<?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 on: Where data goes when it dies and other musings</title>
	<atom:link href="http://factoryjoe.com/blog/2009/02/06/where-does-data-go-when-it-dies/feed/" rel="self" type="application/rss+xml" />
	<link>http://factoryjoe.com/blog/2009/02/06/where-does-data-go-when-it-dies/</link>
	<description>This can all be made better. Ready? Begin.</description>
	<lastBuildDate>Sat, 20 Mar 2010 00:41:35 +0000</lastBuildDate>
	<generator>http://wordpress.org/?v=2.9.2</generator>
	<sy:updatePeriod>hourly</sy:updatePeriod>
	<sy:updateFrequency>1</sy:updateFrequency>
		<item>
		<title>By: Ma.gnolia Is Back &#124; Not So Relevant</title>
		<link>http://factoryjoe.com/blog/2009/02/06/where-does-data-go-when-it-dies/comment-page-1/#comment-110832</link>
		<dc:creator>Ma.gnolia Is Back &#124; Not So Relevant</dc:creator>
		<pubDate>Sun, 04 Oct 2009 11:42:03 +0000</pubDate>
		<guid isPermaLink="false">http://factoryjoe.com/blog/?p=1183#comment-110832</guid>
		<description>[...] anymore and their bookmarks were unavailable. But due to those standards users could find them elsewhere on the internet and restore at least a great part of them. That&#8217;s not replacing a backup strategy, of course. [...]</description>
		<content:encoded><![CDATA[<p>[...] anymore and their bookmarks were unavailable. But due to those standards users could find them elsewhere on the internet and restore at least a great part of them. That&#8217;s not replacing a backup strategy, of course. [...]</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: Chris Messina</title>
		<link>http://factoryjoe.com/blog/2009/02/06/where-does-data-go-when-it-dies/comment-page-1/#comment-103794</link>
		<dc:creator>Chris Messina</dc:creator>
		<pubDate>Fri, 13 Feb 2009 18:06:17 +0000</pubDate>
		<guid isPermaLink="false">http://factoryjoe.com/blog/?p=1183#comment-103794</guid>
		<description>First, thanks all for your comments!

Specific replies:

@Todd: yeah, giving everything permalinks is pretty critical.

@ako: well, it&#039;s not exactly that straight forward. While I agree that we&#039;ll see a shift to more real-time interactions on the web, having a record of interactions and exchanges can prove valuable, in a way that we may not currently be able to imagine. While I don&#039;t mourn the [present] loss of my bookmarks, I still think that it would have been nice to have a collection of them, especially given the metadata that I personally added to them (tags, descriptions, ratings, etc). 

The same holds true for Twitter -- I oftentimes cite Twitter as a source of news. If Twitter didn&#039;t have permalinks per post, that would be like newspapers not keeping back issues of prior papers. Losing that history, to me, would be a crime!

@Tom Gardner: your point is orthogonal to mine. It goes without saying that having a robust, triple-redundant backup system is something that services *should* have. Ma.gnolia didn&#039;t; that can&#039;t be undone. You don&#039;t realize the value of good backups until you need them, and that&#039;s because, as they say, hindsight is 20/20.  

Now, not all services will have a good backup system. Even &lt;a href=&quot;http://betalabs.nokia.com/blog/2009/02/12/contacts-on-ovi-beta-database-failed-my-deepest-and-most-sincere-apologies/&quot; rel=&quot;nofollow&quot;&gt;Nokia got this wrong&lt;/a&gt; recently. You can chide all you want, but if a service goes down, sometimes backups will fail. Sometimes companies will go under. Sometimes sites will be removed from the net by overzealous governments. In those cases, having distributed caches across the web are your only recourse to retrieve public data. Microformats at least help to improve those situations, but certainly don&#039;t replace a regular backup and archive routine.

@Kai: yep, that&#039;s true. I was only discussing &lt;strong&gt;public&lt;/strong&gt; data.

@Bengt: I don&#039;t think Larry made any excuses. He screwed up, admits his mistake, and has taken steps to address the problem. He can&#039;t go back in time, but if and when Ma.gnolia lives again, I&#039;m sure he&#039;ll trust someone else more versed in IT to manage his infrastructure. It&#039;s not about making excuses; it&#039;s about taking responsibility for this situation and accepting that sometimes shit really does happen, but at least if you take a simple step like using microformats (among other things), if and when it does, you have a distributed way of recovering the data.</description>
		<content:encoded><![CDATA[<p>First, thanks all for your comments!</p>
<p>Specific replies:</p>
<p>@Todd: yeah, giving everything permalinks is pretty critical.</p>
<p>@ako: well, it&#8217;s not exactly that straight forward. While I agree that we&#8217;ll see a shift to more real-time interactions on the web, having a record of interactions and exchanges can prove valuable, in a way that we may not currently be able to imagine. While I don&#8217;t mourn the [present] loss of my bookmarks, I still think that it would have been nice to have a collection of them, especially given the metadata that I personally added to them (tags, descriptions, ratings, etc). </p>
<p>The same holds true for Twitter &#8212; I oftentimes cite Twitter as a source of news. If Twitter didn&#8217;t have permalinks per post, that would be like newspapers not keeping back issues of prior papers. Losing that history, to me, would be a crime!</p>
<p>@Tom Gardner: your point is orthogonal to mine. It goes without saying that having a robust, triple-redundant backup system is something that services *should* have. Ma.gnolia didn&#8217;t; that can&#8217;t be undone. You don&#8217;t realize the value of good backups until you need them, and that&#8217;s because, as they say, hindsight is 20/20.  </p>
<p>Now, not all services will have a good backup system. Even <a href="http://betalabs.nokia.com/blog/2009/02/12/contacts-on-ovi-beta-database-failed-my-deepest-and-most-sincere-apologies/" rel="nofollow">Nokia got this wrong</a> recently. You can chide all you want, but if a service goes down, sometimes backups will fail. Sometimes companies will go under. Sometimes sites will be removed from the net by overzealous governments. In those cases, having distributed caches across the web are your only recourse to retrieve public data. Microformats at least help to improve those situations, but certainly don&#8217;t replace a regular backup and archive routine.</p>
<p>@Kai: yep, that&#8217;s true. I was only discussing <strong>public</strong> data.</p>
<p>@Bengt: I don&#8217;t think Larry made any excuses. He screwed up, admits his mistake, and has taken steps to address the problem. He can&#8217;t go back in time, but if and when Ma.gnolia lives again, I&#8217;m sure he&#8217;ll trust someone else more versed in IT to manage his infrastructure. It&#8217;s not about making excuses; it&#8217;s about taking responsibility for this situation and accepting that sometimes shit really does happen, but at least if you take a simple step like using microformats (among other things), if and when it does, you have a distributed way of recovering the data.</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: Bengt</title>
		<link>http://factoryjoe.com/blog/2009/02/06/where-does-data-go-when-it-dies/comment-page-1/#comment-103791</link>
		<dc:creator>Bengt</dc:creator>
		<pubDate>Thu, 12 Feb 2009 22:15:45 +0000</pubDate>
		<guid isPermaLink="false">http://factoryjoe.com/blog/?p=1183#comment-103791</guid>
		<description>Theres really no excuse to not keeping good backups of OTHER peoples data. Atleast if you expect to be taken seriously. I never used magnolia, but I dont have any plans to start now...</description>
		<content:encoded><![CDATA[<p>Theres really no excuse to not keeping good backups of OTHER peoples data. Atleast if you expect to be taken seriously. I never used magnolia, but I dont have any plans to start now&#8230;</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: James O'Keefe</title>
		<link>http://factoryjoe.com/blog/2009/02/06/where-does-data-go-when-it-dies/comment-page-1/#comment-103787</link>
		<dc:creator>James O'Keefe</dc:creator>
		<pubDate>Tue, 10 Feb 2009 19:42:25 +0000</pubDate>
		<guid isPermaLink="false">http://factoryjoe.com/blog/?p=1183#comment-103787</guid>
		<description>Wow.  Thanks for sharing your insights.</description>
		<content:encoded><![CDATA[<p>Wow.  Thanks for sharing your insights.</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: Kai</title>
		<link>http://factoryjoe.com/blog/2009/02/06/where-does-data-go-when-it-dies/comment-page-1/#comment-103785</link>
		<dc:creator>Kai</dc:creator>
		<pubDate>Sun, 08 Feb 2009 19:08:48 +0000</pubDate>
		<guid isPermaLink="false">http://factoryjoe.com/blog/?p=1183#comment-103785</guid>
		<description>If I have private data with a saas site, and I don&#039;t have backups, microformat isn&#039;t going to save me if the saas site goes down.</description>
		<content:encoded><![CDATA[<p>If I have private data with a saas site, and I don&#8217;t have backups, microformat isn&#8217;t going to save me if the saas site goes down.</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: Tom Gardner</title>
		<link>http://factoryjoe.com/blog/2009/02/06/where-does-data-go-when-it-dies/comment-page-1/#comment-103784</link>
		<dc:creator>Tom Gardner</dc:creator>
		<pubDate>Sun, 08 Feb 2009 07:53:05 +0000</pubDate>
		<guid isPermaLink="false">http://factoryjoe.com/blog/?p=1183#comment-103784</guid>
		<description>Microformats are all well and good when it comes to re-constructing data, but wouldn&#039;t it have been a thousand times easier just to advise Magnolia to have an offsite backup in place?

Features like structuring your data with Microformats, and paying consultants to talk to you about them, are useless if you don&#039;t get the basics - like backup - right.

I am sure that if you ask Magnolia now if they would have rather spent the time, money and resources they did on Microformat support instead on backups, they would now choose the later.

Too many developers and consultants spending too much time on the insignificant as core development principals are forgotten.</description>
		<content:encoded><![CDATA[<p>Microformats are all well and good when it comes to re-constructing data, but wouldn&#8217;t it have been a thousand times easier just to advise Magnolia to have an offsite backup in place?</p>
<p>Features like structuring your data with Microformats, and paying consultants to talk to you about them, are useless if you don&#8217;t get the basics &#8211; like backup &#8211; right.</p>
<p>I am sure that if you ask Magnolia now if they would have rather spent the time, money and resources they did on Microformat support instead on backups, they would now choose the later.</p>
<p>Too many developers and consultants spending too much time on the insignificant as core development principals are forgotten.</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: ako</title>
		<link>http://factoryjoe.com/blog/2009/02/06/where-does-data-go-when-it-dies/comment-page-1/#comment-103783</link>
		<dc:creator>ako</dc:creator>
		<pubDate>Sun, 08 Feb 2009 02:28:28 +0000</pubDate>
		<guid isPermaLink="false">http://factoryjoe.com/blog/?p=1183#comment-103783</guid>
		<description>A lot of the writing on the web is more like talking than writing. When you talk, there won&#039;t be an archive where it&#039;s stored. And you don&#039;t care. Same should hold for a lot of the stuff we write on the web. It&#039;s short lived communication, not important enough to be archived.

Twitter is like talking. Backup shouldn&#039;t be that important. Image everything anybody ever said was archived somewhere... Where do your words go after you&#039;ve said them?</description>
		<content:encoded><![CDATA[<p>A lot of the writing on the web is more like talking than writing. When you talk, there won&#8217;t be an archive where it&#8217;s stored. And you don&#8217;t care. Same should hold for a lot of the stuff we write on the web. It&#8217;s short lived communication, not important enough to be archived.</p>
<p>Twitter is like talking. Backup shouldn&#8217;t be that important. Image everything anybody ever said was archived somewhere&#8230; Where do your words go after you&#8217;ve said them?</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: wayne sutton</title>
		<link>http://factoryjoe.com/blog/2009/02/06/where-does-data-go-when-it-dies/comment-page-1/#comment-103782</link>
		<dc:creator>wayne sutton</dc:creator>
		<pubDate>Sun, 08 Feb 2009 02:11:28 +0000</pubDate>
		<guid isPermaLink="false">http://factoryjoe.com/blog/?p=1183#comment-103782</guid>
		<description>Good write up. I too lost a lot of bookmarks. I&#039;m now using Diigo and looking forward to the rebirth of Ma.gnolia</description>
		<content:encoded><![CDATA[<p>Good write up. I too lost a lot of bookmarks. I&#8217;m now using Diigo and looking forward to the rebirth of Ma.gnolia</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: Ari Herzog</title>
		<link>http://factoryjoe.com/blog/2009/02/06/where-does-data-go-when-it-dies/comment-page-1/#comment-103781</link>
		<dc:creator>Ari Herzog</dc:creator>
		<pubDate>Sat, 07 Feb 2009 23:17:20 +0000</pubDate>
		<guid isPermaLink="false">http://factoryjoe.com/blog/?p=1183#comment-103781</guid>
		<description>Neat idea about reinstalling your hard drive. While I&#039;m unsure if I want to follow your steps, you&#039;ll be pleased to learn I&#039;ve sporadically deleted apps running on my PC when I realize I hardly use them. Might as well save the same for PDFs and other stuff I download.</description>
		<content:encoded><![CDATA[<p>Neat idea about reinstalling your hard drive. While I&#8217;m unsure if I want to follow your steps, you&#8217;ll be pleased to learn I&#8217;ve sporadically deleted apps running on my PC when I realize I hardly use them. Might as well save the same for PDFs and other stuff I download.</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: Todd</title>
		<link>http://factoryjoe.com/blog/2009/02/06/where-does-data-go-when-it-dies/comment-page-1/#comment-103780</link>
		<dc:creator>Todd</dc:creator>
		<pubDate>Sat, 07 Feb 2009 09:46:43 +0000</pubDate>
		<guid isPermaLink="false">http://factoryjoe.com/blog/?p=1183#comment-103780</guid>
		<description>&quot;...For example, clearly there is a great deal of movement afoot to advance the state of distributed social networking, as evidenced by XML and JSON-based protocols like Portable Contacts and Activity Streams. But these are primarily transaction-based protocols, and archive poorly.&quot;

Each Twitter status has its own URL, and they are being cached by Google Yahoo with the microformat markup intact. Does that count as a Tantek approved persistence for an activity stream?

Also,

&quot;...It’s simply good data hygiene.&quot;

Ummm...Ew.</description>
		<content:encoded><![CDATA[<p>&#8220;&#8230;For example, clearly there is a great deal of movement afoot to advance the state of distributed social networking, as evidenced by XML and JSON-based protocols like Portable Contacts and Activity Streams. But these are primarily transaction-based protocols, and archive poorly.&#8221;</p>
<p>Each Twitter status has its own URL, and they are being cached by Google Yahoo with the microformat markup intact. Does that count as a Tantek approved persistence for an activity stream?</p>
<p>Also,</p>
<p>&#8220;&#8230;It’s simply good data hygiene.&#8221;</p>
<p>Ummm&#8230;Ew.</p>
]]></content:encoded>
	</item>
</channel>
</rss>
