Herding SharePoint Content Sheep
November 2, 2008
Microsoft may be pushing Fast Search’s ESP into large SharePoint installations, but certified gold partners continue to find opportunities to make money from the 100 million SharePoint installations. Autonomy and Open Text recently rolled out systems that make it easier to keep control of SharePoint content. Why control SharePoint documents? You will learn quickly enough when you get caught in a legal matter and have to figure out which version of each document is the one that is the “right” one. SharePoint offers primitive and clunky controls for herding SharePoint content sheep; that is, the many bits and pieces of modular documents, emails with attachments, and PowerPoint decks with some relevant information but perhaps not the best and final version of the deck. When you get the invoice for collecting documents manually, you will understand why you have to have robust tools for governing information.
Let’s look briefly at two products that herd content sheep:
First, Autonomy has rolled out Controlpoint. You can read more about the product on the Autonomy Web site here and in the Marketwatch write up here. In a nutshell, you install Controlpoint, and you get policy-driven control of all SharePoint content. Autonomy includes a number of content processing functions with Controlpoint; for example, classification of documents.
Next, Open Text dubs its governance solution the sveltly named Open Text Content Lifecycle Management Services for Microsoft SharePoint, eDOCS Edition. The acronym is OTCLMSMSE. Between you and me Autonomy does a better job naming products. The Open Text solution delivers life cycle management, policies, and archiving functions. A licensee can hook SharePoint into Open Text’s other enterprise content management services as well. You can read CMSWire’s write up here.
Let’s step back and think about SharePoint. Autonomy and Open Text have identified a glaring weakness in the Frankenstein SharePoint. SharePoint is, according to some, Microsoft’s next generation operating system. I think that’s pretty wacky. SharePoint is a product that changes with each release. First it was content management. Then it was collaboration. Now it is knowledge management. The Microsoft sale pitch makes SharePoint seem easy, cheap, wonderful, and the cure for what ails a modern organization. In reality, SharePoint lacks the chops to deal with content when a lawyer shows up to collect information as part of the discovery process. SharePoint doesn’t do any single function particularly well. What it does is deliver stub functions that work okay when you have two or three people using a small number of documents. Increase the number of documents and the number of users, and you have a multi million dollar investment to get the system stable and running with acceptable performance.
I am willing to go out on a limb and say that Microsoft will introduce enhanced policy and information governance features. The Microsoft certified professionals will install these extensions, and the company will find that liberal injections of money and technical resources will be needed to get the amalgamation working in an acceptable way.
In the meantime, Autonomy and Open Text should be able to make sales. When Microsoft rolls out its own governance solutions, engineers at these companies will develop a fix for another void in SharePoint.
Stephen Arnold, November 2, 2008
SAP: Trying a Little Too Hard
November 2, 2008
SAP, home of the TREX search system, seems to be trying almost too hard to convince me that the company is humming like a top. In CXO, a publication which ran one of my articles a year or so ago, published here “SAP Helps Weather Economic Storm” by the CXO Staff. The article is short but it hits on themes that, I presume, resonate with beleaguered corporate types who face unprecedented credit challenges and a generally lousy economic climate. SAP is a company with middleware. Increasingly, this middleware runs on Microsoft servers. Microsoft at one time wanted to buy SAP because its midddleware commands multi million price tags and requires considerable hands on configuration, tuning, and customization. Like Baan and other enterprise wide umbrella solutions, SAP is a beast with quite a few moving parts. In fact, in one SAP installation, I recall hearing, “It was easier for us to change our processes to fit what SAP had than try to modify SAP to fit out methods.”
The CXO article makes these points:
- Real time tools to manage in today’s financial climate
- Special incentives are available
- Easy deployment
- Business intelligence available
What is interesting about this write up is that it is almost a brochure. There’s no mention of search; that is, finding what you need in the files and the SAP generated and distributed meta data that makes the SAP integration of information possible. There’s no hint that the the cost and deployment time have changed significantly. Nope. In my opinion, SAP is struggling. The IBM magic that juiced the company when it was founded is gone. The interest Microsoft showed in the company seems to be on the back burner.
The emergence of cloud computing, SharePoint, and enterprise publishing systems make the SAP solution seem out of step with the times. SAP may be the harbinger of change in enterprise applications that integrate disparate systems and data. And what about search? CXO has shifted search to business intelligence. That’s a nice touch as well but I still have to locate the purchase order, the manufacturing order, and the memo about the change to the order.
Can SAP adapt? Unlike the CXO Staff, I am skeptical. Oh, the “X” means any type of chief officer; for example, chief technical officer, chief financial officer, chief strategy officer. A bit of algebra never hurts.
Stephen Arnold, November 2, 2008
Microsoft and Cloud Math
November 2, 2008
Long day in a city with no consonants. Cruising the goodies in my newsreader, I noticed Todd Bishop’s “Microsoft Faces Uncertain Financial Future As It Looks to the Cloud.” You will enjoy the story here. Mr. Bishop reports that Ray Ozzie, Microsoft’s cloud guru, mentioned thinner margins, which means to me “we’ll make it up on volume.” Then came what for me was the most significant passage in the article:
Those thinner margins could hurt Microsoft’s overall business if cloud computing causes companies to make a big shift away from traditional computer servers. Microsoft’s Server & Tools Business — which sells operating systems, database software and other programs for those servers — has been a financial mainstay for the company. However, Ozzie still sees a big role for those traditional servers inside most companies, even as many applications move online. “People will still run most of their big business systems — the back ends of their big business systems — on premises,” he said.
Two assumptions seemed to fuel Mr. Ozzie’s assessment. First, I am no longer convinced that organizations will want servers “on premises”. The work force demographics favor a more fluid approach to organizing work and the work space. Visit a company run by 20 somethings. What I notice is how few people are working “on premises”. Most of the organizations serve clients and some make things, maybe in a remote land, but the gym shoes and the shirts turn up where they are supposed to be.
Second, I used to think I wanted software on machines I controlled. I moved most of my enterprise software to someone else’s servers, and I just “fired” one service provider and moved data and applications to another server provider more quickly than I could have done this with my own gear. I think other folks will make this shift as well.
The larger question in my mind is, “When.” It is okay to announce new initiative and show demonstrations. Delivering a stable, scalable, reliable product is a different type of work. Microsoft has some housecleaning to do to get Tess’s favorite enterprise software to behave. SharePoint requires certified professionals to do too much manual fiddling to activate basic functions to keep her from growling. Even interface consistency in desktop applications needs work. Some Microsoft professionals have insisted that these are details. That may be true. But if the Azure clouds strike me with a lightning bolt, I will be increasingly difficult to convince that Microsoft can deliver on its promises and make money at the same time. And with the police matter clouding enterprise search, how will that help or hinder the FAST ESP push? There is a difference between announcing and delivering across disparate systems and the cloud when it drifts in.
Stephen Arnold, November 1, 2008
German Publishers Are Not Thrilled with Google
November 2, 2008
News and commentary about Google is everywhere I look in central Europe. I can’t reveal my location because I can’t spell the name of the city I’m in. Too many consonants. The buzz at dinner this evening (I am writing this at about midnight on what I think is October 30, 2008) was about German publishers and Google. In the US, Google has entered into a truce with publishers about book scanning. I have been critical of the “dead tree crowd” for many years, and now Google has new diners in the Google cafeteria.
According to Deutsche Welle here, “German publishers [are] accusing Google of controlling culture.” That’s a broad claim, but the new story makes this point, which I found interesting:
“This accord is like a Trojan Horse,” Alexander Skipis, chief executive of the Boersenverein, said in a statement on Thursday, Oct. 30. “Google aims to achieve worldwide control of knowledge and culture. “In the name of cultural diversity, this American model is out of the question for Europe,” he said, adding that it contradicted “the European ideal of diversity through competition.”
Deutsche Welle’s news story is quoting a highly regarded publishing professional, and I think Mr. Skipis is beginning to understand what it means to organize the world’s information. On one hand, I understand the feeling embedded in Mr. Skipis’ comment. On the other hand, the GOOG has been doing its thing for 10 years. The GOOG is a publicly traded company. The GOOG has been using its business model and plumbing like giant earth movers to shift the boundaries of business sectors.
My net net: you haven’t seen anything yet. You will be able to read about it in my new Google and Publishing monograph in December, however. If you want to reserve a copy, navigate to the www.infonortics.com Web site. Contact information is there but a separate order form is not yet online due my heel dragging and the clean up for Martin White’s and my new study Successful Enterprise Search Management due out in November 2008 unless a glitch occurs.
Stephen Arnold, November 1, 2008
Microsoft and Search in a Time Warp
November 2, 2008
In grade school in Illinois, I recall learning the meaning of the word anachronistic. For some reason, the explanation has stuck with me for more than 55 years. The teacher, Miss Chessman, told the class, “Ancient Greeks did not have an alarm clock.” The idea is that if you mix up when things occur, you run the risk of creating the equivalent of a dog’s breakfast.
My newsreader delivered CIO Magazine’s “Search Will Outshine KM” by Mike Altendorf to read. I don’t know Mr. Altendorf, and I have to admit that I disagree with a couple of the points he makes in this two part article, which you can read here. I am a tired goose, and I don’t want to trigger a squabble in the barn yard. I do want to point out where Mr. Altendorf and I part company.
First, the notion of search outshining KM is not something I have thought much about. KM is mostly baloney, one of those “trends” that promise much and deliver little that one can measure. When I watch intelligent people leaving one company for another, no software system captures what that person knows. IBM is trying to prevent a chip designer from leaving Big Blue to join Apple. If KM worked, IBM wouldn’t take such extreme action to prevent a person from changing jobs. That’s KM for you. It doesn’t deliver. And search? It is, in general, not too helpful either. In fact, search is one of the few software systems to engender a dissatisfaction rate among its users of 60 to 70 percent. In my opinion, search outshining KM is a silly assertion, and one that makes it seem that one lousy system can deliver information better than another lousy system. Both search and KM work best when applied to specific problems and bounded by realistic expectations and budgets.
 |
 |
Second, the reference to Microsoft’s acquisition of Fast Search and Transfer and Powerset is startling. First, Mr. Altendorf makes no reference to the police action in Oslo that threatens to undermine the credibility of the Fast Search technology, finances, and executives. Second, Powerset is not complement to Fast Search technology for two reasons: [a] Powerset technology is part of the Live.com service and has not to my knowledge been hooked to the Fast Search system at this time and [b] notion that Microsoft can “tie together” disparate technologies is out of touch with reality. Let me be clear. Microsoft has compatibility issues within its own product families; specifically, SharePoint and the Dynamics range of software. When you toss in the Fast Search conglomeration of original code, the bits of open source Fast Search has used in its system, and the technology from the acquisitions Fast Search made prior to its purchase by Microsoft, you have quite a bit of integrating to do. Now add the Powerset original code with the licensed technology from Xerox Parc, and you have even more work. Microsoft’s units can’t make the ribbon interface consistent across Outlook, Word, and Visio in Office 2007. Mr. Altendorf’s blithely reassures me that Microsoft can work out these incongruities. I beg to differ.
Google: No More Sneakers and T Shirts in DC
November 1, 2008
It’s raining and cold. I am in an Alan Furst novel I think. I found an Internet cafe, but it is full of teens who would prefer an old dude in black not sit amongst them. I did not have time to watch the full Obama infomercial. I was able to read about Google’s CEO appearing in the segment. Based on the tiny pictures that finally resolved in a country with few vowels in its name, Eric Schmidt has on a suit. I was in Washington, DC several years ago when one of the Googlers visited House and Senate members. The Googlers had on, as I recall, T shirts and sneakers. How times have changed. The top Googler looks just like the late Frank Press, former White House science advisor. I can imagine Mr. Schmidt sitting in the OEB with a faux mainframe as a desk and blue diodes winking gently in an infinity mirror. No, I am having a flight of fancy. Mr. Schmidt, as I recall, does not want a White House post. But I have seen Potomac Fever grab by the nose and pull them irrevocably to the power center of America. You can read about the new Google savvy regarding DC power politics here. Definitely no T shirt. Can’t see the sneakers, so I don’t know if those complete the Googler’s wardrobe. Google’s getting with the political program. It only took a decade.
Stephen Arnold, November 1, 2008
Google’s Challenges Defined
November 1, 2008
Jeffrey Rayport’s “Can Anything Derail Google’s Growth Engine?” is a round up write up. Mr. Rayport’s analysis hits the big points about Google and its dominance in Web search. Business Week has put a line in the sand, and Google’s rivals have to do something or get used to breathing the GOOG’s exhaust fumes. You can read this story here. For me, the most interesting comment in the article was this passage:
So how does Google extend its winning streak? In the end, it’s arguably all about amassing the most talent without falling prey to what’s known as Big Company disease, a condition that can leave a once-fleet-footed business complacent, overstaffed, and bureaucratic. In a recession, Google will need to do what “normal” companies do: streamline its workforce, cut budgets, and realign investments.
Mr. Rayport’s article will become a touchstone for the foreseeable future. The focus on ad revenue, the promise of Google Apps, and the reference to “Big Company” disease will define some of the conversation about Google.
There are, in my opinion, several other issues that warrant discussion; for example:
- Regulation. Who will regulate what at Google? The company is a de facto monopoly in advertising and Web search, and it operates globally. Its ad efficiency is different outside the US, but the GOOG is a modern day Standard Oil, US Steel, and Leland Stanford railway in digital form.
- Collateral damage. Google’s success has created opportunities. One new sector is search engine optimization, but that business may be squeezed by Google itself as the company implements some of its newer technologies such as the programmable search engine. Other businesses may find that Google’s ability to slash transactions costs will be highly disruptive, thus exacerbating some firms’ financial problems and opening the door to Google in other business sectors almost “by accident.”
- Culture. Google continues with its math club culture. As Google grows, will the company be able to deal with the loose ends that are becoming more evident; for example, the issues with the Android G 1 phone’s security, the Gmail outage, and company’s legendary indifference to customers who want to know how to make a service work or why a Web page disappeared from the Google index.
Mr. Rayport has defined the Google of late 2008, but more needs to be done to address Google’s strengths and weaknesses. The Business Week article is a good start. Now more is needed.
Stephen Arnold, October 28, 2008
SharePoint and Azure
November 1, 2008
The Microsoft cloud computing initiative is underway and it is early days. My principal concern with cloud computing is that it sounds so compelling. Imagine. Reduce your information technology staff. Get out of the hardware business to some extent. Let someone else worry about upgrading software and stamping out bugs.
According to Arpan Shah’s Web log here, Microsoft SharePoint Services will be part of the Azure Services Platform. You can get more information from Microsoft here. Many details are fuzzy. In the weeks and months ahead, more information will be forthcoming. I have heard that sample code, downloadable executables, and documentation will be provided. Microsoft does a good job of providing information. The challenge for me is finding it.
Before leaving the subject of the link up between Azure and SharePoint, let me capture for my own use the issues that I have with cloud computing in general. (I am an addled goose and this Web log is primarily for me, a digital notepad if you will.)
First, it is expensive to build, deploy, scale, and upgrade cloud infrastructure. I am not sure I can quantify the costs any more clearly than I have in my monographs and studies. CFOs and beancounters don’t understand the non linearity of the costs. I suppose this knowledge will become more widely available after a few of the new entrants in cloud computing explain why their balance sheets are awash in red ink.
Second, the problems of bottlenecking, unexpected faults, and the time required to chase down and figure out problems few or maybe no one has seen before increase over time. Cloud infrastructures are tough to lock down. Every time I turn around a vendor was pushing a firmware update, a point upgrade, or some other piece of software at the data centers in which I was involved. When one of these “fixes” is problematic, I had trouble. What few understand is that the longer the cloud infrastructure operates the more likely there will be unexpected dependencies. These data centers are not workstations, yet many people view them as giant iPods. Crazy.
Third, the client experience is gated by bandwidth and latency. The hottest cloud infrastructure is worthless if the user is trying to access email or some other information via a sluggish network. How stable are network connections in Kentucky? Not too stable. Connectivity is much better in the major European cities. Tokyo is outstanding. But when I have had to work from cities with now vowels in their name, life gets tough. Joburg was okay. Djuma was not. Xian featured great connectivity. Other places in China weren’t so hot, so I used Internet cafes and faced with dial up modem speeds. My guide assured me that the cafe was a broadband hook up. Yeah, right.
What to these three points have to do with Azure and SharePoint. Not much because so far Azure strikes me as a suggestive demo. Down the road, Microsoft will have to deal with SharePoint’s own performance and stability plus the three points I just mentioned.
That is going to be a big, costly job for the folks at Microsoft. The goose issues a gentle quack of support.
Stephen Arnold, November 1, 2008
Google Adds Functionality to Container Documents
November 1, 2008
Chrome is here. Containers are all the rage even though users don’t know what these are. Google’s “controlling Communication within a Container Document” discloses interesting cross boundary functionality. Published by the USPTO on October 23, 2008, this invention by Michael Buerge and a handful of other Googlers says:
A system allows modules associated with different domains to communicate, such as within a container document. To transfer payload data from the first module associated with a first domain to a second module associated with a different domain, the first module adds the payload data as a text string to the URL of a transport module associated with the second domain. This way, the second module may directly access the modified transport module to obtain the payload data from its URL. The second module may likewise add other payload data as a text string to the URL of another transport module associated with the first domain, thereby enabling communication from the second domain to the first.
If a container sounds a bit like a virtual machine, I think you may be interpreting this prose in a similar way. Download US20080263566 here. I am opening this topic in my forthcoming Google and Publishing Monograph that will be available in December 2008 after Martin White’s and my new study Successful Enterprise Search Management from Galatea in the UK.
Stephen Arnold, October 31, 2008
Brainware: Nailing a Big Deal
October 31, 2008
Brainware won’t reveal the name of its client, but the “diversified energy company” bought $2.6 million worth of Brainware technology in October 2007. Brainware explains that this is a “follow on contract”. This size of this deal moves Brainware into Autonomy and Endeca territory, two vendors noted for their ability to land large search and content processing deals. Brainware describes itself as a vendor of “intelligent data capture and enterprise search solutions.” The firm offers what is called “end to end” solutions. Like OpenText and ZyLab, Brainware can put in place scanning and content capture and conversion systems, indexing, content processing, and information access tools. The idea is that paper or digital information goes in at one end of the system and the user can access that content at the other. You can read an interview with one of the Brainware executives here. I profiled the company in Beyond Search, published by the Gilbane Group. More information about that study is here.
According to the Brainware news release here:
This private sector oil company is deploying Brainware Distiller as a front-end data capture solution in conjunction with its global ERP rollout. The Brainware Distiller solution will process millions of invoice pages per year from more than a hundred different countries through their shared services centers located across Europe and the U.S.
A happy quack to the Brainware team. Now, what’s next for the Ashburn, Virginia, company? If you haven’t sold Halliburton, maybe you should aim for Shell or BP next? I found the announcement interesting because Brainware is associated with eDiscovery, not enterprise search, in my mind. Like Recommind, Brainware seems to be making an effort to penetrate new markets with its patented technology for information processing.
Stephen Arnold, October 30, 2008
-
- Subscribe to Beyond Search
Feature archive
News archive
-
