Planet FoxPro

Photo of the Day

2010-03-20T10:01:00Z

A nice winter day in early March in Brickell neighborhood, Miami.

FoxRockXTOC

2010-03-19T16:14:52Z

Editor comments: TOC Fox Rock X March 2010

FoxRockXIssues

2010-03-19T15:33:55Z

Editor comments: Fox Rock X 13 added by Rainer Becker

ChrisProbst

2010-03-19T15:02:09Z

Product Manager for Academic Management Systems in Amherst, NY. Currently managing a web based course evaluation system.

VFPVcxStructure

2010-03-19T13:28:39Z

Here are the fields in the VFP VCX structure, and their meanings:

Photo of the Day

2010-03-19T10:03:00Z

AeroMexico Boeing 737 landing at Miami International

As some of you know, if you've seen my full gallery, I do quite a bit of planespotting, but I do not post many here. This one was special so I thought I'd share.

On St. Pat’s Day what’s better than The Dubliners?

2010-03-19T03:51:52Z

(YouTube link)

TableFileStructure

2010-03-18T17:35:25Z

Don't we need permission to post stuff ripped out of the help file?

FormalFoundation

2010-03-18T12:07:50Z

A formal foundation ensures models are unambiguous and executable (e.g. to automate the storage, verification, transformation and simulation of models). One particular benefit is to allow formal proofs of equivalence and implication between alternative models for the same application. Although ORMs richer graphic constraint notation provides a more complete diagrammatic treatment of schema transformations, use of textual constraint languages can partly offset this advantage. With respect to their data modeling constructs, both UML and ORM have an adequate formal foundation.

Photo of the Day

2010-03-18T10:09:00Z

Chevrolet Bel-Air at a car show in Miami Beach back in January.

Enabling CRUD for SharePoint 2010 external lists using Visual Studio 2010

2010-03-18T08:51:00Z

Recently we have published a blog Walkthrough of enabling CRUD for SharePoint 2010 external lists using Visual Studio 2010. This walkthrough shows you how to pull data from an external database into an external list and enable Create, Read, Update and Delete (CRUD) functions to the external list. I copied some paragraphs for your preview.

“In our last blog of this series Walkthrough of creating a SharePoint 2010 external list using Visual Studio 2010 Beta, we introduced how to create a simple “Hello world” external list in SharePoint 2010 Beta using Business Data Connectivity Designer in Visual Studio 2010 Beta.

In this blog, we will show you how to pull data from an external database into an external list and enable Create, Read, Update and Delete (CRUD) functions to the external list.

First of all, you need to have SharePoint 2010 Public and Visual Studio 2010 installed on your machine in order to complete this walkthrough. We’ll use “Northwind” database as external data source, so if you do not have an existing “Northwind” database available, we’ll walk you through to create a local database using SQL Server Express first.

At the end of this post, we will complete with a BDC model project which has a “Customer” entity connects to “Customer” table in “Northwind” database, and have CRUD operations enabled. "

Open XML SDK 2.0 Released!

2010-03-17T20:50:47Z

Many programs that need to process documents often require manipulation of the file formats directly and doing that through the Office component object model (COM) won’t scale very well. It also requires that Microsoft Office be installed to run. A better route in a lot of these cases is to use the Open XML SDK. Starting with Office 2007, Word documents, Excel spreadsheets and PowerPoint presentations are based on an open standard for packaging XML files called Open XML. You can use the SDK to aid in the development of these types of applications.

I have been using the CTP of the SDK for almost a year so I’m happy to hear that version 2.0 has finally been officially released!

Download the Open XML SDK 2.0 here.

Read the documentation here.

Using Visual Basic’s powerful and simple implementation of LINQ to XML and the Open XML SDK you can work with these new document formats much easier than ever before. Check out some of my samples:

Article: Taking Advantage of LINQ and XML in Microsoft Office 2007

Article: Mail Merging Into a Single Word Document Using XML Literals

Article: Merging Text & Photos into a Word Document using Open XML SDK

Screencast: Showing off the Open XML SDK and LINQ

Screencast: More Fun with Office and XML Literals

Code: Working with Open XML in Visual Basic

What other bloggers are saying:

Also check out the MSDN Open XML Developer Center for more news, downloads and technical content.

Enjoy!

Taking out the Slow Parts, Again

2010-03-17T19:00:21Z

While looking up something else, I came across Brad Martinez’s article on extending the functionality of the TreeView control. One of his points about how to load the TreeView more quickly caught my eye:

Make sure the parent Node's Sorted property is set to False: If Sorted = True, the TreeView must sort every Node as it is added under the parent Node. Make Sorted = True after all child Nodes are added.

I wondered how much of an improvement it would make, so I changed some generic TreeView loading code I use in lots of places to follow his advice. Loading 1,329 nodes dropped from 4.128 to 0.533 seconds, which is almost eight times faster!

As I’ve said before, I love taking out the slow parts!

Ultra Simple Buttons with GDI+X

2010-03-17T12:45:00Z

Not another set of buttons you might say. Actually I developed these buttons for a POS app a while back and finally got round to posting them.The buttons are simple to use with all properties you need to set in the Favourites tab. oFontName - The name of the font to uselShowBell - This gives a bell shape to the gradient. Basically it changes the style of the button. See the image below.nButtonColour1nButtonColour2 - The 2 colours that will be used for the button when nButtonType = 4. Use the colour...(read more)

Photo of the Day

2010-03-17T10:04:00Z

Taking a bath
Duck in Miami Metrozoo

DelLee

2010-03-17T01:56:39Z

Del Lee is Systems Manager for Hahl, Inc., formerly Glassmaster Monofilament Division. Hahl, Inc. is part of Lenzing, a large multinational company based out of Austria. He is currently responsible for documenting Quality, Environmental, and Health & Safety Processes, while at the same time serving as the IT Manager. In addition, he was recently named the ISO 9001 Management Rep as well as OHSAS 18001 Management Rep. He currently maintains a VFP app at Hahl that was written with Visual FoxExpress and accesses a SQL 2008 database. In the last several years, he has created or upsized data from VFP tables to SQL Server (2000, 2005, & 2008) for a number of applications in his consulting gig: Delton Lee Consulting, Inc..

Use a Custom Allocator for your STL container

2010-03-16T23:23:00Z

Last time in Play around with .Net Dictionaries and STL maps, we talked about various containers. Today we’ll write our own allocator for a particular STL container.

A general purpose allocator, like malloc in the C Runtime library or HeapAlloc in the Windows API needs to handle allocation requests of various sizes. The calls to free the memory can come at arbitrary times too. The allocator might have to track the various free blocks somehow, perhaps in a linked list. Fragmentation can occur, causing inefficient use.

A custom allocator can take advantage of various aspects of memory requestors:

· Perhaps every request is identical in size (as in the case of an array or growing a tree)

· Perhaps your usage pattern is to never delete an entry: there are only 12 months in the year, 50 states in the union, even for the forseeable future J

· Perhaps you can preallocate a large enough block for the desired scenario: then free the whole block at once when done with the data (this is the No Release Allocator: you never free the memory til the end (perhaps a destructor can free it at the end for you)). This is what our sample does below.

In some of my memory tracking diagnostic utilities, I use a custom allocator to allocate from a private heap.

The output is just 3 numbers per line, separated by spaces. You can start Excel, paste those numbers in, then choose Data->Text to Columns to convert the data to columns, then graph the data with a few keystrokes:

· Ctrl-A Select All

· Alt-N Insert

· N Line

· Enter 2-D line graph

If you want to add a column to indicate what percent faster the 3^rd is than the 1st: =+(A1-C1)/A1*100

A custom allocator can boost performance. Indeed, using my custom allocator I was able to get around 30% faster performance.

Beware when you do timings that you use a Release build and you’ve read this: You can develop code faster

Examine .Net Memory Leaks

Customize the VS debugger display of your data

Article on allocators: http://www.ddj.com/cpp/184403759

Dim a = New AtlInteropLib.TestInterop

a.Test("From VB")

// TestInterop.cpp : Implementation of CTestInterop

#include "stdafx.h"

#include "TestInterop.h"

#include

// CTestInterop

long g_nCnt = 0; // total # blocks allocated

long g_nTot = 0; // total allocated

template <class T>

class MyCustomAlloc

A custom allocator: given a pool of memory to start, just dole out consecutive memory blocks.

this could be faster than a general purpose allocator.

E.G. it could take advantage of constant sized requests (as in a RedBlack tree)

{

public:

typedef T value_type;

typedef size_t size_type;

typedef ptrdiff_t difference_type;

typedef T* pointer;

typedef const T* const_pointer;

typedef T& reference;

typedef const T& const_reference;

MyCustomAlloc(byte *pool, int nPoolSize)

{

Init();

m_pool = pool;

m_nPoolSize = nPoolSize;

}

MyCustomAlloc(int n)

{

Init();

}

MyCustomAlloc()

{

Init();

}

void Init()

{

m_pool = 0;

m_nPoolSize = 0;

g_nCnt = 0;

g_nTot = 0;

}

MyCustomAlloc(const MyCustomAlloc &obj) // copy constructor

{

Init();

m_pool = obj.m_pool;

m_nPoolSize = obj.m_nPoolSize;

}

private:

void operator =(const MyCustomAlloc &);

public:

byte *m_pool;

unsigned m_nPoolSize;

template <class _Other>

MyCustomAlloc(const MyCustomAlloc<_Other> &other)

{

Init();

m_pool= other.m_pool;

m_nPoolSize = other.m_nPoolSize;

}

~MyCustomAlloc()

{

}

template <class U>

struct rebind

{

typedef MyCustomAlloc other ;

};

pointer

address( reference r ) const

{

return &r;

}

const_pointer

address( const_reference r ) const

{

return &r;

}

pointer

allocate( size_type n, const void* /*hint*/=0 )

{

pointer p;

unsigned nSize = n * sizeof(T);

if (m_pool) // if we have a mem pool from which to allocated

{

p = (pointer) m_pool;// just return the next available mem in the pool

if (g_nTot + nSize > m_nPoolSize)

{

_ASSERT(0);//,"out of mem pool");

return 0;

}

m_pool += nSize; // and bump the pointer

}

else

{

p = (pointer) malloc(nSize);// no pool: just use malloc

}

g_nCnt+=1;

g_nTot += nSize;

_ASSERTE(p);

return p;

}

void

deallocate( pointer p, size_type /*n*/ )

{

if (!m_pool)// if there's a pool, nothing to do

{

free( p);

}

void

construct( pointer p, const T& val )

{

new (p) T(val);

}

void

destroy( pointer p )

{

p->~T();

}

size_type

max_size() const

{

return ULONG_MAX / sizeof(T);

}

};

template <class T>

bool

operator==( const MyCustomAlloc& left, const MyCustomAlloc& right )

{

if (left.m_pool == right.m_pool)

{

return true;

}

return false;

}

template <class T>

bool

operator!=( const MyCustomAlloc& left, const MyCustomAlloc& right)

{

if (left.m_pool != right.m_pool)

{

return true;

}

return false;

}

using namespace std;

typedef pairint> PairStringInt;

// DoTest is a template func that does the test.

// it's a template because 2 diff map types are passed in

#define NSIZE 2

template <typename TMap>

void DoTest(TMap oMap, CComBSTR bstrType)

{

for (int nItem = 0 ; nItem < NSIZE; nItem++) // 2nd time through adds dupes, no prob: ignored

{

// as you single step through here, observe the value of

// oMap in the Locals window!

oMap.insert(PairStringInt (CComBSTR("Jan"),31));

oMap.insert(PairStringInt (CComBSTR("Feb"),28));

oMap.insert(PairStringInt (CComBSTR("Mar"),31));

oMap.insert(PairStringInt (CComBSTR("Apr"),30));

oMap.insert(PairStringInt (CComBSTR("May"),31));

oMap.insert(PairStringInt (CComBSTR("Jun"),30));

oMap.insert(PairStringInt (CComBSTR("Jul"),31));

if (nItem == 0)

{

// declare an iterator

TMap::const_iterator it;

// now iterate through the collection

for (it = oMap.begin() ; it != oMap.end() ; it++)

{ // note they're sorted!

wchar_t buf[1000];

CComBSTR strOutput(bstrType);

strOutput.Append(L" ");

strOutput.Append(it->first);

strOutput.Append(L" ");

_itow_s(it->second, buf,10, 10);

strOutput.Append(buf); // hi Crescens2k

strOutput.Append(L"\n");

OutputDebugString(strOutput);

}

oMap.clear();

}

#define NITERATIONS 1000

#define NumItemsToAdd 10000

#define POOLSIZE (1040208+100)

struct SomeBigData

{

long id;

long dummy[20];

} b1 = {1,{1,1}},

b2 = {2,{2,2}};

typedef pair<long, SomeBigData> mapData; // a pair that holds some data

template <typename TMap>

void DoTestTiming(TMap oMap)

{

// now insert a few big items

for (int nItem = 0 ; nItem < NumItemsToAdd; nItem++) // # items to add to the maps

{

b1.id = nItem;

b2.id = nItem;

oMap.insert(mapData(nItem,b1));

}

void TestTiming(int nIter)

{

char buff[1000];

byte *pool = 0;

for (int iType = 0 ; iType < 3 ; iType++) // the map types

{

int nTotTime = 0;

DWORD nTickStart = GetTickCount();

for (int nIter = 0 ; nIter < NITERATIONS ; nIter ++)

{

g_nTot = 0;

g_nCnt = 0;

switch(iType)

{

case 0:

{

map<long, SomeBigData> oMap0;

// OutputDebugString(L"Using plain map ");

// same as multimap, std::allocator > > oMap0;

DoTestTiming(oMap0);

}

break;

case 1:

{

typedef MyCustomAlloclong, SomeBigData> > MyAllocator;

map<long, SomeBigData, less<long>, MyAllocator> oMap1;

// OutputDebugString(L"custom alloc ");

DoTestTiming(oMap1);

}

break;

case 2:

if (pool == 0)

{

pool = (byte *)malloc(POOLSIZE);

}

{

typedef MyCustomAlloclong, SomeBigData> > MyAllocator;

MyAllocator z = MyAllocator(pool, POOLSIZE);

// OutputDebugString(L"Custom pool ");

map<long, SomeBigData, less<long>, MyAllocator> oMap2(less<long>(),z);

DoTestTiming(oMap2);

}

break;

}

free(pool); // ok to free null (fast too)

pool = 0;

// _snprintf_s(buff,sizeof(buff),"Done nType = %d nLoop = %d mSecs = %7d %d %d\n", iType, nIter, nTickEnd, g_nTot, g_nCnt);

// OutputDebugStringA(buff);

}

DWORD nTickEnd = GetTickCount() - nTickStart;

nTotTime += nTickEnd;

_snprintf_s(buff, sizeof(buff)," %d", nTotTime);

OutputDebugStringA(buff);

}

OutputDebugString(L"\n");

}

STDMETHODIMP CTestInterop::Test(BSTR str, long *pRetval)

{

// single step through and observe the Locals window

//some of this verbosity can be reduced via more typedefs

// first we create a map object

mapint> MonthDict;

DoTest(MonthDict,L"Using a map");

multimapint> MultiMapMonthDict;

DoTest(MultiMapMonthDict, "Using a multimap");

// now we declare our own custom allocator, so the map will use it

// this allocator just uses malloc/free (we don't give it a mem pool to use)

typedef MyCustomAllocint> > MyAllocator;

// we declare an instance of map which uses our custom allocator

mapint,less, MyAllocator > MonthDictCustomAlloc;

DoTest(MonthDictCustomAlloc, L"Custom Alloc");

// now we create a memory pool for our allocator to use

byte *MyMemoryPool = (byte *)malloc(POOLSIZE);

// open a debug window (Debug->Windows->Memory->1)

// and drag the value of MyMemoryPool from the Locals window to it

// single step through so you can see the memory being used for

// the RedBlack tree.

{// scoping so destructors fire before we free memory

// we create an instance of the allocator

MyAllocator z = MyAllocator(MyMemoryPool, POOLSIZE);

// we declare an instance of map which uses our custom allocator using our mem pool

mapint, less, MyAllocator> MonthDictCustomMemoryPool(less(),z);

DoTest(MonthDictCustomMemoryPool,L"Map with mempool");

}

free(MyMemoryPool); // ok to free(0)

for (int i = 0 ; i < 40 ; i++)

{

TestTiming(i);

}

return S_OK;

}

Photo of the Day

2010-03-16T10:03:00Z

Thick-billed ground pigeon at the Miami Zoo Aviary

PhiladelphiaVFPUserGroup

2010-03-15T19:14:25Z

Editor comments: Date change for Advantage Database Server

MadFox 15th Anniversary Par-tay!

2010-03-15T16:26:51Z

It was 15 years ago this month that we first got together. 15 years! That's, like, a really really long time. We should celebrate! We'll start with our normal meeting, Q&A and such. Then Cliff will show some how he's been fooling with a piece of design software called OpenModelSphere and is really liking it. He will do a little demo of what he's learned so far (data design among other things, and both forward and reverse engineering.) I'm sure we'll get to talk about UML and other development related stuff as well. I'll also talk about the "OutFox" meeting and my experience at DAFUG (comparing/contrasting the User Groups so we can steal some of their ideas). Afterwards we'll go to a pizza place and MadFox will blow its entire $68 budget on pizza and drinks for attendees. Bring your memories. Past members we haven't seen in a while are welcome to come! Eric Tuesday, March 16th, 2009 1308 Spring Street, Madison WI (Ring "Office" to Enter) RSVP to Info@MadFox.info

OutlookSecurityWorkaround

2010-03-15T16:15:26Z

Contents:

Photo of the Day

2010-03-15T10:01:00Z

Green-naped pheasant pigeon at the Miami Zoo Aviary

Photo of the Day

2010-03-14T10:00:00Z

Brickell Key

If You Write Visual FoxPro Apps…

2010-03-13T13:03:00Z

If you’re a Visual FoxPro developer,
and you like to install the VFP runtime files in
the Windows\System32 folder,

Please…

STOP.

http://kevinragsdale.net/post/2008/04/27/DLL-Hell-Rears-Its-Ugly-Head.aspx

Photo of the Day

2010-03-13T11:02:00Z

On Washington Ave. - SoBe

BlogWatch

2010-03-13T00:04:29Z

Links to blogs that have something to do with VFP, or by VFP folks

Photo of the Day

2010-03-12T11:02:00Z

The Clevelander on Ocean Dr.

DevTeach 2010 Recap

2010-03-11T21:57:58Z

DevTeach is finally winding down and my sessions are all done. It’s been a great conference, as always. I have been speaking at DevTeach since the second year and it’s always a great time chatting with Canadians about software development. This year I had the pleasure of kicking off the conference on Tuesday morning with the Keynote. I did a Lap around Visual Studio 2010 and I showed fun IDE features and extensibility, showed off a couple of my favorite language enhancements, new Entity Framework 4 & WCF Data Services, new WPF designer and data binding, and Silverlight and SharePoint 2010 tools. It was fun and I think people were pumped.

Here’s some good links to check out for VS2010:

Right after the Keynote I had a session that I’ve spoken about many times before (always a crowd pleaser) on VB10 / C#4 language features which are covered in these walkthroughs, articles and videos, on MSDN:

I also did an off-the-cuff talk yesterday with Jon Aneja at lunch on using Open XML and XML Literals in VB to manipulate Office document formats.

My last session was today on Building Office Productivity Solutions with Visual Studio 2010 and according to the evals it went well just like my language talk. We created an OBA for good old Northwind Traders, bringing them into the 21st century. ;-) I showed the new features of VS2010 that makes Office development easier focusing on RAD data binding (including WPF) and designers, ribbon and the other available designers, as well as SharePoint 2010 tools around workflow and visual web parts.

My main goal was to provide a real-world scenario and architecture but keep the demo code manageable and simple while also introducing folks to all the RAD designers available. This way you can take apart the sample easily and reuse just the pieces you need. Check out the code and all the articles here: http://code.msdn.microsoft.com/OBANorthwind and check out the VSTO Developer Center for more awesome resources: http://msdn.com/vsto

Some of the new features that make Office development and deployment easier are covered in these walkthroughs:

Here are some pictures from the conference which was held at a really nice Microsoft facility in Mississauga.

Now it’s time to get some dinner and good beer. ;-) Goodbye, Canada. Thanks for the awesome hospitality. See you next time!

WPF: Setting Focus in a Custom User Control

2010-03-11T16:50:00Z

When you create a custom user control, it typically contains child user interface elements. To provide a proper "Focus" experience at run time, you need to do the following: 1. Select your user control in design mode and check the Focusable...(read more)

Photo of the Day

2010-03-11T11:05:00Z

Washington at 5th - SoBe

OutlookSecurityWorkaround

2010-03-10T23:59:49Z

Background: I wrote an automated task to sync data from a server outside of the corporate firewall to our FoxPro database on the LAN. The task sends an email to programming staff whenever there are errors during the sync'ing process (such as when there was an attempt to add a duplicate or other data integrity errors)

SelectSQL

2010-03-10T23:43:17Z

Recupera i dati da una o più ´¡¢elle. Visual FoxPro Sintassi:

Southwest Fox 2010: Call for Speakers Deadline Approaches

2010-03-10T22:43:44Z

Southwest Fox 2010 will once again be held in Mesa, AZ in October. Geek Gatherings has issued a call for speakers and the deadline is approaching. Those interested in speaking at Southwest Fox should submit a proposal by March 15. See http://www.swfox.net for all the details.

No Preview Images in File Open Dialogs on Windows 7

2010-03-10T21:55:39Z

I’ve been updating some file uploader code in my photoalbum today and while I was working with the uploader I noticed that the File Open dialog using Silverlight that handles the file selections didn’t allow me to ever see an image preview for image files. It sure would be nice if I could preview the images I’m about to upload before selecting them from a list. Here’s what my list looked like:

This is the Medium Icon view, but regardless of the views available including Content view only icons are showing up.

Silverlight uses the standard Windows File Open Dialog so it uses all the same settings that apply to Explorer when displaying content. It turns out that the Customization options in particular are the problem here. Specifically the Always show icons, never thumbnails option:

I had this option checked initially, because it’s one of the defenses against runaway random Explorer views that never stay set at my preferences. Alas, while this setting affects Explorer views apparently it also affects all dialog based views in the same way. Unchecking the option above brings back full thumbnailing for all content and icon views. Here’s the same Medium Icon view after turning the option off:

which obviously works a whole lot better for selection of images.

The bummer of this is that it’s not controllable at the dialog level – at least not in Silverlight. Dialogs obviously have different requirements than what you see in Explorer so the global configuration is a bit extreme especially when there are no overrides on the dialog interface. Certainly for Silverlight the ability to have previews is a key feature for many applications since it will be dealing with lots of media content most likely.

Hope this helps somebody out. Thanks to Tim Heuer who helped me track this down on Twitter.

Posted in Silverlight Windows

West Wind Html Help Builder 4.55 released

2010-03-10T19:59:12Z

West Wind Technologies has released version 5.55 of West Wind Html Help Builder. Help Builder produces documentation in Html Help, plain HTML and Word format for end user applications as well as developer documentation for FoxPro, .NET, Sql Service, Web Services and other environments. Version 5.55 provides a number of template rendering updates for cleaner HTML output, a host of small UI improvements, Visual Studio 2010 support as well as many fixes around .NET class imports. Help Builder is available as shareware and registered users can update to the latest version through the Check For Latest Version option on the Help Menu.

Top Ten Creationist Arguments - Part 2

2010-03-10T19:37:09Z

YouTube link

(Via The Thinking Atheist)

SouthwestFox

2010-03-10T15:55:41Z

Southwest Fox 2010

SouthwestFox2009

2010-03-10T15:53:39Z

Southwest Fox 2009

Went with the wind

2010-03-10T14:01:44Z

Part One

Part Two

Carol Burnett, Tim Conway, Harvey Korman, Vicki Lawrence, and Dinah Shore destroy Gone With The Wind.

(Via Miss Cellania)

Silverlight: Fixing the "Tag Expander Does Not Exist" Compiler Error

2010-03-10T11:38:00Z

If you create a new Silverlight Navigation project, then compile immediately with no changes, you will get the following compiler error: The tag 'Expander' does not exist in XML namespace 'clr-namespace:System.Windows.Controls;assenbly=System...(read more)

Photo of the Day

2010-03-10T11:04:00Z

Lincoln Rd. - SoBe

Location, Search and Google’s Patent

2010-03-10T00:01:24Z

Google’s patent awarded last week for using location in advertising is interesting news (patent details here via VentureBeat). Not because of the dramatic effect (dun, dun, dah) it might have on the industry . No, the key point here is they submitted the patent for consideration back in September 2003.

This is significant because it shows the foresight Google has.

Sure, the concept of location isn’t new – people have been talking about it (or at least claiming to have talked about it) for more than a decade. And recently the talk has reached mainstream discussion. But most of that talk (myself included) has been about location enabled services.

Google on the other hand – with patents like the one mentioned above – was proactively making monetization plans seven (7) years ago. To put this in perspective time-wise, this is around the same time they launched AdWords.

This is another indication of the gulf between Google and its competitors (Microsoft, Yahoo, Facebook, Apple, etc) when it comes to search. Google has been working on these concepts for so long now, it’s difficult to see how any of their competitors can catch up (especially when we see how primitive Steve Ballmer’s ‘insights’ remain – discussed later).

By the way, don’t get worried about Foursquare, Yelp and other location enabled services, they won’t be affected (contrary to what some blogs misunderstood) since this patent is about using location in advertising delivery. They’re safe. Rather, it’s the companies that use location in their advertising services that should be feeling uncomfortable – for example if Opera uses location information to alter inserted ads in the compressed pages they return.

Aside: Location versus Local Search

BTW, a little clarification on location in search versus local search (since they are often confused). Both are surprisingly complex but here’s a simple overview:

Local search is when a user searches for something specific to a location eg ’sydney airport parking’ and the results that are returned take the location terms (’sydney airport’ in this case) into account.

Using location in search is about search engines using your current location to tailor the results you get eg searching for ‘nearest hospital’ would be a situation where you’d experience this. The search engines have done this for years based on broad location (eg IP address based) and are constantly refining it based on other factors (eg GPS, Wi-Fi and cellular triangulation). And obviously, mobile devices are a particularly appropriate platform for this.

The two can overlap eg when you are physically at Sydney airport and search for ‘nearest parking’ you’ll likely get similar results – often referred to as the ‘where I am versus where I am going’ overlap.

But the big opportunity is when it comes to advertising. Advertisers want maximum control over where and how their ads are shown, and under what circumstances (eg based on the searcher’s specific location, and what they are searching for). Plus in most markets the location will be used for non included keywords (eg searching for ‘nearest dominos pizza’ can return results for other vendor’s locations as well – not just other vendor’s ads).

[BTW it’s not just traditional ads – already we’re seeing Google charge for enhanced organic location results in their 7-packs.]

But back to Google’s foresight…

OK, so we know that location is important, right? Google obviously saw the dollar potential back in 2003.

It’s foresight like this that makes me wonder what Google is working on and patenting now.

By the way, I’ve been following Eric Schmidt a lot more lately. Not that I think he’s particularly charismatic, but I do think he’s fucking smart – even if he still doesn’t get privacy .

No matter what he’s questioned on, he brings an interesting perspective. Take his off-hand comments about Twitter (recorded a year ago) (thanks to Matt Cutts for the link). Most people think about social interaction. Eric thinks about storage and identity. Google is, at it’s core, a data company, and they approach Twitter in terms of data. I wonder what patent they’ve got in play there…

What about Microsoft?

Let’s move on to Microsoft. How will location, mobile advertising and Google’s patent affect them. In fact, how is Microsoft responding in general to the moving target that is Search. Perhaps Steve Ballmer can bring us up to speed…

Unlikely. What a disappointment Steve Ballmer’s keynote at SMX West last week was (embedded below). If you haven’t got time to watch the video (it’s 49 minutes) have a read of Jennita’s summary on this SEOmoz post or Lisa Barone’s excellent coverage on this Outspoken Media post (oh and while you’re there make sure you subscribe to the Outspoken blog – highly recommended).

This could have been *the* opportunity for Microsoft to give us some confidence about their search initiatives; some insights into how they’re going to change the game. But no. Instead Steve trotted out much of the same stuff he spoke about at MIX 08 (yes, two years ago). At the time I’d been impressed with his perspective, today I’m wondering what else he’s going to do… because passing references to personal search, social search, cash back, alluding to game-changer, and an understanding that the ‘search market is still young and we see opportunities’ won’t cut it.

Admittedly, part of the problem was Danny Sullivan’s questions – they were *soft hitting* to say the least. The main problem though is that Microsoft is so far behind the game, and Steve knows they need to focus on growing search share as their highest priority. Perhaps talking about any innovation they’ll be making in search would be construed as them getting ahead of themselves?

But I could hope couldn’t I? Hope to hear about them embracing key growth areas (eg location based search on mobile). Instead we were ‘treated’ to a discourse into regulation, anti-trust issues against Google and advertiser lock-ins before moving onto a few comments about the China situation. Sheesh! At least Steve rightly referred to them as ’sideshows’. I thought we were here to learn about Microsoft’s insights into Search, not their response to agendas set by Google!

Towards the end of the chat, Danny finally gives Steve the gift question when he asks: What is the biggest opportunity in search?

The floor’s yours Steve. Please deliver. Please redeem yourself. Please give us hope.

Despair. Steve answers with a platitude: ‘The biggest opportunity is helping people get done what they need to get done’. Worse, he then outlines some of the problems he personally runs into when searching. Yes, Steve, but what are you doing about it? How are you solving this problem? Give us some substance. Details. Anything that would give us pause to consider moving to Bing. Anything but the stunning insight that you ‘see this as an opportunity’…

This, from the company that will be Google’s biggest competitor in search (as they will be once the Yahoo search deal completes later this year).

Back to that patent

And so we return to Google’s recently awarded patent. Should we be worried? Yes, we should.

When a company has:

The foresight this patent signals
70%+ search market share and online advertising market dominance
Platitudes as the response from their nearest competitor

then we can be reasonably confident/scared they are only going to get stronger.

To this observer it seems that Microsoft is attempting to compete with the Google of 2005. And in the next five years they may even make some in-roads… At which stage they’ll be fighting a Google that is 10 years ahead of them.

Video Ezy Chatswood

2010-03-09T23:58:09Z

Just wanted to give a shout out to our local DVD store: Video Ezy at Chatswood.

They’re just down on the corner of Victoria Avenue and Neridah Street near Chatswood Chase.

Why this post?

Well, two reasons. The first is mentioned in my disclaimer at the bottom, but the second – and most important – is because we want more people to patronise them.

We’ve been happy customers there for more than a decade, and currently take advantage of their $39.95 a month membership. There’s no lock-in, so we can cancel at any time. And we get to hire 4 DVDs at once. Overnight videos need to be returned (or renewed) the next day, but 3 day and weekly hires we can keep as long as we like.

This latter feature is particularly useful for watching TV shows, which we do a lot. Oh, and that’s another thing – they have a great TV show range. Just about every show you can think of (eg The Wire, The Tudors, Life, NCIS, The L-Word, Breaking Bad, Heroes, Bones, Sarah Connor Chronicles, etc as well as all the comedies 2 1/2 men, Seinfeld, Scrubs, How I met Your Mother, etc) are there. In fact they have one of the best ranges of TV shows I’ve seen (and I’ve been to at least one other store ). And bonus, on the older shows hiring the entire series counts as one DVD.

But the best part –> the staff: Matt, Chelsea, Lucy, Jack, Andie, Hannah, Angie, Bec & Adrian – they all rock. Sometimes we go in just to chat to the staff – they should set up a little cafe in there. Flat White with one please!

View Larger Map

[Their other claim to fame is they have the 4th largest blue-ray collection in the state. But this is useless to us since we don’t have a blu ray player. … yet]

(Disclaimer: this post is also part of an experiment)

A day in the life of New York City, in miniature.

2010-03-09T23:47:26Z

The Sandpit, a short time-lapse film by director/photographer Sam O’Hare is wonderful to watch. It was shot on a Nikon D3 with two lenses over 5 days and two nights all over New York and Brooklyn. He did over 35,000 shots that were then blurred and edited in post-processing to give the fake tilt-shift miniature effect. Score is original music by Human, written for the film.

See it in HD in Vimeo.

Photo of the Day

2010-03-09T11:02:00Z

Downtown Miami and Metromover

MasFoxPro

2010-03-08T22:18:50Z

Editor comments: ...and counting

DevTeach Here I Come!

2010-03-08T17:01:06Z

I’m heading up to Toronto this week for the DevTeach conference! I’ve spoken at DevTeach every time except for the very first one. This is a great Canadian conference that really brings a bunch of awesome speakers in our industry into a more intimate setting than say a TechEd. Check out all the sessions here. This year there’s a SharePoint track that I’m really excited about.

I’m doing the keynote this time and I’m going to make an attempt at showing off some of the awesome new features in Visual Studio 2010. I’ve got some IDE tricks up my sleeve and I’ll walk through some language, parallel, data, WPF and SharePoint tools and features. I can’t hit everything because there’s so much packed into Visual Studio 2010 that it would take a full day to explore, but hopefully this gets people excited enough attend the rest of the sessions to learn more. My sessions tend to be pretty demo-heavy so look for less than 15 minutes of Keynote slides and the rest of the time I’ll be wowing the crowd (at least I hope) with real demos. I want to get the crowd PUMPED UP so if I have to put on my tap shoes so be it. ;-)

I’m also doing a session on VB 10 & C# 4 new language features and a session on building on Office and SharePoint that should be a lot of fun.

Hope to see you there!

Office Development with Visual Studio Tutorial Series – Part 4

2010-03-08T16:40:45Z

A couple months ago Robert Green, VSTO MVP, started a series of step-by-step tutorials on building on Office 2007. Part 4 is now published. Thanks Robert!

In this fourth part of the series of tutorials on Office Business Applications, learn how to create an Excel 2007 solution using Visual Studio 2008 that generates reports from a database and allows you to take those reports offline. This tutorial shows you how to cache the set of data directly in the Excel workbook and also shows you how to easily print the data as a PDF. This step-by-step tutorial also includes full source code in Visual Basic & C#. Check out the tutorial on the VSTO Developer Center:

Building an Office Business Application Part 4 – Generating Reports

And if you missed the previous tutorials:

These tutorials are pretty popular so if you’re just getting started with Office development in Visual Studio, this is a great place to start. Download all the code here: http://code.msdn.microsoft.com/ContosoAutoOBA

Enjoy!

CurrentVersion

2010-03-08T15:24:39Z

Current version numbers of software, so that you know if you are up to date.

Photo of the Day

2010-03-08T11:02:00Z

Smew at the Miami Metrozoo aviary

Chicago FUDG - Tuesday 9 March 2010

2010-03-08T02:18:37Z

Chicago FUDG's March meeting will feature presentations by 2 members - Tom Corrigan on Amazon Cloud computing and Jim Wiggins on using the VFPX graphics classes (GDIPLUSX) library. Here is the RSVP url click: http://bit.ly/aQnmD8 Do it now. See you then - 5:30pm on Tuesday the 9th of March - Tech Nexus, 200 S. Wacker Drive, 15th floor meeting room C.

CurrentVersion

2010-03-07T17:56:11Z

Current version numbers of software, so that you know if you are up to date.

Photo of the Day

2010-03-07T13:42:42Z

Great Blue Heron

Loading XAML Workflows in WF4

2010-03-07T08:40:10Z

Note: This blog post is written using the .NET framework 4.0 RC 1

Most of the time I used compiled workflows in Windows Workflow Foundation 4. Its nice and easy, you design the workflow, compile it and at runtime there is a .NET type you use to create and run workflows. The main drawback is that this approach isn’t very flexible, sometimes you want to be able to change your workflow definition at runtime or store it in a database so recompiling isn’t an option.

Fortunately we can also load a workflow from the XAML file itself and execute the resulting workflow activity. This is done using the ActivityXamlServices class that will let us load the XAML file and return an activity, to be exact it returns a DynamicActivity as a wrapper around your definition.

The simplest option is just to call Load() passing in the file name like this:

Activity workflow = ActivityXamlServices.Load("YourWorkflow.xaml");

If you are using activities, or other types, from the local assembly this is going to fail though and you need a slightly more verbose way of doing thing like this:

var settings = new XamlXmlReaderSettings()

    LocalAssembly = typeof(SendForManualApproval).Assembly

};

var reader = new XamlXmlReader("YourWorkflow.xaml", settings);

Activity workflow = ActivityXamlServices.Load(reader);

We need to use the XamlXmlReaderSettings to indicate what the local assembly reference in the XAML is.

Enjoy!

www.TheProblemSolver.nl
Wiki.WindowsWorkflowFoundation.eu

CurrentVersion

2010-03-06T20:53:48Z

Current version numbers of software, so that you know if you are up to date.

SelectSQL

2010-03-06T17:55:13Z

Editor comments: cursorsetprop

UpcomingEvents

2010-03-06T16:52:38Z

A place to list upcoming Visual FoxPro events like conferences, meetings, user groups, open training sessions...

NPR Interviews Cookie Monster

2010-03-06T13:37:01Z

A great interview.

(Via Miss Cellania)

Photo of the Day

2010-03-06T11:01:00Z

Saddle-billed stork
Miami Metrozoo

Generic Co- and Contravariance in Visual Basic 2010

2010-03-05T20:31:13Z

In this month’s issue of MSDN Magazine we’ve got an article on a new language/.NET 4 feature called Generic Co- and Contravariance by Binyam Kelile:

MSDN Magazine: Generic Co- and Contravariance in Visual Basic 2010

I have to admit if I think about this feature too hard my head starts spinning in circles faster than Linda Blair in the Exorcist. But in actuality it’s one of those features that “just works” and probably should have already been available in the CLR. The easiest way for me to think about it is that it enables true inheritance scenarios with generics.

Consider this practical example. Say I have a class called Student that inherits from Person:

Public Class Person
    Property Name As String
    Property Age As Integer
End Class

Public Class Student
    Inherits Person

    Public Property Score As Decimal
End Class

Now say I have a generic List(Of Student) that I want to pass to a method that accepts an IEnumerable(Of Person)):

Sub Main()
    Dim students As New List(Of Student) From
        {New Student With {.Name = "Beth", .Age = 10, .Score = 90.5},
         New Student With {.Name = "Alan", .Age = 11, .Score = 100},
         New Student With {.Name = "Jenn", .Age = 12, .Score = 98.5}}


    PrintNames(students) 'This will not work in VS2008
End Sub

Sub PrintNames(ByVal list As IEnumerable(Of Person))
    For Each p In list
        Console.WriteLine(p.Name)
    Next
End Sub

Even though a List implements IEnumerable and a Student inherits from Person this will not work in Visual Studio 2008 because generic types behave invariantly in the CLR previous to version 4.0. This is now supported. In Visual Basic 10 (and C# 4) you now have the ability to declare covariant (widening) and contravariant (narrowing) generic types with the Out and In modifiers. So they changed the IEnumerable interface in the CLR 4.0 to designate a covariant generic type:

Public Interface IEnumerable(Of Out T) ...

So this means that the method call above will work now because a widening conversion is allowed. Contravariance is the exact opposite. With the In modifier on the generic type a narrowing conversion is allowed. Piece of cake, right?

Check out the Generic Co- and Contravariance article for a deep dive into this feature and how to use it in your programs. Also check out Lucian’s post: Co- and contra-variance: how do I convert a List(Of Apple) into a List(Of Fruit)?

Enjoy!