tbradshaw.net

Recently I’ve been plagued with a series of telemarketing calls to my personal cellphone, none worse than a telemarketing call that I get every several days advertising “free government grants”.

The thing that makes these calls the most annoying are that they are just a recording!  I don’t know how this can possibly be legal, as I know I am within my rights to demand a telemarketer put me on their “do not call list” and stop receiving calls after a two or three week processing time.  In this case, there’s no one to complain to!

So today I had enough, and when I received this recording I clicked “1″ for more information and waited until I received an operator.  She asked my name, I responded politely with “Actually, I’m calling because I keep getting called on my personal cell phone with a telemarketing recording and I want…”

That’s as far as I got.  Because the operator said, “Oookay… click,” and hung up on me.  This is such bullshit.  Even when I get to the point that I can request to never be called again, they just hung up on me.  If I receive the call again this time I’m going to ask for the contact information for this company and contact legal council.  I have a free lawyer here on campus that could at least get me started, and since I know that telemarketing laws are well defined regarding fines, perhaps I could get legal council to just keep the penalty.  I don’t know for sure.

Either way, I’m annoyed enough to think it’s probably worth it to cause this disreputable company a little hassle.

For a long time I’ve been putting off the purchase of a new set of monitors.  While I’m certainly spoiled at work with my two Dell 1901FP flat panel monitors, I’ve just been scraping along at home with a couple of 17″ CRTs that I’ve had for many years now.

It’s been at least five years since I’ve had a new monitor and it’s been showing for several.  For quite some time my left monitor has been flickering as the green (of the red, green, blue signal) cuts out.  My right monitor’s picture shakes violently at times.  Both of these mannerisms are annoying, but I can live through them for the most part.

However, now my left monitor has picked up the most annoying of tendancies.  It emits a high pitched scream almost constantly.  I imagine what this would be like if a dog was trapped near a dog whisle.  It’s horrible.

I don’t have the money now to replace the monitor, but this has got to move up to one of my most important priorities before my head explodes.

In a startling turn of events, our meeting today regarding the roommate situation we actually came to a conclusion.  This was a rather stunning turn around from what I had expected.  It appears that one of our stalemated groups decided that they would rather pick the other guy than see the house come to an end.

I know that I personally was very difficult during this process, and I am moved and appreciative that the other guys felt that the keeping the house going was worth who, in their opinion, was the lesser candidate.  I still stand by my original opinions, but hope that I’ll also get to know the other candidate better and maybe bring him into the fold at a later date.

I still haven’t soaked it all in, but I think that it’s going to be hugely positive experience and I have renewed my dreams to keep this house going for as long as imaginable.  (And that’s a long ass time!)

I’m depressed, there’s no getting around it. It looks like it may be the end of The Bunker. We’ve been having a series of very solid meetings regarding the new roommate situation, and we’ve reached what appears to be a no-win situation.

We have four potentially returning roommates and two candidates to fill the final spot. Very unexpectedly, we have a situation where we ended up split with some of us not wanting to live with one of the roommates and some of us not wanting to live with the other. (This is quite different from a situation where some of us just like one person more, it seems that the two groups have an honest thumbs down on a roommate.) (This is intentionally vague, since the matter is not yet officially resolved. It would be in very poor taste for me to name or number at this point.)

It’s not a situation where there are big fights, or even anything to argue about at all. We have a group of four adults that have a legitmate disagreement of opinion on the future direction of the house and the viability of two candidates to fill the role of a Bunker roommate. The discussion has never been unfriendly, or disrespectful. In fact, I’ve enjoyed the maturity and thoughtfulness that all of my roommates have displayed regarding the matter.

It just seems undeniable that we’ve reached a no-win situation, and I wouldn’t ask any of my friends to “lose”. Although things are not completely forgone, and we are scheduled to meet again tomorrow night to continue our discussion. (We’ve been tabling the issue for 24 hour periods since Monday, where we reconvened from a one week recess.) But barring a complete change of opinion and direction (of which I can think of no rational explanation) from someone in the group, I think we may be dead in the water.

Shutting things down after this lease seems like the responsible and mature way to handle the situation. After all, with no acceptable candidates left and no time to continue to look, the only way that each of us can continue with our personal expectations would be to live seperate. But that rational knowledge of the clear and present “right decision” doesn’t change how much it hurts to think of this house being “over.” My roommates have been my best friends for years now. The times that we’ve had in this house have been absolutely legendary, by any collegiate standard. All of these thoughts make the ending of this era of The Bunker seem so premature and difficult to handle.

I really felt like we were ramping up for a return to greatness and–with a new longterm roommate coming in–a renewed viability. But an honest difference in opinion, instead, may very well sound the death toll. I have no idea how I will be able to concentrate on anything else now with this emotional burden. This is a type of pathetic that I’m not one to frequent, heh heh. Usually it would take some girl to make me this upset. This time it’s a house home.

One of our assignments in den325: Personal and Professional Development (PPD) is to complete a wardrobe checklist. A trivial assignment, not without some merit, but then they cross the line when at the bottom of the assignment they require a five item purchasing plan for improving one’s professional wardrobe.

They ask that one list the next five purchases, including the “type of item” (from the checklist above, no doubt), the “estimated cost” of the item, and the “planned date of purchase”. They also encourage the student to “incorporate this information into your personal plan for improvement due at the end of the semester”.

Well here’s the rub, what if I have no plans what-so-ever to improve my professional wardrobe. It needs to be done, no doubt, but I’ve got several more years of school and I don’t even have an adequate “casual” wardrobe for day-to-day activities. Do I lie on the homework assignment and just make up a plan? (This isn’t really a good solution for me, I hate lying and refuse to do it.) I don’t know. So I sent off an email to one of the TA’s that reads:

I have a question regarding the Wardobe Checklist.

At the bottom it presents a purchasing plan for the next 5 purchases
towards a professional wardrobe, including the type of item, cost, and
planned date of purchase. (With -8, -8, and -4 deductions assigned to
each.)

I have no plans to purchase any of these items in the forseeable
future. I’ll probably end up buying them when I “have” to for a
professional engagement or a funeral. Do I need to lie on the
homework assignment to get full credit? Should I assume that the
homework assignment is “for pretend”?

Or do poor, honest kids just get bad grades? (Mostly kidding.)

Travis Bradshaw

We’ll see what they say in response.

For quite some time I’ve been avoiding the Gmail craze. Not because I didn’t want to participate, no that’s not it at all. In fact, I’ve been relatively certain that Gmail was the best possible way to handle mail currently available to mortal man. So why the hesitation? Well way back when I decided to entire the modern internet community–start blogging and reading blogs, sharing photographs, etc–I had given a lot of thought to the different options out there.

At the time I was primarly considering Drupal versus Wordpress versus Blogger, but the points of contention were exactly the same as my current wrestling with mail providers. While the hosting and feature set of Blogger was(is?) superior to the host-it-yourself solutions of Drupal and Wordpress, I really liked the idea of owning my own data. It would be in a database that I would own, I would back up, I could translate or transfer (which came in handy when I switched from Drupal to Wordpress), that I would control. So I ruled out Blogger because I would lose that control, and I continued to choose between the host-it-yourself options.

With email I came to the same conclusion. I was never interested in a Gmail account because I already have my own web propery with tbradshaw.net and the mail hosting at Dreamhost is absolutely fantastic. I’ve been wrestling with the idea of trying Gmail for quite a while now (there’s actually a “Draft” blog post in WordPress right now called “Gmail Invite” where I never followed through) and last night I decided to take the plunge. I did it for a number of reasons:

• K-State’s email sucks. - After we moved our CIS department email to CNS we realized something very critical that we weren’t expecting. CNS’s email reliability hinged on the fact that hosting just ksu.edu was “easy”. The added complication of our email and especially (perhaps to the point of making CIS insignificant) oznet has mostly destroyed the quality of service from K-State.edu email. Even when it is working, it’s remarkably slow.
• Gmail’s Up-times Approach Infinity - Well, DreamHost’s uptimes are pretty fantastic too, but I have no idea “how many nines” Google pulls. I bet it’s a lot.
• Google has all the sweet toys. - One thing I really miss is not being able to check out all the sweet new toys from Google, a lot of which are starting to be tied together with the Gmail account.
• Gmail establishes Identity - Just as having tbradshaw.net does a lot to establish identity, so goes one’s Gmail address. I have already slumbered long enough to lose the possibility of getting tbradshaw, t.bradshaw, travis.bradshaw, travisbradshaw, or anything similar. ctb is invalid. c.travis.bradshaw is actually a pretty good identifier for me, even if it different than what I’ve been using previously. Starting now landed me c.travis.bradshaw for my Gmail identity.
• I can keep another copy. - With most/all forwarding mechanisms, I can also keep a copy on the IMAP server just like I have all this time. So while the privacy concerns with Google might be a small issue, in the general case I have nothing to lose.

And so that’s it. I’m now a Gmail participant and currently in the process of getting all of my previous k-state email onto gmail so I can really give that search functionality a ride.

I’ve now logged my first dozen or so hours of work with the TurboGears “mega-framework” and the ride has been a bit bumpy. In the general case, I’ve been very pleased with the amount of work that I’ve been able to accomplish in relatively short time. Of course, that’s almost always the case when developing in Python, but I still think that it’s significant. With a minimal amount of fighting, CherryPy (the web development framework) and SQLObject (the object-relational mapper) are just some seriously bad ass projects.

My experience with Kid was a little bit surprising as well. To start with I didn’t even plan on using Kid. I have been using Cheetah in the past and have been very satisfied with it’s ease of use. I noticed that there was a “TurboCheetah” plugin for TurboGears and immediately set about my way to use it. This is when I hit my first “big rub” with TurboGears when I realized that the plugin support for template engines didn’t actually exist yet. I mean, it does exist… in the development version in the source repository, but it doesn’t exist in an easily accessable way described by any of the documentation. Which is kind of difficult, since the documentation already lists a ton of content for the 0.9 release version-that-hasn’t-been-released.

And this would be the definite trend for the TurboGears framework. The documentation is very sparse and most of the “good stuff” you’re looking for is currently unreleased. Of course this isn’t inherently a bad thing, TurboGears is a young project that is making fantastic progress. The only real bummer about it is that I’m wanting a relatively full featured and stable framework to develop a web application on… well… yesterday.

My original plan for satisfying this need was to just use the stable version that has already been released (0.8.9) and later migrate to the newer stuff when it comes out. This turned out to be a bad idea when I spend my first dozen or so hours writing an authentication module, only to find out that the “Identity” package of the next TurboGears does 90% of what I’ve been working on… and does it better than I did.

Clearly a new plan of action is needed. Now it looks like I’ll need to be installing a custom environment that contains the latest nightly of TurboGears that is available from their subversion repository, pray that it’s stable enough to at least work with, and start development with the intention of eventually using the 0.9 or greater project for deployment.

There has also always been a desire of mine to contribute to an open source project. The first project I had previously ear-marked for this involvement was the Drivel Journal Editor that I’ve been using so much. However, inherent barriers include Drivel being written in C using GObject (something that I haven’t done before, and would require quite a bit of study before I could contribute in any meaningful way) and the fact that WordPress 2 has provided a much nicer editor than before. TurboGears on the other hand is a significant project, very young, and written entirely in Python. This really seems like it might be the easy access project that could be a good point for me to contribute to the open source community. Hopefully I can convince my bkms project teammates of the same.

Success. Trac is fully operational and already providing a great organizational service to our project. Overcoming that final obstacle turned out to be a partner deal with a plugin for Trac that used a form submission to check the basic auth of another folder. (Which was acceptable, because I could create that folder and handle the .htaccess requirements there.)

I’m very pleased and impressed with Trac so far. Previously I wasn’t particularly sold on Wiki’s, but the integration of the Wiki to the bug tracker and such is just too slick. I expect this to be a huge boon to continued productivity. So much potential, that I’m rapidly considering moving all of my projects to trac environments. We shall see.

Based on further reading (and an encouraging post by Jeff Weiss) I’ve decided to press on through with getting Trac set up to help with project management on the BKMS project.

After deciding to go for it, it’s been quite a struggle to get Trac fired up on DreamHost. Not necessarily because of problems with DreamHost, but I think the primary problem is the spare documentation that happens to have one key facet in common. All of the documentation assumes that one is an administrator on the server that hosts Trac.

After spending an extended, challenging period getting all of the necessary prerequisites installed, and then another intense stint of time working on getting Trac going with FastCGI support thanks to some creative “wrapping” I’ve finally hit a bit of a wall. It seems that the authentication for Trac hinges on Apache based authentication and requires a snippet like this:



<Location /trac>
... extra directives to invoke trac
... - ie ScriptAlias or mod_python stuff
</Location>
<Location /trac/login>
AuthType Basic
AuthName "trac"
AuthUserFile /var/www/db/passwd
Require valid-user
</Location>

The catch? These particular directives are not valid in an .htaccess file. And since I’m using FastCGI without so much as a ScriptAlias, my importante URI (relative) looks like http://www.tbradshaw.net/trac/bkms.fcgi/login . Note that the “/login” suffix is an argument to the fcgi-script. This means that I can’t just make a login directory and provide an htaccess file to it (I don’t think), and I’m relatively stumped.

I’m going to keep thinking on it and banging on it a little bit tonight, but if anyone out there has any bright ideas be sure to let me know.

Before:

After:

Ever since I first tried to keep up with the math551 (Applied Matrix Theory) taking notes on my laptop in LaTeX have I lusted after a well done Apple tablet. Just the idea of taking math notes seemed like the killer application at the time.

Well Apple has filed for some more tablet related patents, and things are looking all the more likely. I sure hope something like this pans out.

As we get closer to starting actual work on the BKMS project I’m starting to wonder if there is any sort of project management infrastructure that would increase our productivity as we move forward.

It’s not immediately apparent to me if it is really necessary to have a formal project management infrastructure to help us out, but as I look at my various personal websites I find that I’ve spent a noticable amount of time building my own project management infrastructure. Notably, i’ve created the /projects section of my various sites that contains the elements of project management that I think helps me.

Which is fine and dandy, but those static HTML pages that are spiced up a bit with CSS don’t provide any mechanism for collaboration on any of the project management type stuff. As we move forward as a team (Dan, Kuffs, and I), it might be hugely advantageous to be able to work together on the project management documentation.

Well, this just screams wiki at an undeniable volume. I’ve already set up Dan and nearly completed setting up Kuffs with subversion access, and perhaps a bug tracking mechanism would be handy for tracking other things about the project, I’m not sure.

It seems that Trac is the project that fulfills all of those goals. It’s a one stop shop for web-based source code repository access, provides a full wiki for documentation, nice bug tracking facilities, and some report generation for keeping track of things. (I especially like how it can be used to see the comments from changeset commits by date and without a bunch of other svn cruft like the usual logs.)

What seems weird is how uncertain I am about exactly what Trac is “for”, even though I use RT at work extensively. I guess I’ve primarily considered RT as a mechanism for triage and tracking between “end users” and the “developers”, and I haven’t ever really paid attention to the benefits that come from the developers interacting with the developers in this format. (Although just thinking about it now brings a lot of examples to mind.)

I guess I’ll just have to install Trac and see if it’s something that I want to deal with, but if anyone out there has any experience with Trac (or any other project management type software) in the setting of a small team of developers with little end-user interaction I’d really appreciate some advice.

Holy crap this class is painful. Typically I just don’t attend this class for a number of reasons. Most notable are the poor lectures. Perhaps “poor” is a little bold, but all the lectures do is cover the introductory material from each section of the book, nearly verbatim. He doesn’t even change the numbers on the examples, they are literally right out of the book. That leaves me with a class that I can just read the book and learn all of the material faster than attending class.

More painful. He lectures so slow! I just watched him take five minutes asking us to add fractions together. No, seriously. Say you have the list [1/8, 1/8, 1/8, 2/8, 1/8, 1/8, 1/8] and you want to make another list that is a summation of the previous numbers, like this for example: [1/8, 2/8, 3/8, 5/8, 6/8, 7/8, 8/8]. See how that just adds up the fractions? He waited for someone to answer him for every single freaking question. The class has already went to sleep, no one in the engineering curriculum needs to listen to the teacher to understand that 1/8 + 1/8 is 2/8. I mean holy shit!

So today I have a little presentation for a K-Slug luncheon regarding a LAN party that I’ve agreed to coordinate. However there’s also a “quiz” today in stat510. I come to stat510 so that I can take the quiz, stay for a maximum of 30 minutes, and make it to the lunch. Instead he start lecturing, and it’s so painful I can’t take it. I can hardly stand having my time wasted like this when I have better things, more productive things, to be doing instead.

I just can’t stand it. I’m leaving. I’m out!

My research into web application frameworks lead me many directions. My latest stop has been TurboGears. I must say that my examination of the framework so far has been dreamy.

The general idea is that they have stacked (front to back) MochiKit on top of Kid on top of CherryPy on top of SQLObject, while providing all of the glue necessary to call the conglomeration “cohesive”. MochiKit is a nice JavaScript library for AJAX style fun, Kid is a templating system (although I’m tempted to use Cheetah instead, since I’m already used to it), CherryPy is a Python-based application server, and SQLObject is an object-relational mapper for database access.

So far the installation was trivial, the “built-in” application server is great (and more than adequate for our intranet needs), and using it has been an absolute joy! The entire package already maps how I “prefer” to find open source solutions, stacking tools on top of tools to create an entire framework. Only whenever I realize I’m ready to use another tool, TurboGears already has that tool lined up in the stack and ready to go.

I’ve never really done this kind of web development, web development that isn’t about the glue and instead is about the actual application. Any web application I’ve ever written was just as much about form validation and authentication and serving pages as it was about accomplishing something or actually providing a service.

I have to say that I’m absolutely thrilled to be working with this framework for developing the majority of the BKMS project. I think that Dan and I are really going to zoom through this application with a lot of measurable progress in most areas right off the bat. With previous technologies it even took a while before we had established a tool chain and development environment and were ready to actually talk about the application. (On our first false start we spent a good planning session on designing a quality relational database and was never even ready to create code before realizing our tool chain was a bit tedious.)

I’m stoked to start making progress already, it’s time to start getting a class diagram created and plan this baby out.