UPS manages to deliever a package before it arrives

I seriously wonder how hard it is to track a package in this day and age…

1. The boolean as anything but a boolean

This happens frequently when people are moving from an access database to anything else. Access wasn’t exactly friendly with boolean values (thou, it was possible to make a boolean field as an int(1) ). At any rate, most people who designed access databases used a char(1) where the values where either ‘Y’ or ‘N’. It’s an obvious mapping to a boolean. For whatever reason, people just leave these fields as the single character fields, instead of migrating it to a more friendly boolean field. So, that’s the first no-no of database design. Using anything but a boolean (or I’ll accept a bit field) for boolean values

2. The complete avoidance of enumeration fields

There are many cases when a field represents something like the status of an order: new, pending_payment, processed, shipped, complete. Often, I see these sorts of status as varchar fields. This is fine and dandy, except I have no idea what are acceptable values to put in the that column. I could do a select unique on it and hope that there is atleast one record representing each possible value, but that isn’t a reliable way to do it. One could argue the DBA would have documentation on what should go in there, or that the business logic would dictate it; but, most projects skip those points. Really, only the orgional programmer has any hope of remembering what can go in those fields, making them somewhat problematic to work with for future work.

In comes enumerations. Because you specify what the heck is valid in the field, you don’t have to guess what can possibly go in them. It’s actually their intended use. It also provides a form of feedback when you try to set that field to something your not expecting. Further preventing people from doing stuff like “New” vs “new” (which I have seen…).

So, being the solutions sort of person I am, I ventured forth to get a programmable thermostat. Nothing fancy, just something that would let me set a cooler temprature at night and heat the place up when I got home. I went with the good old Honeywell junk, because apparently that’s all home depot sells (shutter, I have to stop going there…) They had three models with no price ( I lied, one did have a price), one looked too fancy, so I ignored it. The last two seemed identical, but one had a blue illuminator. I take both options, vowing to buy the cheaper of the two, even if that ment a life with no blue light on my thermostat. Turns out that the blue light model was $10 cheaper, so my life is atlast fulfilled!

I get home and liberate the thermostat from it’s plastic clam shell, only to find the instructions to be useless. See, They’e all about colors of wires, except not. They want you to use your old thermostat to figure out what wire does what. The problem with this is that two wire manual thermostats don’t care what wire is what; but, digital ones do. I figured I had a 50/50 chance that I would guess wrong. I traced the wires from the source to the wall and really, there was no indication of what color could be what. Black isn’t a normal HVAC color, and red is apparently not used in preference for more flashy colors. Basically, who ever installed this, didn’t follow any standard setout at all… Of course, after careful reading, it appears that the industry standard is the mix the colors up as much as possible to ensure you have repeat business replacing thermostats.

So I did something incredibly clever. I calculated that I could run a single blue LED off the 24VDC wire with a 1.2Kohm resistor. With the clever application of a diode, I could prevent myself from hooking it up with the wrong polarity and damaging the LED. I just twisted and taped this all together (I would of taken a picture, but I took it appart right away…) At any rate, it was a simple as hooking the wires up to either lead. Once was all it took to confirm RED was hot and BLACK was the furance relay. EASY!

Gift from Rogers Wireless

Always providing top notch service…

The UPS package was shipped on March 3rd and the air mail package was shipped on March 4th.

The UPS package traveled 2178km and the air mail package traveled 10,240 km.

The UPS package arrived at 9:48AM March 10th and the air mail package arrived at 11:25AM March 10th.

The UPS package cost $59.42 USD and the air mail package cost $35.00 CDN.

Average speed: 13km/h VS 71km/h

Approximate cost over distance and time: $4.58/ km/h VS $0.49 /km/h

• The difference between MSI project types is nearly meaningless. “Basic MSI” can only literally handle just a straight up wizard created project. Anything remotely useful requires you to migrate to the InstallScript related projects. My point here, why even bother having it as an option.
• Documentation included within the app is sparse at best. Documentation online is marginally better. $250 manuals, ever so slightly more useful than online docs. Would it kill them to keep these manuals up to date as well? I’ve already come across a few cases where the manual for 2008 does not reflect changes made since 12.
• Complete lack of a useful Dialog building GUI. Was an issue in 10.5 and still is an issue in 2008. Seriously people, if you going to charge that much for your product… give me something I can actually see as an improvement.
• Did I mention the entire interface for the application has basically remained the same? Well it has, all the slow loading pages and forms are still here in 2008.
• Want to test custom script actions? You’ll have to rebuild your who MSI project to do that. In my case that’s a 20 minute or so wait between tests.
• They have this nifty feature called release flags, but fail to implement it everywhere it could be useful. Results in you hacking the install scripts to get behaviors based on the release flag anyways.
• Skinning support is poor to horrible. Let’s not forget that it’s documentation is poor as well, making it all that more entertaining to use.

Well I should also point out what this software does well:

• Dependency scanning for crap old VB applications is amazing. It also includes (or can be downloaded with the update manager) all the MSI merge modules for those dependencies that are difficult to install manually. It’s actually why we updated to 2008.
• Multi language support is half-descent
• Phone support has so far been very good
• It has a lot of nifty built in features that nearly work exactly has you want them to, majority of the time. XML and INI file editing comes to mind.
• It has a GUI for what it’s worth…

Some of you are probably thinking to your selves, “But James, there’s plenty of free alternatives out there.” And you sir/madam would be correct and very smart. Let me tell you why I couldn’t use InnoSetup or NSIS. They don’t dependency scan (even with the addon tools) nearly well enough to catch everything a complex VB application uses to run. At one point I actually tried to duplicate what Install Shield had determined the dependencies as, but that didn’t work either.

I do plan on using one of those two alternatives in the future (most likely NSIS). Both options built the installation package much faster than InstallShield can. They both are very flexible and extend able; and, they’re free and well supported. NSIS comes with an amazing set of examples that cover nearly any installation situation you would come across. InnoSetup also has a very good community behind it. Until the software is rewritten to have less dependencies I don’t believe I’ll be able to escape the InstallShield monster; but, for now I do have a good handle on how to work with it and get it to do my whims (most of the time).

1. Code that works in VB.Net and should in theory work in VB6 (with the appropriate changes) never does.
2. Modules are static classes and everyone uses them way too much.
3. Dependencies with OCX and DLL files are cached in some queer system that is supposed to save me time. More often than not it buggers the project up on release.
4. The Intellesense line complete is retarded 99% of the time.
5. Building your project requires you to specify where to save; and, what the file name is every time.
6. The errors VB passes along are generic to the point where it may as well say, “Something bad happend. Insert MsgBox someplace random.”
7. The only way to debug an app with any degree of accuracy is with lots of MsgBox’s.
8. Add a new Module, Class or User Control. Try to compile without saving your project first. Watch it burn in flames.
9. Doesn’t auto save before debug. If you get some sort of long msgbox loop from debugging there is no way to stop it other than killing Visual Studio 6. Redo all work.
10. Pretend enums, even thou we have enums in VB6. Apparently they decided not to use them for anything remotely intellegent.
11. Take two integers (whole numbers) and divide them. Usually you would expect an integer as a result, but with VB if the result is not a whole number it returns a float. If it is a whole number it returns an integer. Makes you wrap all your integer math in CInt()/Int() if you want it to be sane. Added August 22nd, 2007
12. VS VB6 seems to forget where referenced DLL’s are; or, just ignores where it knows they are and pretends they’re not there. The result of this is that any “object” that used the reference is removed from the code, along with any attributes customized for it. Added August 22nd, 2007
13. There is no undo for moving form elements. If you had an exactly placed element and accidentally moved it your hooped.Added August 24th, 2007

Note on August 24th, 2007: I realize many of these issues are fixed in later editions. My point here is that I shouldn’t be developing software in VB6… it’s wasting my time.

You have a drop down list on a page that is DataBound to a list of items. These items all appear on the page, but you may have to scroll to them. You want to let the use select the item from the drop down and scroll to it (with the magic of anchor tags). Typically you add an “onchange” event to call some javascript that does this. Something simple like: window.location = “#item_id”;

Unfortunately, ASP.NET 2.0 doesn’t want you to do client side scripting with their controls, so you left with these options:
1. Render the control manually in the ASP page. This is the method we’ve been doing for years, creating large unreadable source pages. It’s what all this ASP control jazz is supposed to fix.
2. Extend or rewrite a new control that accepts this property with no complaints. This may already be done by someone out there. Down side is you add a dependency to your project and usually get a bunch of other crap you didn’t want. Extending it consumes some time, but if done write you can probably reuse it.
3. Add the ‘onchange’ attribute to the control with some code: controlName.Attributes.Add(“onChange”,”javascript: return confirm(“Are you sure?”);”); This works, but you end up combining your view and control code, which once again is something ASP was supposed to fix…

None of these solutions are particularly good. When I worked on a site that used Castle I found many similar problems with their “Helpers”. I submitted updates to them, so everyone benefited. With Microsofts ASP.NET, we have to bitch about it. Send in a request. Wait. Wait some more. Then maybe they integrate the requested fix/feature in the next version. The whole process taking years usually.

Note: I’m not endorsing the use of Castle. It’s relatively mature, but doesn’t provide much of an advantage over a well designed asp.net site. Also, to do anything moderately well in ASP.NET, you should understand their DataSource stuff. For larger sites, be prepared to write your own datasource handlers, because the ones they provide are limited in their usefulness on large sets of data (e.g., they retrieve full sets of data before pagination or sorting).

Remember, you saw it here first (well actually you read it hear first, but perhaps I’ll post a photo for your amusement later on).

Right now I have my 300mm, 35mm and flash in the bag. I store the battery charge in the top, along with an extra usb cable and some cleaning supplies. Having the ability to swing the bag under my arm and swap lens is very liberating. Getting used to the way the bag opens up is a bit of a learning experience, so practice on your bed or something. Dropping a lens is never an amusing situation…

This bag has enabled me to causally bring my camera with me places I wouldn’t normally consider taking a camera. In that sense, I feel more opportunities to take pictures, even if they are of non-sense stuff that no one cares about. Experience composing pictures is experience composing pictures. Anyways, I give it an arbitrary two thumbs up.