<?xml version="1.0" encoding="UTF-8"?><!-- generator="wordpress/1.5.2" -->
<rss version="2.0" 
	xmlns:content="http://purl.org/rss/1.0/modules/content/">
<channel>
	<title>Airs - Ian Lance Taylor Comments</title>
	<link>http://www.airs.com/blog</link>
	<description>Ian Lance Taylor</description>
	<pubDate>Sat, 17 May 2008 10:43:39 +0000</pubDate>
	<generator>http://wordpress.org/?v=1.5.2</generator>

	<item>
 		<title>Comment on Monotheism by: ncm</title>
		<link>http://www.airs.com/blog/archives/193#comment-13707</link>
		<pubDate>Sat, 17 May 2008 09:05:43 +0000</pubDate>
		<guid>http://www.airs.com/blog/archives/193#comment-13707</guid>
					<description>A reduction from N to one, on the way to zero, is progress, presuming you don't stop before you get there.

At the time the advantage was that you needed only to assuage one deity, and did not risk offending each by acknowledging jurisdiction of another, or a dozen others.  There were direct financial consequences, too: how many priesthoods must you support?  (Arguably these are two restatements of the same problem.)  Polytheism makes for a very complicated life.  Arguably that's a big part of the appeal of occultism among the idle rich -- life seems too predictable without it.

It's amusing how many centuries it took to settle on the &quot;died for our sins&quot; line.  It seems to me that any of the other early contenders could easily have won out, but for unspecifiable fluctuations of the ectoplasmic ether.</description>
		<content:encoded><![CDATA[	<p>A reduction from N to one, on the way to zero, is progress, presuming you don&#8217;t stop before you get there.</p>
	<p>At the time the advantage was that you needed only to assuage one deity, and did not risk offending each by acknowledging jurisdiction of another, or a dozen others.  There were direct financial consequences, too: how many priesthoods must you support?  (Arguably these are two restatements of the same problem.)  Polytheism makes for a very complicated life.  Arguably that&#8217;s a big part of the appeal of occultism among the idle rich &#8212; life seems too predictable without it.</p>
	<p>It&#8217;s amusing how many centuries it took to settle on the &#8220;died for our sins&#8221; line.  It seems to me that any of the other early contenders could easily have won out, but for unspecifiable fluctuations of the ectoplasmic ether.
</p>
]]></content:encoded>
				</item>
	<item>
 		<title>Comment on Layered Programming by: ncm</title>
		<link>http://www.airs.com/blog/archives/192#comment-13667</link>
		<pubDate>Fri, 16 May 2008 04:54:25 +0000</pubDate>
		<guid>http://www.airs.com/blog/archives/192#comment-13667</guid>
					<description>As in all early Spolsky columns, there's a bit of nonsense right in the foundation.  I think we call this one &quot;selection bias&quot;: the abstractions Joel notices are the ones that leak.  The overwhelming majority of abstractions are airtight and invisible, at least to Joel.  (You haven't noticed any iron nuclei decaying lately.)  Joel has perhaps more experience with leakiness than most of us, from his time at Microsoft.</description>
		<content:encoded><![CDATA[	<p>As in all early Spolsky columns, there&#8217;s a bit of nonsense right in the foundation.  I think we call this one &#8220;selection bias&#8221;: the abstractions Joel notices are the ones that leak.  The overwhelming majority of abstractions are airtight and invisible, at least to Joel.  (You haven&#8217;t noticed any iron nuclei decaying lately.)  Joel has perhaps more experience with leakiness than most of us, from his time at Microsoft.
</p>
]]></content:encoded>
				</item>
	<item>
 		<title>Comment on Layered Programming by: Ian Lance Taylor</title>
		<link>http://www.airs.com/blog/archives/192#comment-13659</link>
		<pubDate>Thu, 15 May 2008 21:50:22 +0000</pubDate>
		<guid>http://www.airs.com/blog/archives/192#comment-13659</guid>
					<description>Thanks for the link.  You're right, and that was six years ago.</description>
		<content:encoded><![CDATA[	<p>Thanks for the link.  You&#8217;re right, and that was six years ago.
</p>
]]></content:encoded>
				</item>
	<item>
 		<title>Comment on Layered Programming by: omkar</title>
		<link>http://www.airs.com/blog/archives/192#comment-13642</link>
		<pubDate>Thu, 15 May 2008 06:12:32 +0000</pubDate>
		<guid>http://www.airs.com/blog/archives/192#comment-13642</guid>
					<description>Thats exactly the concern of Joel Spolsky in his beautiful article named &quot;The Law of Leaky Abstractions&quot;

http://www.joelonsoftware.com/articles/LeakyAbstractions.html</description>
		<content:encoded><![CDATA[	<p>Thats exactly the concern of Joel Spolsky in his beautiful article named &#8220;The Law of Leaky Abstractions&#8221;</p>
	<p><a href='http://www.joelonsoftware.com/articles/LeakyAbstractions.html' rel='nofollow'>http://www.joelonsoftware.com/articles/LeakyAbstractions.html</a>
</p>
]]></content:encoded>
				</item>
	<item>
 		<title>Comment on Peer Review by: Ian Lance Taylor</title>
		<link>http://www.airs.com/blog/archives/191#comment-13637</link>
		<pubDate>Thu, 15 May 2008 00:51:09 +0000</pubDate>
		<guid>http://www.airs.com/blog/archives/191#comment-13637</guid>
					<description>No feedback is a huge problem.  But lettimg any other write-after-approval maintainer approve the change doesn't seem like the right answer to me.  We give out write-after-approval pretty freely.

Note that there generally isn't a person in charge.  There are several people, which makes it easier to ignore patch pings, especially for complicated patches.</description>
		<content:encoded><![CDATA[	<p>No feedback is a huge problem.  But lettimg any other write-after-approval maintainer approve the change doesn&#8217;t seem like the right answer to me.  We give out write-after-approval pretty freely.</p>
	<p>Note that there generally isn&#8217;t a person in charge.  There are several people, which makes it easier to ignore patch pings, especially for complicated patches.
</p>
]]></content:encoded>
				</item>
	<item>
 		<title>Comment on Peer Review by: Samuel Tardieu</title>
		<link>http://www.airs.com/blog/archives/191#comment-13615</link>
		<pubDate>Wed, 14 May 2008 07:58:59 +0000</pubDate>
		<guid>http://www.airs.com/blog/archives/191#comment-13615</guid>
					<description>About GCC: It would be great if &quot;write after approval&quot; maintainers could get approval from other w-a-a maintainers if the persons in charge of a subsystem don't answer in a reasonable delay. The problem as I see it is not that patches from w-a-a maintainers are rejected by the person in charge, which is totally reasonable if there are issues with the proposed change, but that there is sometimes no feedback at all for months despite repeated &quot;PING&quot; requests.

Forcing the person in charge to step up if he doesn't want a change to go in (rather than staying mute) would probably be more productive in the case of some subsystems (the Ada front-end comes to mind). It would either increase the number of approved patches or at least improve communication and feedback.</description>
		<content:encoded><![CDATA[	<p>About GCC: It would be great if &#8220;write after approval&#8221; maintainers could get approval from other w-a-a maintainers if the persons in charge of a subsystem don&#8217;t answer in a reasonable delay. The problem as I see it is not that patches from w-a-a maintainers are rejected by the person in charge, which is totally reasonable if there are issues with the proposed change, but that there is sometimes no feedback at all for months despite repeated &#8220;PING&#8221; requests.</p>
	<p>Forcing the person in charge to step up if he doesn&#8217;t want a change to go in (rather than staying mute) would probably be more productive in the case of some subsystems (the Ada front-end comes to mind). It would either increase the number of approved patches or at least improve communication and feedback.
</p>
]]></content:encoded>
				</item>
	<item>
 		<title>Comment on Linker relro by: drow</title>
		<link>http://www.airs.com/blog/archives/189#comment-13508</link>
		<pubDate>Sun, 11 May 2008 21:53:58 +0000</pubDate>
		<guid>http://www.airs.com/blog/archives/189#comment-13508</guid>
					<description>That description (minor security feature, not an optimization, wastes most of a page) matches the description Jakub gave me when I asked him about relro support.</description>
		<content:encoded><![CDATA[	<p>That description (minor security feature, not an optimization, wastes most of a page) matches the description Jakub gave me when I asked him about relro support.
</p>
]]></content:encoded>
				</item>
	<item>
 		<title>Comment on GCC in C++ by: ncm</title>
		<link>http://www.airs.com/blog/archives/187#comment-13460</link>
		<pubDate>Sun, 11 May 2008 00:01:55 +0000</pubDate>
		<guid>http://www.airs.com/blog/archives/187#comment-13460</guid>
					<description>pinskia: string isn't part of the STL.  It's also an unfortunate textbook example of a standard-committee orphan.  The original draft standard string was simple and clean.  Then some busybody decided it was too simple, and made a proposal to mess it up badly.  After that was accepted (at the meeting before my first) the original designer gave up.  String got very little attention for yeas after that, except what I could spare to keep it from getting even worse.

Stepanov has said that if he were starting over, STL components would have no public named member functions at all.  Before template function overloading that wasn't possible, but then they were added after it was too late to take out the member functions.</description>
		<content:encoded><![CDATA[	<p>pinskia: string isn&#8217;t part of the STL.  It&#8217;s also an unfortunate textbook example of a standard-committee orphan.  The original draft standard string was simple and clean.  Then some busybody decided it was too simple, and made a proposal to mess it up badly.  After that was accepted (at the meeting before my first) the original designer gave up.  String got very little attention for yeas after that, except what I could spare to keep it from getting even worse.</p>
	<p>Stepanov has said that if he were starting over, STL components would have no public named member functions at all.  Before template function overloading that wasn&#8217;t possible, but then they were added after it was too late to take out the member functions.
</p>
]]></content:encoded>
				</item>
	<item>
 		<title>Comment on Israel and Palestine and Annapolis by: avjo</title>
		<link>http://www.airs.com/blog/archives/107#comment-13437</link>
		<pubDate>Sat, 10 May 2008 10:50:22 +0000</pubDate>
		<guid>http://www.airs.com/blog/archives/107#comment-13437</guid>
					<description>I agree, we can't afford waiting. But it just seems inevitable now.
I hope the current talks with Mahmoud Abbas will yield something but I honestly doubt that. Even if we do reach an agreement with him, it will not be accepted by Hamas (Iran), and they will keep shooting. violence will not stop. So it really depends on whether he (== his government, laws, police forces...) can control his people or not.
Otherwise there's nothing much to do than literally wait for another strong Palestinian leader who wants peace.

I wasn't clear on the Gaza thing. Of course we didn't expect Gaza to turn into Paris in a year. But we did hoped they would WANT to. And it was a big issue here between the  right and left political wings. Right-wing people claimed that by leaving Gaza it will turn into a violent, Hamas state. Left-wing people had a different vision: they hoped that without any further land claims, the Palestinian in Gaza will put their guns down and instead think of building a state. This dream was shattered. Listen to them - now they really think they are closer than even to destroy Israel and gain control of the complete territory. They see our peace attempts as a weakness.

No logic problem here. Israel's political map is splitted (roughly) fifty fifty. half right, half left. Actually after the GAZA failure people are going right a bit. Anyway, such a split has huge support for settlements. Even if it's 30%, it's still &quot;a lot of Israelis&quot;. Nevertheless, if a real oppurtunity would come, people will vote for peace. Just like they voted for evacuating Sinai (in exchange for peace with Egypt), S. Lebanon (in hope the hezbollah will put down its weapons once they will be left out of land claims), GAZA,..  So - deep inside - people know that the vast majority of settlements will be evacuated one day...

I think I got the newlines better this time. Have a good weekend !</description>
		<content:encoded><![CDATA[	<p>I agree, we can&#8217;t afford waiting. But it just seems inevitable now.<br />
I hope the current talks with Mahmoud Abbas will yield something but I honestly doubt that. Even if we do reach an agreement with him, it will not be accepted by Hamas (Iran), and they will keep shooting. violence will not stop. So it really depends on whether he (== his government, laws, police forces&#8230;) can control his people or not.<br />
Otherwise there&#8217;s nothing much to do than literally wait for another strong Palestinian leader who wants peace.</p>
	<p>I wasn&#8217;t clear on the Gaza thing. Of course we didn&#8217;t expect Gaza to turn into Paris in a year. But we did hoped they would WANT to. And it was a big issue here between the  right and left political wings. Right-wing people claimed that by leaving Gaza it will turn into a violent, Hamas state. Left-wing people had a different vision: they hoped that without any further land claims, the Palestinian in Gaza will put their guns down and instead think of building a state. This dream was shattered. Listen to them - now they really think they are closer than even to destroy Israel and gain control of the complete territory. They see our peace attempts as a weakness.</p>
	<p>No logic problem here. Israel&#8217;s political map is splitted (roughly) fifty fifty. half right, half left. Actually after the GAZA failure people are going right a bit. Anyway, such a split has huge support for settlements. Even if it&#8217;s 30%, it&#8217;s still &#8220;a lot of Israelis&#8221;. Nevertheless, if a real oppurtunity would come, people will vote for peace. Just like they voted for evacuating Sinai (in exchange for peace with Egypt), S. Lebanon (in hope the hezbollah will put down its weapons once they will be left out of land claims), GAZA,..  So - deep inside - people know that the vast majority of settlements will be evacuated one day&#8230;</p>
	<p>I think I got the newlines better this time. Have a good weekend !
</p>
]]></content:encoded>
				</item>
	<item>
 		<title>Comment on Israel and Palestine and Annapolis by: Ian Lance Taylor</title>
		<link>http://www.airs.com/blog/archives/107#comment-13422</link>
		<pubDate>Sat, 10 May 2008 00:55:26 +0000</pubDate>
		<guid>http://www.airs.com/blog/archives/107#comment-13422</guid>
					<description>The newlines don't look too bad to me; I wouldn't worry about it.

I think that if Israel just waits, it will lose in the end.  That would be unfortunate.  I touched on that in my original blog entry.

The expectation that Gaza would simply clean itself up in a couple of years was wildly unrealistic.  Did people really think that?  Gaza did not and does not have a functioning independent economy.  It is entirely dependent on Israel, and Israel remains in complete control of its borders which precludes developing any other possibilities.  It will take many years to change that even under good circumstances.

I simply don't believe that simultaneously a lot of Israelis support expanding settlements and that &quot;everybody&quot; also believes that they will be evacuated eventually.  That defies both logic and common sense.</description>
		<content:encoded><![CDATA[	<p>The newlines don&#8217;t look too bad to me; I wouldn&#8217;t worry about it.</p>
	<p>I think that if Israel just waits, it will lose in the end.  That would be unfortunate.  I touched on that in my original blog entry.</p>
	<p>The expectation that Gaza would simply clean itself up in a couple of years was wildly unrealistic.  Did people really think that?  Gaza did not and does not have a functioning independent economy.  It is entirely dependent on Israel, and Israel remains in complete control of its borders which precludes developing any other possibilities.  It will take many years to change that even under good circumstances.</p>
	<p>I simply don&#8217;t believe that simultaneously a lot of Israelis support expanding settlements and that &#8220;everybody&#8221; also believes that they will be evacuated eventually.  That defies both logic and common sense.
</p>
]]></content:encoded>
				</item>
</channel>
</rss>
