jr conlin's ink stained banana

Well, this is annoying.

So, sitting in the living room a few days ago, i noticed that my Compaq v6000 took forever to come out of hibernation and that it was running really slow after it did. That was to be somewhat expected since i had a bunch of network stuff that was previously running when i put it to sleep. i proceed to try to shut things down, but it was taking forever. So i decide to do a bit of brute force. i shut down the various editors and then cold boot the box.

On reboot, it came up, but it wasn't connecting to the network. That's when i noticed the little orange dot of doom. (Got a v6000? You know what i mean.) i tried resetting the switch, nothing. Waiting 20 seconds, nothing. Cursing like a drunken sailor who just smashed a fine bottle of rum against his own head, nothing. i was not happy.

Well, turns out that somewhere around August of 2007, HP released a new Broadcom driver. i popped that driver onto a thumb drive, ran it on the V6000, (then found the actual setup file in C:/SWSetup/SP36684A) and proceeded to upgrade the drivers for my wirelss card. Things work, i'm happy, and then… well… they stop working.

Now, i'm really getting into the spittle driven anger.

Well, turns out that apparently the Broadcom wireless card in said V6000 decided it doesn't want to play anymore. No problem, i think, because this will be a nice opportunity for me to swap out the mini-ide card for one that may be a tad more advanced. i could pop in an intel 802.11bgn. Cool beans! There was even a online shop not that far away that sold 'em. Two days later, it arrives, i plug it in, restart and immediately see "INVALID WIRELESS DEVICE DETECTED. REMOVE IT AND REBOOT." Yeah, that'd be HPs fun way of saying " You Buddy! Gotta buy the crappy $20 card we stuffed in there. Mwah-ha-ha!"

Ok, so i probably could have tried another 802.11b/g card.

Still, darn annoying.

As it is, i'm using an old $20 80211g dongle i had left over until that one shows up next week.

So, basically, laptop broke and it's going to cost me something like $35 to fix it.

Ok, it's a principal thing.

With sincere apologies to Mr. Freidl, i'm having a hard time rationalizing spending $300 for Lightroom.

About a week ago, i downloaded a trial copy to try it out. i had been shooting a lot of raw images and wanted to see what sort if imagey magic i might be able to do with them. i tend to stay reasonably organized when dealing with images (even if i shoot a crapload of them) so the library wasn't really that interesting to me. Instead, i turned my attention to the "develop" function.

i dunno. For the most part, i was kinda unimpressed.

Mind you, i've been using Paint Shop Pro 8 (Photoshop's mutant clone created from some DNA samples scraped from a shower wall) and recently Gimp with UFRaw (which… well… let's just hope they're from the same general sector in space as Photoshop) to monkey with images, so i was kinda expecting a bit more. Most of what Lightroom really gives you is some of the stuff you could do in a Darkroom before images went digital. Along with not some stuff you could do. i've done my time in darkrooms, so i know what's possible. It's kinda weird.

i did learn a few things about histogram modifications i was able to use on some images, so it wasn't completely worthless, and i'm also damn sure that i don't fit the photo professional profile this was obviously targeted at. There are some features i'll miss in 20 days when the demo expires, but not enough to separate myself with three Benjamins. Mostly, i'll contend myself with taking what i've learned and using the bone saws and flint axes that i'm currently working with to do image corrections and enhancements.

Or maybe i'm just incredibly cheap.

For reasons outside of my control, several individuals have informed me that they wish to use DocBook for document generation. Mind you, i'm more than comfortable with "traditional" HTML layout for these sorts of things, but yes, i can see where having a nice generic SGML layout for a document does make a great deal of sense. That part i'm very comfortable with.

Where things break down for me is with the various consumers of said docbook generated content. i hateses them. i hateses them so much.

You see, the problem is that the docbook generation system uses a combination of XSL and a pile of still warm chicken entrails to convert the "raw" docbook formatted XML into something less XML-y. This system is designed to generate one layout of all documents. i know this because of delightful little details like how elements store styling information for the HTML constructs. (What? You DON'T want <H2> tags to have "clear:both"? Are you MAD!? What possible thing could you ever want to put next to a document!? i mean, aside from a table of contents, images, a sidebar, example code, or really anything other than the loving document WE generated for YOU?)

Honestly, i've had to replicate vast tracts of XSL in order to do silly things like make pages semantic so i can apply simple CSS to them or not use the kludgey, pre-baked navigation systems. …or change context for sections to NOT be the current one and instead pull from the top, or other various things you'd really, seriously, think someone would have converted to a parameter. Don't even get me started on what an incredible pain it is to generate a full table of contents on output page without resorting to an external script to "fix" several pages.

Oh, and those bits about DocBook being "simple" and "easy to configure"? Don't you believe them.

i signed up for yet another social content aggregation site. This time, Friendfeed, (which seems to have a pretty high number of yahoo's on it). The funny thing is, it got me thinking about the problem at hand.

Here's the crux of it. There's a lot of services out there where i can contribute content, (like delicious, flickr, my blog, etc.). There's a growing number of sites that are also trying to organize that (like MyBlogLog, Facebook, FriendFeed, etc.) While those services are interesting, the problem is that i don't "live" there.

i really need to figure out how to let folks follow what i do, and ideally, let folks be able to install it.

Kind of a "feed of me", hosted here where i live, controlled by me. Kind of like how you can publish your own information as part of your OpenID profile and let properties pull that info in.

Hmm….

Over the past few days (sorry, too many other things have been cropping up), i've been spending time learning about Python. i think i finally see what's gotten folks so darn hyped up about it.

For me, Python is what happens when you take COBOL, SmallTalk, Javascript and Perl, toss them in a box, shake really hard and pull out what stuck together. i could argue and point out what came from where, but that really doesn't matter. If someone wasn't exposed to those languages and only saw Python, they'd be right enamored too.

Python definitely seems to be targeted at text protocols like XML, HTTP and the like. Perl's a bit easier dealing with stuff that's Regex heavy like straight text processing, but Python obviously takes advantage of later developments in document models. Bits seem a bit overblown (seriously, Lambda functions are simply anonymous functions. Javascript, Perl, and a host of other languages have 'em. Likewise, SmallTalk has had the "everything is an object" thing for just about ever. Including operators. Meaning that yes, you could do things like //Convert '15' to a number and add 10 limiting to one byte
value = '15'.asNumber +.inOneByte 10 if you really felt like giving the folks that may inherit your code reasons to hate you forever.) i'm still not horribly convinced that building structure around whitespace dependencies is a horribly good idea (sorry, dealt with too many bits of code poorly transported to not appreciate the concept of delineated code blocks), but i'll get over that.

Still, it's interesting to see what one can do with this. Definitely going to have to play with it a lot more.