Last night, due to the graces of On Demand video, Jenn and I watched The Master of Disguise.

Hoooo boy.

How do movies like this get made? I mean, who reads a script like that and says, “Hey, this will be great on film! Let’s do it!” I understand that there’s a market for children’s films, but even kids are going to wonder what the makers of this pile of shit were smoking.

I think I laughed like four times. Mostly it was at the quips the grandfather character used to tell Dana Carvey to shut up. Stuff like “curb your yammering skullcave,” which I will now attempt to work into daily conversation somehow. It cracked me up because it was so out of nowhere… but those were about the only things that were any good. (Do you remember when Dana Carvey was actually funny? I do. I miss those days.)

That’s an hour and a half of my life I will never reclaim. (And I’m not the only one who thinks so.) I am actually dumber for having watched that. I award myself no points, and may God have mercy on my soul.

When it comes to programming, I will admit I am a spoiled, whiny baby. I hate low-level crap. I am not a C++ programmer. Sure, it may be powerful and flexible and yadda yadda yadda, but having to deal with all of the little things in there is such a pain. I like higher level languages that abstract that away. I’m a big fan of C#, Perl, Smalltalk, and whatnot. (You may or may not like any or all of those. Too bad. They’re my personal preferences… even if I haven’t written in Smalltalk for years.)

I’m working on this project where I want to put the Windows “Send To” menu inside the Solution Explorer of Visual Studio .NET. I’m finding that in order to do that, you have to do some pretty low-level shit. As such, I started browsing around in the Shell Programming section of MSDN.

Combining the knowledge I was gleaning from that and some code snippets and things I found online, I started compiling a library of imported shell programming related objects into a big library - shell programming gone C#, right?

The problem I’m having is that I can’t get it to quite work right. The marshaling of parameters is befuddling me because there doesn’t seem to be a consistent translation from C# managed types to C++ unmanaged types (of course, this is probably my lack of education speaking). Where I can get some methods to work correctly, others don’t seem to want to cooperate.

I decided to set about converting the various sample code snippets found on the MSDN shell programming site to my C# library, figuring that’d be a good way to test - code that supposedly works, just translated to the new language. No problem, right?

Huge problem.

I’ve been fighting this thing for quite some time now and I just can’t seem to get it to work. Now, here’s where you’re going to smack me: It’s making me think I should just learn the C++ way and do it in C++.

And you say, “You don’t already know how to do it in C++?” To which I respond, “Well… no.” Okay, I admit, that’s probably pretty fundamental, but if I can read how it’s done in C++, and I can see how that works in the documentation, I don’t see how copying and pasting example code into a sample project is going to make it any easier to convert the stuff.

Which leaves me thinking I should just leave the conversion of shell stuff to the folks who do shell programming. I’ll DllImport my way to freedom and leisure where necessary, and/or maybe just write the thing in managed C++ that I’ll call from my VS add-in. I’ve seen the managed Send To example out there; maybe I’ll start with that. Regardless, I can only beat my head against something for so long before I decide it’s not worth the effort and leave it to the professionals. My bag is web apps, not COM interop and low-level shell programming.

I’m what you could consider an “entry level” user when it comes to financial services. I don’t know all the terminology. I don’t know what everything means. I’m not intuitive when it comes to figuring financial things out. What I know is simple stuff - what accounts I have, what I want to do, what I want to see. Very simple.

I used to have an account with a stock brokerage in Portland called “Bidwell & Co.” I liked them. The customer service representatives were friendly, my monthly statement from them was easy to understand (big print, small words, lots of pictures), and when I wanted to do something on their web site, it was easy to figure out. If I wanted to see my portfolio, I clicked the “Portfolio” button. If I wanted to sell stock, I clicked the “Sell” button. Even more key, when I got to my destination (“Portfolio” or “Sell” or what have you), the interface was very simple

• I wasn’t overwhelmed with buttons and text boxes and options and words I didn’t understand. For things I didn’t get, there was help all over the page that I could click on and read about the tough stuff.

Bidwell was bought out by Ameritrade recently. My account was transferred over, and now I have to manage my stock via the Ameritrade site.

Oh

My

God.

I don’t get it. Seriously, I don’t. There are far too many options to know which one I need to click to get to where I want to go. Want to see your portfolio? Click a link (hidden in a list of like 50 other links) along the side of the page marked “Positions.” Positions? What the hell is that supposed to mean? Click on that, though, and you see an abbreviated version of your portfolio. Want to see a more detailed view? I did. I wanted to see my portfolio broken out by individual lots of stock. That’s under “Gain/Loss Tracker” - a whole new link along the left of the screen - then you have to select the “Detail” view from a dropdown box, then click the “Update” button. Isn’t that a portfolio thing? Maybe not. It is from the uninitiated perspective. (I actually had to call them to figure this out.)

Got a Capital One credit card? Their site rocks. There are four tabs at the top of the page you can choose from. One gets you a view of your current activity, one gets you your statement, one lets you pay your bill online, and one gets you a list of everything else you can do. Click on one and you know what? You get exactly what you’re expecting. Very simple.

Got a BankOne credit card? Ever try their site? The print is tiny tiny tiny, and they try to cram way too many options onto the page. Do I really need to be able to get from any page to any other page in one click? No. I need the interface to make sense. More mouse clicks is not the problem; being able to find what I want to do and understand what I’m looking at is the problem.

I think financial institutions might want to think about that. My grandparents, for example, who do own stocks and bonds and all that, will never ever be able to manage any of that online because there’s too much going on. Stop listening to these so-called interface rating services. I’ve seen the sites they rate highly; it’s scary what they think is easy to use. Stop ranking things on how many mouse clicks it takes to get somewhere. Instead, think about how things are logically. Sure, you may have the technical ability to enable the most flexible web site ever - offer 2000 ways to transfer money from one account to the next or whatever - but in all honesty, how many do customers really want?

Here’s an idea: Take a page from the search engines and offer two versions of the user interface. One version is the “basic” version - for the 95% of people who want to just get the job done. The other version is the “advanced” version - for the 5% of people who need that ultra flexibility.

Or, better, here’s another idea: Get a beginning banking user (not unlike myself) to help design the interface. Listen to him (or her) when he tells you that the interface is too complicated. Consider the fact that what you’re hearing is probably more important than what you heard from the interface rating institution because if you can’t get a beginning banking user to understand how your product works, they aren’t going to use it. Get grandparents in there to try it out. Get eighth graders in there, fresh out of personal finance class, to try it out. If they can all figure it out, that’s where you need to be.

[Important Note: This is all definitely my personal opinion and absolutely not [necessarily] my employer’s. (Seeing as how I work for a company that makes financial web sites and all…)]

This weekend was Father’s Day, which meant carting around town visiting family.

Saturday night we hung out at my parents’ condo and visited with them, my sisters, mom’s sisters, and mom’s dad. Not sure where my dad’s parents were.

Granddad got going on telling a story and had us rolling. Apparently his dog likes to sleep under the footrest of his recliner while Granddad watches TV. One time there was a power failure and Granddad got up to get a lantern so he could see, but when he came back, he couldn’t find his dog. He looked all over the house and came to the conclusion that, in his words, “The devil done shut off my lights and stole my dog!” (How hilarious is that? The best writers in Hollywood would never come up with “the devil done shut off my lights and stole my dog.” Too funny.) Turns out when he stood up, he pushed the dog under the recliner with the footrest and trapped the dog under the chair. Heh.

Sunday afternoon we had lunch with Jenn’s family at Marie Callender’s. Pretty tasty, and we all ended up with leftovers.

All in all, a good weekend hanging out with the family. Not particularly restful, but a good weekend nonetheless.

Went to Home Depot to check out paint colors for the bathrooms. Our entire house is the same color of white inside - trim, doors, walls, etc. - so we need a little color. Found some great stuff and will have to get on that as soon as we get money.

Spent a little at Bed, Bath, and Beyond, too, picking up some new bathroom rugs to replace the ones the cats have torn up and have seen better days. Plus, we had to get something that goes with the shower curtain we bought last weekend.

In other news, I think I have a little athlete’s foot - minor, but annoying - on my right foot. Not sure how I got it since I don’t go tromping around in public showers or really ever go around barefoot, but it’s there and it’s created a weird sort of dry spot that occasionally itches. At least, I think it’s athlete’s foot. I dunno. The point is, it’s resistant to the over-the-counter Lotrimin I’ve been using, so I figured I should go see a dermatologist to make sure I’m even dealing with it correctly. Made the appointment this morning for three weeks from now. That’s the soonest they could get me in! Seems dermatology’s in high demand around here. I’d say I should have been a dermatologist, but I’m sure you see some pretty disgusting shit in that line of work. I have issues picking up cat puke; no way am I dealing with skin diseases.

If you work with SharePoint Portal Server 2003, you’ve probably messed around a bit with creating web parts or maybe automating some of the routine administration tasks, but have you tried accessing the search web service?

There are articles out there telling how to integrate the search web service with the Office 2003 Research Pane, there’s documentation on the schemas for querying the search service, but if you look at the SDK documentation, there’s really no example for what the query itself should look like. I mean, I can figure out how SQL Server full-text queries work, but what’s the source I’m querying? When you put everything together, what are you really sending to the web service, and, more importantly, what’s coming back?

To answer this, you could use something like SOAPscope and start watching packets that way, but you’re still going to be putting things together in a sort of trial-and-error fashion, hoping your packet is formed right, etc.

Instead, I created a little app specifically for this: Give it the URL of your server’s search web service, enter your search terms (the keywords for a “keyword” search; the SQL full-text query SELECT statement for a “SQL Fulltext” search) and click “Execute Query” - the program goes out to your search service, registers to search, gets the list of available catalogs and search scopes, and executes your query. You can see the form the query packet takes and the results you received from that query. You also have the option of changing search settings to different limits - start at a particular record number in the results, retrieve a certain number of results, only retrieve certain types of results, etc.

Note: While this is a very helpful tool to test out queries and see if they work, I still don’t have any real documentation about the form of full-text queries for SPS. I’m working on getting some information from Microsoft on this, to find out which fields I can query for/against and how to determine that for different servers, as well as figuring out what different errors mean. When I have more, I’ll update this entry. (The only queries I’ve gotten to work are variations on SELECT "DAV:href" FROM Non_Portal_Content..SCOPE() WHERE size > 0 so… I’ll keep you posted. It doesn’t seem to work like the SPS 2001 queries did, though.)

UPDATE: 6/18/04 11:00 AM - It turns out they embed the entire SQL query that gets run on the Advanced Search page right in the source.

Also, I’ll be updating the app as I learn more about the search functionality so I can offer more flexibility in the querying.

Download SPSSearchTest 1.1.0 (MSI)

Download SPSSearchTest Source 1.1.0 (ZIP)

Version History 1.1.0:

• Added syntax highlighting for SQL queries using the ICSharpCode.TextEditor control.
• Moved view and query options into menus.
• Now showing the Registration Request packet. 1.0.1 - First public release. 1.0.0 - Internal/unreleased version (still needed more functionality for prime time).