| 19 Apr 2021 |
 Yorick Peterse | lots of work to do :) | 21:10:54 |
| 25 Apr 2021 |
 tsmall | Hey Yorick! I noticed today that the manual uses the "class" keyword to define objects/classes, but the 0.9.0 release notes and compiler use "object". Is the manual behind or ahead of changes? | 23:55:17 |
| 26 Apr 2021 |
| Yorick Peterse | tsmall: Ah yeah, on master the object keyword is replaced with class | 19:27:13 |
| 28 Apr 2021 |
 kiti_nomad | Both class and type are translated as 类型 in Chinese, but class is usually a template for object.Type refers to the type system | 16:22:28 |
| kiti_nomad | When we declare an object or a variable with a type, we usually mark it with a type signature.The code block of an object is labeled as Object, which I think is more unified and harmonious | 16:24:29 |
| Yorick Peterse | The keyword object is confusing though, because both classes and their instances are objects | 20:00:45 |
| Yorick Peterse | it's also different from most conventional languages, and not necessarily for the better | 20:00:59 |
| Yorick Peterse | Using class also makes it easier to talk about, at least in English. e.g. "the class of a string is String", whereas otherwise you have to say something like "a string is an instance of the String object" | 20:01:36 |
| 30 Apr 2021 |
|  Conrad Steenberg joined the room. | 20:50:50 |
| 5 May 2021 |
| kiti_nomad | In some languages, object refers to the creation of a singleton object, and class refers to the creation of a type template | 20:02:59 |
| 12 May 2021 |
| kiti_nomad | Inko's pattern matching is the same as kotlin. I think this is not powerful enough. I think inko should adopt a combination of Scala and kotlin's pattern matching, that is, the state of an object can be extracted directly in inko, which is a field | 20:48:19 |
| kiti_nomad | In inko, operators are functions, which is great. I think the traditional oop syntax for sending messages can be used. Maybe you can also use infix expressions like Scala and omit the dots. | 20:51:34 |
| 13 May 2021 |
| Yorick Peterse | In reply to @hellonomad:matrix.org
Inko's pattern matching is the same as kotlin. I think this is not powerful enough. I think inko should adopt a combination of Scala and kotlin's pattern matching, that is, the state of an object can be extracted directly in inko, which is a field This makes pattern matching far too complicated, both for the user and the compiler. Pattern matching in Inko is largely meant for graph traversal, parsing, and similar tasks; for that it works fine | 20:12:40 |
| Yorick Peterse | In reply to @hellonomad:matrix.org
In inko, operators are functions, which is great. I think the traditional oop syntax for sending messages can be used. Maybe you can also use infix expressions like Scala and omit the dots. What do you mean? You don't need dots for operators such as foo + bar. Or do you mean that foo bar should translate into foo.bar()? If so, that won't work syntax wise; the resulting syntax would be too ambiguous | 20:13:44 |
| Yorick Peterse | In reply to @yorickpeterse:matrix.org
This makes pattern matching far too complicated, both for the user and the compiler. Pattern matching in Inko is largely meant for graph traversal, parsing, and similar tasks; for that it works fine The single ownership branch introduces some assignment pattern matching (e.g. you can dolet (a, b) = Pair.new(10, 20), but that really only exists to make for loops more pleasant to use | 20:14:39 |
| Yorick Peterse | I'm specifically avoiding all the complexity and boilerplate that Scala has. I think it's one of several reasons why Scala isn't all that popular, and why Kotlin managed to take away many of its users | 20:15:23 |
| 22 May 2021 |
| kiti_nomad | I thought about it. My idea is meaningless. Usually an object calls an instance method to use dot. However, some languages do not need dot, such as dog eat: meat bone | 22:16:11 |
| kiti_nomad | Usually dog.eat(meat, bone); | 22:17:30 |
| kiti_nomad | This is just a personal preference and has no practical meaning | 22:19:10 |
| kiti_nomad | Because I think it takes a long time for a language to become popular, and the grammar of natural language that looks like humans will attract those young people, because people who like c-like grammar may be dead. | 22:21:36 |
| kiti_nomad | In China, even primary school students are learning python | 22:24:34 |
| 29 May 2021 |
| Yorick Peterse changed the room topic to "The Inko programming language." from "General purpose chat room about the Inko programming language.". | 20:25:47 |
| Yorick Peterse changed the room name to "Inko" from "Inko (lang)". | 21:44:11 |
| 30 May 2021 |
|  brendanzab joined the room. | 02:22:34 |
| 2 Jun 2021 |
| Yorick Peterse | error: aborting due to 404 previous errors; 9 warnings emitted seems there's still some work to be done before I get the VM working again :P | 23:12:21 |
| Yorick Peterse | that number hovers between 380 and 400, depending on what state of refactoring I'm in | 23:12:37 |
| 3 Jun 2021 |
 codec_abc | Ahah a feeling we have all experimented at some point | 07:59:08 |
| codec_abc | Bon courage :) | 07:59:49 |
| 6 Jun 2021 |
| Yorick Peterse | Quite excited about all this refactoring. It will be a while before I can run any Inko code again, but progress is there | 13:17:01 |
| Yorick Peterse | less excited about having to adjust the Ruby compiler to support this new setup. Very tempting to rewrite it, but that will be a huge time sink | 13:17:21 |
