<?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"
	>
<channel>
	<title>Comments on: release 0.7.5 and Thunderbird Google address sync</title>
	<atom:link href="http://www.zindus.com/blog/2008/04/30/release-075-and-thunderbird-google-address-sync/feed/" rel="self" type="application/rss+xml" />
	<link>http://www.zindus.com/blog/2008/04/30/release-075-and-thunderbird-google-address-sync/</link>
	<description>Google and Zimbra contact sync for Thunderbird.</description>
	<pubDate>Sat, 31 Jul 2010 20:43:00 +0000</pubDate>
	<generator>http://wordpress.org/?v=2.5.1</generator>
		<item>
		<title>By: mfc</title>
		<link>http://www.zindus.com/blog/2008/04/30/release-075-and-thunderbird-google-address-sync/#comment-1082</link>
		<dc:creator>mfc</dc:creator>
		<pubDate>Thu, 29 Apr 2010 04:49:57 +0000</pubDate>
		<guid isPermaLink="false">http://www.zindus.com/blog/2008/04/30/release-075-and-thunderbird-google-address-sync/#comment-1082</guid>
		<description>Leni, thanks.

For me no one way sync is fine.  But for your consideration let me make the argument why some care about it a lot.  The problems is syncs are risky.  No matter what you are syncing, every now and then, someting goes wrong.  And the less you understand the details of what is happening, the more likely you are to have problems, and the harder it is to recover.   With very large addressbooks, this can be serious.   The solution for many people is simple:  one-way sync.  The advantage for you of course is:  the less that can go wrong, the lower the support burden.  

Anyway, just something to think about.  Either way it makes sense to do nothing until the interface has re-settled.</description>
		<content:encoded><![CDATA[<p>Leni, thanks.</p>
<p>For me no one way sync is fine.  But for your consideration let me make the argument why some care about it a lot.  The problems is syncs are risky.  No matter what you are syncing, every now and then, someting goes wrong.  And the less you understand the details of what is happening, the more likely you are to have problems, and the harder it is to recover.   With very large addressbooks, this can be serious.   The solution for many people is simple:  one-way sync.  The advantage for you of course is:  the less that can go wrong, the lower the support burden.  </p>
<p>Anyway, just something to think about.  Either way it makes sense to do nothing until the interface has re-settled.</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: leni</title>
		<link>http://www.zindus.com/blog/2008/04/30/release-075-and-thunderbird-google-address-sync/#comment-1076</link>
		<dc:creator>leni</dc:creator>
		<pubDate>Wed, 28 Apr 2010 22:07:10 +0000</pubDate>
		<guid isPermaLink="false">http://www.zindus.com/blog/2008/04/30/release-075-and-thunderbird-google-address-sync/#comment-1076</guid>
		<description>@mfc re: your two feature requests above:
on 1, I think it makes sense to wait until the Google postal feature goes into production then see how the two-way sync works.  If it works fine then there'd be no point in a semi-one-way sync feature.
on 2, that's a great idea, will file a feature request for it.</description>
		<content:encoded><![CDATA[<p>@mfc re: your two feature requests above:<br />
on 1, I think it makes sense to wait until the Google postal feature goes into production then see how the two-way sync works.  If it works fine then there&#8217;d be no point in a semi-one-way sync feature.<br />
on 2, that&#8217;s a great idea, will file a feature request for it.</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: mfc</title>
		<link>http://www.zindus.com/blog/2008/04/30/release-075-and-thunderbird-google-address-sync/#comment-1066</link>
		<dc:creator>mfc</dc:creator>
		<pubDate>Tue, 27 Apr 2010 10:38:26 +0000</pubDate>
		<guid isPermaLink="false">http://www.zindus.com/blog/2008/04/30/release-075-and-thunderbird-google-address-sync/#comment-1066</guid>
		<description>Thanks very much for the response.

1.  After file comparison I found the 1 error was in fact in my input data.
So given this and the fields created in the exported csv,
my assumption is that google is maintaining this internally
but just not yet exposing these fields via api.
I certainly agree that it would be bootless to try and infer semantics
and you should wait until google exposes the fields.

2.  One other benefit to wait for:  
when I import a .csv into google, it also keeps fields such as custom1 to custom4
and I can see them on the webpage.  
So I would hope these might be exposed soon also so they can be synced.
-----------------------------

Testing for the last 2 days has led me to two feature requests.
I hope this is the right place to put them - if not let me know.

1.  Same as David L's.
Tb -&#62; Google:  combine the separate Tb address lines into the single line in the Google address field.
No need to sync back to Tb if a change is made in G:  Tb always overwrites G
(unless the Tb field is empty, in which case it does not overwrite G)..
This seems easy and a no-brainer.

Why Important:
when you are on the road, one of the key pieces of info you need is the address of someone you are meeting.
If you have added the address in G, it will be there.
But if you have added the address in Tb, you suddenly find yourself stuck on the road with no address.

2.  Feedback
ie how many records have been added / altered.  

Why Important:
Knowing this is very important to knowing if something has gone wrong.
While testing I discovered (by detailed inspection) almost all records got altered on 1 sync.
Sometimes on syncs, the unexpected happens, and you need to know asap.
One should be able to see this after it has happened. 

Maybe something in the status bar like:
  Tb:  6/3   G:  2/1
meaning
for Tb 6 altered 2 added, Google 2 altered, one added on last sync.

Hope this is helpful.
mc</description>
		<content:encoded><![CDATA[<p>Thanks very much for the response.</p>
<p>1.  After file comparison I found the 1 error was in fact in my input data.<br />
So given this and the fields created in the exported csv,<br />
my assumption is that google is maintaining this internally<br />
but just not yet exposing these fields via api.<br />
I certainly agree that it would be bootless to try and infer semantics<br />
and you should wait until google exposes the fields.</p>
<p>2.  One other benefit to wait for:<br />
when I import a .csv into google, it also keeps fields such as custom1 to custom4<br />
and I can see them on the webpage.<br />
So I would hope these might be exposed soon also so they can be synced.<br />
&#8212;&#8212;&#8212;&#8212;&#8212;&#8212;&#8212;&#8212;&#8212;&#8211;</p>
<p>Testing for the last 2 days has led me to two feature requests.<br />
I hope this is the right place to put them - if not let me know.</p>
<p>1.  Same as David L&#8217;s.<br />
Tb -&gt; Google:  combine the separate Tb address lines into the single line in the Google address field.<br />
No need to sync back to Tb if a change is made in G:  Tb always overwrites G<br />
(unless the Tb field is empty, in which case it does not overwrite G)..<br />
This seems easy and a no-brainer.</p>
<p>Why Important:<br />
when you are on the road, one of the key pieces of info you need is the address of someone you are meeting.<br />
If you have added the address in G, it will be there.<br />
But if you have added the address in Tb, you suddenly find yourself stuck on the road with no address.</p>
<p>2.  Feedback<br />
ie how many records have been added / altered.  </p>
<p>Why Important:<br />
Knowing this is very important to knowing if something has gone wrong.<br />
While testing I discovered (by detailed inspection) almost all records got altered on 1 sync.<br />
Sometimes on syncs, the unexpected happens, and you need to know asap.<br />
One should be able to see this after it has happened. </p>
<p>Maybe something in the status bar like:<br />
  Tb:  6/3   G:  2/1<br />
meaning<br />
for Tb 6 altered 2 added, Google 2 altered, one added on last sync.</p>
<p>Hope this is helpful.<br />
mc</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: leni</title>
		<link>http://www.zindus.com/blog/2008/04/30/release-075-and-thunderbird-google-address-sync/#comment-1064</link>
		<dc:creator>leni</dc:creator>
		<pubDate>Mon, 26 Apr 2010 22:06:30 +0000</pubDate>
		<guid isPermaLink="false">http://www.zindus.com/blog/2008/04/30/release-075-and-thunderbird-google-address-sync/#comment-1064</guid>
		<description>@mfc, thanks for the feedback!  Couple of comments:
1. agree that the downsides of the xml tag approach are pretty serious
2. as you point out, inferring semantics from position can work most of the time - 1 glitch out of 467 ain't bad.  But if every time someone had a problem, they made a request for support (because one of their contacts postal addresses got mangled), that translates to a 0.5% increase in addon support, which means that as the user-base scales, the addon becomes unsustainable.  If there was a revenue model for the addon, it'd be possible to absorb an increase in support costs.  But as things stand, there's not one feature in the addon that doesn't/shouldn't work 100% of the time and until there's a revenue model for the addon that's unlikely to change.
3. also as you point out, google has a bunch of smarts around this, and last year committed to exposing these smarts via API.  They still haven't implemented to follow through on their commitment, but neither have they withdrawn their commitment.  So the plan for the addon is to use that API for postal addresses when it comes available.  Follow the comments in this post to track any change in status:
http://www.zindus.com/blog/2008/06/17/thunderbird-google-postal-address-sync-part-two</description>
		<content:encoded><![CDATA[<p>@mfc, thanks for the feedback!  Couple of comments:<br />
1. agree that the downsides of the xml tag approach are pretty serious<br />
2. as you point out, inferring semantics from position can work most of the time - 1 glitch out of 467 ain&#8217;t bad.  But if every time someone had a problem, they made a request for support (because one of their contacts postal addresses got mangled), that translates to a 0.5% increase in addon support, which means that as the user-base scales, the addon becomes unsustainable.  If there was a revenue model for the addon, it&#8217;d be possible to absorb an increase in support costs.  But as things stand, there&#8217;s not one feature in the addon that doesn&#8217;t/shouldn&#8217;t work 100% of the time and until there&#8217;s a revenue model for the addon that&#8217;s unlikely to change.<br />
3. also as you point out, google has a bunch of smarts around this, and last year committed to exposing these smarts via API.  They still haven&#8217;t implemented to follow through on their commitment, but neither have they withdrawn their commitment.  So the plan for the addon is to use that API for postal addresses when it comes available.  Follow the comments in this post to track any change in status:<br />
<a href="http://www.zindus.com/blog/2008/06/17/thunderbird-google-postal-address-sync-part-two" rel="nofollow">http://www.zindus.com/blog/2008/06/17/thunderbird-google-postal-address-sync-part-two</a></p>
]]></content:encoded>
	</item>
	<item>
		<title>By: mfc</title>
		<link>http://www.zindus.com/blog/2008/04/30/release-075-and-thunderbird-google-address-sync/#comment-1060</link>
		<dc:creator>mfc</dc:creator>
		<pubDate>Mon, 26 Apr 2010 06:42:01 +0000</pubDate>
		<guid isPermaLink="false">http://www.zindus.com/blog/2008/04/30/release-075-and-thunderbird-google-address-sync/#comment-1060</guid>
		<description>spellcheck apology:  4th text line:  "because" should be "became".</description>
		<content:encoded><![CDATA[<p>spellcheck apology:  4th text line:  &#8220;because&#8221; should be &#8220;became&#8221;.</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: mfc</title>
		<link>http://www.zindus.com/blog/2008/04/30/release-075-and-thunderbird-google-address-sync/#comment-1059</link>
		<dc:creator>mfc</dc:creator>
		<pubDate>Mon, 26 Apr 2010 06:39:16 +0000</pubDate>
		<guid isPermaLink="false">http://www.zindus.com/blog/2008/04/30/release-075-and-thunderbird-google-address-sync/#comment-1059</guid>
		<description>I realize this is sometime after the discussion above,
but just had an interesting experience: 

I just exported a .csv of my contacts and imported into google contacts.
Each address field in Tb because a new address line in the Google Contacts address field - perfect.  

I then exported a csv from Google contacts into the Outlook csv format (no Tb format provided).  
Google did an excellent job of separating the address lines into the correct Tb address fields.
Of 467 contacts, only one v minor misplacement. 

The output fields created were:
.  Business Street		might have multiple lines but correct
.  Business Street  2		unused
.  Business Street  3		unused
.  Business City		correct		
.  Business State		correct
.  Business Postal Code	correct
.  Business Country		correct

Given that there were multiple countries, with multiple address formats  (US, HK, Japan, China, Aust, Sng, UK,...)
with many missing fields, it was remarkable, suggesting either a truly remarkable algorithm,
or Google internally keeps multiple address fields but displays as one field.

So my base suggestion would be similar to David L's:
Tb -&#62; Google:  combine the Tb address lines into separate lines in the Google address field.  
Where a Tb address field is empty, create an empty line (or a line with say a ".") in the google address field.
Given that most people using Tb would be using it as the main address book, this is probably acceptable.

But I would add that, given the above, it might also be possible to do the reverse.
Google somehow do it when creating the csv.
However I would not regard this as critical.

I think the downsides of the [zindus] tag or xml tag approach in the top comment are too serious,
esp having these tags appear on my android phone.</description>
		<content:encoded><![CDATA[<p>I realize this is sometime after the discussion above,<br />
but just had an interesting experience: </p>
<p>I just exported a .csv of my contacts and imported into google contacts.<br />
Each address field in Tb because a new address line in the Google Contacts address field - perfect.  </p>
<p>I then exported a csv from Google contacts into the Outlook csv format (no Tb format provided).<br />
Google did an excellent job of separating the address lines into the correct Tb address fields.<br />
Of 467 contacts, only one v minor misplacement. </p>
<p>The output fields created were:<br />
.  Business Street		might have multiple lines but correct<br />
.  Business Street  2		unused<br />
.  Business Street  3		unused<br />
.  Business City		correct<br />
.  Business State		correct<br />
.  Business Postal Code	correct<br />
.  Business Country		correct</p>
<p>Given that there were multiple countries, with multiple address formats  (US, HK, Japan, China, Aust, Sng, UK,&#8230;)<br />
with many missing fields, it was remarkable, suggesting either a truly remarkable algorithm,<br />
or Google internally keeps multiple address fields but displays as one field.</p>
<p>So my base suggestion would be similar to David L&#8217;s:<br />
Tb -&gt; Google:  combine the Tb address lines into separate lines in the Google address field.<br />
Where a Tb address field is empty, create an empty line (or a line with say a &#8220;.&#8221;) in the google address field.<br />
Given that most people using Tb would be using it as the main address book, this is probably acceptable.</p>
<p>But I would add that, given the above, it might also be possible to do the reverse.<br />
Google somehow do it when creating the csv.<br />
However I would not regard this as critical.</p>
<p>I think the downsides of the [zindus] tag or xml tag approach in the top comment are too serious,<br />
esp having these tags appear on my android phone.</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: Alf</title>
		<link>http://www.zindus.com/blog/2008/04/30/release-075-and-thunderbird-google-address-sync/#comment-607</link>
		<dc:creator>Alf</dc:creator>
		<pubDate>Tue, 13 Oct 2009 10:06:21 +0000</pubDate>
		<guid isPermaLink="false">http://www.zindus.com/blog/2008/04/30/release-075-and-thunderbird-google-address-sync/#comment-607</guid>
		<description>First, thanks for your work.
I agree with DavidL and Julien23.
I precise that Google Sync application on Blackberry smartphone use exactly this rule (one line for each field in Gmail contact postal adress).
When I synchronize Gmail with my phone, the fields are filled correctly if I respect this simple rule exactly like describ by DavidL !
Why does not simply to propose this option ?
Best regards
Alain LE FLOCH</description>
		<content:encoded><![CDATA[<p>First, thanks for your work.<br />
I agree with DavidL and Julien23.<br />
I precise that Google Sync application on Blackberry smartphone use exactly this rule (one line for each field in Gmail contact postal adress).<br />
When I synchronize Gmail with my phone, the fields are filled correctly if I respect this simple rule exactly like describ by DavidL !<br />
Why does not simply to propose this option ?<br />
Best regards<br />
Alain LE FLOCH</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: Ivan Lujan</title>
		<link>http://www.zindus.com/blog/2008/04/30/release-075-and-thunderbird-google-address-sync/#comment-251</link>
		<dc:creator>Ivan Lujan</dc:creator>
		<pubDate>Fri, 21 Nov 2008 21:48:40 +0000</pubDate>
		<guid isPermaLink="false">http://www.zindus.com/blog/2008/04/30/release-075-and-thunderbird-google-address-sync/#comment-251</guid>
		<description>Hello,

I am very interested in syncing my Live Mail account with ThunderBird. Thank you for all your tech support.

Be Well,
Ivan Lujan</description>
		<content:encoded><![CDATA[<p>Hello,</p>
<p>I am very interested in syncing my Live Mail account with ThunderBird. Thank you for all your tech support.</p>
<p>Be Well,<br />
Ivan Lujan</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: Zindus &#187; Blog Archive &#187; Thunderbird Google Postal Address sync - Part Two</title>
		<link>http://www.zindus.com/blog/2008/04/30/release-075-and-thunderbird-google-address-sync/#comment-104</link>
		<dc:creator>Zindus &#187; Blog Archive &#187; Thunderbird Google Postal Address sync - Part Two</dc:creator>
		<pubDate>Tue, 17 Jun 2008 05:47:20 +0000</pubDate>
		<guid isPermaLink="false">http://www.zindus.com/blog/2008/04/30/release-075-and-thunderbird-google-address-sync/#comment-104</guid>
		<description>[...] earlier blog entry on postal addresses began by noting the underlying difficulty of syncing Thunderbird and Google [...]</description>
		<content:encoded><![CDATA[<p>[...] earlier blog entry on postal addresses began by noting the underlying difficulty of syncing Thunderbird and Google [...]</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: Eric</title>
		<link>http://www.zindus.com/blog/2008/04/30/release-075-and-thunderbird-google-address-sync/#comment-102</link>
		<dc:creator>Eric</dc:creator>
		<pubDate>Mon, 09 Jun 2008 14:15:57 +0000</pubDate>
		<guid isPermaLink="false">http://www.zindus.com/blog/2008/04/30/release-075-and-thunderbird-google-address-sync/#comment-102</guid>
		<description>Shoot hit return too soon!

Why not encode in the comment field what the format used for the address field was? This works fine as long as Google isn’t the source of the info. So for standard US addresses, you might have


1234 Main St.
Shipping Dept.
Anywhere, FL 123456 

and in the comment field something like this

Work Address: %A1 %N A%2 %N %CITY, %STATE %POSTAL (%N marks the new lines)</description>
		<content:encoded><![CDATA[<p>Shoot hit return too soon!</p>
<p>Why not encode in the comment field what the format used for the address field was? This works fine as long as Google isn’t the source of the info. So for standard US addresses, you might have</p>
<p>1234 Main St.<br />
Shipping Dept.<br />
Anywhere, FL 123456 </p>
<p>and in the comment field something like this</p>
<p>Work Address: %A1 %N A%2 %N %CITY, %STATE %POSTAL (%N marks the new lines)</p>
]]></content:encoded>
	</item>
</channel>
</rss>
