Brandon's Blog
7/9/2012 #
Standing on One's Own
Effecting a vision is hard in meatspace. Think about October Sky. A crew of plucky kids with a dream. A dream that could not be a reality without some serious resource commitment and craftsmanship. I recall a machinist character who snuck around to steal tool time from a company to fabricate a metal part of the rocket’s body.
The human and material capital required to produce something quite simple, let alone some reasonably complex system like a battleship, is profoundly humbling. Just the tooling required to build a small table can bear a retail cost multiples higher than the cost of a mass-produced equivalent, including materials.
The “tooling” required for a lot of intellectual and technical development is often free, or at least quite cheap. I was thinking last night about Efendi, my own creation that runs my blog, and Clarity, which does the job of Picasa or Flickr, but just the way I want it to work, and on my own (rented) iron and on my terms. Cluster 2 runs on Python and was my creation. Cluster 1 limped its way through on PHP and was the work of one slow afternoon.
I wandered away for a moment’s nostalgia, but my point is directed toward the idea of feeling a sense of delight and pride at watching one’s creations doing good work. Standing on the shoulders of giants, of course, I effectively breathed life into these things, installed them on a platform of my provision, and they now provide a service to me and people I know. It’s a good feeling, even though they are trivial things.
I get a similar pleasure when I install a light fixture, or hang a bookshelf, or wire a network, or hook up a ceiling fan, but it’s just so darn expensive to do anything like that, let alone to build something up nearly from scratch. It’s a fascinating turn of events that innovation in computer science is so comparatively cheap.
It was an interesting feeling last night to hold Claire up by the armpits just above my lap. She pumped her legs outward as is her wont, but this time - for the first time - she straightened them out and pushed against my legs. The force borne by my hands reduced as she lifted herself up a little bit taller. She did that on legs that, not three months ago, were somewhat alarmingly thin and relatively underdeveloped at the time of her birth.
That action was the product of cumulative days of seemingly random experimental muscle movements. She did it burning high-density calories off of food created by her mother’s own body. And there we were again, seeing one’s own creation do a little something on its own.
7/7/2012 #
Clarity is Here!
I finally put the hackish finishing touches on Clarity. Not my favorite build from a design standpoint, but it’s getting the job done. Let me know if it breaks!
By the way, one of my design compromises is a lack in dating precision for new pictures, so they will insert in the blog feed somewhat randomly and apparently below this post. This doesn’t make sense when I’m adding a bunch of pictures for the launch, but once it’s more one-at-a-time it’s going to work fine.
7/3/2012 #
Progress
I underestimated the templating requirements of the Clarity admin panel, but a dirty implementation is coming along. I swore to myself I wouldn’t over-engineer this thing, and I’m keeping my promise as I hammer through it.
7/3/2012 #
Compounding the Last Few Posts
In the spirit of breaking out of ruts but conserving time, I resolve to implement Clarity in a first-round attempt that will be technically lacking but functionally adequate. If I ever want it to be better, I will rewrite it in a language new or different to me. Maybe node.js or Lisp or something.
7/2/2012 #
New Thing
I’m reaching a point where I feel like I’m really good at the things I’m good at, I tend to default to using the things I’m good at, and don’t know much about stuff I don’t use. There are some nasty Venn diagrams that come from that statement, but the essence of it is pretty accurate.
I’m not sure I like that completely. I almost wrote Clarity in node.js just for giggles, just to bust into something new and not-Python. If I had a little more time to devote to the project I might have. But that’s kind of what it comes down to: I want a lightweight, customized photoblog more than I want to learn something new. And I didn’t want to hose my freshly-unhosed server putting it together. That was enough to set me back to copying boilerplate out of Efendi’s git repository.
I just did some VBA coding for a friend here, and it kind of felt like playing tee ball. It felt good to swing the bat, but there was nothing much new, aside from incidental learning. I’m resolutely looking for chances to break that rut.
7/2/2012 #
New 'Design'
In preparation for Clarity, I decided to kill off my colorful site and go basic for a while. Every time I say a design is temporary it lasts a really long time, and whenever I say I’m happy with a design it goes quickly. I don’t know what to say about this. I’ll miss Gaudi’s columns, but I won’t miss the shoddy CSS and kind of dated color palette. We’ll see.
7/2/2012 #
Gaining Clarity
In between lots of family engagements and a heightened battle with fussiness, Clarity is taking shape. It’s a stunningly simple read-only front-end (I’ve elected to forgo commenting), a one-function backend (import and catalog incoming pictures), and a lightweight hook to Efendi so I can get linked thumbnail pictures onto the front page of the blog.
My “stack” is Apache for the basic serving, Python for teh codez, Werkzeug (a Python library) to handle the interface between Python and Apache, and a file-based “database” which is really just an encoded list of strings in Python.
Werkzeug is my kind of tool. What it lacks in detailed documentation is compensated for by its extreme libertarianism. You pretty much pull in hunks of Werkzeug as you want them, and write your own or omit if you don’t. Turns out the glue that stuck Werkzeug to Efendi could really have been reused in this app (and probably in Brightlamp as well), so there’s a bit of copy-paste coding going on as I rig up the basics of my app. In many ways, Clarity is running on a very thin skeleton of what makes Efendi work.
The productivity of Python makes the coding part of this exercise, especially with Efendi providing what might be considered a silver-standard boilerplate, almost trivial. My work pattern has been mostly thinking, then more thinking, then reversing previous thoughts, then a little blast of coding and copy-paste that implements an entire section of the app.
The front-end is essentially complete (did the HTML side over breaks at work), and the data model is basically there. This pretty much leaves the photo-importing backend as the major work to do. That’s not too bad, but it’s the major work of the project. I’ve just paper-and-pencil’ed it to only consume a single webpage, so I’m excited to code it in tonight, time permitting. This may mean I could have a working version by today or tomorrow, if I’m lucky!
6/28/2012 #
Ruling
I honestly don’t think I am willing to read enough to form an airtight opinion on what happened today in the Court, but I do think there is something at least relatively sound about what happened.
The majority opinion, as far as I understand, did rule that the mandate does not stand up using the Commerce Clause. That’s a big deal, as the misuse of this clause is a true travesty of legislative cowardice thrust upon this nation. If the federal government wants to expand its power to regulate, it should do so by constitutional amendment, not by expanding and “evolving” a clause that clearly was meant to avoid tariff and duty wars across state lines, which was a huge problem in the Articles of Confederation era.
What Roberts did, which I don’t think I fully support, was to say, “Look, what you have here is an expansion of authority by taxation. It’s not a regulation, it’s not a mandate. If you have a problem with the tax system in this country, solve it at the ballot box, but don’t look to us to toss out tax bills because they restrict your freedom. You gave this government the power to tax.”
If SCOTUS holds fast to this halt on expansion of the Commerce Clause, this is a landmark decision. If the Commerce Clause gets expanded further within the next 10 years, this is a joke.
In the case of ObamaCare itself, we’ve already seen how this can change behaviors. Obama denied up and down that this was a tax. It’s clearly easier to sell something as a mandate versus a tax. So, the Supreme Court has effectively forced Congress to be accountable to citizens by restricting its power to wielding a club that nobody likes.
It’s not a great day to be a libertarian, but the reasoning applied has some sense to it, even from a conservative perspective.
6/25/2012 #
SNAPpy
I don’t understand why SNAP (the fancy new name for the food stamps program) uses almost entirely open-ended debit cards, yet WIC generally sticks with vouchers geared toward specific types of food (milk, cereal, etc.).
The very acronym of SNAP, which begins with “Supplemental Nutrition,” seems to belie what should be the real intent of the program. I would prefer it to be called ASAP, Anti-Starvation Assistance Program. Why am I spending $80 billion a year to supplement the diets of millions? I would rather spend $40 billion, or preferably much less, a year to provide baseline nutrition for those in emergency need. They can do their own supplementing as it is possible.
If you establish the spirit of the program to provide bread, milk, beans, rice, vegetables, fruit, and nutritional and inexpensive proteins, and bottled water when necessary, you can essentially set up SNAP to be the same thing as WIC, but for everyone in need. In fact, just merge the two and have infants be an additional rider on an existing assistance contract.
I don’t see how SNAP needs to be so broad, if WIC is supposed to be good enough to keep infants, young children, and mothers alive. What’s sauce for the goose is sauce for the gander, as dear Mittens would say.
If FLOTUS can dictate to me what I should eat, I don’t see why SNAP can’t be a pretty strict plan. Maybe in earlier eras it seemed okay to be prescriptive only about what infants and children should be eating, but in the Bloombergian era of banning soft drink sizes for all, why can’t we go ahead and get a little bossier about what adults should be getting on the dole? If the free market is getting limited by government, why can’t the public sector food racket get the same, or preferably much stricter, treatment?
Sliding toward the absurdity of Freakonomics and such, what if we cut people’s SNAP benefits if they and their families are found to be gaining weight? The way it is now, SNAP is just another injection of cash to construct a “true” “living wage” for those who are working, and to subsidize higher market wages by keeping a large set of people out of the workforce and on the dole. It’s not overly concerned with exactly how close somebody is to being truly in need. And it’s certainly not watching out for what people are doing with the money, beyond the most basic of controls.
These are luxury expenditures on a societal level, just like baking contraceptives into health plans legislatively. We can certainly afford luxury in our society, even on a grand scale, but with the debt outlook we have, it’s going to get necessary to ask if we’re accomplishing the actual goals of these programs. Which luxuries do we want to fund?
Political rhertoric as it is, you hear “Paul Ryan wants to shut X down, cut Y totally, etc.” This is not good. We don’t want people starving on the streets, but we need to be able to talk over how we avoid this without accusations of total malice immediately flying. The means are a valid discussion, and maybe the same - or similar - ends can be met in different ways.
> Newer Posts
< Older Posts