<div dir="ltr"><div>Hi, Dave --</div><div><br></div><div>Thanks for the question, this is actually a really fascinating topic :)</div><div><br></div><div>First, we wrote up some general information about it (that let's you</div> <div>interactively try out things ;) here:</div><div><a href="http://lively-kernel.org/repository/webwerkstatt/documentation/Serialization-Overview.xhtml">http://lively-kernel.org/repository/webwerkstatt/documentation/Serialization-Overview.xhtml</a></div> <div>and here:</div><div><a href="http://lively-kernel.org/repository/webwerkstatt/documentation/Serialization-Interface.xhtml">http://lively-kernel.org/repository/webwerkstatt/documentation/Serialization-Interface.xhtml</a></div> <div><br></div><div>I guess the "hairy" part was/is how to deal with "native" objects. JS browser</div><div>environments introduce functions and state that are not implemented /</div><div>represented in the JS context but hidden. The DOM and DOM nodes are an example</div> <div>for that -- you cannot get or modify all the state that would be necessary to</div><div>capture or restablish a document / world.</div><div><br></div><div>The solution that we came up with and that works very well is to implement a</div> <div>general JS serializer that walks an object graph starting from root objects.</div><div>When certain objects are encountered - e.g. DOM nodes - we make an exception</div><div>(this is what the serialization plugins that are mentioned in the worlds above</div> <div>are for) and store not their full object representation but just "what we need</div><div>to know".</div><div><br></div><div>The creation of objects from a serialization works accordingly</div><div>create/instantiate objects + run custom init code for the "exceptions".</div> <div><br></div><div>The shortcomings of this approach are the following:</div><div>- On the application development level you still need to be a bit careful what</div><div> objects you reference. Direct pointers to DOM nodes for example won't break</div> <div> the serialization but when you deserialize you need custom init logic to</div><div> make things work as expected again.</div><div>- The stored representations become big (x-xxx MBs) really quickly.</div><div> Implementing optimizations using the plugin approach is possible but</div> <div> requires additional work.</div><div><br></div><div>This deals with the "state" of a JS application / Lively world. Another point</div><div>that you mention is to capture running computations. From a certain level of</div> <div>abstraction this is actually the same thing but since JS has incomplete</div><div>metprogramming capabilities (you are not able to reflect on closures, e.g.)</div><div>the "hidden state" problem comes up again. For Lively practically this has</div> <div>little impact since in the "reactive" browser environment Lively don't have to</div><div>implement a "main" function. Anyway, we dealt with the problem and came up</div><div>with a solution. I will describe that in an upcoming post.</div> <div><br></div><div>Please let me know if you have questions or want a more technical answer.</div><div><br></div><div>Best,</div><div>Robert</div><div><br></div></div><div class="gmail_extra"><br><br><div class="gmail_quote"> On Mon, Mar 11, 2013 at 5:51 PM, David Paola <span dir="ltr"><<a href="mailto:dpaola2@gmail.com" target="_blank">dpaola2@gmail.com</a>></span> wrote:<br><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex"> Hi lively kernel folks,<br> <br> I've spent the past month or so digging around in several language VMs -- CPython, Rubinius, Topaz, Pypy, etc in an attempt to add the equivalent of the original Smalltalk "snapshot" VM primitive. Obviously I have been naive.<br> <br> I've learned a lot, above all else that I'm not giving up. I have a decent, academic understanding of compilers, interpreters, VMs (and a foggy understanding of JITs), and was curious if anyone could clarify how the lively kernel serializes the world into JSON. Was this hairy? What were the hardest parts?<br> <br> I realize everyone has a full time job and can't hand-hold a newbie, so any direction at all would be appreciated. I tried to pick apart the Squeak source code but without a background in the Squeak architecture, it was fruitless.<br> <br> Thanks so much for your energy on lively kernel, I'm looking forward to hearing more and possibly contributing in the future.<br> <br> -dave<br> <br> More info:<br> <br> I realize that the "high level" idea of snapshotting a running VM basically involves serializing the object memory, bytecode, and instruction pointer, and then deserializing that on "resume". Most of the issues I'm encountering lead me to believe I have an incomplete understanding.<br> <br> --<br> Dave Paola<br> <br> _______________________________________________<br> lively-kernel mailing list<br> <a href="mailto:lively-kernel@hpi.uni-potsdam.de">lively-kernel@hpi.uni-potsdam.de</a><br> <a href="http://lists.hpi.uni-potsdam.de/listinfo/lively-kernel" target="_blank">http://lists.hpi.uni-potsdam.de/listinfo/lively-kernel</a><br> </blockquote></div><br></div>