FreeTrack - How to Make a 3-Point clip

Mon, 24 Sep 2007 18:45:15 GMT

I've been toying witht he FreeTrack program and I must say it is very good. Considering that TrackIRs cost almost $200US with a headset clip is really outrageous. I went and picked up a $40 webcam and some parts to build one for myself. My goal was to see how fast I can put one together so that I can get my friends to use it at merely the cost of a webcam.

So far I found these IR LEDs from Marlin P Jones Online work better than any others I tried (including RadioShacks)

http://www.mpja.com/prodinfo.asp?number=16798+OP

I paired them in combo with a 1/4A 47Ohm resistor and they work awesome. Now if I could only get my camera FOV increased....

Anyways, I documented my construction technique and I wish I would have taken a photo of the LED assembly before it was wrapped to show the new guys that the Flat Side of the LED is connected to the ground wire and the positive side is connected to a resistor.

Here is a PDF of my construction:

FreeTrackHowTo.pdf

The IDE concept

Mon, 24 Sep 2007 18:44:22 GMT

I remember the days of line-by-line programming in BASIC. So for me, the introduction to Visual Studio as an IDE when I started work in C# was like going from horse carriages to spaceships.

Now, I am tasking myself in creating a full-blown IDE for a game (sim) called Armed Assault. Usually, games only seriously interest me when they have one of two (or both) qualities: A) good editing capability to extend the entertainment beyond the original program's design B) good simulation of real-life problems. Now, the latter sounds very broad and very “realistic”, but that is not my intention. Not many games/simulations properly reflect a decision system which mimics the dynamics of real life. I could dive into a huge pit of discussion about how ineffectual I think realistic strategy games (like chess or modern day computer strategy games) can be, but I won’t! ;)

No, the focus was about IDEs. So I am building this IDE based around this game. The game actually is very complex and has many components involved when developing for it. For example, there are two script languages utilizing similar commands but completely different operational capacity. There are special class files utilizing yet a third structure. This does not include the hundreds of types of content one could use when developing. All this adds up to the very reason I am making this IDE: how can we cohesively put everything together into one central program/system for faster productivity?

During development I have come across a problem which has altered my vision (only slightly) of how this centralized system should operate. For example, most content files are developed using more advanced technology than I could possibly muster (like Photoshop or 3D-Max) or they are developed using proprietary tools which I cannot touch. This makes keeping things “central” very difficult. While I could ignore content files and dive strictly into the source files like scripts and FSM files, I feel I would be short changing my system. I noticed problems among even the mighty Visual Studio when using a large amount of content files. The real work of the IDEs become as manipulation of the mechanics of the development process. Even Microsoft has realized this (go figure) and developed a third product called Expression or something like that. The idea is that the designers of the content can manipulate the product without seeing or touching the source code of the mechanics.

Complete separation of form from function.

SCREEEEEEECH!!!!!…..(this is me applying brakes btw)

Since when did form and function become separate again? I thought that other than a few purists, we were striving to maintain form and function within software to help make it… I don’t know.. more human? …more comforting?

I am partially confused by this concept but the logic in me is saying “this will work” along with all the others. I can understand from the perspective of development of large applications where a design studio is in London and the programmers are in Utah (or some other odd place where geeks like to congregate). This would only make sense. However, what about lone developers? What about the developers that are both the programmer AND the designer? It seems counter-intuitive in that situation. But I assure you… Microsoft probably doesn’t care about that aspect too much ;)

Ok, so back to my IDE. Content and source… content and source… how to handle it all?

I am beginning to wonder about the future of the all-in-one IDE. Sounds more like the OS should be the IDE. Several Applications working together to perform many tasks… sounds like an OS to me! So, does this render my IDE application obsolete before it is even completed? Not in the least. In fact, what we need from an IDE is more centralization and less application. A good IDE should have access to any possible application to complete a project. Whether it is Adobe Photoshop or MS Paint… whether it is Notepad or Notepad++… seems like the IDE should simply tie the content and source together and not try and reinvent the wheel in regards to apps.

Note: In many ways IDEs (such as Visual Studio or SharpDevelop in my case) perform duties that no other application can: Intellisense, compilation, and debugging. In that respect, they ARE the application. However, as programming languages are becoming more evolved, I do not see why they cannot evolve into complete GUI related models. If and when that happens, it simply reinforces my construct that IDEs should only be the glue to hold development together – whether it is one or many.

Web Services and Serializable Objects

Mon, 24 Sep 2007 18:44:22 GMT

I've been currently working on getting one database syncronized with another. The task is hard to maintain two databases, so I thought about another way to approach the problem. Since everything eventually gets down to business logic, why maintain two databases? For reference purposes, one database is MS SQL and contains everything important for our main application. The other one is provided by a third-party via a Linux box running MySQL. Well, I would connect via ODBC and Read/Write data to the MS SQL Database for ease of programming (less ODBC drivers to handle on 30+ PCs). You may ask why I did such a thing? Well, the ODBC drivers are unreliable and as I found out, don't even work with Vista. So our client PCs would have been screwed if they relied on those drivers. Our one server, however, managed just fine so it did all the work for the clients.

Enter Web Services and XML.

The server, rather than updating a client database, now retrieves rows and provides business level objects via XML to the clients for direct use. It is a no brainer. Works far better and has less CPU usage on our main server. I define a single namespace with my objects and serialize them between web service and client app. On another note, it is also scalable but then again so was my other solution (provided the SQL Server was detached and moved with the sync service). However, this eliminates the need to detach anything. On the downside, I do not have a SQL Server method of backing up the data with my other data. I now need to backup seperately.

MS Office sucks...

Mon, 24 Sep 2007 18:44:22 GMT

I hate to be so blatant and cruel, but it's the little things that hurt. MS Office 2007 (like previous versions) is full of them. They spent 4 years between Office 2003 and Office 2007 and you STILL cannot print a label or envelope without going through 14 billion clicks. Even so, making one label versus a dozen or so labels is an entirely different process???

I am TRYING very hard to convince our sales department that Business Contact Manager is Good, Act is bad, and MS CRM is too expensive. The whole time it is blowing up in my face. The only real problem here is that even if I bailed and went with current version of Act!, it would STILL blow up in my face. I just figure Office is easier to control. Seriously, how can you develop a Business contact manager software package and NOT include the ability to write a contact a letter.... seriously... it's beyond stupid.

And if you are thinking of proclaiming "Go to Open Office, it is better and *Free*!" - about the only part about that statement I believe is the word "it". "It" is not free (time invested in setup, learning, etc..). "It" is not better - in fact it is downright a bad copy of an already bad product (Office). Finally, "It" is buggy as all hell. More so than Office is.

Stereotypes and Immigration

Mon, 24 Sep 2007 18:44:22 GMT

I just finished reading several studies done in regards to individuality versus collectivism of Chinese natives, Chinese immigrants, and European non-immigrants. There was an emerging pattern of difference between immigrants and Chinese natives. The hypothesis stated that Chinese natives were more collectivistic than immigrants and that non-immigrants were the least collectivistic. It was revealed the hypothesis was completely false and there were very interesting patterns. One pattern which I noted was that immigrants were typically more collectivistic than Chinese natives and that a possible reason might be the lack of friends and family in a strange country might induce more collectivistic behavior. Alot of stereotypes are usually based on immigrant perception. This is mostly due to the fact that immigrants are like a first impression for most people. It is not like an entire nation can visit the native areas to deduce stereotypical behavior. Are therefore stereotypes actually based on behavior of a culture shock? is it possible we can form the orgination of most stereotypes to behavior which is not only unrepresentative of native individuals, but also is representative of a cultural response to adaptation?

I am not usually involved in cultural behavior or social science like this, so it is merely my thoughts spewing out of my head as I read these articles.

Some of the articles I read:

Social Values and Self-Disclosure: A Comparison of Chinese Native, Chinese Resident (in U.S.) and North American Spouses by JACKI FITZPATRICK, SHULIANG, DU FENG, DUANE CRAWFORD, GWENDOLYN TSORELL, and BARBARA MORGAN-FLEMING (Journal of Comparitive Family Studies)

Rethinking Culture and Self-Construal: China as a Middle Land by Han Z. Li, Zhi Zhang, Gira Bhatt, and Young-Ok Yum (Journal of Social Psychology)

What .. you want me to NOT use Caps Lock consistantly?

Mon, 24 Sep 2007 18:44:22 GMT

This is partially a rant and partially a duck. It's like that odd animal with a duck-bill and looks like a beaver... what was that again? oh yeah.. Dana Carvey.

Ok back to my rant.... as a software programmer, I get alot of UI questions. The most horrendous and annoying questions ever asked of me usually involve typing. For some reason, somewhere along the line, the idea of "typing" words into the computer became standardized somehow to include the following:

ONLY CAPITAL LETTERS
Occasional Missing Vowls
Excessive - uses of the - dash instead of - oh I don't know - using perhaps a paranthesis - or a colon - or some other punctuation mark

It might be that every person who works for my company types with two fingers. It might be the fact they have never heard that one can change a font size, shape, color, etc. Whatever the reason might be, it gives me great pain in my head region whenever I hear things like:

"Can you make it so everything I type is programmatically & automatically capitalized? Regardless of whether I have Caps Lock on or not?"

"Can I just get a list of all 1000+ options for our two-letter-defined location areas and just click the one I want rather than type it in?"

"Why do I have to enter the name of the person on one field and the product in another field? Can I just type them in all in one line and seperate them with a dash?"

and my ALL-TYME PERSONAL FAVRITE:

"Can I get a spell checker for that name field? Oh, and I want all the letters to be captialized all the time."

These, I will admit are not 'actual' quotes but rather summations of things I encounter... a lot. What prompted this entry was actually a request by someone to have an entry field for our inventory locations (Made up of one letter and one number, e.g. A1) converted automatically to uppercase despite the way it is entered. Now, I am all for this idea because of two things. One, if you enter it in lowercase, it is treated as a different location (bad design of some other brilliant programmer and not I). Two, it is generally a good thing to correct things you can easily find out are mistakes of the user. What really bites at me though... deep down inside... is the fact it is a SINGLE letter. Apparently, it IS too difficult for people to grasp that they need to hit the "shift" key when they enter that letter. Not intuitive from a design point-of-view, I understand that and that is why I will try and see if I can implement this. From an ethical point-of-view, I find this demoralizing that people cannot complete the simplist of tasks. If this were occasional errors, that would be one thing. What makes this so demoralizing to me is that people will purposefully or routinely enter lowercase letters for inventory locations. it just shows how much they really care....

Archive

FreeTrack - How to Make a 3-Point clip

The IDE concept

Web Services and Serializable Objects

MS Office sucks...

Stereotypes and Immigration

What .. you want me to NOT use Caps Lock consistantly?