<?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/"
	xmlns:georss="http://www.georss.org/georss" xmlns:geo="http://www.w3.org/2003/01/geo/wgs84_pos#" xmlns:media="http://search.yahoo.com/mrss/"
		>
<channel>
	<title>Comments on: Cubes and Universes</title>
	<atom:link href="http://blog.srlanger.com/2006/05/24/cubes-and-universes/feed/" rel="self" type="application/rss+xml" />
	<link>http://blog.srlanger.com/2006/05/24/cubes-and-universes/</link>
	<description>Just another Knowledge Worker</description>
	<lastBuildDate>Tue, 29 Dec 2009 09:01:52 +0000</lastBuildDate>
	<generator>http://wordpress.com/</generator>
	<sy:updatePeriod>hourly</sy:updatePeriod>
	<sy:updateFrequency>1</sy:updateFrequency>
		<item>
		<title>By: Stéphane-Robert Langer</title>
		<link>http://blog.srlanger.com/2006/05/24/cubes-and-universes/#comment-2890</link>
		<dc:creator>Stéphane-Robert Langer</dc:creator>
		<pubDate>Tue, 29 Dec 2009 09:01:52 +0000</pubDate>
		<guid isPermaLink="false">http://srlanger.wordpress.com/2006/05/24/cubes-and-universes/#comment-2890</guid>
		<description>Michael: Thank you for your kind words. Of course there&#039;s a lot more to SAP&#039;s BI tools than the somewhat disappointing implementation of Business Objects cubes. I&#039;m also not sure if/how that is being merged into SAP Business Warehouse. A strong case could certainly be made for both platforms depending on the situation.

I hope you can sort things out soon. As I think you suggest, it might (to some extent) be more important at some point to choose a direction, stick to it and deliver rather than to select one tool over another.

Merry Christmas and a Happy New Year!</description>
		<content:encoded><![CDATA[<p>Michael: Thank you for your kind words. Of course there&#8217;s a lot more to SAP&#8217;s BI tools than the somewhat disappointing implementation of Business Objects cubes. I&#8217;m also not sure if/how that is being merged into SAP Business Warehouse. A strong case could certainly be made for both platforms depending on the situation.</p>
<p>I hope you can sort things out soon. As I think you suggest, it might (to some extent) be more important at some point to choose a direction, stick to it and deliver rather than to select one tool over another.</p>
<p>Merry Christmas and a Happy New Year!</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: Michael Schmalz</title>
		<link>http://blog.srlanger.com/2006/05/24/cubes-and-universes/#comment-2889</link>
		<dc:creator>Michael Schmalz</dc:creator>
		<pubDate>Mon, 28 Dec 2009 21:46:09 +0000</pubDate>
		<guid isPermaLink="false">http://srlanger.wordpress.com/2006/05/24/cubes-and-universes/#comment-2889</guid>
		<description>Nice read...as the Supervisor of Mfg. Development at a Disk Drive Technology company, this debate is near and not-so-dear to my heart. We have been debating the vritues of Microsoft vs. SAP&#039;s BI tools for way too long.....and Donovan (above) nailed it with the comment &quot;...if I were to choose a platform from the developer perspective, SSAS definitely wins hand-down. However, when IT Ops weighs in, BO probably wins 9 out of 10 times.&quot;
I hope someone can someday cut thru all of the non-technical politics and make a decision on which direction our company should go....
Thks again for your insight....mjs</description>
		<content:encoded><![CDATA[<p>Nice read&#8230;as the Supervisor of Mfg. Development at a Disk Drive Technology company, this debate is near and not-so-dear to my heart. We have been debating the vritues of Microsoft vs. SAP&#8217;s BI tools for way too long&#8230;..and Donovan (above) nailed it with the comment &#8220;&#8230;if I were to choose a platform from the developer perspective, SSAS definitely wins hand-down. However, when IT Ops weighs in, BO probably wins 9 out of 10 times.&#8221;<br />
I hope someone can someday cut thru all of the non-technical politics and make a decision on which direction our company should go&#8230;.<br />
Thks again for your insight&#8230;.mjs</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: Stéphane-Robert Langer</title>
		<link>http://blog.srlanger.com/2006/05/24/cubes-and-universes/#comment-2888</link>
		<dc:creator>Stéphane-Robert Langer</dc:creator>
		<pubDate>Fri, 18 Dec 2009 20:35:46 +0000</pubDate>
		<guid isPermaLink="false">http://srlanger.wordpress.com/2006/05/24/cubes-and-universes/#comment-2888</guid>
		<description>Thank you Santhosh for taking the time to update this page with your findings.</description>
		<content:encoded><![CDATA[<p>Thank you Santhosh for taking the time to update this page with your findings.</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: Santhosh Shetty</title>
		<link>http://blog.srlanger.com/2006/05/24/cubes-and-universes/#comment-2887</link>
		<dc:creator>Santhosh Shetty</dc:creator>
		<pubDate>Fri, 18 Dec 2009 16:57:18 +0000</pubDate>
		<guid isPermaLink="false">http://srlanger.wordpress.com/2006/05/24/cubes-and-universes/#comment-2887</guid>
		<description>I see this blog is initiaed in 2007. There are few new features added to the latest release (3.1 sp2 ). Universe now has MDX editor and parsing.
Some of the fixes are added for SAP BW which have fixed some SSAS integration issues as well. But, there are differences between MDX generted by Excel and universe(from WEBI).
Never understood why universe creates sets for every measures in the universe when report is just using one of the measures.
Eventhough it is helping subsequent MDX queries where it is getting the data from session cache, but initial set creation causing performance issues.</description>
		<content:encoded><![CDATA[<p>I see this blog is initiaed in 2007. There are few new features added to the latest release (3.1 sp2 ). Universe now has MDX editor and parsing.<br />
Some of the fixes are added for SAP BW which have fixed some SSAS integration issues as well. But, there are differences between MDX generted by Excel and universe(from WEBI).<br />
Never understood why universe creates sets for every measures in the universe when report is just using one of the measures.<br />
Eventhough it is helping subsequent MDX queries where it is getting the data from session cache, but initial set creation causing performance issues.</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: Stéphane-Robert Langer</title>
		<link>http://blog.srlanger.com/2006/05/24/cubes-and-universes/#comment-2886</link>
		<dc:creator>Stéphane-Robert Langer</dc:creator>
		<pubDate>Mon, 14 Sep 2009 06:58:18 +0000</pubDate>
		<guid isPermaLink="false">http://srlanger.wordpress.com/2006/05/24/cubes-and-universes/#comment-2886</guid>
		<description>James,

Thank you for your feedback. I stand corrected - your suggestions for attribute sorting make perfect sense. There also seems to be &lt;a href=&quot;http://blogs.hexaware.com/business-intelligence/business-objects-way/business-objects-security.html&quot; rel=&quot;nofollow&quot;&gt;several options for fine-grained security&lt;/a&gt; which I really should have researched initially.

The object-to-object filtering you&#039;re describing seems like a useful option. There are Excel plugins on the market that would allow SSAS end users to apply any kind of complex filter but not without some exposure to MDX - which I agree isn&#039;t as easy as drag-and-drop.</description>
		<content:encoded><![CDATA[<p>James,</p>
<p>Thank you for your feedback. I stand corrected &#8211; your suggestions for attribute sorting make perfect sense. There also seems to be <a href="http://blogs.hexaware.com/business-intelligence/business-objects-way/business-objects-security.html" rel="nofollow">several options for fine-grained security</a> which I really should have researched initially.</p>
<p>The object-to-object filtering you&#8217;re describing seems like a useful option. There are Excel plugins on the market that would allow SSAS end users to apply any kind of complex filter but not without some exposure to MDX &#8211; which I agree isn&#8217;t as easy as drag-and-drop.</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: James Foreman</title>
		<link>http://blog.srlanger.com/2006/05/24/cubes-and-universes/#comment-2884</link>
		<dc:creator>James Foreman</dc:creator>
		<pubDate>Mon, 14 Sep 2009 02:37:31 +0000</pubDate>
		<guid isPermaLink="false">http://srlanger.wordpress.com/2006/05/24/cubes-and-universes/#comment-2884</guid>
		<description>Stephane,

Nice article, but a few points that nobody has seen to correct:

Attribute sorting - the date issue is a bit of a non-starter - that&#039;s to do with particular universe design, rather than a flaw in BO itself.  If you store your months as string data, you&#039;ll only get an alphabetical sort.  If you store your months as datetime data, and apply formatting in the report to make it read April, May, etc then it will sort in chronological fashion.  Now when it comes to arbitrary sorting of other column values, I feel you have a point, but that&#039;s something that has to be worked around in the DB layer, regardless of whether you&#039;d be using cubes or universes - in which case you could generate another column for ordering and use that in BO too.

Your security point is not correct; since (at least) 2000 BO has had the ability to enforce security down to row level in tables; it&#039;s not the case that one has to give all users complete access to the database.  From what I remember (it&#039;s been a good 7 years now since I last did admin work on a BO install) this should integrate with Windows authentication &amp; groups, if you want it to.

Named sets are nice, and that is a failing of BO.  On the other hand, a constant frustration for me in using SSAS via Excel is that I don&#039;t get BO&#039;s nice canned filters (eg dynamically generating last year) which gets you to a similar position.  What&#039;s more convenient (and maybe MS get round it elsewhere in the stack, but if you&#039;ve seen it then it would be nice to be enlightened) is the object-to-object filtering in BO.  If I&#039;m working for an airline and I want to get all the flights we sold where the destination country is the same as the departure country, in BO that&#039;s easy: you have object Destination Country = object Departure Country.  In Excel I seem to be stuck with building a big pivot and taking the leading diagonal.  I suppose if the predefined set existed, I could use that, but if it didn&#039;t then being able to define these relations on the fly in BO seems more beneficial to analysts who aren&#039;t deep in the IT department.</description>
		<content:encoded><![CDATA[<p>Stephane,</p>
<p>Nice article, but a few points that nobody has seen to correct:</p>
<p>Attribute sorting &#8211; the date issue is a bit of a non-starter &#8211; that&#8217;s to do with particular universe design, rather than a flaw in BO itself.  If you store your months as string data, you&#8217;ll only get an alphabetical sort.  If you store your months as datetime data, and apply formatting in the report to make it read April, May, etc then it will sort in chronological fashion.  Now when it comes to arbitrary sorting of other column values, I feel you have a point, but that&#8217;s something that has to be worked around in the DB layer, regardless of whether you&#8217;d be using cubes or universes &#8211; in which case you could generate another column for ordering and use that in BO too.</p>
<p>Your security point is not correct; since (at least) 2000 BO has had the ability to enforce security down to row level in tables; it&#8217;s not the case that one has to give all users complete access to the database.  From what I remember (it&#8217;s been a good 7 years now since I last did admin work on a BO install) this should integrate with Windows authentication &amp; groups, if you want it to.</p>
<p>Named sets are nice, and that is a failing of BO.  On the other hand, a constant frustration for me in using SSAS via Excel is that I don&#8217;t get BO&#8217;s nice canned filters (eg dynamically generating last year) which gets you to a similar position.  What&#8217;s more convenient (and maybe MS get round it elsewhere in the stack, but if you&#8217;ve seen it then it would be nice to be enlightened) is the object-to-object filtering in BO.  If I&#8217;m working for an airline and I want to get all the flights we sold where the destination country is the same as the departure country, in BO that&#8217;s easy: you have object Destination Country = object Departure Country.  In Excel I seem to be stuck with building a big pivot and taking the leading diagonal.  I suppose if the predefined set existed, I could use that, but if it didn&#8217;t then being able to define these relations on the fly in BO seems more beneficial to analysts who aren&#8217;t deep in the IT department.</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: Stéphane-Robert Langer</title>
		<link>http://blog.srlanger.com/2006/05/24/cubes-and-universes/#comment-2749</link>
		<dc:creator>Stéphane-Robert Langer</dc:creator>
		<pubDate>Sat, 20 Sep 2008 07:29:02 +0000</pubDate>
		<guid isPermaLink="false">http://srlanger.wordpress.com/2006/05/24/cubes-and-universes/#comment-2749</guid>
		<description>Donovan: thanks for chiming in and for understanding the limited scope of the comparison. You&#039;re also raising a good point here with regards to the reasons for organizations adopting a particular solution over another.

IBM/Cognos is certainly a strong contender. Starting from scratch, I guess I&#039;d also be looking at Oracle/Hyperion.</description>
		<content:encoded><![CDATA[<p>Donovan: thanks for chiming in and for understanding the limited scope of the comparison. You&#8217;re also raising a good point here with regards to the reasons for organizations adopting a particular solution over another.</p>
<p>IBM/Cognos is certainly a strong contender. Starting from scratch, I guess I&#8217;d also be looking at Oracle/Hyperion.</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: Donovan Hsieh</title>
		<link>http://blog.srlanger.com/2006/05/24/cubes-and-universes/#comment-2748</link>
		<dc:creator>Donovan Hsieh</dc:creator>
		<pubDate>Thu, 18 Sep 2008 23:38:45 +0000</pubDate>
		<guid isPermaLink="false">http://srlanger.wordpress.com/2006/05/24/cubes-and-universes/#comment-2748</guid>
		<description>Having developed &amp; used both BO Universe &amp; SSAS cubes in the past, your comparison is fair from the BI modeling perspective. I agree with your assessment that BO Universe is not as powerful as SSAS cubes functionality-wise. However, BO offers other reasons for consideration that SSAS doesn&#039;t have, i.e., scalability running on non-Windows boxes. Many enterprises wouldn&#039;t even consider Windows as production-worthy due to security and performance risks. That said, if I were to choose a platform from the developer perspective, SSAS definitely wins hand-down. However, when IT Ops weighs in, BO probably wins 9 out of 10 times.

Alternatively, I would choose Cognos (now acquired by IBM) since it has both strength of BO Universe &amp; SSAS cubes albeit with some compromise.</description>
		<content:encoded><![CDATA[<p>Having developed &amp; used both BO Universe &amp; SSAS cubes in the past, your comparison is fair from the BI modeling perspective. I agree with your assessment that BO Universe is not as powerful as SSAS cubes functionality-wise. However, BO offers other reasons for consideration that SSAS doesn&#8217;t have, i.e., scalability running on non-Windows boxes. Many enterprises wouldn&#8217;t even consider Windows as production-worthy due to security and performance risks. That said, if I were to choose a platform from the developer perspective, SSAS definitely wins hand-down. However, when IT Ops weighs in, BO probably wins 9 out of 10 times.</p>
<p>Alternatively, I would choose Cognos (now acquired by IBM) since it has both strength of BO Universe &amp; SSAS cubes albeit with some compromise.</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: Stéphane-Robert Langer</title>
		<link>http://blog.srlanger.com/2006/05/24/cubes-and-universes/#comment-2302</link>
		<dc:creator>Stéphane-Robert Langer</dc:creator>
		<pubDate>Sat, 15 Dec 2007 09:58:50 +0000</pubDate>
		<guid isPermaLink="false">http://srlanger.wordpress.com/2006/05/24/cubes-and-universes/#comment-2302</guid>
		<description>Mike: It is worth repeating that I know much less about BO than about Microsoft&#039;s BI stack. According to information from BO&#039;s forums however, there&#039;s a productivity pack for Business Objects Enterprise XI R2 (Service Pack 2) that allows you to access SSAS 2005 through a new OLAP tool called Voyager. That would tend to confirm your understanding that you need to invest in a new version, but please bring that up with BO if you&#039;re currently a customer.</description>
		<content:encoded><![CDATA[<p>Mike: It is worth repeating that I know much less about BO than about Microsoft&#8217;s BI stack. According to information from BO&#8217;s forums however, there&#8217;s a productivity pack for Business Objects Enterprise XI R2 (Service Pack 2) that allows you to access SSAS 2005 through a new OLAP tool called Voyager. That would tend to confirm your understanding that you need to invest in a new version, but please bring that up with BO if you&#8217;re currently a customer.</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: Mike Day</title>
		<link>http://blog.srlanger.com/2006/05/24/cubes-and-universes/#comment-2297</link>
		<dc:creator>Mike Day</dc:creator>
		<pubDate>Fri, 14 Dec 2007 14:46:25 +0000</pubDate>
		<guid isPermaLink="false">http://srlanger.wordpress.com/2006/05/24/cubes-and-universes/#comment-2297</guid>
		<description>&#039;There is no reason, by the way, why you couldn’t build SSAS cubes and make them available both directly for certain purposes and via a BO universe for other purposes...&#039; 

I&#039;m told that Business Objects Universes can&#039;t use SSAS 2005 cubes, short of investing in new versions of Business Objects.  Is this true?</description>
		<content:encoded><![CDATA[<p>&#8216;There is no reason, by the way, why you couldn’t build SSAS cubes and make them available both directly for certain purposes and via a BO universe for other purposes&#8230;&#8217; </p>
<p>I&#8217;m told that Business Objects Universes can&#8217;t use SSAS 2005 cubes, short of investing in new versions of Business Objects.  Is this true?</p>
]]></content:encoded>
	</item>
</channel>
</rss>
