Smalltalk › English

Splashcon 18

_‹ Previous Topic Next Topic _›

Classic

List

Threaded

3 messages Options

melkyades

Splashcon 18

Hi Stefan, it's been a long time since we last talked. There's been a lot things going on here this year, so where should I start... Maybe you know we sent a paper to DLS this year with Guido Chari et al about SqueakNOS, so I hope to see you in Boston in November. At the same time I got inscribed to the phd with Diego Garbervetsky as my director, and the work in Bee smalltalk has continued through all the year. I was not planning to show anything from Bee at splash, but then I got an idea for vmil or meta workshops and you may want to participate. Diego G is in, and maybe Guido will also join, so here is the title and a short abstract to let you see what it is about. There is little time until the deadline but I think it's enough. Here it goes.

Cheers,

Javier

---

The tension of metacircular virtual machines

Dynamic languages with high-level abstractions like JavaScript, Python or Ruby commonly make use of language virtual machines. While they provide their users with useful mechanisms like dynamic dispatch or garbage collection, their most popular implementations are implemented in lower-level languages like C or C++, which do not have such helpful mediums. On the other hand, metacircular virtual machines have existed for decades. It could be argued that VMs are still written in C because of performance reasons, but in cases like python or ruby the mainstream implementations, written in C, are orders of magnitude slower than alternative metacircular implementations. What is the reason of such anomaly? are VMs written in C simpler than others written in high-level languages?

Setting away from subjective evaluations, we provide details of objective advantages and disadvantages of implementing metacircular VMs, found out by the authors of this work through their experience in implementing such VMs.

--
You received this message because you are subscribed to the Google Groups "Smalltalk Research" group.
To unsubscribe from this group and stop receiving emails from it, send an email to [hidden email].
For more options, visit https://groups.google.com/d/optout.

Stefan Marr-6

Re: Splashcon 18

Hi Javier:

Good to hear from you.
But we probably want to take this off-list.

Though, I am not opposed to discuss ideas here :)

Best regards
Stefan

> On 3 Aug 2018, at 18:11, Javier Pimás <[hidden email]> wrote:
>
> Hi Stefan, it's been a long time since we last talked. There's been a lot things going on here this year, so where should I start... Maybe you know we sent a paper to DLS this year with Guido Chari et al about SqueakNOS, so I hope to see you in Boston in November. At the same time I got inscribed to the phd with Diego Garbervetsky as my director, and the work in Bee smalltalk has continued through all the year. I was not planning to show anything from Bee at splash, but then I got an idea for vmil or meta workshops and you may want to participate. Diego G is in, and maybe Guido will also join, so here is the title and a short abstract to let you see what it is about. There is little time until the deadline but I think it's enough. Here it goes.
>
> Cheers,
> Javier
>
> ---
>
> The tension of metacircular virtual machines
>
> Dynamic languages with high-level abstractions like JavaScript, Python or Ruby commonly make use of language virtual machines. While they provide their users with useful mechanisms like dynamic dispatch or garbage collection, their most popular implementations are implemented in lower-level languages like C or C++, which do not have such helpful mediums. On the other hand, metacircular virtual machines have existed for decades. It could be argued that VMs are still written in C because of performance reasons, but in cases like python or ruby the mainstream implementations, written in C, are orders of magnitude slower than alternative metacircular implementations. What is the reason of such anomaly? are VMs written in C simpler than others written in high-level languages?
>
> Setting away from subjective evaluations, we provide details of objective advantages and disadvantages of implementing metacircular VMs, found out by the authors of this work through their experience in implementing such VMs.
>
>
> --
> You received this message because you are subscribed to the Google Groups "Smalltalk Research" group.
> To unsubscribe from this group and stop receiving emails from it, send an email to [hidden email].
> For more options, visit https://groups.google.com/d/optout.

--
Stefan Marr
School of Computing, University of Kent
http://stefan-marr.de/research/

--
You received this message because you are subscribed to the Google Groups "Smalltalk Research" group.
To unsubscribe from this group and stop receiving emails from it, send an email to [hidden email].
For more options, visit https://groups.google.com/d/optout.

melkyades

Re: Splashcon 18

In reply to this post by melkyades

Oops!

El vie., 3 ago. 2018 2:11 p. m., Javier Pimás <[hidden email]> escribió:

Hi Stefan, it's been a long time since we last talked. There's been a lot things going on here this year, so where should I start... Maybe you know we sent a paper to DLS this year with Guido Chari et al about SqueakNOS, so I hope to see you in Boston in November. At the same time I got inscribed to the phd with Diego Garbervetsky as my director, and the work in Bee smalltalk has continued through all the year. I was not planning to show anything from Bee at splash, but then I got an idea for vmil or meta workshops and you may want to participate. Diego G is in, and maybe Guido will also join, so here is the title and a short abstract to let you see what it is about. There is little time until the deadline but I think it's enough. Here it goes.

Cheers,
Javier

---

The tension of metacircular virtual machines

Dynamic languages with high-level abstractions like JavaScript, Python or Ruby commonly make use of language virtual machines. While they provide their users with useful mechanisms like dynamic dispatch or garbage collection, their most popular implementations are implemented in lower-level languages like C or C++, which do not have such helpful mediums. On the other hand, metacircular virtual machines have existed for decades. It could be argued that VMs are still written in C because of performance reasons, but in cases like python or ruby the mainstream implementations, written in C, are orders of magnitude slower than alternative metacircular implementations. What is the reason of such anomaly? are VMs written in C simpler than others written in high-level languages?

Setting away from subjective evaluations, we provide details of objective advantages and disadvantages of implementing metacircular VMs, found out by the authors of this work through their experience in implementing such VMs.