text stringlengths 0 2.35k |
|---|
**Ronna Steinberg:** Well, I kind of agree... \[laughs\] But then I went into the rabbit hole, because what I remembered from my university was my professor, Jeff Rosenschein, explaining that object-oriented is about being able to send messages to objects. And I checked the origins of the idea and it appears to be the ... |
In the workshop we do things very Java-like. Exactly the kinds of things that you're not supposed to do... Or it's not really that you're not supposed to do, but we really try to express the same types of ideas, but keeping loyal to, I guess, the Go way of doing things still. So yeah, at the very least it's gonna be a ... |
I also think that there is a lot of room for people to decide within their teams their own best practices, their own practices, how they want to do things. I do think there should be a discussion, and I also think that it should be a little bit more lively than it is now, where I think the veterans sort of dictate how ... |
You asked me about quotes that I got... Somebody quoted Rob Pike to me. I tried, but I couldn't find the quote... But somebody quoted Rob Pike to me, so apparently Rob Pike did say something about object-oriented \[unintelligible 00:16:06.22\] or something of this sort... But yeah, I didn't get to the source. |
\[16:17\] What I do see is a lot of people -- this is where I think this is coming from... How people write APIs, how people write their handlers - it's all pretty much functional programming. And the way that this is designed, most people don't actually use an interface in these situations, and even if they do, they k... |
I think that's where it's coming from, because it became a language that is so strong in that space. And in that space you really don't need a lot of running objects. Whatever objects you have are probably going to die soon anyways. And if we think also, those APIs - it sounds funny, but objects don't live long in APIs... |
Their structs are limited to fields, like configurations, and they don't have many methods. But then if you look for instance at how they set up their repositories, or how they set up their models, or how they set up these things, because we didn't have generics up until very recently, they don't really use objects the... |
**Break:** \[18:21\] to \[20:13\] |
**Ian Lopshire:** So you going through that, what you've just said, made me think maybe my idea of what OOP is is pretty heavily influenced and biased by writing Go. I haven't written anything but Go in a long time. Probably two or three years at this point. |
So going through that - you're right, I think a lot of the OOP stuff that was taught in school I don't use anymore, but I still think of what I'm doing as object-oriented. And maybe I'm wrong. In my mind, object-oriented is just encapsulating a set of data and having methods on it that -- it's like self-contained, righ... |
**Ronna Steinberg:** That's interesting... You said it. You actually said it, "encapsulation". So it's about encapsulation, abstraction, and later came in generalization. So to encapsulate, you don't need hierarchy. Or you can have a composition, essentially, and that would work pretty well too, to encapsulate. But ess... |
If we have encapsulation, which I think we do in Go, and we are able to define abstractions, which we have interfaces for... And by the way, there is a case - it's kind of interesting, I've found this code snippet that was exactly what I needed for this workshop at the time, of C++ code. I really loved it, because it s... |
By the way, Go always had generics -- actually, that would be my unpopular opinion. Go always had generics, because map is a generic type. It maps from a key that is a generic type to a value that is a generic type... I see you're sipping, you're drinking, so I know that you disagree... |
**Ian Lopshire:** No, I a hundred percent agree. |
**Ronna Steinberg:** Oh, you do. Okay. \[laughs\] The slide says that we're always generic... We just couldn't define our own, so it was like good for the 80% of the use cases, but then we had our own use cases that we needed it, and we just didn't have anything. So you will talk now, and I will sip a drink. |
**Ian Lopshire:** \[laughs\] No, I definitely agree with you that the maps and the slices are generic, literally since day one. And like you said, it is that 80% use case, right? I don't know how to tie this back to OOP stuff now... |
**Ronna Steinberg:** Why I think it plays a role in this is because what people did in those cases is they would write a lot of functional code, procedural code etc. and did not use necessarily generic types. I think we've had a lot of workarounds. I think now we can eliminate them. I think the language is very mature ... |
**Ian Lopshire:** That makes sense to me. |
**Ronna Steinberg:** Natalie, \[unintelligible 00:25:21.26\] |
**Natalie Pistunovich:** No... One thing is on a Zoom call, another thing is on a podcast. It's a whole new level. I've found the quote from Rob Pike about Go being yes or no object-oriented. So it's a bit of a thread... So it starts with "Whenever someone from Java, or C++, or C\# (I'm looking at you, Ronna) comes to ... |
This is the insight (which goes all the way back to SmallTalk) on which the object-oriented type system, including the interface model, is built. Stopping at "struct == class" misses much of what makes Go work." So what are your thoughts? |
**Ronna Steinberg:** I agree... It sort of works very well with everything that I learned recently about this, honestly. It feels like Go is more naturally object-oriented than languages that I worked with before. I had this discussion with my boyfriend, because of the workshop and the amount of resistance that I was g... |
**Natalie Pistunovich:** Does he think Go is object-oriented? |
**Ronna Steinberg:** Yes. So his answer was "Well, of course it is." It wasn't even a question in this mind. And then he told me "Well, you know how the JavaScript people say that a prototype is the best way to define object-oriented." And my take-away from this is that every night I go to sleep next to a man who quote... |
**Ian Lopshire:** The quote there, where it says it separates the idea of data and behavior - I think that's the important part of that area... But I'm having a hard time articulating exactly why in my head. |
**Ronna Steinberg:** Do you think that that is a good way to express object-oriented, abstracting away the data from the functionality? |
**Ian Lopshire:** Yeah, I think going to what you were saying, where Go is almost a more pure version of object-orientation, I think that's what leads to it; the idea that -- I don't know, I'd have to really sit down and think about it, but that's where my mind is going. |
**Ronna Steinberg:** \[27:57\] I think that I've found something that a person that wrote me about Go not being an object-oriented language might be related to... So I searched for Rob Pike and object-oriented and I've found Wikipedia. On Wikipedia he is criticizing object-oriented. So it's possible that that was what ... |
**Natalie Pistunovich:** Do you want to read this out and share this later in our show notes, so everybody can go back to this, too? |
**Ronna Steinberg:** Yeah. I think it's funny... \[laughs\] I mean, clearly... Yes, I will send it to you, and we can include that. So he criticized object-oriented for being incredibly heavy, I guess... And that sort of works pretty well with those quotes about objects being lightweight in Go. I think I agree. I think... |
By the way, something that I really love is we don't have enumerators, but we do have -- like, it's very easy for me when I look at pseudo-code with those red/black trees, and coloring notes in a graph, and stuff like that - this is the kind of stuff that I teach to people who don't have (let's say) the classical backg... |
**Ian Lopshire:** I definitely could not code up a red/black tree without googling something right now. |
**Ronna Steinberg:** Yeah, obviously. No, I still do these exercises with people so they get used to the idea. And then if they can do this, they can do anything... Stuff like that, essentially. But yeah, there are way too many edge cases in a red/black tree. It's not okay. \[laughs\] \[unintelligible 00:30:56.00\] |
**Ian Lopshire:** Someone should write a generic library for it once, and then never touch it again. \[laughter\] |
**Ronna Steinberg:** That sounds very good. So that's a rant about people who give impossible tests to complete beginners in coding. |
**Natalie Pistunovich:** Or tell us how you do this right, when you're hiring for your team. |
**Ronna Steinberg:** So I don't know that I do this right, let's start with that... But I try to decide what I want, and then I don't test people on stuff that I don't want. What I will do - for instance, if I want to know how thorough somebody is going to be, then I will dive with them on something that they do know, ... |
Let's start with the very basics - if somebody tells me that they have worked with MySQL, then I will test them heavily on MySQL until I reach something that either they don't know, or... To get a sense of how far into the rabbit hole they went. Or any other topic. This one is kind of normal. |
\[32:05\] But it all depends on their profile. It really all depends on their profile. We said earlier that I am the sum of the opportunities that were given to me... And I don't think I'm a bad developer. We all want to hire somebody who cannot do what we can't do. \[laughs\] Essentially, I feel like we are raising th... |
For instance, if I can't do something and I am going to search for somebody who's going to be able to do it, that's a little bit unfair. And I feel like we are doing this all the time; we're sort of like raising the bar, raising the bar, raising the bar, to essentially maybe make up for our own disadvantages. |
So I try to not be unreasonable when I interview somebody. I do try to see if they can learn, there in the interviewing process, for instance. So I will present them with something that they will have to think about, internalize, and then spit out some information that could be wrong, could be right... It could be very... |
For instance, I work with billing people, and I'm not necessarily familiar with those concepts, so I will introduce something that has something to do with what we do in the domain... And it's fine that they don't know this. And then I will ask them "How would you do it, given those constraints? What would you do?" So ... |
So yeah, I am the sum of the opportunities that were given to me, and I am a huge believer that if you give people an opportunity, they're not gonna let you down, especially if they've been begging for work since forever... And I do think that you can bank on people's loyalty in those cases. I think for a lot of people... |
Even now, if I go for a job interview, I am never going to apply for something that I already know. I want to grow just as much as everyone else, so I'm always going to go and try to convince somebody that they need to invest in my potential. |
I am very, very similar to a complete beginner, because we all have to go around and essentially convince people to give us a chance to prove that we can do something that we have not shown before that we can do. It's so obvious to people that "Okay, Ronna after 20 years is going to be able to do this." But it shouldn'... |
**Break:** \[35:23\] to \[38:26\] |
**Natalie Pistunovich:** You mentioned briefly about your background, starting with C++, and also a little bit on how you've come to the idea of doing this workshop. Do you want to elaborate a little bit on giving the context of what brought you on the shorter-term and on the longer-term to this idea? |
**Ronna Steinberg:** I don't know, there was this jobs fair in April where I sat down with the organizer of GopherCon Europe, and I asked, point-blank, "Why have you never asked me to do a workshop? You know I do this..." \[laughs\] And then she told me "Well, it's a lot of work..." and I said "Well, yeah, but this is ... |
**Natalie Pistunovich:** Why did you want to do it? How did you even come to think of it? |
**Ronna Steinberg:** Because that's what they do, and I actually don't think that those patterns that they are using serve my team very well. Actually, it's not really my team, it's my former team, because since about a month ago we reorganized... We used \[unintelligible 00:39:26.05\] in a way that was not really work... |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.