Hello, there is an explosion of different kinds of VM we're building today.. and i think it would be nice to use different names for them. Currently i am using following naming scheme: - VMs based on StackInterpreter: StackVM - VMs with JIT Cog - VMs with JIT + MT CogMT i am open for suggestions of giving better/proper names. Maybe for end-user it is not important to use different names (squeak.. yeah.. this is squeak.. who cares).. but when testing VMs and/or opening older images with proper VMs it turns into a puzzle game. Also, i think that Cog is good name and executables deserve to carry this name (instead of squeak or croquet). That's why i'm not questioning whether new VMs should use different names. I am only questioning which names we should use :) -- Best regards, Igor Stasenko AKA sig. |
what means MT? Stef On Mar 19, 2011, at 11:03 PM, Igor Stasenko wrote: > > Hello, > > there is an explosion of different kinds of VM we're building today.. > and i think it would be nice to use different names for them. > Currently i am using following naming scheme: > > - VMs based on StackInterpreter: > StackVM > > - VMs with JIT > Cog > > - VMs with JIT + MT > CogMT > > i am open for suggestions of giving better/proper names. > > Maybe for end-user it is not important to use different names > (squeak.. yeah.. this is squeak.. who cares).. > but when testing VMs and/or opening older images with proper VMs it > turns into a puzzle game. > Also, i think that Cog is good name and executables deserve to carry > this name (instead of squeak or croquet). > That's why i'm not questioning whether new VMs should use different > names. I am only questioning which names we should use :) > > > -- > Best regards, > Igor Stasenko AKA sig. |
On Sun, Mar 20, 2011 at 9:38 AM, stephane ducasse <[hidden email]> wrote:
Multi Threaded or similar
|
Yes I got it after :) On Mar 20, 2011, at 10:57 AM, Mariano Martinez Peck wrote: > > > On Sun, Mar 20, 2011 at 9:38 AM, stephane ducasse <[hidden email]> wrote: > > what means MT? > > Multi Threaded or similar > > > Stef > On Mar 19, 2011, at 11:03 PM, Igor Stasenko wrote: > > > > > Hello, > > > > there is an explosion of different kinds of VM we're building today.. > > and i think it would be nice to use different names for them. > > Currently i am using following naming scheme: > > > > - VMs based on StackInterpreter: > > StackVM > > > > - VMs with JIT > > Cog > > > > - VMs with JIT + MT > > CogMT > > > > i am open for suggestions of giving better/proper names. > > > > Maybe for end-user it is not important to use different names > > (squeak.. yeah.. this is squeak.. who cares).. > > but when testing VMs and/or opening older images with proper VMs it > > turns into a puzzle game. > > Also, i think that Cog is good name and executables deserve to carry > > this name (instead of squeak or croquet). > > That's why i'm not questioning whether new VMs should use different > > names. I am only questioning which names we should use :) > > > > > > -- > > Best regards, > > Igor Stasenko AKA sig. > > |
In reply to this post by Igor Stasenko
On Sat, 19 Mar 2011, Igor Stasenko wrote: > > Hello, > > there is an explosion of different kinds of VM we're building today.. > and i think it would be nice to use different names for them. Sounds good. > Currently i am using following naming scheme: > > - VMs based on StackInterpreter: > StackVM > > - VMs with JIT > Cog > > - VMs with JIT + MT > CogMT > > i am open for suggestions of giving better/proper names. What about sticking VM to the end of each of them? Like here: StackVM, CogVM, CogMTVM, and InterpreterVM > > Maybe for end-user it is not important to use different names > (squeak.. yeah.. this is squeak.. who cares).. It also helps identifying the VM in bug reports. Levente > but when testing VMs and/or opening older images with proper VMs it > turns into a puzzle game. > Also, i think that Cog is good name and executables deserve to carry > this name (instead of squeak or croquet). > That's why i'm not questioning whether new VMs should use different > names. I am only questioning which names we should use :) > > > -- > Best regards, > Igor Stasenko AKA sig. > |
On Sun, Mar 20, 2011 at 8:18 PM, Levente Uzonyi <[hidden email]> wrote:
I like the VM suffix. But what would really float my boat would be if InterpreterVM was called ContextVM. Taking Igor's list anew -Interpreted VMs using contexts based on Interpreter ContextVM
- Interpreted VMs using a stack organization based on StackInterpreter: StackVM - Co-Interpreter+JIT VMs using a stack organization based on CoInterpreter: CogVM - Co-Interpreter+JIT VMs using a stack organization and with multi-threading FFI support based on CoInterpreterMT CogMTVM
|
On 21 March 2011 05:45, Eliot Miranda <[hidden email]> wrote: > > > > On Sun, Mar 20, 2011 at 8:18 PM, Levente Uzonyi <[hidden email]> wrote: >> >> On Sat, 19 Mar 2011, Igor Stasenko wrote: >> >>> >>> Hello, >>> >>> there is an explosion of different kinds of VM we're building today.. >>> and i think it would be nice to use different names for them. >> >> Sounds good. >> >>> Currently i am using following naming scheme: >>> >>> - VMs based on StackInterpreter: >>> StackVM >>> >>> - VMs with JIT >>> Cog >>> >>> - VMs with JIT + MT >>> CogMT >>> >>> i am open for suggestions of giving better/proper names. >> >> What about sticking VM to the end of each of them? Like here: >> StackVM, CogVM, CogMTVM, and InterpreterVM > > I like the VM suffix. But what would really float my boat would be if InterpreterVM was called ContextVM. Taking Igor's list anew > > -Interpreted VMs using contexts based on Interpreter > ContextVM > - Interpreted VMs using a stack organization based on StackInterpreter: > StackVM > > - Co-Interpreter+JIT VMs using a stack organization based on CoInterpreter: > CogVM > > - Co-Interpreter+JIT VMs using a stack organization and with multi-threading FFI support based on CoInterpreterMT > CogMTVM Good. except that we don't have configs to build ContextVM. -- Best regards, Igor Stasenko AKA sig. |
Well... I need to make a paraphrasis of one of my childhood heroes here: "I fight for the users!" :) I think so much names and versions are misleading for end users (of course, it is useful for us)... but I think end users will have a hard travel if we start to rename everything. I think all the new vms are "Cog", not matter if stack, jit, or jit-mt. More important: jit-mt is the evolution of jit version, so nobody will use the older once mt is tested enough. So, I rather name the vms: Cog Stack (or something more appealing for end users like "Cog Universal") Cog 1.0 Cog 2.0 ... or something like that (I know, I'm minority in this, but I needed to express my point of view, just for the record :) ) best, Esteban El 21/03/2011, a las 4:48a.m., Igor Stasenko escribió: > > On 21 March 2011 05:45, Eliot Miranda <[hidden email]> wrote: >> >> >> >> On Sun, Mar 20, 2011 at 8:18 PM, Levente Uzonyi <[hidden email]> wrote: >>> >>> On Sat, 19 Mar 2011, Igor Stasenko wrote: >>> >>>> >>>> Hello, >>>> >>>> there is an explosion of different kinds of VM we're building today.. >>>> and i think it would be nice to use different names for them. >>> >>> Sounds good. >>> >>>> Currently i am using following naming scheme: >>>> >>>> - VMs based on StackInterpreter: >>>> StackVM >>>> >>>> - VMs with JIT >>>> Cog >>>> >>>> - VMs with JIT + MT >>>> CogMT >>>> >>>> i am open for suggestions of giving better/proper names. >>> >>> What about sticking VM to the end of each of them? Like here: >>> StackVM, CogVM, CogMTVM, and InterpreterVM >> >> I like the VM suffix. But what would really float my boat would be if InterpreterVM was called ContextVM. Taking Igor's list anew >> >> -Interpreted VMs using contexts based on Interpreter >> ContextVM >> - Interpreted VMs using a stack organization based on StackInterpreter: >> StackVM >> >> - Co-Interpreter+JIT VMs using a stack organization based on CoInterpreter: >> CogVM >> >> - Co-Interpreter+JIT VMs using a stack organization and with multi-threading FFI support based on CoInterpreterMT >> CogMTVM > > Good. > except that we don't have configs to build ContextVM. > > > > -- > Best regards, > Igor Stasenko AKA sig. |
On 21 March 2011 13:15, Esteban Lorenzano <[hidden email]> wrote: > > Well... I need to make a paraphrasis of one of my childhood heroes here: "I fight for the users!" :) > > I think so much names and versions are misleading for end users (of course, it is useful for us)... but I think end users will have a hard travel if we start to rename everything. > I think all the new vms are "Cog", not matter if stack, jit, or jit-mt. More important: jit-mt is the evolution of jit version, so nobody will use the older once mt is tested enough. > > So, I rather name the vms: > > Cog Stack (or something more appealing for end users like "Cog Universal") > Cog 1.0 > Cog 2.0 > > ... or something like that > > (I know, I'm minority in this, but I needed to express my point of view, just for the record :) ) > Not at all. At a time of release, we can always rename VM to Squeak/Cog/Pharo/Croquet/whatever and ship it with corresponding image(s) in 1-click package. As you said, we need it internally (and especially during transition) because it is much nicer to have distinct names for every kind of VMs we're using/testing/developing. > best, > Esteban > -- Best regards, Igor Stasenko AKA sig. |
On Mon, Mar 21, 2011 at 02:27:53PM +0100, Igor Stasenko wrote: > At a time of release, we can always rename VM to > Squeak/Cog/Pharo/Croquet/whatever and ship it with corresponding > image(s) in 1-click package. > As you said, we need it internally (and especially during transition) > because it is much nicer to have distinct names for every kind of VMs > we're using/testing/developing. Ooh. internal code names. fancy -- Matthew Fulmer (a.k.a. Tapple) |
Free forum by Nabble | Edit this page |