現在のIllustratorデスクトップ版で「選択範囲に整列」を設定し、複数のオブジェクトを対象に「垂直方向中央に整列」または「水平方向中央に整列」を実行すると、それぞれのオブジェクトの中央の座標を平均した位置で揃います。

一方で、Illustrator iPad版で同じ操作を行うと異なる結果になります。iPad版の「中央に整列」では、座標の平均ではなく、選択範囲（バウンディングボックス）の中央でオブジェクトが揃うしくみです。
自分が確認した限りではPhotoshopやInDesignもこちらの仕様です。

• 「選択範囲に整列」というオプションの名称からも、座標の平均で揃うことは推測しにくい
• iPad版のように、選択範囲の中央に揃うほうが直感的でイメージしやすい
• デスクトップ版とiPad版の同じ機能がそれぞれ違う動きをするのは避けるべき
• 選択範囲の中央に揃う仕様はその他のAdobe製品でも採用されている

上記の理由から、Illustratorデスクトップ版でも「中央に整列」では選択範囲の中央に揃える仕様に変更してほしいです。

---------（used the translation tool）---------

In Illustrator on desktop, select multiple objects, and in the "Align to Selection" setting, execute "Vertical ( or Horizontal ) Align Center".

The objects will then be aligned at the average of their respective center coordinates.

However, the same operation in Illustrator on iPad produces a different result. The iPad's "Align Center" aligns the object at the center of the selection (bounding box), rather than at the average of the coordinates.
As far as I have confirmed, Photoshop and InDesign also have this specification.

• It is difficult to infer from the name of the "Align to Selection" option that it will be aligned with the average of the coordinates.
• It is more intuitive and easier to imagine the result if it is aligned to the center of the selection, as in the iPad version.
• It is not good to have the same functionality on desktop and on iPad, but with different results on each.
• Alignment to the center of the selection is also used in other Adobe products.

For the above reasons, I would like the Illustrator desktop version to change the specification to align to the center of the selection.

We have to work with a strange legacy dichotomy when making a selection in Layers is NOT equal to making a selection on canvas and vice versa.

No other vector application does it like this:
XD, Affinity Designer, Corel Draw, Figma, Xara, Inkscape...

Other apps which have layered structures don’t do this also:
Photoshop, Blender, 3DS Max...

InDesign and Illustrator are lonely black sheep in the flock of apps.

We need a mode (as an option) to be able to sync both selection methods.
Clicking an object on canvas should highlight the name of it in Layers.
Clicking a name of an object in Layers should select this object on canvas.

Extended reading:

These two selections are like the dark matter and the normal matter — they don’t influence each other and can’t be converted into one another.
Here is the request about it: https://illustrator.uservoice.com/forums/333657/suggestions/32505679

When an object is selected on canvas, Layers does not get expanded to display it, and we have to click the tiny button to see it in the document structure.
Here is the request about it: https://illustrator.uservoice.com/forums/333657/suggestions/39104935

Yeah, we can wrap it an an action and spend a precious F-keys to robotically press it after each selection, which is twice dumb, because an action which has only one operation in it is a bad action. Every command in Illustrator should be assignable to a hotkey.
Here is the request about it: https://illustrator.uservoice.com/forums/333657/suggestions/31609237

And also we have three places to click in Layers to select an object:
1. Names, largest and most easy to click areas — clicking does nothing to select objects though
2. Target buttons, small circles, which are in fact just indicators for the custom appearance applied — clicking them both targets and selects an object, all their children (except locked and hidden), and it’s the only way to select objects on canvas via the panel. Also it support selecting ranges with Shift held — after we begged for it for 3 years (here is the request about it: https://illustrator.uservoice.com/forums/333657-illustrator-desktop-feature-requests/suggestions/19196674-select-more-than-one-layer-at-once).
3. Selection markers, which are indicators too, but again can be used as buttons... but without the ability to select ranges with Shift. In fact, Shift here works as Ctrl/Cmd should, and Ctrl/Cmd does nothing. I have no hypothesis why. Markers can be displayed as small and large, meaning partial (some children are selected) and full selection (all children selected).

You can read about these columns here: https://helpx.adobe.com/in/illustrator/using/layers.html
Note that this article says nothing about the dichotomy.

Finally, I can see why the current approach can be useful. This allows us to put some objects in an existing group, to reorder, to move to another layer, to switch visibility or lock in mass — via a stupid dialog... without messing up with what is selected on canvas.

Is it worth the hustle though? This can be done with a context menu for Layers panel (here is the request about it: https://illustrator.uservoice.com/forums/333657-illustrator-desktop-feature-requests/suggestions/31956982-contextual-menu-with-layers-palette-same-as-phot) with 'Lock selection', 'Hide selection', 'Isolate selection' (for objects from different groups, here is the request about it: https://illustrator.uservoice.com/forums/333657-illustrator-desktop-feature-requests/suggestions/31540384-isolate-selected-objects-without-having-to-group-t), Copy, Group/Ungroup, etc.

There are standards used throughout computing that inform how field values work and Illustrator screws this up terribly.

When a field is not in focus, clicking on the field or the field value brings the field and value into focus. This automatically should select the entire user-editable contents of the field value. This allows you to click and type a new value to change the value.

Illustrator instead changes the cursor to an insertion point when hovering over a field value. Clicking in the field value, places the insertion point at the location the pointer was at. If you then type the value you want, you will NEVER get the result that you want. This is not okay.

Illustrator requires you to click in a field (places the insertion point in the middle of the field value), then click again at the beginning of the field value (which is super problematic when the field value contents are bigger than the field input space) and drag to the end of the field value BEFORE you can enter your new value. Most often, you have to select part of the field value, delete it and then press the del or delete key until you get the full contents gone so you can enter your new value.

And what is more infuriating is when you enter a value with units, for example, without looking you press tab to advance to the next field and get a modal dialog box stopping you from doing anything else just to tell you that 300em4px is not a valid value. This is infuriating.

And this is problematic through Illustrator's numerical field value selection. If you have just entered one value and press tab, the next field value should be selected by default. Here, Illustrator handles the focusing of the field with autoselecting the field value correctly. But that's problematic if you just went through the reprogramming to convince yourself that because of Illustrator's failure to understand the focus event, you need to manually select the field contents, now you are doing extra steps you don't need to do. Unbelievable that an app could ***** it up no matter how it is used.

Current Problem: Illustrator places the insertion point on a newly focused field instead of selecting the entire contents, only when selecting from the mouse or touch interface. It should, instead, follow the same procedure that it uses to focus a field when tabbing from one field to another: Select the entire contents of the field value when focusing.

Ai can display triangular meshes alright, but not edit.

We can generate one by expanding modern 3D object with the Render as Vector option turned on, we can have one generated after expanding an object filled with Freeform Gradient (which is just the live container with control points for creating one, with no control over the density).

But once they are created, we get 'Non-Native Art' bricks we can do nothing with — no Recolor, no direct edit with mesh tools, only scale and rotate and distort.

Not enough! We need you to support these fully.
And make some good tools for editing meshes. Sure, we have meshtormentor.com — but how many users are aware of it?

There is a problem with global gradient swatches, described at https://illustrator.uservoice.com/forums/908050/suggestions/44461230

Illustrator applies changes to all copies of a gradient in a document when this gradient is changed via a script. It makes sense from a practical point of view, but is devastating for scripts in general.

We can’t duplicate gradients with a script properly, like we can do with any other swatch, because when it's done, the gradient swatch get forcibly added to Swatches and then corrupt the art which uses them when deleted... and it does not break this global link anyway.

At the moment we copy a gradient into the swatch, it will be associated with the first object. If we repaint the swatch to use it for another object, the first object will change its fill momentarily. Just like the video from the issue above shows — https://illustrator.uservoice.com/assets/235626175/linked-gradients-bug.mp4

So we 'instance' gradients instead of 'duplicating' them.
Creating a new gradient via script also creates a new swatch. Deleting either the new gradient or the new swatch also deletes the other. If the need is to get rid of the swatch but not the gradient, that doesn’t look to be possible via with the current APIs. Perhaps we need a new API to delete just the swatch and not the associated global (gradient, pattern etc.).

Having this will allow us to keep gradient changes on one object from affecting other objects. To duplicate, a new gradient will be created and a copy of all gradient properties (type, number of stops, stops) will be performed, via existing app.activeDocument.gradients.add()
Then we will be able to delete the swatch with a new API without crashing Illustrator.

To edit a brush stroke, we have to do the following:
1. Edit the panel
2. Grad the brush and drag it onto the canvas
3. Edit
4. Alt+drag it over the original brush
5. Think about applying changes to existing art if the brush is already used

Meanwhile, editing a symbol looks like this:
1. Double-click a symbol
2. Pass the warning (or disable it once)
3. Edit a symbol
4. Double-click on the canvas or hit Escape to finish

Looks like almost the same length... but clicking is faster than dragging and aiming. Why no use the same approach for brushes?
Double-click the brush stroke in the panel, get a Brush Option dialog that is not modal and allows to work with the canvas below, give a brush stroke in the isolation mode, finish on clicking OK/Cancel or Back, with Save confirmation.

I work in printing, we just got 15 files in using freeform gradients where the majority of the gradient is 80% opacity and the edges of the shapes are 10% opacity. After printing samples, the customer has decided to change the 80% to 100%. Normally, this is very quickly executed with solid colors and within gradient meshes by selecting one object or point and choosing "select same opacity" from the menu. But this feature is not available in the freeform gradients. I now have the task of individually changing the opacity on over 1500 points across all of the files. With how popular the freeform gradients are becoming, I expect this issue may arise more often as many artists will need to perform minor tweaks which should take literally seconds, but will turn into multiple hours of tedious work. Not to mention that the opacities are similar enough that it leaves a large margin of error if any of the color points are missed. because of human error. Please let the computer do the dirty work!