Inert Detritus The Internet's dust bunnies

New MacBook Pros: antiCAPSLOCK

Jonathan Rentzsch showed us the anti-caps lock bias of the new Apple alu­minum key­boards. With the arrival of my new Mac­Book Pro today, I’ve observed the same bias with the built-in key­board: the caps lock keys have to be pressed and held down to turn on caps lock, but will turn off caps lock with the slight­est press. Apple to blog­gers, IRC denizens, and email authors every­where: STOP YELLING.

CPU History — An Activity Monitor Replacement

Today marks the release of my first appli­ca­tion for Mac OS X: CPU His­to­ry. CPU His­to­ry is a 10.5‑compatible shame­less ripoff of Apple’s orig­i­nal CPU Mon­i­tor, which I used in 10.2 to 10.4, but stopped work­ing in 10.5.

CPU His­to­ry is designed to do one thing, and do it well: graph your CPU usage his­to­ry. It graphs the user, sys­tem, nice, and idle usage in the Dock icon, and option­al­ly in a win­dow which can be set to float above all oth­ers. CPU His­to­ry lets you change the update fre­quen­cy and the bar width, along with set­ting col­ors for each type of CPU usage.

Activ­i­ty Mon­i­tor is a resource hog for this sim­ple Dock icon graph­ing: between Activ­i­ty Mon­i­tor and its pmTool back­ground process, the two con­sumed 15–20% CPU on my iBook G4 in 10.5. By com­par­i­son, CPU His­to­ry con­sumes 0.8–1.2%.

CPU His­to­ry 1.0.1 is avail­able over at its new home on cbowns.com, along with a git repos­i­to­ry of the source code.

Much thanks to Bern­hard Baehr and Peter Hosey for their appli­ca­tions, Mem­o­ry Mon­i­tor and CPU Usage. Thanks to their source code, I was able to graft pieces of each onto the oth­er and come up with a use­ful new application.

A Dissection of Roxio Toast 8’s Disc Spanning

Toast 8, a CD/DVD burn­ing util­i­ty for the Mac, intro­duced a new fea­ture: when you add files to a data CD above what a sin­gle disc can hold, it offers to burn them onto sep­a­rate discs. When I went to back up (and remove from my hard dri­ve) about 40 GB of ripped DVDs, this looked like a fan­tas­tic fea­ture. How­ev­er, I want­ed to make sure that, should I need to use the discs on a PC, on a Unix machine, or on a new­er or old­er Mac. I could­n’t let myself be tied to a plat­form, OS, or application.

For­tu­nate­ly, Rox­io has ensured that, short of los­ing one of the discs, you’ll be able to get at your data somehow.

First, Toast copies a text-editable Disc Index.plist to each disc, hid­den at the root lev­el of the dri­ve, gzipped to save space. It’s iden­ti­cal across all discs in the ses­sion, so any disc has a com­plete copy. It con­tains a dic­tio­nary of all the files in the ses­sion, includ­ing their size, whether they’re invis­i­ble, their date mod­i­fied, whether to copy them onto a PC, and for a spanned file, the discs on which to find the file.

I first test­ed the includ­ed Mac ver­sion of their disc span­ning util­i­ty: it sim­ply copies the pieces of the files and con­cate­nates them togeth­er (more on that in a minute). The discs also include a Win­dows exe, though I haven’t test­ed it. The spanned and remerged files emerg­ing from the util­i­ty are, as you would hope, bina­ry iden­ti­cal to the original.

Ini­tial­ly, I was wary of using disc span­ning: a pro­pri­etary util­i­ty to reassem­ble files? An unknown for­mat to recov­er them from? But it turns out that their util­i­ty is a very pret­ty inter­face for a cou­ple of sim­ple Unix com­mands. By con­cate­nat­ing togeth­er the two (or more) pieces of the files, in order, you pro­duce an exact copy of the orig­i­nal file. I test­ed this on a split AVI: cat file1.spanned > testfile; cat file2.spanned >> testfile; md5sum testfile originalfile; will pro­duce iden­ti­cal MD5 checksums.

So, at the end of the day, I went ahead and burned a 10 disc ses­sion of movies to DVDs. If I lose any of the discs, sure, the data is lost and the remain­der is unre­cov­er­able, but I have a list of what was lost in the Disc Index.plist, and can Bit­Tor­rent as need­ed to replace it. Would I span a disk image with a back­up across two pieces of media? Prob­a­bly not with­out burn­ing two or three copies of the ses­sion, and test­ing each. But for movies, it’s a great way to back them up.

A Technology Saying for the Technophobe

Inspired by warped­vi­sions:

“It’s all geek to me.”

(Dear read­er: I’m sor­ry. It’s finals, and I’m scrap­ing the bar­rel for some­thing to post here. I swear, I’ll make it up to you after this is all done.)

Today’s Stupid Thing

We have our PHP error log­ging turned back to E_ERROR | E_PARSE_ERROR. Why, you ask, if it’s in the log, and it’s good to know what’s going on? Why, because if we turn on E_ALL, the log­file grows very fast (!) and the serv­er goes down when it’s over 2 GB.

*shakes head* WHAT?!? That’s the dumb­est thing I’ve heard today. Either fix your damn errors so a log­file does­n’t kill things, or have a cron­job take care of log rotation/compression/deletion. Good god.

What “Why Spatiality is Nonsense” Missed

Rixstep just pub­lished an arti­cle about why spa­tial file nav­i­ga­tion does­n’t work in today’s file sys­tems: there’s just too many files. And if you assume you’re try­ing to assign a unique loca­tion on screen to each fold­er, you’re absolute­ly right: you’d have to be insane. But nav­i­gat­ing in (for the moment, we’ll use the OS 9 Find­er as our bench­mark, see­ing as OS X’s bro­ken­ness) Ye Olde Find­er, I nev­er have more than a few fold­ers open at once, and when I do, they relate to each oth­er. What are the odds that I’ll be open­ing ~/Documents/Desktop Pic­tures at the same time as /usr/share/lib? Or, for that mat­ter, why in the world would I nav­i­gate the Unix under­pin­ning (/bin, /dev, /var) with Find­er at all? Any manip­u­la­tions I want to per­form on the con­tents of those fold­ers are only avail­able in a shell to begin with, so why both­er nav­i­gat­ing through them spatially?

I sup­pose that very flaw may be part of their argu­ment, but, if you sep­a­rate out nav­i­ga­tion into “Duh, use zsh” for the Unix under­side of OS X, and, “Let me drag and drop my files!” in ~/Documents or ~/Desktop or ~/Pictures, spa­tial nav­i­ga­tion becomes more use­ful. I’ll say it again: spa­tial nav­i­ga­tion makes blind­ing­ly obvi­ous sense in cer­tain cas­es, and just because it does­n’t work every­where does­n’t mean it nev­er works anywhere.

Panic Sans: A Monospaced Font

No thanks to @duncan, @gruber, and oth­er Twit­ter mis­cre­ants, I’ve spent the last 15 min­utes play­ing with Pan­ic Sans, a mod­i­fied ver­sion of DejaVu Sans, as a replace­ment to my Incon­so­la­ta.

Pan­ic Sans fix­es the under­scores and hyphens of DejaVu Sans. I’m going to try it for a day or two and see what I think. It’s avail­able if you tear open the .app pack­age of Coda, their web design application.

Update: this post nev­er real­ly made it up when I first start­ed using Pan­ic Sans. For the most part, I’m a fan of it, though there is one part that’s hold­ing me back from using Tin­ker­Tool to replace the sys­tem mono­spaced font with it: the line-spac­ing on Pan­ic Sans is too small, so lines with descen­ders (like g or p) get over­lapped with the ascen­ders of the fol­low­ing line, cut­ting them off. I had to adjust the ver­ti­cal spac­ing set­ting in Ter­mi­nal to make it work properly.

Technology and Content

A tech­nol­o­gy gen­er­al­iza­tion, stem­ming from a Bri­jit arti­cle sum­ma­ry I just read.

To be use­ful and suc­cess­ful, tech­nol­o­gy has to have access to con­tent. Tech­nol­o­gy with­out con­tent is use­less: what’s an iPod with­out MP3s? A DVD play­er with no discs? A Wii with no games? This is espe­cial­ly impor­tant to new and emerg­ing tech­nol­o­gy: with­out con­tent to feed it and dri­ve both invest­ment and inno­v­a­tive uses, the tech­nol­o­gy will die.

Let This Serve As A Lesson…

I’ve just spent the past two hours today, and two hours on Wednes­day, hunt­ing down a bug that popped up when we migrat­ed PEAR data­base objects from DB to MDB2. Turns out, two sin­gle quote marks: ’ ’ in a select clause is com­plete­ly fine with DB, but MDB2 won’t even parse an SQL state­ment with those in them for pre­pared state­ment placeholders.

Good god. That was painful.

Google Announces OpenSearch Alliance: Helping Search Also-Rans

Fake Steve on Google’s OpenSo­cial project and hand­set alliance:

Com­pa­nies don’t form alliances and con­sor­tia when they’re winning.

…When­ev­er you see com­pa­nies start talk­ing about being “open,” it means they’re get­ting their ass kicked. You think Google will be form­ing an OpenSearch alliance any time soon, to help also-rans in search get a share of the spoils? Me neither.