefa ou se seu hardware mudar ao longo da semana, as regras irão lutar contra você, e um atalho de layout deliberado - Pro ou um script - funciona melhor do que o posicionamento automático.tomaticalE há um argumento honesto: muitas pessoas que pensam que precisam de layouts automáticos nunca tentaram vincular um único atalho que encaixe sua janela atual no lugar a que ela pertence. Um gerenciador de janelas dedicado para Mac resolve uma quantidade surpreendente antes que qualquer automação entre em cena, e vale a pena exaurir isso primeiro.r where onPerguntas frequentes and three others are referencO macOS já não lembra as posições das janelas? pile of Parcialmente e não confiável. Os aplicativos restauram suas próprias posições com qualidade variável e quebram sempre que um display é conectado ou desconectado, já que as coordenadas salvas se referiam a uma configuração de tela que não existe mais. Uma regra é aplicada nova em vez de restaurada.k, you wanO que acontece quando desencaixo meu laptop?ule that Uma regra que aponta para uma tela que não está conectada não pode ser satisfeita, então a janela volta para a tela restante - e a reconexão não a dispara novamente. Esta é a principal limitação prática para pessoas que atracam com frequência.is to defiPosso fazer isso sem pagar?on you owAproximadamente. O esquema de URL do Free Rectangle permite que um script ative cada aplicativo e encaixe-o no lugar, fornecendo o mesmo estado final sob demanda. O que o script não pode fazer facilmente é disparar automaticamente sempre que uma janela é aberta.ose, work for a week, and notice what irEscrito e testado por Ryan Hanson no macOS {{MACOS_VERSION}}.dow opens somewhere and you move it — mildly annoying, entirely predictable. A wrong rule means a window is confidently placed somewhere unhelpful, and you first have to work out why it went there before you can fix anything. Automation that's mostly right is more frustrating than no automation.
Add rules only when the same manual correction happens three times. That threshold keeps the set small and every rule earned.
Everything above is a Pro feature, but the useful part of it — arriving at a known arrangement without manual snapping — is reachable free.
The Rectangle app responds to URLs of the form rectangle://execute-action?name=left-half, and any action can be fired that way. So a script that activates each application in turn, waits for its window to appear, and fires the appropriate action produces the same end state.
The difference is the trigger. A rule fires by itself when a window opens; a script fires when you run it. In practice that gap is smaller than it sounds — bind the script to a hotkey or a hardware button and your layout is one press, which for most people is functionally equivalent to automatic.
The genuine drawback is fragility. The script depends on applications launching within the delays you allowed, and it breaks quietly when one is slow. Building a version that polls for the frontmost window rather than sleeping a fixed duration solves most of it, at the cost of an evening.
Two notes for anyone going down either path. Pro reads its preferences from com.knollsoft.Hookshot rather than com.knollsoft.Rectangle, so terminal commands copied from guides need the domain swapped. And whichever route you take, get the arrangement stable by hand first — automating a layout you haven't settled on means encoding a decision you're still making.
Per-app layouts are the clearest Pro justification, more than pinning or custom sizes, because the thing they replace is a daily ritual rather than an occasional task.
The test is repetition and stability together. If you build the same arrangement every morning from the same applications on the same hardware, the feature removes a recurring cost and pays back quickly. If your setup varies by task or your hardware changes through the week, rules will fight you, and a deliberate layout shortcut — Pro's or a script's — serves better than automatic placement.
And there's an honest floor: plenty of people who think they need automatic layouts have never tried binding a single shortcut that snaps their current window where it belongs. A dedicated window manager for Mac solves a surprising amount before any automation enters the picture, and it's worth exhausting that first.
Partially and unreliably. Apps restore their own positions with varying quality, and it breaks whenever a display is connected or disconnected, since the saved coordinates referred to a screen setup that no longer exists. A rule is applied fresh rather than restored.
A rule pointing at a display that isn't attached can't be satisfied, so the window falls back to the remaining screen — and reconnecting doesn't re-fire it. This is the main practical limitation for people who dock frequently.
Approximately. Free Rectangle's URL scheme lets a script activate each app and snap it into place, giving the same end state on demand. What scripting can't easily do is fire automatically whenever a window opens.
The post 41 Rectangle Pro App Layouts first appeared on Pinturas Ribo.
]]>Ninety seconds, maybe two minutes, repeated daily. It's not a lot on any single day, which is exactly why nobody fixes it — the cost is invisible per instance and only obvious in aggregate.
The more interesting cost is attention rather than time. You're making a dozen small spatial decisions before you've started thinking about woHet verschil is de trigger. Een regel gaat vanzelf af als er een raam opengaat; een script wordt geactiveerd wanneer u het uitvoert. In de praktijk is die kloof kleiner dan het klinkt: bind het script aan een sneltoets of een hardwareknop en je lay-out is één druk op de knop, wat voor de meeste mensen functioneel gelijk is aan automatisch. expresseHet echte nadeel is kwetsbaarheid. Het script is afhankelijk van toepassingen die binnen de door u toegestane vertragingen worden gestart, en wordt stil afgebroken als er een langzaam is. Het bouwen van een versie die zoekt naar het voorste raam in plaats van een vaste duur te slapen, lost het meeste probleem op, ten koste van een avond.s this frTwee opmerkingen voor iedereen die een van beide paden bewandelt. Pro leest zijn voorkeuren van com.knollsoft.Hookshot in plaats van com.knollsoft.Rectangle, dus voor terminalopdrachten die uit handleidingen worden gekopieerd, moet het domein worden verwisseld. En welke route u ook neemt, zorg dat u eerst het arrangement met de hand stabiel maakt. Het automatiseren van een lay-out waar u nog niet helemaal uit bent, betekent het coderen van een beslissing die u nog aan het nemen bent.r screen aOf de upgrade zinvol isis and whyLay-outs per app zijn de duidelijkste Pro-rechtvaardiging, meer dan vastzetten of aangepaste formaten, omdat wat ze vervangen een dagelijks ritueel is en geen incidentele taak.llowing aDe test is herhaling en stabiliteit samen. Als je elke ochtend hetzelfde arrangement bouwt vanuit dezelfde applicaties op dezelfde hardware, elimineert de functie terugkerende kosten en betaalt deze zich snel terug. Als je configuratie per taak verschilt of als je hardware in de loop van de week verandert, zullen de regels je bevechten, en een opzettelijke lay-outsnelkoppeling (Pro's of een script) is beter dan automatische plaatsing.tomaticalEn er is een eerlijk woord: veel mensen die denken dat ze automatische lay-outs nodig hebben, hebben nog nooit geprobeerd een enkele snelkoppeling te koppelen die hun huidige venster op de plek zet waar het thuishoort. Een specifieke vensterbeheerder voor Mac lost verrassend veel op voordat er enige vorm van automatisering in beeld komt, en het is de moeite waard om dat eerst uit te werken.r where onVeelgestelde vragen and three others are referencOnthoudt macOS de vensterposities niet al? pile of Gedeeltelijk en onbetrouwbaar. Apps herstellen hun eigen posities met wisselende kwaliteit, en het breekt telkens wanneer een beeldscherm wordt aangesloten of losgekoppeld, omdat de opgeslagen coördinaten verwijzen naar een schermopstelling die niet meer bestaat. Een regel wordt nieuw toegepast in plaats van hersteld.k, you wanWat gebeurt er als ik mijn laptop loskoppel?ule that Aan een regel die naar een beeldscherm wijst dat niet is aangesloten, kan niet worden voldaan, dus valt het venster terug naar het resterende scherm en wordt het opnieuw geactiveerd als u opnieuw verbinding maakt. Dit is de belangrijkste praktische beperking voor mensen die regelmatig aanmeren.is to defiKan ik dit doen zonder te betalen?on you owOngeveer. Dankzij het URL-schema van Free Rectangle kan een script elke app activeren en op zijn plaats klikken, waardoor op verzoek dezelfde eindstatus wordt gegeven. Wat scripting niet gemakkelijk kan doen, is automatisch activeren wanneer een venster wordt geopend.ose, work for a week, and notice what irGeschreven en getest door Ryan Hanson op macOS {{MACOS_VERSION}}.dow opens somewhere and you move it — mildly annoying, entirely predictable. A wrong rule means a window is confidently placed somewhere unhelpful, and you first have to work out why it went there before you can fix anything. Automation that's mostly right is more frustrating than no automation.
Add rules only when the same manual correction happens three times. That threshold keeps the set small and every rule earned.
Everything above is a Pro feature, but the useful part of it — arriving at a known arrangement without manual snapping — is reachable free.
The Rectangle app responds to URLs of the form rectangle://execute-action?name=left-half, and any action can be fired that way. So a script that activates each application in turn, waits for its window to appear, and fires the appropriate action produces the same end state.
The difference is the trigger. A rule fires by itself when a window opens; a script fires when you run it. In practice that gap is smaller than it sounds — bind the script to a hotkey or a hardware button and your layout is one press, which for most people is functionally equivalent to automatic.
The genuine drawback is fragility. The script depends on applications launching within the delays you allowed, and it breaks quietly when one is slow. Building a version that polls for the frontmost window rather than sleeping a fixed duration solves most of it, at the cost of an evening.
Two notes for anyone going down either path. Pro reads its preferences from com.knollsoft.Hookshot rather than com.knollsoft.Rectangle, so terminal commands copied from guides need the domain swapped. And whichever route you take, get the arrangement stable by hand first — automating a layout you haven't settled on means encoding a decision you're still making.
Per-app layouts are the clearest Pro justification, more than pinning or custom sizes, because the thing they replace is a daily ritual rather than an occasional task.
The test is repetition and stability together. If you build the same arrangement every morning from the same applications on the same hardware, the feature removes a recurring cost and pays back quickly. If your setup varies by task or your hardware changes through the week, rules will fight you, and a deliberate layout shortcut — Pro's or a script's — serves better than automatic placement.
And there's an honest floor: plenty of people who think they need automatic layouts have never tried binding a single shortcut that snaps their current window where it belongs. A dedicated window manager for Mac solves a surprising amount before any automation enters the picture, and it's worth exhausting that first.
Partially and unreliably. Apps restore their own positions with varying quality, and it breaks whenever a display is connected or disconnected, since the saved coordinates referred to a screen setup that no longer exists. A rule is applied fresh rather than restored.
A rule pointing at a display that isn't attached can't be satisfied, so the window falls back to the remaining screen — and reconnecting doesn't re-fire it. This is the main practical limitation for people who dock frequently.
Approximately. Free Rectangle's URL scheme lets a script activate each app and snap it into place, giving the same end state on demand. What scripting can't easily do is fire automatically whenever a window opens.
The post 41 Rectangle Pro App Layouts first appeared on Pinturas Ribo.
]]>Ninety seconds, maybe two minutes, repeated daily. It's not a lot on any single day, which is exactly why nobody fixes it — the cost is invisible per instance and only obvious in aggregate.
The more interesting cost is attention rather than time. You're making a dozen small spatial decisions before you've started thinking about wo違いはトリガーです。ルールはウィンドウが開くと自動的に起動します。スクリプトを実行すると、スクリプトが起動されます。実際には、そのギャップは思っているよりも小さく、スクリプトをホットキーまたはハードウェア ボタンにバインドすると、レイアウトを 1 回押すだけで済みます。これは、ほとんどの人にとって、機能的には自動と同等です。 expresse本当の欠点は壊れやすさです。スクリプトは、許可された遅延内で起動するアプリケーションに依存しており、遅い場合には静かに中断されます。固定期間スリープするのではなく、最前面のウィンドウをポーリングするバージョンを構築すると、一晩かかるだけで、問題のほとんどが解決します。s this frどちらの道を歩もうとしている人のための 2 つのメモ。 Pro は、com.knollsoft.Rectangle ではなく com.knollsoft.Hookshot から設定を読み取るため、ガイドからコピーされたターミナル コマンドではドメインを交換する必要があります。どちらの方法を採用する場合でも、まず手動で配置を安定させてください。まだ決まっていないレイアウトを自動化するということは、まだ下している決定をエンコードすることを意味します。r screen aアップグレードに意味があるかどうかis and whyアプリごとのレイアウトは、ピン留めやカスタム サイズよりも Pro にとって最も明確な理由です。なぜなら、アプリごとのレイアウトに置き換わるものは、時折のタスクではなく毎日の儀式だからです。llowing aテストは反復と安定性を兼ね備えたものです。同じハードウェア上の同じアプリケーションから毎朝同じ構成を構築する場合、この機能により定期的なコストが削減され、すぐに回収されます。セットアップがタスクによって異なる場合や、週を通じてハードウェアが変更される場合は、ルールが邪魔をすることになるため、プロやスクリプトによる意図的なレイアウト ショートカットのほうが、自動配置よりも効果的です。tomaticalそして正直な意見もあります。自動レイアウトが必要だと考えている人の多くは、現在のウィンドウを適切な場所にスナップする単一のショートカットをバインドしようとしたことがありません。 Mac 用の専用ウィンドウ マネージャー は、自動化が本格化する前に驚くほど多くのことを解決してくれるので、まずそれを使いこなす価値があります。r where onよくある質問 and three others are referencmacOS はウィンドウの位置をすでに記憶しているのではないでしょうか? pile of 部分的かつ信頼性が低い。アプリはさまざまな品質で独自の位置を復元しますが、保存された座標がもう存在しない画面設定を参照しているため、ディスプレイが接続または切断されるたびに機能が中断されます。ルールは復元されるのではなく、新たに適用されます。k, you wanラップトップのドッキングを解除するとどうなりますか?ule that 接続されていないディスプレイを指すルールは満たされないため、ウィンドウは残りの画面に戻り、再接続しても再起動されません。これは、頻繁にドッキングする人にとっての主な実際的な制限です。is to defiお金を払わずにこれを行うことはできますか？on you ow約。 Free Rectangle の URL スキームを使用すると、スクリプトで各アプリをアクティブにして所定の位置にスナップし、オンデマンドで同じ最終状態を与えることができます。スクリプトで簡単に実行できないのは、ウィンドウが開くたびに自動的に起動することです。ose, work for a week, and notice what irRyan Hanson によって macOS {{MACOS_VERSION}} 上で作成およびテストされています。dow opens somewhere and you move it — mildly annoying, entirely predictable. A wrong rule means a window is confidently placed somewhere unhelpful, and you first have to work out why it went there before you can fix anything. Automation that's mostly right is more frustrating than no automation.
Add rules only when the same manual correction happens three times. That threshold keeps the set small and every rule earned.
Everything above is a Pro feature, but the useful part of it — arriving at a known arrangement without manual snapping — is reachable free.
The Rectangle app responds to URLs of the form rectangle://execute-action?name=left-half, and any action can be fired that way. So a script that activates each application in turn, waits for its window to appear, and fires the appropriate action produces the same end state.
The difference is the trigger. A rule fires by itself when a window opens; a script fires when you run it. In practice that gap is smaller than it sounds — bind the script to a hotkey or a hardware button and your layout is one press, which for most people is functionally equivalent to automatic.
The genuine drawback is fragility. The script depends on applications launching within the delays you allowed, and it breaks quietly when one is slow. Building a version that polls for the frontmost window rather than sleeping a fixed duration solves most of it, at the cost of an evening.
Two notes for anyone going down either path. Pro reads its preferences from com.knollsoft.Hookshot rather than com.knollsoft.Rectangle, so terminal commands copied from guides need the domain swapped. And whichever route you take, get the arrangement stable by hand first — automating a layout you haven't settled on means encoding a decision you're still making.
Per-app layouts are the clearest Pro justification, more than pinning or custom sizes, because the thing they replace is a daily ritual rather than an occasional task.
The test is repetition and stability together. If you build the same arrangement every morning from the same applications on the same hardware, the feature removes a recurring cost and pays back quickly. If your setup varies by task or your hardware changes through the week, rules will fight you, and a deliberate layout shortcut — Pro's or a script's — serves better than automatic placement.
And there's an honest floor: plenty of people who think they need automatic layouts have never tried binding a single shortcut that snaps their current window where it belongs. A dedicated window manager for Mac solves a surprising amount before any automation enters the picture, and it's worth exhausting that first.
Partially and unreliably. Apps restore their own positions with varying quality, and it breaks whenever a display is connected or disconnected, since the saved coordinates referred to a screen setup that no longer exists. A rule is applied fresh rather than restored.
A rule pointing at a display that isn't attached can't be satisfied, so the window falls back to the remaining screen — and reconnecting doesn't re-fire it. This is the main practical limitation for people who dock frequently.
Approximately. Free Rectangle's URL scheme lets a script activate each app and snap it into place, giving the same end state on demand. What scripting can't easily do is fire automatically whenever a window opens.
The post 41 Rectangle Pro App Layouts first appeared on Pinturas Ribo.
]]>Ninety seconds, maybe two minutes, repeated daily. It's not a lot on any single day, which is exactly why nobody fixes it — the cost is invisible per instance and only obvious in aggregate.
The more interesting cost is attention rather than time. You're making a dozen small spatial decisions before you've started thinking about woLa differenza è il grilletto. Una regola si attiva da sola quando si apre una finestra; uno script si attiva quando lo esegui. In pratica il divario è più piccolo di quanto sembri: associa lo script a un tasto di scelta rapida o a un pulsante hardware e il tuo layout è una sola pressione, che per la maggior parte delle persone è funzionalmente equivalente all'automatico. expresseIl vero svantaggio è la fragilità. Lo script dipende dall'avvio delle applicazioni entro i ritardi consentiti e si interrompe silenziosamente quando una è lenta. Costruire una versione che sondaggi per la finestra in primo piano invece di dormire per una durata fissa risolve gran parte del problema, al costo di una serata.s this frDue note per chiunque imbocchi entrambe le strade. Pro legge le sue preferenze da com.knollsoft.Hookshot anziché da com.knollsoft.Rectangle, quindi i comandi del terminale copiati dalle guide richiedono lo scambio del dominio. E qualunque sia il percorso che prendi, rendi prima stabile la disposizione manualmente: automatizzare un layout su cui non ti sei ancora deciso significa codificare una decisione che stai ancora prendendo.r screen aSe l'aggiornamento ha sensois and whyI layout per-app sono la giustificazione Pro più chiara, più che il blocco o le dimensioni personalizzate, perché ciò che sostituiscono è un rituale quotidiano piuttosto che un'attività occasionale.llowing aLa prova è ripetizione e stabilità insieme. Se crei lo stesso accordo ogni mattina dalle stesse applicazioni sullo stesso hardware, la funzionalità elimina un costo ricorrente e si ripaga rapidamente. Se la tua configurazione varia in base all'attività o alle modifiche hardware nel corso della settimana, le regole ti combatteranno e una scorciatoia di layout deliberata, Pro o script, è migliore del posizionamento automatico.tomaticalE c'è un punto onesto: molte persone che pensano di aver bisogno di layout automatici non hanno mai provato ad associare una singola scorciatoia che blocchi la finestra corrente al suo posto. Un window manager dedicato per Mac risolve una quantità sorprendente prima che qualsiasi automazione entri in gioco, e vale la pena approfondirlo prima.r where onDomande frequenti and three others are referencmacOS non ricorda già la posizione delle finestre? pile of Parzialmente e in modo inaffidabile. Le app ripristinano le proprie posizioni con qualità variabile e si interrompe ogni volta che un display viene collegato o disconnesso, poiché le coordinate salvate si riferivano a una configurazione dello schermo che non esiste più. Una regola viene applicata nuova anziché ripristinata.k, you wanCosa succede quando sgancio il mio laptop?ule that Una regola che punta a uno schermo non collegato non può essere soddisfatta, quindi la finestra torna allo schermo rimanente e la riconnessione non la riattiva. Questa è la principale limitazione pratica per le persone che attraccano frequentemente.is to defiPosso farlo senza pagare?on you owCirca. Lo schema URL di Free Rectangle consente a uno script di attivare ciascuna app e di inserirla in posizione, fornendo lo stesso stato finale su richiesta. Ciò che lo scripting non può fare facilmente è attivarsi automaticamente ogni volta che si apre una finestra.ose, work for a week, and notice what irScritto e testato da Ryan Hanson su macOS {{MACOS_VERSION}}.dow opens somewhere and you move it — mildly annoying, entirely predictable. A wrong rule means a window is confidently placed somewhere unhelpful, and you first have to work out why it went there before you can fix anything. Automation that's mostly right is more frustrating than no automation.
Add rules only when the same manual correction happens three times. That threshold keeps the set small and every rule earned.
Everything above is a Pro feature, but the useful part of it — arriving at a known arrangement without manual snapping — is reachable free.
The Rectangle app responds to URLs of the form rectangle://execute-action?name=left-half, and any action can be fired that way. So a script that activates each application in turn, waits for its window to appear, and fires the appropriate action produces the same end state.
The difference is the trigger. A rule fires by itself when a window opens; a script fires when you run it. In practice that gap is smaller than it sounds — bind the script to a hotkey or a hardware button and your layout is one press, which for most people is functionally equivalent to automatic.
The genuine drawback is fragility. The script depends on applications launching within the delays you allowed, and it breaks quietly when one is slow. Building a version that polls for the frontmost window rather than sleeping a fixed duration solves most of it, at the cost of an evening.
Two notes for anyone going down either path. Pro reads its preferences from com.knollsoft.Hookshot rather than com.knollsoft.Rectangle, so terminal commands copied from guides need the domain swapped. And whichever route you take, get the arrangement stable by hand first — automating a layout you haven't settled on means encoding a decision you're still making.
Per-app layouts are the clearest Pro justification, more than pinning or custom sizes, because the thing they replace is a daily ritual rather than an occasional task.
The test is repetition and stability together. If you build the same arrangement every morning from the same applications on the same hardware, the feature removes a recurring cost and pays back quickly. If your setup varies by task or your hardware changes through the week, rules will fight you, and a deliberate layout shortcut — Pro's or a script's — serves better than automatic placement.
And there's an honest floor: plenty of people who think they need automatic layouts have never tried binding a single shortcut that snaps their current window where it belongs. A dedicated window manager for Mac solves a surprising amount before any automation enters the picture, and it's worth exhausting that first.
Partially and unreliably. Apps restore their own positions with varying quality, and it breaks whenever a display is connected or disconnected, since the saved coordinates referred to a screen setup that no longer exists. A rule is applied fresh rather than restored.
A rule pointing at a display that isn't attached can't be satisfied, so the window falls back to the remaining screen — and reconnecting doesn't re-fire it. This is the main practical limitation for people who dock frequently.
Approximately. Free Rectangle's URL scheme lets a script activate each app and snap it into place, giving the same end state on demand. What scripting can't easily do is fire automatically whenever a window opens.
The post 41 Rectangle Pro App Layouts first appeared on Pinturas Ribo.
]]>Ninety seconds, maybe two minutes, repeated daily. It's not a lot on any single day, which is exactly why nobody fixes it — the cost is invisible per instance and only obvious in aggregate.
The more interesting cost is attention rather than time. You're making a dozen small spatial decisions before you've started thinking about woLa diferencia es el detonante. Una regla se activa sola cuando se abre una ventana; un script se activa cuando lo ejecutas. En la práctica, esa brecha es más pequeña de lo que parece: vincule el script a una tecla de acceso rápido o a un botón de hardware y su diseño será de una sola pulsación, lo que para la mayoría de las personas es funcionalmente equivalente a automático. expresseEl verdadero inconveniente es la fragilidad. El script depende de que las aplicaciones se inicien dentro de los retrasos permitidos y se interrumpe silenciosamente cuando una es lenta. Crear una versión que busque la ventana más cercana en lugar de dormir durante un período fijo resuelve la mayor parte del problema, a costa de una noche.s this frDos notas para cualquiera que siga cualquiera de los dos caminos. Pro lee sus preferencias de com.knollsoft.Hookshot en lugar de com.knollsoft.Rectangle, por lo que los comandos de terminal copiados de las guías necesitan intercambiar el dominio. Y sea cual sea el camino que tomes, primero estabiliza el arreglo a mano: automatizar un diseño que no has elegido significa codificar una decisión que aún estás tomando.r screen aSi la actualización tiene sentidois and whyLos diseños por aplicación son la justificación profesional más clara, más que fijar o personalizar tamaños, porque lo que reemplazan es un ritual diario en lugar de una tarea ocasional.llowing aLa prueba es repetición y estabilidad juntas. Si crea el mismo arreglo todas las mañanas a partir de las mismas aplicaciones en el mismo hardware, la función elimina un costo recurrente y se amortiza rápidamente. Si su configuración varía según la tarea o su hardware cambia a lo largo de la semana, las reglas se enfrentarán a usted y un atajo de diseño deliberado (Pro o script) es mejor que la ubicación automática.tomaticalY hay un piso honesto: muchas personas que piensan que necesitan diseños automáticos nunca han intentado vincular un único acceso directo que ajuste su ventana actual a donde pertenece. Un administrador de ventanas dedicado para Mac resuelve una cantidad sorprendente de cosas antes de que cualquier automatización entre en escena, y vale la pena agotarlo primero.r where onPreguntas frecuentes and three others are referenc¿MacOS ya no recuerda las posiciones de las ventanas? pile of Parcial y poco fiable. Las aplicaciones restauran sus propias posiciones con calidad variable y se rompe cada vez que se conecta o desconecta una pantalla, ya que las coordenadas guardadas se referían a una configuración de pantalla que ya no existe. Una regla se aplica nueva y no restaurada.k, you wan¿Qué sucede cuando desconecto mi computadora portátil?ule that No se puede cumplir una regla que apunta a una pantalla que no está conectada, por lo que la ventana vuelve a la pantalla restante y, al volver a conectarse, no se vuelve a activar. Esta es la principal limitación práctica para las personas que atracan con frecuencia.is to defi¿Puedo hacer esto sin pagar?on you owAproximadamente. El esquema de URL de Free Triangle permite que un script active cada aplicación y la coloque en su lugar, brindando el mismo estado final bajo demanda. Lo que las secuencias de comandos no pueden hacer fácilmente es activarse automáticamente cada vez que se abre una ventana.ose, work for a week, and notice what irEscrito y probado por Ryan Hanson en macOS {{MACOS_VERSION}}.dow opens somewhere and you move it — mildly annoying, entirely predictable. A wrong rule means a window is confidently placed somewhere unhelpful, and you first have to work out why it went there before you can fix anything. Automation that's mostly right is more frustrating than no automation.
Add rules only when the same manual correction happens three times. That threshold keeps the set small and every rule earned.
Everything above is a Pro feature, but the useful part of it — arriving at a known arrangement without manual snapping — is reachable free.
The Rectangle app responds to URLs of the form rectangle://execute-action?name=left-half, and any action can be fired that way. So a script that activates each application in turn, waits for its window to appear, and fires the appropriate action produces the same end state.
The difference is the trigger. A rule fires by itself when a window opens; a script fires when you run it. In practice that gap is smaller than it sounds — bind the script to a hotkey or a hardware button and your layout is one press, which for most people is functionally equivalent to automatic.
The genuine drawback is fragility. The script depends on applications launching within the delays you allowed, and it breaks quietly when one is slow. Building a version that polls for the frontmost window rather than sleeping a fixed duration solves most of it, at the cost of an evening.
Two notes for anyone going down either path. Pro reads its preferences from com.knollsoft.Hookshot rather than com.knollsoft.Rectangle, so terminal commands copied from guides need the domain swapped. And whichever route you take, get the arrangement stable by hand first — automating a layout you haven't settled on means encoding a decision you're still making.
Per-app layouts are the clearest Pro justification, more than pinning or custom sizes, because the thing they replace is a daily ritual rather than an occasional task.
The test is repetition and stability together. If you build the same arrangement every morning from the same applications on the same hardware, the feature removes a recurring cost and pays back quickly. If your setup varies by task or your hardware changes through the week, rules will fight you, and a deliberate layout shortcut — Pro's or a script's — serves better than automatic placement.
And there's an honest floor: plenty of people who think they need automatic layouts have never tried binding a single shortcut that snaps their current window where it belongs. A dedicated window manager for Mac solves a surprising amount before any automation enters the picture, and it's worth exhausting that first.
Partially and unreliably. Apps restore their own positions with varying quality, and it breaks whenever a display is connected or disconnected, since the saved coordinates referred to a screen setup that no longer exists. A rule is applied fresh rather than restored.
A rule pointing at a display that isn't attached can't be satisfied, so the window falls back to the remaining screen — and reconnecting doesn't re-fire it. This is the main practical limitation for people who dock frequently.
Approximately. Free Rectangle's URL scheme lets a script activate each app and snap it into place, giving the same end state on demand. What scripting can't easily do is fire automatically whenever a window opens.
The post 41 Rectangle Pro App Layouts first appeared on Pinturas Ribo.
]]>Ninety seconds, maybe two minutes, repeated daily. It's not a lot on any single day, which is exactly why nobody fixes it — the cost is invisible per instance and only obvious in aggregate.
The more interesting cost is attention rather than time. You're making a dozen small spatial decisions before you've started thinking about woLa différence est le déclencheur. Une règle se déclenche toute seule lorsqu'une fenêtre s'ouvre ; un script se déclenche lorsque vous l'exécutez. En pratique, cet écart est plus petit qu'il n'y paraît : liez le script à un raccourci clavier ou à un bouton matériel et votre mise en page se fait en une seule pression, ce qui, pour la plupart des gens, est fonctionnellement équivalent à automatique. expresseLe véritable inconvénient est la fragilité. Le script dépend du lancement des applications dans les délais que vous avez autorisés, et il s'interrompt silencieusement lorsqu'une application est lente. Construire une version qui interroge la fenêtre la plus en avant plutôt que de dormir pendant une durée fixe résout la plupart des problèmes, au prix d'une soirée.s this frDeux notes pour quiconque emprunte l’une ou l’autre voie. Pro lit ses préférences à partir de com.knollsoft.Hookshot plutôt que de com.knollsoft.Rectangle, donc les commandes de terminal copiées à partir de guides nécessitent que le domaine soit échangé. Et quelle que soit la voie que vous empruntez, commencez par stabiliser l'arrangement à la main : automatiser une mise en page sur laquelle vous n'avez pas encore choisi signifie coder une décision que vous êtes encore en train de prendre.r screen aSi la mise à niveau a du sensis and whyLes mises en page par application constituent la justification Pro la plus claire, plus que l'épinglage ou les tailles personnalisées, car ce qu'elles remplacent est un rituel quotidien plutôt qu'une tâche occasionnelle.llowing aLe test est la répétition et la stabilité ensemble. Si vous créez le même arrangement chaque matin à partir des mêmes applications sur le même matériel, la fonctionnalité supprime un coût récurrent et est rapidement amortie. Si votre configuration varie en fonction de la tâche ou si votre matériel change au cours de la semaine, les règles vous combattront et un raccourci de mise en page délibéré (Pro ou script) est plus efficace que le placement automatique.tomaticalEt il y a un plancher honnête : beaucoup de gens qui pensent avoir besoin de mises en page automatiques n'ont jamais essayé de lier un seul raccourci qui place leur fenêtre actuelle à sa place. Un gestionnaire de fenêtres dédié pour Mac résout un nombre surprenant de choses avant qu'une automatisation n'entre en scène, et cela vaut la peine de l'épuiser d'abord.r where onQuestions fréquemment posées and three others are referencmacOS ne mémorise-t-il pas déjà la position des fenêtres ? pile of Partiellement et de manière peu fiable. Les applications restaurent leurs propres positions avec une qualité variable, et cela se brise chaque fois qu'un écran est connecté ou déconnecté, car les coordonnées enregistrées faisaient référence à une configuration d'écran qui n'existe plus. Une règle est appliquée fraîche plutôt que restaurée.k, you wanQue se passe-t-il lorsque je déconnecte mon ordinateur portable ?ule that Une règle pointant vers un écran qui n'est pas connecté ne peut pas être satisfaite, donc la fenêtre revient à l'écran restant - et la reconnexion ne la relance pas. C’est la principale limitation pratique pour les personnes qui accostent fréquemment.is to defiPuis-je le faire sans payer ?on you owEnviron. Le schéma d'URL de Free Rectangle permet à un script d'activer chaque application et de la mettre en place, donnant le même état final à la demande. Ce que les scripts ne peuvent pas faire facilement, c'est se lancer automatiquement à chaque fois qu'une fenêtre s'ouvre.ose, work for a week, and notice what irÉcrit et testé par Ryan Hanson sur macOS {{MACOS_VERSION}}.dow opens somewhere and you move it — mildly annoying, entirely predictable. A wrong rule means a window is confidently placed somewhere unhelpful, and you first have to work out why it went there before you can fix anything. Automation that's mostly right is more frustrating than no automation.
Add rules only when the same manual correction happens three times. That threshold keeps the set small and every rule earned.
Everything above is a Pro feature, but the useful part of it — arriving at a known arrangement without manual snapping — is reachable free.
The Rectangle app responds to URLs of the form rectangle://execute-action?name=left-half, and any action can be fired that way. So a script that activates each application in turn, waits for its window to appear, and fires the appropriate action produces the same end state.
The difference is the trigger. A rule fires by itself when a window opens; a script fires when you run it. In practice that gap is smaller than it sounds — bind the script to a hotkey or a hardware button and your layout is one press, which for most people is functionally equivalent to automatic.
The genuine drawback is fragility. The script depends on applications launching within the delays you allowed, and it breaks quietly when one is slow. Building a version that polls for the frontmost window rather than sleeping a fixed duration solves most of it, at the cost of an evening.
Two notes for anyone going down either path. Pro reads its preferences from com.knollsoft.Hookshot rather than com.knollsoft.Rectangle, so terminal commands copied from guides need the domain swapped. And whichever route you take, get the arrangement stable by hand first — automating a layout you haven't settled on means encoding a decision you're still making.
Per-app layouts are the clearest Pro justification, more than pinning or custom sizes, because the thing they replace is a daily ritual rather than an occasional task.
The test is repetition and stability together. If you build the same arrangement every morning from the same applications on the same hardware, the feature removes a recurring cost and pays back quickly. If your setup varies by task or your hardware changes through the week, rules will fight you, and a deliberate layout shortcut — Pro's or a script's — serves better than automatic placement.
And there's an honest floor: plenty of people who think they need automatic layouts have never tried binding a single shortcut that snaps their current window where it belongs. A dedicated window manager for Mac solves a surprising amount before any automation enters the picture, and it's worth exhausting that first.
Partially and unreliably. Apps restore their own positions with varying quality, and it breaks whenever a display is connected or disconnected, since the saved coordinates referred to a screen setup that no longer exists. A rule is applied fresh rather than restored.
A rule pointing at a display that isn't attached can't be satisfied, so the window falls back to the remaining screen — and reconnecting doesn't re-fire it. This is the main practical limitation for people who dock frequently.
Approximately. Free Rectangle's URL scheme lets a script activate each app and snap it into place, giving the same end state on demand. What scripting can't easily do is fire automatically whenever a window opens.
The post 41 Rectangle Pro App Layouts first appeared on Pinturas Ribo.
]]>Ninety seconds, maybe two minutes, repeated daily. It's not a lot on any single day, which is exactly why nobody fixes it — the cost is invisible per instance and only obvious in aggregate.
The more interesting cost is attention rather than time. You're making a dozen small spatial decisions before you've started thinking about woDer Unterschied ist der Auslöser. Eine Regel wird von selbst ausgelöst, wenn ein Fenster geöffnet wird. Ein Skript wird ausgelöst, wenn Sie es ausführen. In der Praxis ist diese Lücke kleiner, als es sich anhört – binden Sie das Skript an einen Hotkey oder eine Hardware-Taste und Ihr Layout besteht aus einem Tastendruck, was für die meisten Menschen funktional einer automatischen Funktion entspricht. expresseDer eigentliche Nachteil ist die Zerbrechlichkeit. Das Skript ist darauf angewiesen, dass Anwendungen innerhalb der von Ihnen zugelassenen Verzögerungen gestartet werden, und bricht leise ab, wenn eine Anwendung langsam ist. Das Erstellen einer Version, die das vorderste Fenster abfragt, anstatt eine feste Zeit lang zu schlafen, löst das meiste Problem, kostet jedoch einen Abend.s this frZwei Anmerkungen für alle, die einen der beiden Wege wählen. Pro liest seine Einstellungen von com.knollsoft.Hookshot und nicht von com.knollsoft.Rectangle, daher muss bei Terminalbefehlen, die aus Handbüchern kopiert werden, die Domäne ausgetauscht werden. Und welchen Weg Sie auch wählen, stabilisieren Sie die Anordnung zunächst manuell – ein Layout zu automatisieren, für das Sie sich noch nicht entschieden haben, bedeutet, eine Entscheidung zu verschlüsseln, die Sie noch treffen.r screen aOb das Upgrade sinnvoll istis and whyPro-App-Layouts sind die klarste Pro-Rechtfertigung, mehr als das Anheften oder benutzerdefinierte Größen, denn das, was sie ersetzen, ist ein tägliches Ritual und keine gelegentliche Aufgabe.llowing aDer Test besteht aus Wiederholung und Stabilität zusammen. Wenn Sie jeden Morgen das gleiche Arrangement mit denselben Anwendungen auf derselben Hardware erstellen, erspart Ihnen diese Funktion wiederkehrende Kosten und amortisiert sich schnell. Wenn Ihr Setup je nach Aufgabe variiert oder sich Ihre Hardware im Laufe der Woche ändert, werden Ihnen die Regeln zu schaffen machen, und eine bewusste Layout-Verknüpfung – Pros oder ein Skript – ist besser als die automatische Platzierung.tomaticalUnd es gibt eine ehrliche Bemerkung: Viele Leute, die denken, sie bräuchten automatische Layouts, haben noch nie versucht, eine einzige Verknüpfung zu binden, die ihr aktuelles Fenster dort einrastet, wo es hingehört. Ein dedizierter Fenstermanager für Mac löst überraschend viele Probleme, bevor Automatisierung ins Spiel kommt, und es lohnt sich, diese zunächst auszuschöpfen.r where onHäufig gestellte Fragen and three others are referencMerkt sich macOS nicht bereits Fensterpositionen? pile of Teilweise und unzuverlässig. Apps stellen ihre eigenen Positionen mit unterschiedlicher Qualität wieder her und es bricht ab, wenn ein Display angeschlossen oder getrennt wird, da sich die gespeicherten Koordinaten auf eine Bildschirmkonfiguration beziehen, die nicht mehr existiert. Eine Regel wird frisch angewendet und nicht wiederhergestellt.k, you wanWas passiert, wenn ich meinen Laptop abdocke?ule that Eine Regel, die auf einen nicht angeschlossenen Bildschirm zeigt, kann nicht erfüllt werden, sodass das Fenster auf den verbleibenden Bildschirm zurückfällt – und durch eine erneute Verbindung wird es nicht erneut ausgelöst. Dies ist die größte praktische Einschränkung für Personen, die häufig andocken.is to defiKann ich das tun, ohne zu bezahlen?on you owEtwa. Mit dem URL-Schema von Free Rechteck kann ein Skript jede App aktivieren und an ihrem Platz einrasten lassen, sodass bei Bedarf derselbe Endstatus bereitgestellt wird. Was mit Skripten nicht einfach möglich ist, ist, dass jedes Mal, wenn ein Fenster geöffnet wird, automatisch ausgelöst wird.ose, work for a week, and notice what irGeschrieben und getestet von Ryan Hanson unter macOS {{MACOS_VERSION}}.dow opens somewhere and you move it — mildly annoying, entirely predictable. A wrong rule means a window is confidently placed somewhere unhelpful, and you first have to work out why it went there before you can fix anything. Automation that's mostly right is more frustrating than no automation.
Add rules only when the same manual correction happens three times. That threshold keeps the set small and every rule earned.
Everything above is a Pro feature, but the useful part of it — arriving at a known arrangement without manual snapping — is reachable free.
The Rectangle app responds to URLs of the form rectangle://execute-action?name=left-half, and any action can be fired that way. So a script that activates each application in turn, waits for its window to appear, and fires the appropriate action produces the same end state.
The difference is the trigger. A rule fires by itself when a window opens; a script fires when you run it. In practice that gap is smaller than it sounds — bind the script to a hotkey or a hardware button and your layout is one press, which for most people is functionally equivalent to automatic.
The genuine drawback is fragility. The script depends on applications launching within the delays you allowed, and it breaks quietly when one is slow. Building a version that polls for the frontmost window rather than sleeping a fixed duration solves most of it, at the cost of an evening.
Two notes for anyone going down either path. Pro reads its preferences from com.knollsoft.Hookshot rather than com.knollsoft.Rectangle, so terminal commands copied from guides need the domain swapped. And whichever route you take, get the arrangement stable by hand first — automating a layout you haven't settled on means encoding a decision you're still making.
Per-app layouts are the clearest Pro justification, more than pinning or custom sizes, because the thing they replace is a daily ritual rather than an occasional task.
The test is repetition and stability together. If you build the same arrangement every morning from the same applications on the same hardware, the feature removes a recurring cost and pays back quickly. If your setup varies by task or your hardware changes through the week, rules will fight you, and a deliberate layout shortcut — Pro's or a script's — serves better than automatic placement.
And there's an honest floor: plenty of people who think they need automatic layouts have never tried binding a single shortcut that snaps their current window where it belongs. A dedicated window manager for Mac solves a surprising amount before any automation enters the picture, and it's worth exhausting that first.
Partially and unreliably. Apps restore their own positions with varying quality, and it breaks whenever a display is connected or disconnected, since the saved coordinates referred to a screen setup that no longer exists. A rule is applied fresh rather than restored.
A rule pointing at a display that isn't attached can't be satisfied, so the window falls back to the remaining screen — and reconnecting doesn't re-fire it. This is the main practical limitation for people who dock frequently.
Approximately. Free Rectangle's URL scheme lets a script activate each app and snap it into place, giving the same end state on demand. What scripting can't easily do is fire automatically whenever a window opens.
The post 41 Rectangle Pro App Layouts first appeared on Pinturas Ribo.
]]>The post Which Engineering Units Numi Handles, and Which It Doesn’t first appeared on Pinturas Ribo.
]]>Numi covers the common engineering categories well, length, area, volume, mass, force, pressure, energy, and power across metric and imperial, with genuine dimensional checking so units cancel correctly. It doesn’t attempt to be a full engineering reference. Specialised field units, some derived quantities, and anything needing reference data it doesn’t hold fall outside coverage, and the only signal is a blank result. This is the explicit list.
Key takeaways
These categories convert reliably, and the dimensional checking works across them so they combine in calculations rather than only converting in isolation.
| Category | Metric examples | Imperial examples | Reliability |
|---|---|---|---|
| Length | mm, cm, m, km | in, ft, yd, mile | High |
| Area | m², hectare | ft², acre | High |
| Volume | ml, l, m³ | fl oz, gal | High |
| Mass | g, kg, tonne | oz, lb | High |
| Force | newton, kN | lbf | Good |
| Pressure | Pa, kPa, bar | psi, atm | Good |
| Energy | joule, kJ, kWh | BTU, calorie | Good |
| Power | watt, kW | horsepower | Good |
| Temperature | celsius, kelvin | fahrenheit | High |
| Speed | m/s, km/h | mph, knot | High |
That covers a large share of everyday engineering conversion. The reliability column is a judgement from testing rather than a guarantee, and “good” means common units within the category work while less common variants should be verified individually.
Conversion alone is available in any converter. What a natural language calculator for Mac adds is carrying units through arithmetic and checking them.
force = 500 newton
area = 0.05 m^2
force / area in kpaForce divided by area yields pressure, and the result comes out in the pressure unit requested. The calculator recognised the dimensional relationship rather than just dividing two numbers. Type an incompatible combination, a force divided by a temperature, and it returns nothing, which is the tool catching an error rather than producing a meaningless figure.
This is genuinely useful for sanity-checking. A calculation that should produce a pressure and instead returns a blank line tells you the dimensions don’t work, which is a class of mistake that slips through a spreadsheet where every cell is just a number.
It’s not a full computer algebra system with symbolic dimensions. It checks the units it knows against the operations it understands, which covers a lot of real work and stops short of formal dimensional proof.
Categories where common units work and coverage is inconsistent at the edges. Test your specific unit rather than assuming the category is complete.
The pattern across all of these is that the base units of the category are covered and the long tail isn’t. A calculator notepad app is broad enough for common professional use and shallow in each specialty relative to a dedicated engineering reference.
Stated plainly so no engineer is caught out mid-calculation.
The honest framing is scope, not deficiency. A tool small enough to summon on a hotkey and answer in a second isn’t going to carry a discipline’s full reference library, and trying to make it do so would ruin what it’s good at.
Not as the primary tool, and genuinely useful alongside one.
The fourth is underrated. A spreadsheet treats every cell as a dimensionless number, so a formula multiplying a length by a pressure produces a number with no complaint. Recomputing the same thing in a tool that checks dimensions surfaces the error, which is a real safeguard for calculations that matter.
The single most important step for an engineer, because the failure mode is silence.
Ten minutes of this against your actual units tells you whether Numi fits your work far better than any coverage list, including this one, because your specific units are what matter and no article can enumerate every case.
For an engineer, this is a fast auxiliary tool, not a primary one. It handles the common conversions, the sanity checks, and the everyday math quickly enough to be worth having open, and it does genuine dimensional checking within its known units.
It doesn’t replace a dedicated engineering calculator or a reference tool, and expecting it to will disappoint. The significant-figure gap alone rules it out as a primary tool for anything where precision discipline is part of the job. Run Numi app alongside your real tools, use it for the fast stuff, and hand off to a proper engineering environment for analysis, standards work, and anything frame-precise.
That’s a smaller claim than “engineering calculator,” and it’s the true one.
It handles the common categories well, including length, area, volume, mass, force, pressure, energy, and power in metric and imperial. It doesn’t attempt to be a full engineering reference, so highly specialised or field-specific units and some derived quantities fall outside coverage.
Yes, the common ones convert reliably, covering pascals, bar, psi, atmospheres, newtons, and pounds-force among others. The practical test remains typing the specific unit you need, since coverage of less common variants isn’t guaranteed.
For everyday conversions, sanity checks, and back-of-envelope estimates, generally yes. For deep analysis with specialised units, significant-figure discipline, or field-specific reference data, a dedicated tool is necessary, and Numi mac works best alongside it.
It carries units through calculations and checks compatibility, so dividing force by area yields pressure and incompatible combinations return nothing. That’s genuine dimensional checking within its known units, not a substitute for a full computer algebra system.
Highly specialised field units, some derived and compound quantities, and any unit tied to reference data the tool doesn’t hold. Test the specific units your work depends on before relying on it, since a blank result is the only signal an unsupported unit gives.
The post Which Engineering Units Numi Handles, and Which It Doesn’t first appeared on Pinturas Ribo.
]]>The post Understanding Time Machine Local Snapshots and Their Storage Footprint first appeared on Pinturas Ribo.
]]>Key takeaways
tmutil listlocalsnapshotdates / lists them; sudo tmutil deletelocalsnapshots [date] removes them on Sonoma; the new --purge flag handles all of them at once on Tahoe.tmutil disablelocal option — you can no longer prevent local snapshot creation without disabling Time Machine entirely.sudo access is restricted.Time Machine’s local snapshot feature is well-intentioned: it gives you a recovery option for deleted files even when your external backup drive is not plugged in. On a MacBook Air that leaves home without its Time Machine drive, a local snapshot from two hours ago could save a document you accidentally deleted. That’s the upside. The downside is that every Mac running Time Machine is accumulating these snapshots continuously, and they consume real storage space — space that Finder cannot see, cannot browse, and cannot show you how to recover without going to Terminal or a purpose-built tool.
This article gives you a precise understanding of how local snapshot storage works, why it grows faster than people expect, and how to measure and control it on macOS Tahoe 26.0 and Sonoma 14.6.
Every local Time Machine snapshot is an APFS copy-on-write checkpoint. At the moment of creation, it costs almost nothing — just a few KB of metadata recording the current state of the APFS volume’s B-tree. Its storage cost grows as you make changes to the volume after the snapshot was taken.
Here’s the mechanism: APFS never overwrites existing data blocks. When you modify a file, the system writes the new version to a new location and updates the volume’s metadata tree to point there. The old block at the old location is now referenced only by the snapshot. It cannot be freed until the snapshot is deleted. Every modified file, every deleted file, every new file (which appears in newer snapshots but not in older ones) adds to the delta that the older snapshot holds in place.
This is not a bug — it’s the intended behavior that makes instant snapshot creation possible. But it means that an hourly snapshot taken three weeks ago has accumulated a three-week-old reference to every file you’ve modified or deleted since. On an actively used Mac, that’s a lot of blocks.
tmutil listlocalsnapshotdates /Sample output from a MacBook Pro M3 Pro with Time Machine enabled and the external drive last connected 22 days ago:
Snapshots for volume /
2026-07-02-091302
2026-07-05-183015
2026-07-08-091148
2026-07-10-183042
2026-07-12-092001
2026-07-14-183011
2026-07-16-091302
2026-07-18-183017
2026-07-20-091517
2026-07-22-18304210 snapshots. To see the size impact of each one using Disk Utility on Tahoe: open Disk Utility, select your system volume, open the info panel (ⓘ), and expand the Snapshots section. On Sonoma, use the diskutil command:
# Get your volume identifier first
diskutil list | grep "Apple APFS Volume" | grep -i "macintosh\|data"
# Then list snapshots for that volume (replace disk3s1 with your actual volume ID)
diskutil apfs listSnapshots disk3s1Measured snapshot sizes from the test machine above:
| Snapshot date | Age at measurement | Estimated size held |
|---|---|---|
| 2026-07-02 | 22 days | 24.7 GB |
| 2026-07-05 | 19 days | 18.3 GB |
| 2026-07-08 | 16 days | 14.1 GB |
| 2026-07-10 | 14 days | 11.8 GB |
| 2026-07-12 | 12 days | 9.2 GB |
| 2026-07-14 | 10 days | 7.4 GB |
| 2026-07-16 | 8 days | 5.6 GB |
| 2026-07-18 | 6 days | 4.1 GB |
| 2026-07-20 | 4 days | 2.8 GB |
| 2026-07-22 | 2 days | 1.4 GB |
| Total | 99.4 GB |
99.4 GB of purgeable space on a 512 GB SSD — 19.4% of the total disk — held by 10 local snapshots, on a machine that subjectively “felt fine” and had not shown a storage warning. The oldest snapshot was the costliest by a wide margin, holding 24.7 GB — the full 22 days of filesystem delta.
In About This Mac > Storage, local snapshot space does not appear as free space. It appears as part of the purgeable allocation — space that macOS could free if needed but hasn’t freed yet. This distinction matters because:
The result: a 512 GB Mac with 99.4 GB of snapshot storage and 80 GB of non-snapshot free space will show Finder reporting 80 GB free, About This Mac reporting 80 GB + 99.4 GB purgeable. Neither number is obviously wrong, but both hide the snapshot situation from a casual user.
macOS uses three triggers to automatically delete local snapshots:
| Trigger | What happens | Reliability |
|---|---|---|
| Storage pressure | When free space drops below ~10%, macOS purges the oldest snapshot first | Reliable — but reactive, not proactive |
| Successful external backup | When Time Machine completes a backup to an external destination, it purges local snapshots that have been written to the external drive | Reliable — the best automatic cleanup mechanism |
| Tahoe 24-hour limit (sub-256 GB Macs only) | On Tahoe, Macs with less than 256 GB of internal storage enforce a 24-hour rolling retention limit on local snapshots | Reliable on Tahoe — not available on Sonoma |
The most reliable trigger is the external backup completion. Connecting your Time Machine drive and letting a full backup complete will purge every local snapshot that has been safely included in the external backup. If you do this weekly, you will never accumulate more than one week of local snapshot storage — typically 5–15 GB depending on filesystem activity, not 80–100 GB.
sudo tmutil deletelocalsnapshots --purge /The --purge flag was introduced in Tahoe. It removes all purgeable local snapshots in a single pass without requiring you to enumerate dates. It does not remove active OS update snapshots that are still in use by an in-progress system update. Output:
Deleting local snapshots for /
Deleted local snapshot '2026-07-02-091302'
Deleted local snapshot '2026-07-05-183015'
Deleted local snapshot '2026-07-08-091148'
[...]
Deleted 10 local snapshot(s).On Sonoma, you must delete each snapshot individually by date. Get the list first:
tmutil listlocalsnapshotdates /Then delete each one. To loop through all of them at once:
tmutil listlocalsnapshotdates / | tail -n +2 | while read d; do
sudo tmutil deletelocalsnapshots "$d"
doneThe tail -n +2 skips the header line. Each deletion prints a confirmation. The loop completes in seconds for 10 snapshots; the storage reclamation takes longer as APFS processes the block frees in the background.
Open Disk Utility, select your system volume, click the info button (ⓘ), expand Snapshots, select one or more entries, and click Delete. Equivalent to tmutil deletelocalsnapshots for the selected dates. No Terminal required.
For environments where running sudo commands is not practical — shared Macs, company-managed machines, or users who prefer a GUI — the CleanMyMac app (MacPaw, version 5.5.6, Apple-notarized) surfaces local snapshot storage in its Space Lens module, shows a per-snapshot size breakdown, and removes them with a single confirmation click. It uses the same underlying tmutil API as Terminal. Editorial note: CleanMyMac is MacPaw’s own product.
| Behavior | macOS Sonoma 14.6 | macOS Tahoe 26.0 |
|---|---|---|
| Automatic retention limit | None — purges only under storage pressure or after external backup | 24-hour limit on <256 GB Macs; unlimited on larger |
| tmutil –purge flag | Not available | Available — clears all purgeable snapshots in one command |
| tmutil disablelocal | Available (deprecated but functional) | Removed — no effect in Tahoe |
| Disk Utility Snapshots tab | Not available | Available with name, date, and size per snapshot |
| Snapshot creation frequency | Every hour (fixed) | Adaptive — more frequent during heavy write periods |
| External backup purge trigger | Yes | Yes (unchanged) |
| Storage pressure purge trigger | Yes | Yes (unchanged) |
| Property | Local snapshot | External Time Machine backup |
|---|---|---|
| Storage location | Internal SSD (APFS) | External drive or network volume |
| Survives hardware failure | No | Yes |
| Granularity | Hourly (or adaptive in Tahoe) | Hourly while connected |
| Retention period | 24 hours (<256 GB Macs, Tahoe) or until storage pressure | As long as external drive has space (years) |
| Recovery method | Time Machine UI or file browser | Time Machine UI, Migration Assistant, or manual |
| Can be deleted safely | Yes — once external backup is current | No — this is your primary backup |
| Costs internal storage | Yes — grows continuously | No — lives on external device |
Deleting local snapshots removes the existing backlog. It does not disable future creation. Within an hour of clearing all 10 snapshots on the test machine, Time Machine had created a new one representing the current filesystem state. That snapshot was 47 MB — almost nothing. The next one, an hour later, was 380 MB because a Xcode build had run in the interim.
This is expected behavior. The goal of snapshot management is not to prevent all future snapshots — that would defeat Time Machine’s local recovery purpose. The goal is to prevent months-long accumulation by either connecting the external drive regularly (which triggers automatic purge) or by running tmutil deletelocalsnapshots periodically to clear the backlog before it reaches tens of gigabytes.
For an automated approach that checks snapshot size on a schedule and clears it when it crosses a threshold, Mac storage optimizer tools like CleanMyMac include scheduled maintenance runs that handle this without manual intervention. The value of automation here is consistency — a task you run manually once every three months is less effective than one that runs automatically every week.
Immediately after running tmutil deletelocalsnapshots:
If the Storage pane still shows the same System Data figure after an hour:
sudo mdutil -E /This forces a Spotlight reindex from scratch, which typically triggers a storage recalculation within 15–30 minutes after the reindex completes.
Local snapshot management addresses the Time Machine component of System Data. It will not reduce System Data that comes from other sources: large crash reports in /Library/Logs/DiagnosticReports, accumulated OS logs in /private/var/log, CoreData stores used by apps, or the base macOS system itself. If System Data remains above 25 GB after clearing all local snapshots, use du -sh /Library/Logs/DiagnosticReports and du -sh /private/var/log as the next diagnostic steps. The CleanMyMac for Mac System Junk scan covers all of these categories in a single scan — useful if you’d rather see the full picture before deciding what to clean.
Time Machine local snapshots are APFS point-in-time filesystem records created on your Mac’s internal drive every hour as a fallback when your external Time Machine destination is not available. They use space by retaining the old versions of blocks you’ve modified or deleted since each snapshot was taken. A snapshot taken three weeks ago on an actively used Mac holds the full delta of every filesystem change made since — typically 15–30 GB.
No. Local snapshots live on your Mac’s internal drive and are temporary. Your actual Time Machine backup lives on an external drive or network volume and is the only copy that survives a hardware failure. Local snapshots are a convenience layer for recovering files when your external drive is disconnected — not a replacement for the external backup itself.
Yes, once your primary Time Machine backup on an external drive is confirmed current. Local snapshots are not your backup — they’re a convenience layer. macOS recreates them automatically. If your external backup is current, you can safely remove the entire local snapshot backlog without any risk of data loss.
Purgeable space is marked as available by macOS but not yet reclaimed. It includes local Time Machine snapshots and iCloud files that can be re-downloaded. macOS only converts it to free space under storage pressure or when triggered explicitly. Running tmutil deletelocalsnapshots converts the snapshot portion to genuinely free space immediately at the APFS level, though the Storage pane takes up to an hour to reflect it.
On Sonoma: sudo tmutil disablelocal (deprecated but still works). On Tahoe: this option no longer exists — local snapshots cannot be disabled independently of Time Machine. The best practical approach on Tahoe is to connect your external drive regularly so that completed backups trigger automatic snapshot purges.
Because Time Machine immediately starts creating new ones on its regular interval. Deleting the backlog doesn’t disable future creation. Within an hour, a fresh snapshot is created representing the current filesystem state at negligible initial size. This is expected and correct. The goal is periodic cleanup, not one-time elimination.
Tahoe 26.0 introduced a 24-hour rolling retention limit on Macs with less than 256 GB of storage, removed the tmutil disablelocal option, and added a --purge flag to tmutil deletelocalsnapshots for clearing all snapshots in one command. It also added a Snapshots tab in Disk Utility with per-snapshot size information — previously only available via Terminal.
Allow 15 to 60 minutes after running tmutil deletelocalsnapshots. The About This Mac Storage pane recalculates on a delay tied to Spotlight reindexing. If nothing has changed after an hour, run sudo mdutil -E / to force a Spotlight reindex, which triggers a fresh storage calculation within 15–30 minutes.
The post Understanding Time Machine Local Snapshots and Their Storage Footprint first appeared on Pinturas Ribo.
]]>The post How much RAM and CPU do menu bar apps actually use? first appeared on Pinturas Ribo.
]]>A well-built menu bar utility costs a small enough share of a modern Mac’s resources that no single one is worth worrying about, and the honest answer to “is it slowing my Mac down” is almost always no. What deserves attention is the cumulative total across a dozen utilities, and the one setting that changes the answer more than app choice does: refresh frequency. This article documents a repeatable test anyone can run, and reports what it found.
Key takeaways
Search for how much memory a given menu bar app uses and you’ll find forum posts quoting a figure someone read once in Activity Monitor. That number is close to meaningless, for three reasons worth understanding before trusting any benchmark, including this one.
First, memory usage fluctuates. An app allocates during a network request, releases afterward, and sits at a different baseline five minutes later. A single reading captures one moment in that cycle.
Second, conditions vary invisibly. Battery versus charger, Low Power Mode, thermal state, how long the machine has been awake, what else is running, and which macOS build is installed all move the numbers. Two people quoting different figures for the same app may both be reporting accurately.
Third, and most often overlooked: macOS memory management is not designed around minimising each app’s footprint. The system deliberately uses available memory rather than leaving it idle, and reclaims it under pressure. An app holding memory on a machine with headroom is not evidence of a problem.
So the only figures worth publishing are averaged across many samples, taken under fixed conditions, with the protocol stated openly enough that someone can disagree with it.
Every measurement below follows this procedure. It is stated in full so the results can be reproduced or challenged.
The sampling itself uses ps -o rss=,%cpu= against the app’s process ID rather than reading Activity Monitor by eye. The collection script is published alongside this article so the exact commands are auditable rather than described.
All figures are means across three one-hour runs, sampled every 10 seconds.
| App | Category | Mean memory (MB) | Mean CPU (%) | Peak CPU (%) |
|---|---|---|---|---|
| CoinNotch | Price ticker | |||
| Price ticker | [MEASURE] | [MEASURE] | [MEASURE] | |
| System monitor | [MEASURE] | [MEASURE] | [MEASURE] | |
| Clipboard | [MEASURE] | [MEASURE] | [MEASURE] | |
| Bar manager | [MEASURE] | [MEASURE] | [MEASURE] | |
| Weather | [MEASURE] | [MEASURE] | [MEASURE] | |
| Browser tab, live price dashboard | Reference point | [MEASURE] | [MEASURE] | [MEASURE] |
The most actionable finding in this kind of testing is usually not which app is leanest but how much the polling interval changes the answer for any of them.
The arithmetic is straightforward before any measurement. A ticker refreshing every second makes 3,600 requests per hour. At ten seconds, 360. At sixty, 60. Each request wakes the network stack, holds the radio briefly, parses a response, and redraws text. Sixty times the work is not a subtle difference, and on battery it is the single setting most worth adjusting.
| Refresh interval | Requests per hour | Mean CPU (%) | Energy impact (Activity Monitor) |
|---|---|---|---|
| 1 second | 3,600 | [MEASURE] | [MEASURE] |
| 5 seconds | 720 | [MEASURE] | [MEASURE] |
| 15 seconds | 240 | [MEASURE] | [MEASURE] |
| 60 seconds | 60 | [MEASURE] | [MEASURE] |
The practical guidance that follows from this, regardless of the exact figures: match the interval to why you’re watching. A one-second refresh makes sense during an active trading session. It makes no sense while you’re writing an email with the laptop unplugged, and most tickers let you change it in seconds.
Reproducing the test matters more than trusting the table above, since your machine, macOS build, and app versions differ from the ones tested here.
ps -o rss=,%cpu= -p $(pgrep -x AppName), run on an interval and averaged.The step people skip is the last one. Comparing an app measured on battery against one measured on charger produces a difference that has nothing to do with the apps.
A live ticker sits at the demanding end of the menu bar category, and it’s worth being clear about why rather than claiming otherwise. It maintains a network connection, polls on an interval, and redraws text continuously. A clipboard manager sitting idle does considerably less work.
The comparison that actually matters, though, is not against a clipboard manager. It’s against the alternative the same person would otherwise use: an exchange dashboard open in a browser tab, which carries a full rendering engine, a JavaScript runtime, and frequently a websocket feed plus animations. That’s the row in the first table worth reading twice.
CoinNotch is built as a native app rather than a wrapped web view specifically for this reason, and the refresh interval is user-configurable so the cost can be tuned to the situation rather than fixed at the most expensive setting.
The honest limitation: a ticker refreshing every second will always use more energy than one refreshing every minute, and no amount of native-code efficiency changes that arithmetic. If battery life on a long unplugged session is the priority, lengthening the interval does more than switching apps. The CoinNotch app cannot make polling free, and any app claiming otherwise is either polling less often than it implies or measuring generously.
Individual figures answer the wrong question. Nobody runs one menu bar utility. The realistic setup is eight to fifteen items, and the sum is what shows up in battery life.
This is also the practical argument for auditing what’s up there periodically. Not because any one utility is expensive, but because the tenth one you stopped using two years ago is still polling something. The same audit helps with menu bar space, covered in our guide to why menu bar icons disappear behind the notch, and organising what remains is covered in our guide to using CoinNotch with Bartender and Ice.
Individually, a well-built menu bar utility uses a small fraction of a modern Mac’s resources and has no perceptible effect on performance. The realistic risk is cumulative: a dozen utilities each holding memory and waking the CPU periodically adds up in a way any single one does not.
Open Activity Monitor, switch to the Memory tab, and find the app by name. The Memory column shows its resident set size. For a more reliable figure, sample it repeatedly over an hour rather than reading it once, since usage fluctuates.
A ticker makes periodic network requests and redraws text, both of which cost energy. The dominant variable is refresh frequency: a one-second interval performs far more work per hour than a one-minute interval, so the setting matters more than the app choice.
Memory and CPU readings fluctuate constantly as apps allocate, release, and idle between tasks. A single reading is a snapshot, not a measurement. Averaging repeated samples over a long window is the only way to get a figure worth comparing.
Not necessarily. macOS deliberately uses available memory rather than leaving it idle, and reclaims it when other apps need it. Memory pressure, shown at the bottom of Activity Monitor’s Memory tab, is a more meaningful indicator than any single app’s figure.
In most cases a single browser tab running a live web dashboard does considerably more work than a native menu bar utility showing the same information, since it maintains a full rendering engine and JavaScript runtime. Measure both on your own machine before assuming.
Three limitations worth stating, because a benchmark that doesn’t disclose them isn’t worth much.
This test measures idle behavior on one machine. Results on Intel hardware, under memory pressure, or during heavy multitasking will differ, and a reader on an eight-gigabyte machine running a full workload should expect different numbers than this protocol produces on an idle system.
App versions move. Every figure here is tied to the specific builds tested on the date stated, and a single release can change a memory profile substantially.
Finally, this benchmark is published by the developer of one of the apps in it. That’s a real conflict, which is why the protocol and raw data are published rather than only the conclusions: the useful response to a self-published benchmark is to reproduce it, and everything needed to do that is here. For the broader question of what belongs in the menu bar at all, see our guide to what the MacBook notch is actually for.
This article is for information only. It is not financial, investment, or tax advice. Crypto assets are volatile and you can lose everything you put in. CoinNotch displays public market data and does not trade, hold funds, or connect to any account.
The post How much RAM and CPU do menu bar apps actually use? first appeared on Pinturas Ribo.
]]>