a primary platform
126 views
Skip to first unread message
Mark Tarver
Apr 5, 2013, 6:49:26 AM4/5/13
to Qilang
> I, personally, think it would strengthen Shen to pick a primary platform(I
> know it would never happen, I made this irrelevant comment anyway).
> Jacob
I yanked this out of the thread it was posted to because it was an
important remark which was a little out of the context of the topic of
the OP.
Jacob, I thoroughly agree with that view. However this being a
virtually OS project in most essential ways, one is constrained by the
human element. I'd put my position as rather being like a captain in
the days of sail, who sets a port, but whose course is influenced by
the prevailing winds. The winds do not always blow towards the goal;
in some cases the personal desires for advancment may blow a
contributor in a different direction.
In contrast, a manager in Google or Sun is rather more like a captain
in the age of steam, because he can issue directives and require them
to be carried out within a time frame. This means that the passengers
know when they will arrive. In the days of sail, one arrived, but the
time and sometimes the exact place were determined by the weather!
As you know the chosen platform was Java, for fairly obvious reasons,
and (vide my notice) we now have two downloadables, however little has
emerged re Joel's port (Java A) this year. Hence your scepticism, I
guess, which I understand. However vide my post on the SECD
machine . This is a long term project, but will deliver I hope a high
performance version that can be linked to C, LLVM or JVM etc.
To answer your point then; it will happen; i.e. there will be a
primary platform, but like a sailing captain I can only tell people
that they will get there. Not precisely when. However you don't have
to be a passenger. You can climb up the rigging too.
Mark
> know it would never happen, I made this irrelevant comment anyway).
> Jacob
I yanked this out of the thread it was posted to because it was an
important remark which was a little out of the context of the topic of
the OP.
Jacob, I thoroughly agree with that view. However this being a
virtually OS project in most essential ways, one is constrained by the
human element. I'd put my position as rather being like a captain in
the days of sail, who sets a port, but whose course is influenced by
the prevailing winds. The winds do not always blow towards the goal;
in some cases the personal desires for advancment may blow a
contributor in a different direction.
In contrast, a manager in Google or Sun is rather more like a captain
in the age of steam, because he can issue directives and require them
to be carried out within a time frame. This means that the passengers
know when they will arrive. In the days of sail, one arrived, but the
time and sometimes the exact place were determined by the weather!
As you know the chosen platform was Java, for fairly obvious reasons,
and (vide my notice) we now have two downloadables, however little has
emerged re Joel's port (Java A) this year. Hence your scepticism, I
guess, which I understand. However vide my post on the SECD
machine . This is a long term project, but will deliver I hope a high
performance version that can be linked to C, LLVM or JVM etc.
To answer your point then; it will happen; i.e. there will be a
primary platform, but like a sailing captain I can only tell people
that they will get there. Not precisely when. However you don't have
to be a passenger. You can climb up the rigging too.
Mark
jo...@mentics.com
Apr 5, 2013, 9:58:40 AM4/5/13
to qil...@googlegroups.com
> As you know the chosen platform was Java, for fairly obvious reasons,
> and (vide my notice) we now have two downloadables, however little has
> emerged re Joel's port (Java A) this year.
Zzzzz... <snort> wha.... what? Huh? Oh, why are my ears burning? :)
> and (vide my notice) we now have two downloadables, however little has
> emerged re Joel's port (Java A) this year.
Sorry about that. I have been working on simplifying the packaging/use
of it, and it's been upgraded to Shen
9.0... I just need to release it :) I was going to switch it to use
the eclipse compiler (much faster than javac) and finish up interop
first. Though maybe I should release it now and do those things for
the next release.
Mark Tarver
Apr 12, 2013, 7:27:23 AM4/12/13
to Qilang
Thanks for that; the experiments with the SECD are very useful.
Greg is asking a lot of good questions, but we are using that
uncertain medium of conversation, English, to answer them. The SECD
machine provides a precise operational semantics for KLambda and also,
a means of linking the implementation to the architecture of the
digital computer. I think perhaps it might be worth looking at some
of these question through the SECD.
Mark
On Apr 12, 8:48 am, Artella Coding <artella.cod...@googlemail.com>
wrote:
>
> https://groups.google.com/d/msg/clojure/PsgKVlWZjOw/5YwxV2rSe8UJ<https://groups.google.com/forum/#!topic/clojure/PsgKVlWZjOw/discussion>
Greg is asking a lot of good questions, but we are using that
uncertain medium of conversation, English, to answer them. The SECD
machine provides a precise operational semantics for KLambda and also,
a means of linking the implementation to the architecture of the
digital computer. I think perhaps it might be worth looking at some
of these question through the SECD.
Mark
On Apr 12, 8:48 am, Artella Coding <artella.cod...@googlemail.com>
wrote:
> >This is a long term project, but will deliver I hope a high
>
> performance version that can be linked to C, LLVM or JVM etc
>
> This link (for clojure) might be useful :
>
> performance version that can be linked to C, LLVM or JVM etc
>
>
> https://groups.google.com/d/msg/clojure/PsgKVlWZjOw/5YwxV2rSe8UJ<https://groups.google.com/forum/#!topic/clojure/PsgKVlWZjOw/discussion>
Reply all
Reply to author
Forward
0 new messages