<?xml version="1.0" encoding="UTF-8"?><rss version="2.0"
	xmlns:content="http://purl.org/rss/1.0/modules/content/"
	xmlns:dc="http://purl.org/dc/elements/1.1/"
	xmlns:atom="http://www.w3.org/2005/Atom"
	xmlns:sy="http://purl.org/rss/1.0/modules/syndication/"
	>
<channel>
	<title>Comments for Out of Hanwell</title>
	<atom:link href="http://blog.outofhanwell.com/comments/feed/" rel="self" type="application/rss+xml" />
	<link>http://blog.outofhanwell.com</link>
	<description></description>
	<pubDate>Wed, 07 Jan 2009 06:58:41 +0000</pubDate>
	<generator>http://wordpress.org/?v=MU</generator>
	<sy:updatePeriod>hourly</sy:updatePeriod>
	<sy:updateFrequency>1</sy:updateFrequency>
		<item>
		<title>Comment on JavaScript: Property Names vs. Variables by Jenny</title>
		<link>http://blog.outofhanwell.com/2006/01/10/javascript-property-names-vs-variables/#comment-4329</link>
		<dc:creator>Jenny</dc:creator>
		<pubDate>Tue, 23 Dec 2008 17:07:23 +0000</pubDate>
		<guid isPermaLink="false">http://blog.outofhanwell.com/2006/01/10/javascript-property-names-vs-variables/#comment-4329</guid>
		<description>Interesting article. I found some more information &lt;a href="http://www.sourcearticle.info/?id=Mzk5ODA1LEphdmFTY3JpcHQgUG9wdXAgV2luZG93LDA=" rel="nofollow"&gt;here&lt;/a&gt;</description>
		<content:encoded><![CDATA[<p>Interesting article. I found some more information <a href="http://www.sourcearticle.info/?id=Mzk5ODA1LEphdmFTY3JpcHQgUG9wdXAgV2luZG93LDA=" rel="nofollow">here</a></p>
]]></content:encoded>
	</item>
	<item>
		<title>Comment on Khoros: A Lua Operating System by Matthias Miller</title>
		<link>http://blog.outofhanwell.com/2008/08/16/khoros-a-lua-operating-system/#comment-4322</link>
		<dc:creator>Matthias Miller</dc:creator>
		<pubDate>Tue, 25 Nov 2008 12:26:34 +0000</pubDate>
		<guid isPermaLink="false">http://outofhanwell.wordpress.com/?p=164#comment-4322</guid>
		<description>Ken, yes, I've been using VirtualBox for development (except I haven't gotten VirtualBox running on my Eee yet).</description>
		<content:encoded><![CDATA[<p>Ken, yes, I&#8217;ve been using VirtualBox for development (except I haven&#8217;t gotten VirtualBox running on my Eee yet).</p>
]]></content:encoded>
	</item>
	<item>
		<title>Comment on Khoros: A Lua Operating System by Ken Yap</title>
		<link>http://blog.outofhanwell.com/2008/08/16/khoros-a-lua-operating-system/#comment-4321</link>
		<dc:creator>Ken Yap</dc:creator>
		<pubDate>Sat, 22 Nov 2008 15:08:12 +0000</pubDate>
		<guid isPermaLink="false">http://outofhanwell.wordpress.com/?p=164#comment-4321</guid>
		<description>This is interesting. I was just reading a blurb about the benefits of Lua for constrained platforms like cell phones, linked to from Lua's about page. It sounds like your project might be just the thing for embedded platforms. The idea just came to me from the conjunction of your article and the aforementioned blurb. So I haven't pondered it in depth, but you could dispense with a lot of the baggage associated with the PC hardware architecture.

Also is your OS bootable on something like Virtualbox?</description>
		<content:encoded><![CDATA[<p>This is interesting. I was just reading a blurb about the benefits of Lua for constrained platforms like cell phones, linked to from Lua&#8217;s about page. It sounds like your project might be just the thing for embedded platforms. The idea just came to me from the conjunction of your article and the aforementioned blurb. So I haven&#8217;t pondered it in depth, but you could dispense with a lot of the baggage associated with the PC hardware architecture.</p>
<p>Also is your OS bootable on something like Virtualbox?</p>
]]></content:encoded>
	</item>
	<item>
		<title>Comment on Nested CSS Blocks by jdunkelberger</title>
		<link>http://blog.outofhanwell.com/2005/09/19/nested-css-blocks/#comment-4320</link>
		<dc:creator>jdunkelberger</dc:creator>
		<pubDate>Thu, 20 Nov 2008 20:31:28 +0000</pubDate>
		<guid isPermaLink="false">http://blog.outofhanwell.com/2005/09/19/nested-css-blocks/#comment-4320</guid>
		<description>So I've been using CSS for about 10 minutes now (I've done mostly some backend work until now), and I tried nesting CSS in the way that you have now made clear to me does not exist D: That would really make this stuff easier to keep track of.

Maybe I'll write a conversion utility where you write your "source" CSS files in that nonexistant nested way. And it'll just dump everything out in the disgustingly long and verbose list that every site has because there's no other way(?).

I'm still &lt;em&gt;pretty&lt;/em&gt; new to CSS, so my naivete may be getting the best of me at this point. Still cool to see that someone else out there thought of this too :)</description>
		<content:encoded><![CDATA[<p>So I&#8217;ve been using CSS for about 10 minutes now (I&#8217;ve done mostly some backend work until now), and I tried nesting CSS in the way that you have now made clear to me does not exist D: That would really make this stuff easier to keep track of.</p>
<p>Maybe I&#8217;ll write a conversion utility where you write your &#8220;source&#8221; CSS files in that nonexistant nested way. And it&#8217;ll just dump everything out in the disgustingly long and verbose list that every site has because there&#8217;s no other way(?).</p>
<p>I&#8217;m still <em>pretty</em> new to CSS, so my naivete may be getting the best of me at this point. Still cool to see that someone else out there thought of this too <img src='http://s.wordpress.com/wp-includes/images/smilies/icon_smile.gif' alt=':)' class='wp-smiley' /> </p>
]]></content:encoded>
	</item>
	<item>
		<title>Comment on The window.onload Problem Revisited by Vlad Kout</title>
		<link>http://blog.outofhanwell.com/2006/06/08/the-windowonload-problem-revisited/#comment-4310</link>
		<dc:creator>Vlad Kout</dc:creator>
		<pubDate>Sat, 11 Oct 2008 18:33:13 +0000</pubDate>
		<guid isPermaLink="false">http://blog.outofhanwell.com/2006/06/08/the-windowonload-problem-revisited/#comment-4310</guid>
		<description>Hi, Matthias and all!

This is a fully validated and verified version, it does not need
conditional hacks, iframe hacks, https hacks!
It has NO memory leaks and allows multiple event functions
to be executed BEFORE the document.body or document.images are fully loaded, 
as well there is no need 
to write those functions in a separate script and/or to put inside the  
ending BODY tag! - 
instead, all should be in ONE script before ending HEAD tag:

see the working test page  * http://arieslink.sitebar.org/onload-1.html 
(the script there is internal, look at the source code, minimal changes 
were done to Jesse Skinner's work, based in its turn on yours, 
but all errors and warnings were eliminated)

Best regards,
Vlad</description>
		<content:encoded><![CDATA[<p>Hi, Matthias and all!</p>
<p>This is a fully validated and verified version, it does not need<br />
conditional hacks, iframe hacks, https hacks!<br />
It has NO memory leaks and allows multiple event functions<br />
to be executed BEFORE the document.body or document.images are fully loaded,<br />
as well there is no need<br />
to write those functions in a separate script and/or to put inside the<br />
ending BODY tag! -<br />
instead, all should be in ONE script before ending HEAD tag:</p>
<p>see the working test page  * <a href="http://arieslink.sitebar.org/onload-1.html" rel="nofollow">http://arieslink.sitebar.org/onload-1.html</a><br />
(the script there is internal, look at the source code, minimal changes<br />
were done to Jesse Skinner&#8217;s work, based in its turn on yours,<br />
but all errors and warnings were eliminated)</p>
<p>Best regards,<br />
Vlad</p>
]]></content:encoded>
	</item>
	<item>
		<title>Comment on Khoros: A Lua Operating System by Christian</title>
		<link>http://blog.outofhanwell.com/2008/08/16/khoros-a-lua-operating-system/#comment-4308</link>
		<dc:creator>Christian</dc:creator>
		<pubDate>Tue, 07 Oct 2008 11:35:10 +0000</pubDate>
		<guid isPermaLink="false">http://outofhanwell.wordpress.com/?p=164#comment-4308</guid>
		<description>The idea of writing an OS in Lua is great. I'm currently thinking about creating a project on a "higher level" than an OS. It would be a mixture of middleware and application. The first step would be to have the project working on a "host OS". But having a kernel in Lua would make it possible to exchange data between both. A fascinating idea.
I subscribed to your blog and would be happy about some updates on your project. 
Personally, I think I would implement some filesystem functionality to be able to read and write data. I'm still a big fan of the Apple Newtons way of storing data: in a database. Having a system that relies on DBs instead of filesystems.
As Tim C mentioned above using some parts of C code from other projects could be a good idea. Lua is great for interfacing with C code, and FreeDOS http://www.freedos.org is the right place to look at.

Just drop me a mail if you like.

Christian</description>
		<content:encoded><![CDATA[<p>The idea of writing an OS in Lua is great. I&#8217;m currently thinking about creating a project on a &#8220;higher level&#8221; than an OS. It would be a mixture of middleware and application. The first step would be to have the project working on a &#8220;host OS&#8221;. But having a kernel in Lua would make it possible to exchange data between both. A fascinating idea.<br />
I subscribed to your blog and would be happy about some updates on your project.<br />
Personally, I think I would implement some filesystem functionality to be able to read and write data. I&#8217;m still a big fan of the Apple Newtons way of storing data: in a database. Having a system that relies on DBs instead of filesystems.<br />
As Tim C mentioned above using some parts of C code from other projects could be a good idea. Lua is great for interfacing with C code, and FreeDOS <a href="http://www.freedos.org" rel="nofollow">http://www.freedos.org</a> is the right place to look at.</p>
<p>Just drop me a mail if you like.</p>
<p>Christian</p>
]]></content:encoded>
	</item>
	<item>
		<title>Comment on The window.onload Problem Revisited by coupon</title>
		<link>http://blog.outofhanwell.com/2006/06/08/the-windowonload-problem-revisited/#comment-4307</link>
		<dc:creator>coupon</dc:creator>
		<pubDate>Tue, 30 Sep 2008 18:18:02 +0000</pubDate>
		<guid isPermaLink="false">http://blog.outofhanwell.com/2006/06/08/the-windowonload-problem-revisited/#comment-4307</guid>
		<description>I'am getting this error "operation aborded", then i have using this code

window.onload = function(){
}

all is ok now, but i must wait that all the webpage is loaded for javascript execute my script...

Is there any other way ?</description>
		<content:encoded><![CDATA[<p>I&#8217;am getting this error &#8220;operation aborded&#8221;, then i have using this code</p>
<p>window.onload = function(){<br />
}</p>
<p>all is ok now, but i must wait that all the webpage is loaded for javascript execute my script&#8230;</p>
<p>Is there any other way ?</p>
]]></content:encoded>
	</item>
	<item>
		<title>Comment on The window.onload Problem Revisited by Son of Suckerfish dropdowns in jQuery :: CreateOpen</title>
		<link>http://blog.outofhanwell.com/2006/06/08/the-windowonload-problem-revisited/#comment-4306</link>
		<dc:creator>Son of Suckerfish dropdowns in jQuery :: CreateOpen</dc:creator>
		<pubDate>Mon, 22 Sep 2008 17:58:46 +0000</pubDate>
		<guid isPermaLink="false">http://blog.outofhanwell.com/2006/06/08/the-windowonload-problem-revisited/#comment-4306</guid>
		<description>[...] This is much neater and uses a nice regular expression to remove the class name. It also uses Microsoft&#8217;s attachEvent, which is much easier to make non-distructive. There is still a problem with this: the code only executes when the page is fully loaded. What we really need is to wait until the DOM is ready to be manipulated, then run the code. Unfortunately, as we&#8217;re aiming this code at IE6, we can&#8217;t just do a document.addEventListener(&#8221;DOMContentLoaded&#8221;, sfHover). The only way we can get a DOM ready event is by creating an empty script with defer=&#8221;defer&#8221; and listen for it to load, then run our code. This was thought up by Matthias Miller. [...]</description>
		<content:encoded><![CDATA[<p>[...] This is much neater and uses a nice regular expression to remove the class name. It also uses Microsoft&#8217;s attachEvent, which is much easier to make non-distructive. There is still a problem with this: the code only executes when the page is fully loaded. What we really need is to wait until the DOM is ready to be manipulated, then run the code. Unfortunately, as we&#8217;re aiming this code at IE6, we can&#8217;t just do a document.addEventListener(&#8221;DOMContentLoaded&#8221;, sfHover). The only way we can get a DOM ready event is by creating an empty script with defer=&#8221;defer&#8221; and listen for it to load, then run our code. This was thought up by Matthias Miller. [...]</p>
]]></content:encoded>
	</item>
	<item>
		<title>Comment on Khoros: A Lua Operating System by Tim C</title>
		<link>http://blog.outofhanwell.com/2008/08/16/khoros-a-lua-operating-system/#comment-4300</link>
		<dc:creator>Tim C</dc:creator>
		<pubDate>Sun, 24 Aug 2008 16:54:53 +0000</pubDate>
		<guid isPermaLink="false">http://outofhanwell.wordpress.com/?p=164#comment-4300</guid>
		<description>Yes Lua is interesting and I too am ambivalent, Lua is superb and horrible such as poor documentation omitting vital information. A lot of needless frustration results.
(just gone through hell over something undocumented, for-goodness-sake don't just say it returns a thing, what exactly does it return because it doesn't do what is implied. Turned out to be broken, missing release resource and a different not so recommended way does work but only if other things not mentioned are done, logical and easy once you know) The Lua code is very good, just hard to use.

ANSI C was a very good move and some libs are portable too.

Bringing up say IBM clone might not be so hard and there are examples of amateur OS.

My historic background is embedded systems where in effect your code is all there is and might be a whole language. Never done this with massive systems though.

Lua is large, not small in the embedded context. It will however compile straight and just about run under DOS 16, done as an experiement and just needed careful compiler and linker use.

There is an enormous amount I could say but leaving out a lot of justifications and reasons why...

Either you accept Unix and do it all in Unix of some kind making Unix of some kind or accept reality and take the DOS route. Few Unix people ever understand what DOS is/was, wrong thinking.
It's actually very clever and powerful, done in a small space across a lot of hardware, and simple, KISS principle. It is a personal computer not a holy mainframe, live with it.

One way is this.

Find a simple old computer, good IBM clone with IDE and straghtforward graphics, well known network card. That is your working system.

Bring that up using FreeDOS. You can also see if you can get wattcp working on the network card as a precursor for later use. (I could send you a copy of Lua which will run under DOS but is of very restricted capability, too litte stacks space, did run factorial.lua first time though, not useful only a proof of concept)

Back in the 1980s I did several things which here give some clues.
I ported a specialist Forth from CPM/80 to DOS. This was written (by me) in assembler. Made some minor changes and passed the source though an American Automation cross assembler (cost a lot of money). Because i86 and 8080/Z80 are very similar the result was a DOS executable.

This ran first time complete with disk access. Understand, CPM/80 and DOS and to at least XP there are a common set of BIOS calls, (first 16 I think) which are identical. Even the DTA is in both (disk transfer area).

Another thing I did was write the first 32 bit floating point Forth. This was in 68k assembler and aimed at embedded systems but ran on personal computers too.

Essentially there is a relocatable chunk which can be loaded into memory by a loader, go. (or work out of eprom etc.) At a definite place is a jump table, these are the hooks to the system.
Things like initialise, where is memory, readkey, write char, graphics... and so on, all basic and only a very few are needed to have a text system. The correct OS addresses are set there and off you go.
Some very fundamental code has to be somewhere BIOS is part which is unique to that particular hardware, not a lot.

The next stage I will mention is perhaps look at Open Watcom, ex full commerical and a very good toolchain. This is actively in serious use... as a hint can target
DOS 16, DOS 32, DOS extender, QNX, Novell, Win 16, Win 32, Linux, OS/2

Will compile Lua and libs.

Lua is not yet really suitable for full embedded without a formal OS, got to get rid of the C OS libraries and calls are littered all over the source. Others are I suspect working on fixing this.

Compiling a say DOS 32 Lua and maybe it would work very usefully from FreeDOS.

Go to DOS extender and there is megabytes available and the start of crossing the barrier between real mode and protected mode to get at the BIOS calls for things like disk access.

wattcp should allow getting up TCP/IP, just needs a new personality source in Luasocket, currently usocket.c and wsocket.c
I guess it would work by crossing the barrier into wattcp running in real mode. There is an API of some kind, maybe you write it.

Graphics? Can start under DOS and go upwards from there. Libs if you know where to look. Rest assured that Linux and BSD go through hell handling graphics hardware, often needing to but can't cross the barrier into real mode. Use VESA, life is too short to invent the lot.
Widget lib, try FLTK. Rumour it has been used under DOS. And see murgalua.

Then work on eliminating FreeDOS. Easiest is just accept it as a core boot loader. FreeBSD for example uses a Forth as a boot loader. Debian uses some kind of embedded VM to boot. And so on.
That Forth is embeddable and I guess boots in effect on BIOS.

Layers tend to be everywhere. What else is X but a layer sitting on a command line system?

Yes Watcom can run (gui) debug over a serial port or whatever, debugging the app on your new box.

Currently the Watcom DOS distribution is without a maintainer and is probably broken. So it might be difficult to avoid win32. Doubt that it works well enough under Linux yet to be useful.

I recently released a simple tool in C which helps in making static linked Lua, including Lua scripts. This might be useful before you have disk working (but feeding source via a terminal channel is old hat in the embedded world) Lua in fact uses this simple streaming of basic channels, io. does just that.

We were writing dll for DOS way back so that is no big deal.

Maybe I misunderstand.

You got my email address.

Tim</description>
		<content:encoded><![CDATA[<p>Yes Lua is interesting and I too am ambivalent, Lua is superb and horrible such as poor documentation omitting vital information. A lot of needless frustration results.<br />
(just gone through hell over something undocumented, for-goodness-sake don&#8217;t just say it returns a thing, what exactly does it return because it doesn&#8217;t do what is implied. Turned out to be broken, missing release resource and a different not so recommended way does work but only if other things not mentioned are done, logical and easy once you know) The Lua code is very good, just hard to use.</p>
<p>ANSI C was a very good move and some libs are portable too.</p>
<p>Bringing up say IBM clone might not be so hard and there are examples of amateur OS.</p>
<p>My historic background is embedded systems where in effect your code is all there is and might be a whole language. Never done this with massive systems though.</p>
<p>Lua is large, not small in the embedded context. It will however compile straight and just about run under DOS 16, done as an experiement and just needed careful compiler and linker use.</p>
<p>There is an enormous amount I could say but leaving out a lot of justifications and reasons why&#8230;</p>
<p>Either you accept Unix and do it all in Unix of some kind making Unix of some kind or accept reality and take the DOS route. Few Unix people ever understand what DOS is/was, wrong thinking.<br />
It&#8217;s actually very clever and powerful, done in a small space across a lot of hardware, and simple, KISS principle. It is a personal computer not a holy mainframe, live with it.</p>
<p>One way is this.</p>
<p>Find a simple old computer, good IBM clone with IDE and straghtforward graphics, well known network card. That is your working system.</p>
<p>Bring that up using FreeDOS. You can also see if you can get wattcp working on the network card as a precursor for later use. (I could send you a copy of Lua which will run under DOS but is of very restricted capability, too litte stacks space, did run factorial.lua first time though, not useful only a proof of concept)</p>
<p>Back in the 1980s I did several things which here give some clues.<br />
I ported a specialist Forth from CPM/80 to DOS. This was written (by me) in assembler. Made some minor changes and passed the source though an American Automation cross assembler (cost a lot of money). Because i86 and 8080/Z80 are very similar the result was a DOS executable.</p>
<p>This ran first time complete with disk access. Understand, CPM/80 and DOS and to at least XP there are a common set of BIOS calls, (first 16 I think) which are identical. Even the DTA is in both (disk transfer area).</p>
<p>Another thing I did was write the first 32 bit floating point Forth. This was in 68k assembler and aimed at embedded systems but ran on personal computers too.</p>
<p>Essentially there is a relocatable chunk which can be loaded into memory by a loader, go. (or work out of eprom etc.) At a definite place is a jump table, these are the hooks to the system.<br />
Things like initialise, where is memory, readkey, write char, graphics&#8230; and so on, all basic and only a very few are needed to have a text system. The correct OS addresses are set there and off you go.<br />
Some very fundamental code has to be somewhere BIOS is part which is unique to that particular hardware, not a lot.</p>
<p>The next stage I will mention is perhaps look at Open Watcom, ex full commerical and a very good toolchain. This is actively in serious use&#8230; as a hint can target<br />
DOS 16, DOS 32, DOS extender, QNX, Novell, Win 16, Win 32, Linux, OS/2</p>
<p>Will compile Lua and libs.</p>
<p>Lua is not yet really suitable for full embedded without a formal OS, got to get rid of the C OS libraries and calls are littered all over the source. Others are I suspect working on fixing this.</p>
<p>Compiling a say DOS 32 Lua and maybe it would work very usefully from FreeDOS.</p>
<p>Go to DOS extender and there is megabytes available and the start of crossing the barrier between real mode and protected mode to get at the BIOS calls for things like disk access.</p>
<p>wattcp should allow getting up TCP/IP, just needs a new personality source in Luasocket, currently usocket.c and wsocket.c<br />
I guess it would work by crossing the barrier into wattcp running in real mode. There is an API of some kind, maybe you write it.</p>
<p>Graphics? Can start under DOS and go upwards from there. Libs if you know where to look. Rest assured that Linux and BSD go through hell handling graphics hardware, often needing to but can&#8217;t cross the barrier into real mode. Use VESA, life is too short to invent the lot.<br />
Widget lib, try FLTK. Rumour it has been used under DOS. And see murgalua.</p>
<p>Then work on eliminating FreeDOS. Easiest is just accept it as a core boot loader. FreeBSD for example uses a Forth as a boot loader. Debian uses some kind of embedded VM to boot. And so on.<br />
That Forth is embeddable and I guess boots in effect on BIOS.</p>
<p>Layers tend to be everywhere. What else is X but a layer sitting on a command line system?</p>
<p>Yes Watcom can run (gui) debug over a serial port or whatever, debugging the app on your new box.</p>
<p>Currently the Watcom DOS distribution is without a maintainer and is probably broken. So it might be difficult to avoid win32. Doubt that it works well enough under Linux yet to be useful.</p>
<p>I recently released a simple tool in C which helps in making static linked Lua, including Lua scripts. This might be useful before you have disk working (but feeding source via a terminal channel is old hat in the embedded world) Lua in fact uses this simple streaming of basic channels, io. does just that.</p>
<p>We were writing dll for DOS way back so that is no big deal.</p>
<p>Maybe I misunderstand.</p>
<p>You got my email address.</p>
<p>Tim</p>
]]></content:encoded>
	</item>
	<item>
		<title>Comment on Don&#8217;t Steal My Parameters by Mr. Shiny &#38; New</title>
		<link>http://blog.outofhanwell.com/2007/10/17/dont-steal-my-parameters/#comment-4299</link>
		<dc:creator>Mr. Shiny &#38; New</dc:creator>
		<pubDate>Tue, 19 Aug 2008 19:01:28 +0000</pubDate>
		<guid isPermaLink="false">http://blog.outofhanwell.com/2007/10/17/javascript-mind-reading/#comment-4299</guid>
		<description>If you have trojan Javascript replacing functions called by your validateUser function you're already screwed, since trojan code can just as easily replace validateUser itself.  So there isn't really a security issue here.</description>
		<content:encoded><![CDATA[<p>If you have trojan Javascript replacing functions called by your validateUser function you&#8217;re already screwed, since trojan code can just as easily replace validateUser itself.  So there isn&#8217;t really a security issue here.</p>
]]></content:encoded>
	</item>
</channel>
</rss>
