What a Microsoftie Learned at MSFT
December 14, 2009
I wish this were a hypothetical. A “hypothetical” is one of those law school or business school conventions. Essentially, the players discuss an imaginary scenario, usually anchored tenuously to facts. “Stuff I’ve Learned at Microsoft” by Sriram Krishnan is not hypothetical. My hunch is that the blog post is a version of reality through the eye balls and other senses of the author. The core idea is that whilst working at Microsoft, practical knowledge moved from the company to the author of the Web log.
Just for fun, let’s take the learnings and then map them to some recent Microsoft products, actions, and services. This, of course, is a hypothetical, and I want you to enter into the spirit of the exercise. Put out of your mind the realities that make up * your * learnings about Microsoft. In the table below, the Sriram Krishnan’s learnings are in the left hand column and the addled goose’s learnings in the right hand column:
| Krishnan’s Learnings | Goose’s Learnings |
| Ask for forgiveness, not for permission | At least try it at European Union hearings. |
| (Most) Screw ups are OK | Consider Bob. Consider Vista. |
| Look for the line at your door | What if the person is In when she is out and out when she is in? |
| Code is king | What about the auto numbering feature in Word? |
| Lone wolf syndrome | Group-think produces products like SharePoint |
| Try out stuff | Hard to do when Apple products are not in favor |
| New team? Pick people over products | What if the people you want now work at Google? |
| Get out of your comfort zone | Create the Xbox and not address hardware failure rates |
| Ask the uncomfortable questions | Why did MSFT pay $1.23 billion for Fast Search & Transfer when actual revenues were in question |
| Go say ‘Hi!’ | If people are “in” |
| Praise in public, pull down pants in private | Comments about killing Google in the Kai Fu Lee affair |
| Best things are taken, not given | STAC compression |
| Don’t be an a**hole | See my write up about MSFT PowerShell cmdlets for Fast Search which is a dead link. |
I like these hypotheticals. We need more of them in search and content processing. For example, Microsoft’s enterprise search system scales in a cost effective manner.
Stephen Arnold, December 14, 2009
I wish to disclose to the Department of Commerce that I was not paid to write this goosely article. The commerce associated with products that do not work at advertised does generate a lot of dough. Too bad the goose does not know how to ride a gravy train without becoming the main course.
Governments, Data, Transparency, Threats, and Common Sense
December 14, 2009
A happy quack to the reader (one of two or three sad to say) who sent me a link to The Register’s “Gov Slams Critical Database Report as Opaque, Flawed, Inaccurate”. The idea is that the UK government has a bit of a tussle underway with an outfit called Joseph Rowntree Reform Trust. The Trust published a report. The UK government says, according to the Register, that the consultants got its facts wrong. In my experience, this is the pot calling the kettle discolored.
Here are some links provided by my colleague in the Eastern Mediterranean basin:
- http://www.jrrt.org.uk/uploads/Database%20State.pdf see especially “Developing Effective Systems” pdf
- http://www.justice.gov.uk/publications/docs/government-response-rowntree-illegal-databases-report.pdf
I think consultants get stuff wrong and I think governments get stuff wrong as well. This is the norm. The reason is that consultants don’t see government efforts from the government’s point of view. The government, on the other hand, has a tough time seeing consultants as much more than reasons to have another meeting. By definition, citizen facing data will be assembled with intent. By definition, consultants will be able to find fault with almost any data a government entity produces. When consultants produce data for the government and then the government makes those data available to citizens, then other consultants will rise to the occasion. In short, data, transparency, threats to the nation state, and common sense collide. Part of the landscape. Live with it, opines this addled goose.
Stephen E. Arnold, December 14, 2009
I wish to report to the manager of the US government’s Recovery.gov Web site that I was not paid to write this paragraph pointing out what seems obvious to geese living in Harrod’s Creek. Real humans may have another viewpoint. No problemo. I disclosed, didn’t I?
Trouble Looms for Enterprise Database Crowd
December 13, 2009
One of the developments I await in 2010 is the impact of open source databases in the enterprise. Most organizations don’t question the costs of their commercial databases. The big numbers paid to the IBMs, the Microsofts, and the Oracles of the data management world are grandfathered. A new president or CFO doesn’t try to cut these costs. The received wisdom is that the organization could not turn on lights or flush toilets without these blue chip, wind powered digital clipper ships of software. The reference to clipper ships is appropriate. When steam poked its smoking stacks into the fair wind, the end of the wind powered era had arrived. Codd databases are like these clipper ships. The RDBMS is clever but it cannot compete with newer technologies.
When I read “The New FOSS Frontier: The Database Market”, I thought about how sleek sailing ships ended up as scrap. The era of ugly, smoke belching steamers got the job done somewhat better, almost faster, and certainly cheaper. Those clipper ships could not stuff as much stuff in their holds as the chunky steamers. I have seen some slick RDBMS implementations in my time, but that time is drawing to a close.
As Tech News World’s article makes clear, the threat is not from the Google, although Google will almost certainly chose an inopportune moment to destabilize the database market. The threat is from open source databases. One of the comments in the article I found interesting was:
With Oracle dominating the commercial DBMS market, there is ample motivation for a community to create a challenger. Postgres has the breadth and depth of features to rival Oracle, and with commercial vendors (including EnterpriseDB) offering services, support, and the all-important one throat to choke, the database market is poised to be commoditized.
Then I noted this passage:
With a viable product available, a thriving community in place, and a market ready for commoditization, it is a safe bet that the database will be the next component in the enterprise to embrace open source, and it will likely see the success shared by Linux and JBoss. This is good news for all enterprise architects and project managers who have applications to build and a budget to balance.
In my view, the financial realities will force companies to look at grandpa and make some hard decisions. Anyone who has had to deal with an aging parent knows how difficult the decision to put grandpa in the assisted living facility. I think large organizations will treat their IBM, Microsoft, and Oracle RDBMS systems like grandpa. There will be some hand wringing but then it’s off to Sylvan Acres or whatever the facility is called.
With greater economic pressure causing more organizations to look at open source databases, I think the Google will use some of its nifty data management technology both in open source and more proprietary packaging to push the IBMs, Microsofts, and Oracles into more adrenaline pumping situations.
Should be interesting for this goose to watch from the cold, gray hollow in Harrod’s Creek. My pond will be more hospitable that the CFO’s office when she informs one of the legacy vendors that the good times are over.
Stephen Arnold, December 13, 2009
I wish to disclose to the National Park Service that I was not paid to point out that the article was written for free. My approach is similar to those national parks which allow a person to enter without charge. When one needs to stay overnight, then the cash register rings.
Editor & Publisher: Trade Publication Goes Dark
December 11, 2009
You have to admire the magazine sector. On one hand, the mainstream magazine folks in the US are creating an iTunes for magazine articles. On the other hand, according to CIO, “108 Year Old Editor and Publisher Going Out of Business.” The subhead is clear:
A going concern since 1901, Editor & Publisher — “America’s Oldest Journal Covering the Newspaper Industry” — is going away a time when that industry continues to get smaller by the day.
I wonder if Google could have saved this publication. I wonder if a mainstream magazine publisher could have saved this publication. I wonder if News Corp. could have saved this publication?
The answer must be “no” since the story said:
From a memo issued this morning by the publication’s owner, Nielsen Business Media: “We’ve made the decision to cease operations for Editor & Publisher and Kirkus Reviews. This move will allow us to strengthen investment in our core businesses – those parts of our portfolio that have the greatest potential for growth – and ensure our long-term success.”
The author of the story seemed to speak directly to the addled goose when he wrote:
Those media bashers who gloat over these losses are sadly misinformed, and, in some cases, cruelly oblivious to the pain being suffered by journalists and their families. Those who believe these losses won’t matter are simply wrong.
There are blogging opportunities galore in my opinion.
Stephen Arnold, December 11, 2009
I wish to report to the US Forest Service that I was not paid to write this short recycled news item. The trees in my backyard seemed to breathe a sigh of relief. Nah, just my imagination.
Burn and Churn in Content Processing 2010
December 10, 2009
The addled goose is awash in rumors, claims, and counter claims. We are now trying to unravel the following bits and pieces of information:
Is Siderean Software alive and well or in hibernatino? The Beyond Search team heard that one of the founders of Siderean, a developer of semantic content systems and methods, has moved to a big European publisher? The news on the site has been frozen at February 2008. We liked the firm’s approach to semantic content processing.
What are the properties owned by Mondo, a Danish company that was the subject of a Computerworld story? We have conflicting reports about who owns what software properties and the financial health of the various companies with similarly named products. We anticipate more local color from Hamlet’s home and the “to be or not to be” of these firms.
How can reports of a certain software company’s record breaking year match statements by a former employee? This vendor is in the womb-to-tomb business and operates from a location near Washington, DC? This is particularly baffling because public information reports a firm with the heart of a decathalon winner. The former employee is not on the same athletic field which is a crowded one when it comes of scanning and indexing paper documents.
What is the state of the acquisition of a search technology company that may be the object of desire by one of the world’s largest software companies? Sorry, the goose cannot point to the wooer or the wooee. This is a particularly troublesome item because it surfaces each year and then fades away. Baffling because the * big name * vendor’s search system is quite a challenged table tennis player.
Who is running the search and content processing show at Microsoft? The addled goose thought he heard one of the firm’s senior mangers reference a new sheriff in town? One rumor is that the new guru hails from the country trying to put Google executives in prison. We are chasing our pin tails on this one.
What’s going on in search and content processing? The addled goose is more confused than ever.
Stephen Arnold, December 11, 2009
Alert the US Postal Service! The goose buys stamps. No one at the USPS or anywhere else pays for this marketing Web log.
Reed Elsevier and Trade Newspaper Paywall
December 10, 2009
Reed Elsevier is trying to deal with the digital avalanche that is sweeping down Mount Information. I read in the Straits Times’s “Variety to Begin Charging.”
We fundamentally believe that the readers should pay one price and get all or any of our content,’ said Neil Stiles, president of Variety Group, a unit of London and Amsterdam-based Reed Elsevier Group PLC. ‘If you don’t pay, you don’t get anything.’ While the 104-year-old newspaper expects to lose many of its roughly 2.5 million monthly online visitors, it values more highly the 25,000 subscribers of its daily printed version and 30,000 subscribers of its weekly printed version.
The question is, “Will there be enough Hollywood hungry folks to make the content generate enough revenue to keep the lights on?” My hunch is that there will be some people who will pay, but the margins of the print publication from 10 years ago are not going to be achievable.
What will happen? I anticipate these events:
- Big splash.
- Lousy numbers
- Regrouping
- Relaunch
- Sale of the property.
Don’t get me wrong. Silobreaker’s consumer service is generating cash. That service uses smart software, not humans. AOL and Yahoo offer entertainment sites. I can create a Hollywood feed on Congoo.com with a few mouse clicks. These competitors are not performing equally well. That’s not the point. There are lots of sites that generate Hollywood content. You can download a podcast from KCRW that delivers “the Business.”
Something more than a paywall will be needed to keep Variety healthy. I have some ideas, but these are not for this free, Web log. Get my drift?
Stephen Arnold, December 10, 2009
I feel compelled by the imperative of a 40 page movie script to report to the custodial contractor for the Old Executive Office Building that I was not paid to write this opinion piece. Wow, confession cleans out the doubt.
Silobreaker Applies Intelligence Technology to Consumer Topics
December 9, 2009
Silobreaker, http://www.silobreaker.com, started out as a intelligence service for government agencies and competitive intelligence professionals. Its search function targets news and current events in Global Issues, Technology, Science, Business, Energy, and World topics. But it doesn’t just return a list of results, it aggregates a collection of information around the key words entered and presents it in a visual interface. Now Silobreaker is expanding into a more consumer-targeted market and offering its search aggregator’s services in the highly popular fields of sports and entertainment. http://sports.silobreaker.com/ and http://entertainment.silobreaker.com/ offer fans the chance to create their own targeted Silobreaker page with widgets to keep track of top stories, news, blogs, reports and research, audio and visual material, trends, quotes, and even material at YouTube.com. You can start with the standard generated page and filter results, or the customized user page can be shared with others. I think sports and entertainment fanatics who are willing to spend the time to set up the widgets would really enjoy this search service. Silobreaker likes the new services as well. Ads on the sites are generating “real money”, the managing director told Beyond Search.
Jessica Bratcher, December 9, 2009
I paid Ms. Bratcher to write this article. Silobreaker did not pay me one red cent, but I was assured of herring 12 ways when I am in Stockholm in May 2010. Yummy!
Search, Its Biggest Change, and Yawns
December 8, 2009
I try to steer clear of the search engine optimization crowd. A reader sent me a link to a write up called “Google’s Personalized Results: The “New Normal” That Deserves Extraordinary Attention”. The idea is that Google can personalize search results for every user in the world. Search Engine Land slaps the word “biggest” on this Google announcement. The idea is that users should be revved up, excited, concerned, involved, etc.
I suppose I should be excited, but the personalization can be turned off. I have noticed shaped search results for quite a while. The scale interests me. Personalization is one consequence of Google’s adaptive functions. Newly visible to users, not new.
Stephen Arnold, December 8, 2009
Oyez, oyez, I want to disclose to the Geological Survey (USGS) that this new world has been explored already. I did my write up without any payment. Tough to charge money to state the obvious.
Libreka: Belgium Builds a Google Killer for the Europeans
December 8, 2009
I am not sure if this headline is accurate, but that’s what I wrote in my notes after I saw a demo of Libreka. This is not a new service. Ars Technica wrote about it in October 2007. You may find this write up useful: “German Publishers Challenge Google Book Search with Libreka.” The article said:
The new program, called Libreka, has attracted plenty of German publishers who like its “opt-in” approach. Publishers who don’t want to make snippets or sample pages of their works available have that option, unlike with Google, which shows tiny snippets of text even from copyrighted works. Those who want to offer sample pages and make their books searchable can do that too.
I saw a document on Wikileaks that suggested that the Libreka service was not scoring hat tricks at the cash register. You can find that document here.
Libreka has hit my radar a number of times. I have in my files a copy of presentation given a year ago. You can find the document online here. The Libreka business model is quite ambitious, maybe too ambitious:
The document asserts:
- 100,000+ books fully searchable online
- 30m+ book pages online
- 11,000+ E-Books for sale
- 1,200 participating publishers
- 600 participating booksellers.
According to my source in London last week, Libreka is software built by Bureau Marcel van Dijk. You can run queries on the system. Notice that it has a number of Google Books features. The service offers a “wish list”, which allows me to “reserve a book”. I am not sure I will use this feature. User query terms are highlighted in the page displays. I set up a user account, and then I was able to run a query and display a number of pages. I installed the Adobe Digital Editions software. (Once I ran a couple of queries, I uninstalled this software. Yuck. Adobe.) The system also supports PDF “flavors”. (Yuck. Adobe.) My estimate is that the pages displayed are dictated by the publisher participating in the program. If you read German, there’s an FAQ at http://www.libreka.de/help#faq. If you are a Google Translate fan, the Google system cannot parse this particular url, so you will be on your own or you will need to find a German reading friend to assist you.
Too little, too late in my opinion.
Stephen Arnold, December 8, 2009
Oyez, oyez, Library of Congress. I wish to disclose that I am writing this article for free and in English. My German is rusty but I figured out enough to conclude that German booksellers are pushing the wrong way in the digital data flows.
Government 2.5: Traditional Information Technology Evolves
December 7, 2009
I have just returned from my endnote at the International Online Conference in London. On December 14, 2009, I will be taking one of the 10 trends for 2010 from my London UK talk and expanding on the idea of dataspaces, not databases. Most governmental entities are anchored in traditional database technology. Although state of the art in the 1970s, the RDBMS framework is ill suited for the rigors of Government 2.5 information.
I will be attending the CoolBlue Government 2.5 conference in Washington, DC, on December 14 and 15, 2009. You can get full details about the conference from the program’s Web site.
You can get a glimpse of what’s in my talk. Just search this Web log for the term “dataspace”, and you will get some background information. The dataspace technology is one of Google’s crown jewels, and it a core capability little known outside of a small circle of wizards. You can see a tiny fragment of the dataspace technology in action if you navigate to the Google Wave information page and do some exploration.
My remarks created quite a stir in London on Thursday, December 3, 2009, and I anticipate a similar reaction in Washington on December 14, 2009. Googlers are largely unaware of the dataspace technology, how it embraces the Google programmable search engine, and the company’s push to become the Semantic Web.
I will be linking these technologies to likely government use cases. If you want to talk after the event, just write me at seaky2000 at yahoo dot com. I will make time to visit with Government 2.5 attendees.
Stephen Arnold, December 7, 2009
Oyez, oyez, I want to alert the mayor of Washington, DC, that I was not paid to write this blatant self promotion or mention the CoolBlue conference. I think the conference’s PR manager will buy me a Diet Pepsi. I have my Web feet crossed.
-
- Subscribe to Beyond Search
Feature archive
News archive
-
