# Saturday Ada visibility puzzle
**URL:** https://forum.ada-lang.io/t/saturday-ada-visibility-puzzle/4793
**Category:** General
**Created:** [October 3, 2026, 10:45am UTC](https://forum.ada-lang.io/t/saturday-ada-visibility-puzzle/4793 "2026-10-03T10:45:05Z")
**Posts on this page:** 6
**Page:** 1
### Author:  [@dmitry-kazakov](https://forum.ada-lang.io/u/dmitry-kazakov)
#### Post date: [October 3, 2026, 10:45am UTC](https://forum.ada-lang.io/t/saturday-ada-visibility-puzzle/4793/1 "2026-10-03T10:45:05Z")
Is this legal?
```ada
with Standard.Ada;
procedure Foo is
begin
null;
end Foo;
```
---
### Author:  [@Nordic\_Dogsledding](https://forum.ada-lang.io/u/Nordic_Dogsledding)
#### Post date: [October 3, 2026, 5:56pm UTC](https://forum.ada-lang.io/t/saturday-ada-visibility-puzzle/4793/2 "2026-10-03T17:56:17Z")
If there is a user defined package `Standard` with child unit `Ada`, then yes, of course.
The fully expanded name of this package `Standard` woult be `Ada.Standard`.
Thou shalt not name an Ada unit Standard.
---
### Author:  [@dmitry-kazakov](https://forum.ada-lang.io/u/dmitry-kazakov)
#### Post date: [October 3, 2026, 6:33pm UTC](https://forum.ada-lang.io/t/saturday-ada-visibility-puzzle/4793/3 "2026-10-03T18:33:46Z")
> [@Nordic\_Dogsledding](#):
>
> Thou shalt not name an Ada unit Standard.
God forbid, that blashpemy.
But what would you say about this:
```ada
with Ada;
package W renames Standard.Ada;
```
(The question is about the extent of the rule that all top-level packages are children of the package Standard.)
---
### Author:  [@Nordic\_Dogsledding](https://forum.ada-lang.io/u/Nordic_Dogsledding)
#### Post date: [October 4, 2026, 5:19pm UTC](https://forum.ada-lang.io/t/saturday-ada-visibility-puzzle/4793/4 "2026-10-04T17:19:36Z")
> [@Nordic\_Dogsledding](#):
>
> The fully expanded name of this package `Standard` woult be `Ada.Standard`.
Excuse me for writing nonsense. Such a package’s full name would be `Standard.Standard`, but I doubt this could be written. (Any library unit’s `X` full name is `Standard.X`.)
---
### Author:  [@Nordic\_Dogsledding](https://forum.ada-lang.io/u/Nordic_Dogsledding)
#### Post date: [October 4, 2026, 5:22pm UTC](https://forum.ada-lang.io/t/saturday-ada-visibility-puzzle/4793/5 "2026-10-04T17:22:20Z")
> [@dmitry-kazakov](#):
>
> ```ada
> with Ada;
> package W renames Standard.Ada;
>
> ```
>
> (The question is about the extent of the rule that all top-level packages are children of the package Standard.)
Since a user defined package `Standard` is not visible, declaring `W` should be well defined. (I did not ask the compiler.)
---
### Author:  [@dmitry-kazakov](https://forum.ada-lang.io/u/dmitry-kazakov)
#### Post date: [October 4, 2026, 8:09pm UTC](https://forum.ada-lang.io/t/saturday-ada-visibility-puzzle/4793/6 "2026-10-04T20:09:56Z")
> [@Nordic\_Dogsledding](#):
>
> I did not ask the compiler.
I did. GNAT rejects with `Standard.Ada`, but accepts `rename Standard.Ada`.
The argument of **with** is _library\_unit_\_name. The argument of **rename** is _package_\_name.
My theory is that the name of a library item must be in “canonical” form. The package Standard is a library item without a library item name, yet it has a package name.