Comments on: ARGON http://www.snell-pym.org.uk/archives/2009/06/28/argon/ Sarah and Alaric Snell-Pym living in interesting times Tue, 16 Mar 2010 05:35:37 +0000 http://wordpress.org/?v=2.7 hourly 1 By: alaric http://www.snell-pym.org.uk/archives/2009/06/28/argon/comment-page-1/#comment-98011 alaric Tue, 30 Jun 2009 09:38:04 +0000 http://www.snell-pym.org.uk/?p=1542#comment-98011 <p>HYDROGEN's not meant for (normal) users to see - it's more the assembly language of a VM, that core bits of the system can be written in but will mainly be the target for an HLL compiler (that will provide sandboxing through pointer safety etc. for a start!)</p> <p>It is indeed way low level :-)</p> <p>However, this means that little else can really be implemented without it (sure, we could experiment with writing HLL implementations, but that'd be wasted effort in the longer run).</p> HYDROGEN's not meant for (normal) users to see - it's more the assembly language of a VM, that core bits of the system can be written in but will mainly be the target for an HLL compiler (that will provide sandboxing through pointer safety etc. for a start!)

It is indeed way low level :-)

However, this means that little else can really be implemented without it (sure, we could experiment with writing HLL implementations, but that'd be wasted effort in the longer run).

]]> By: Faré http://www.snell-pym.org.uk/archives/2009/06/28/argon/comment-page-1/#comment-97968 Faré Mon, 29 Jun 2009 16:50:43 +0000 http://www.snell-pym.org.uk/?p=1542#comment-97968 <p>Of course, my "solution" would be to start standardizing, not the base level of the implementation tower, but the protocol of implementation/virtualization of a computing system with another.</p> Of course, my "solution" would be to start standardizing, not the base level of the implementation tower, but the protocol of implementation/virtualization of a computing system with another.

]]> By: Faré http://www.snell-pym.org.uk/archives/2009/06/28/argon/comment-page-1/#comment-97964 Faré Mon, 29 Jun 2009 16:26:33 +0000 http://www.snell-pym.org.uk/?p=1542#comment-97964 <p>I'd like to "me too" at a lot of what you said. However, what you actually propose in the end leaves me dubious.</p> <p>HYDROGEN looks way low-level. Why would anyone want to use it? And fixed stack sizes only makes it worse.</p> <p>A modern system would provide users with abstractions in terms no lower-level than some well-typed high-level graph. How are you going to reconcile that with the low-level aspect of HYDROGEN?</p> I'd like to "me too" at a lot of what you said. However, what you actually propose in the end leaves me dubious.

HYDROGEN looks way low-level. Why would anyone want to use it? And fixed stack sizes only makes it worse.

A modern system would provide users with abstractions in terms no lower-level than some well-typed high-level graph. How are you going to reconcile that with the low-level aspect of HYDROGEN?

]]>