10:02 <achamber1> What is the thought on renaming net/http Logger to HTTPLogger?

10:02 <achamber1> it seems kind of annoying I can't use http and logger in the same file

10:03 <achamber1> as it is I need to use an import alias

10:03 <achamber1> but its kind of silly to type something like

10:03 <achamber1> log.Logger, or http.HTTPServer

11:32 <vaninwagen> achamber1, i would in favor of renaming here. the same issue exists with Set, Map, List in collections and collections/persistent.

12:16 <SeanTAllen> i would not be in favor of renaming the persistent collections without some serious convincing. renaming net/http logger seems somewhat reasonable to me, although i'd want to read more about the pain in the RFC.

12:18 <vaninwagen> SeanTAllen totally agree! just wanted to mention a place where i stumbled over equal type names before.

13:02 <Candle> Should the core libraries have a target to not have duplicate basic names? I still have flashbacks from when I was learning Java; and encountering java.util.List and java.awt.List!

13:04 <Candle> Another approach might be to enable aliasing of types; though that's a change to how the 'use' statement works.

13:11 <vaninwagen> i like this target i have to admit

13:13 <SeanTAllen> i dont like the target

13:14 <SeanTAllen> what do you mean by "Another approach might be to enable aliasing of types; though that's a change to how the 'use' statement works." Candle ?

13:15 <SeanTAllen> do you mean something like use new_module_name = "actual_module_name" ?

13:17 <Candle> SeanTAllen: 'use h = "http/client"' becomes something more like 'use HTTPLogger = "http/client.Logger"'

13:17 <Candle> So you're just aliasing that type from the http library.

13:17 <SeanTAllen> im quite happy with h.Logger

13:18 <SeanTAllen> feel free to open an RFC for the other

13:18 <SeanTAllen> this has never been an issue for me in our codebase

13:18 <Candle> Personally, as soon as I thought of it I was talking myself out of it - it complicates the language even more, and obsfucates code that uses it.

13:26 <vaninwagen> my only argument for avoiding equal typenames in stdlib is to reduce possible friction/surprise for newcomers - i don't know if it is actually worth the changes. actually i am also living with the same situation in scala right now and it never really bugged me.

13:45 <Candle> vaninwagen: On the flip side; it introduces people to aliasing in a use statement!

13:46 <vaninwagen> :) good point

16:21 <jemc> it can be hard to choose whether to prefix types with the library name or not - `use "http"` works better if the type name is `HTTPServer`, and `use http = "http"` works better if the type name is just `Server` - it can be hard to know what users of your library will want to do

16:21 <jemc> this is one reason why I wish pony's include system was a bit more like some other languages with nested namespaces that you can import in various ways

16:23 <jemc> for example, something like `HTTP::Server` or `HTTP.Server` representing a `Server` type inside an `HTTP` namespace

16:23 <jemc> this would give the best of both worlds, I think