That’s because true messaging is only done by Ruby, Eiffel and ObjC. Ironic that most edgelords hate it for being associated with Apple, but it fulfills the initial ideas that Kay el al. had. I think that’s because the current generation of devs is completely clueless about these things, so they don’t appreciate the mentality.
@@gareginasatryan6761 You can still program this way in most OO languages AFAIK. I did some Java and Javascript, and it looks very familiar. Ofcourse a lot of shit was added to Java later (like functional programming/lambdas), but you don't HAVE to use that new stuff. You can still keep it strictly OO.
@@gareginasatryan6761 It's because they were closer to the source of the pain points which happend before OOP. Similar to how geographically americans makes the best tech, it might be because the US created so many tech before it. Similar to European super cars or fashion.
In case you didn't notice the reference at 37:45, David Ungar is one of the creators of the Self language. So, we have one of the fathers of classical inheritance in an interview with one of the fathers of prototypal inheritance.
Now I understand why people back then and still today get confused between abstract data programming and true Alan-Kay-style Object Oriented Programming. They commingled and overloaded many terms and skipped over key concepts
This presentation assumed a ton of tacit knowledge, context and key phrases used sloppily Looks like this was an internal demo but was sold as holy gospel which was open to tremendous strange interpretation and obfuscation
Did he foresee the problem of AbstractFactoryGeneratorBuilderBuilder???? But seriously speaking, Erlang implements the concepts of message passing by treating processes as mutable objects. Within a Erlang process, variables are immutable.
If messages are the most important part of “object oriented” programming, why isn’t “message” in the name? We need poets to name things for engineers… I’m creating a new theme in keynote for my presentations to match this style.
Most modern takes of "OOP" are not actually OOP. The most important part of OOP is message passing and none of the top 10 programming languages to this day stay true to that.
Glorious presentation. This is conceptually so elegant that it hurts to go back and see what we have today
That’s because true messaging is only done by Ruby, Eiffel and ObjC. Ironic that most edgelords hate it for being associated with Apple, but it fulfills the initial ideas that Kay el al. had.
I think that’s because the current generation of devs is completely clueless about these things, so they don’t appreciate the mentality.
@@gareginasatryan6761 You can still program this way in most OO languages AFAIK. I did some Java and Javascript, and it looks very familiar. Ofcourse a lot of shit was added to Java later (like functional programming/lambdas), but you don't HAVE to use that new stuff. You can still keep it strictly OO.
@@gareginasatryan6761 It's because they were closer to the source of the pain points which happend before OOP. Similar to how geographically americans makes the best tech, it might be because the US created so many tech before it. Similar to European super cars or fashion.
This perhaps is a must-watch for every programmer starting on OOP paradigm, thank you!
Or any programmers for that matter ;)
I have been looking for this presentation for years! Especially around minute 43....
In case you didn't notice the reference at 37:45, David Ungar is one of the creators of the Self language. So, we have one of the fathers of classical inheritance in an interview with one of the fathers of prototypal inheritance.
Interesting Dan defines the property “simulation” @27:25 which is known to me as the Liskov Substitution Principle. Essential!
Now I understand why people back then and still today get confused between abstract data programming and true Alan-Kay-style Object Oriented Programming.
They commingled and overloaded many terms and skipped over key concepts
This presentation assumed a ton of tacit knowledge, context and key phrases used sloppily
Looks like this was an internal demo but was sold as holy gospel which was open to tremendous strange interpretation and obfuscation
Thank you so much for sharing this!
very articulate
Mind opening
Gosh, thanks Napoleon! Now I understand true OOP and not that C++ / Java bastardization of it.
Alan Kay tried to tell everyone but was shouted down for “performance”
Did he foresee the problem of AbstractFactoryGeneratorBuilderBuilder????
But seriously speaking, Erlang implements the concepts of message passing by treating processes as mutable objects. Within a Erlang process, variables are immutable.
That man looks like an AI composite of Barrack Obama and Napoleon Dynamite
I was about to make a comment about this guy at parties, but then I noticed the paisley tie and think he secretly hits the disco every saturday.
Ok
ty for sharing
Where can I get the full series
If messages are the most important part of “object oriented” programming, why isn’t “message” in the name? We need poets to name things for engineers…
I’m creating a new theme in keynote for my presentations to match this style.
Most modern takes of "OOP" are not actually OOP. The most important part of OOP is message passing and none of the top 10 programming languages to this day stay true to that.
Alan Kay himself said that he regretted not calling it Message-Oriented programming or some variant of it.
The amount of freely available information is directly opposed to the level of common sense in a given society.
the f is this ?
It's a lecture about the object-oriented paradigm for programming.