For whatever reason (I smell lawyers), DRM-free purchases (“iTunes Plus”, cheesy har-har) are not enabled by default in iTunes 7.2. In fact, they’re positively buried:
hat tip: SteveX Compiled
addendum: Looks like they placed iTunes Plus links more prominently in the store interface, so you no longer have to dig for them. Thanks, Apple.
If someone sends you an “HTML” mail from Outlook, even Tidy will run away screaming unless you strip out some of the gunk manually before trying to fix it.
If it’s Quoted-Printable, you have a bit more work to do first [maybe this (web service) or this (sed script).], though you probably have even more work to do if the original document used a non-Western encoding. Not tested.
sed -e "s/\<o\:p\>/\<p\>/g" | sed -e "s/\<\/o\:p\>/\<\/p\>/g" | /usr/local/bin/tidy -c
broken into two
sed invocations for
readability’s (hah!) sake…
Of course, it’s all very brute-force, but usually good enough for government work.
A couple of weeks ago, my third el-cheapo inkjet in as many years turned into a pile of useless plastic crap.
I’m not #@%ing around anymore.
I bought it at a computer show for approximately what 3 cartridges would have cost me for my old inkjet. It sits in a back room, where its bulk and noise don’t bother me, produces crisp text and solid color graphics, fast, and has a network print server and Adobe Postscript built in. It uses more electricity than I’d like, so it’ll be turned off most of the time.
(any modern OS) Instiki: Cool Wiki engine. Yet another stop along the road in my quest for the perfect personal task manager / to do list / project space. The notable things about this one: nicely self-contained (runs on its own included webserver), supports Markdown and Textile, runs on Ruby (Ruby built into OS X is too old, grab the prebuilt version that embeds the Ruby runtime or install a current Ruby dist via Darwinports or Fink).
(web) Memigo: A headline clipping service that learns what you like, can export feeds.
(mac) iEatBrainz: Uses the MusicBrains metadata engine to help you clean up your iTunes library’s metadata. Great for doing something about those “mystery tracks” that you collected from who knows where.
Oh é, ya sexy beast.
I never consciously planned to do it, but two months after trying Markdown for the first time, I find I’m using it for substanially all of my web-based writing. I’d played around with other human-centered web markup formats before (WikiText, Textile, etc.), but with every one I always found myself having to unlearn too much HTML markup to feel efficient. Markdown succeeds, for me, in a few crucial ways where the others always failed.
I’m not storing my blog text in Markdown’s syntax. I’m writing in Markdown, then running the finished text through the parser, rendering it to HTML and filtering it through Tidy. Still, I find I’m producing substantially stupid-free (i.e. no open or improperly nested tags) text faster than I ever did writing my posts in raw HTML.
Rather than using a physical TV listings guide or viewing TV listings on a website, I prefer to use a specialized application to see what’s on. For quite a while, I’ve been using Tom Talbott’s MyTelly, which was always pretty nice. It’s written in Java, but it’s acceptably responsive even on my G3. The only drawback was that it retrieved its listings by screen scraping the listings at the Zap2it site. It did a good job of it, but of course it was handicapped by the inherent limitations of that technique — it had to grab entire web pages and labouriously filter and extract the programming from them.
Well, Zap2it has deployed a SOAP interface, and the newest MyTelly release talks directly to it. Wow. Fetching 8 days worth of programming data with the old screen-scraping MyTelly engine took 10-15 minutes over a cable modem connection. The new release fetches the same data in about 45-60 seconds. I suspect the amount of bandwidth consumed is a couple of orders of magnitude less than with the old version. Sweet.
...reasons why everyone who works with this stuff eventually ends up bald.
Unfortunately, the old escape/unescape two-step goes haywire when you view the post's title from the sidebar on another page of his weblog:
When viewed in an aggregator, an entirely different bit of the title goes awry:
I think I'm going to take up gardening, or ASCII flat files.
A long time ago I realized that, given half a chance, I tend to rattle on a bit. (at this point, you sit back, say, "Duh, you're a freakin' blogger, numnuts...", and continue to sip your coffee) I made a conscious decision to strive for a bit of (false?) economy in my writing, a nip here, a tuck there, and eventually I'm only boring the world 50% as much as before. The danger with doing this is that sometimes you forget the people reading aren't sitting inside your brain, participating in the editing process, so you omit or elide things that you really should have made a bit more explicit.
My last entry was a bit of a vent in which I talked about the hazards of interface customization. The first line was:
The mantra of moderately-to-very experienced computer users is "customization customization customization"which Sven quite rightly calls me out on. Looking back at my post I see something that was very clear in my mind as I was writing it but entirely absent from the actual post: that I personally am not a fan of completely customizable interfaces -- all too often, a completely customizable interface is a cop-out for developers failing to deliver a usable default interface at all. This was completely clear in my head, but of course it never actually shows up in the actual entry. I'm not an interface tweaker at all -- I'm much more likely to want to mod an application's underlying functionality via wild-eyed patching and plugins than I am to ever want to muck with the interface. More clearly stated, what you often run into in places like Slashdot, Ars Technica, Mozillazine, OSNews, and on technical folks weblogs are people clamoring for more tweakable interfaces. Throwing more toolbars and widgets at these people (who, quite sadly, are quite often the same people responsible for writing project reviews, which only fuels the vicious cycle) may shut them up (temporarily), but it thoroughly screws the pooch for the novice user.
As unfashionable as it is to suggest a public sector solution to a problem that is (allegedly) being handled by the private sector, I think that personal certificates, at least, are something that governments, particularly at the state or province level, are well-positioned to provide. There's already a level of institutional trust when it comes to these agencies, particularly drivers' license bureaus, when it comes to identity verification. At least in the USA, a state-issued driver's license is accepted as proof of individual identity virtually everywhere, as is a federally issued passport. Since the state and federal governments already have identity verification mechanisms (via birth records, etc.) in place, the most obnoxious part of trying to get a certificate validated (all the various dicking around with notaries and the like) can be avoided. Wouldn't it be great if you could get a CDROM with a state-signed certificate (in the various necessary formats at the same time you got your drivers' license or passport? You've already done all the legwork of providing identity documentation to these agencies. For businesses, processes like filing formal incorporation papers or sales tax licenses could serve a similar purpose. Why not leverage this? It's too late for this to happen, though. There are already entrenched private firms with a business model to protect, and, as we've seen with the record companies, an industry with even a demonstrably broken business model will fight like a cornered animal to protect its turf.
It's well known that my current favorite text editor is
Hydra SubEthaEdit. I can't use it on my office PC, but I have found something to use there that I like nearly as much. It's called jEdit, and it runs as a Java desktop application. I know that sets off alarm bells in some people's heads, but it's really quite nice. It's very responsive (even as a non-native application), and the interface isn't particularly jarring (I'm notoriously not picky about interface consistency when I'm in Windows, I'll admit.) Since it's Java, it runs in most every modern OS. I've run it in OSX, Windows, and I'm going to try it on FreeBSD later. It does everything you'd expect from a modern programmer's editor, and benefits from one of the coolest plugin architectures I've seen in any application. The plugin manager is fully web-integrated: it connects to a server and presents a fully up-to-date list of modules, complete with descriptions, that you can download and install with a single click. It handles dependencies automatically, too. The only negative is that plugins require a restart before they become active, which is subobptimal, but hardly a deal-breaker.
Unconnected observation -- the Delgados are really, really good. They've been around a while, too -- I wonder how I missed them.
I originally tried posting this as a comment on another blog, but
it didn't go through for some reason. I'll post it here because a full
clipboard is A Terrible Thing To Waste...
Sven has a problem with hard drive performance on his Powerbook. Well, this doesn't directly address his issue, but I found this article (via Slashdot) to be very interesting -- a reminder that cheaper isn't always better, sometimes it's just... cheaper. Sigh. For many years Macs shipped with SCSI drives by default, but ironically enough (in an Alanis sense) they did it when their machines were shipping a single-tasking largely unthreaded operating system that didn't exercise the storage susbsystem's potential. Now they ship modern OS with server tasks and real virtual memory and they ship all their machines (except the custom configured ones) with good-old brain dead (but cheap cheap cheap!) polled-I/O IDE drives. Why? Because the beige box vendors realized they could shave dollars off their bottom lines by bundling in cheaper (and in this case clearly less performant) hardware. When you have to compete on price with "the cheapest possible hardware that will work", you don't have a lot of options.
After playing around with VoodooPad, which I quite like, I decided I'd register it and start using it as my personal organizer, so to speak. Then I thought a bit more and realized that, no matter how much I liked the app, that wasn't going to work out very well. I don't have a PowerBook (sob), and I'm primarily limited to Windows machines at work, and if I end up going mobile with a Hiptop or some other PDA then it's inaccessible there, too. Then I realized -- I have a webserver!
I looked at Alex King's Tasks, which looks really good. But really, it's not really the sort of thing I'd really use. I need something more freeform -- I don't really need all the alarms and "project 50% done" indicators and all that. What I really need is a virtual scratchpad where I can record semi-random stuff:
and a million other of the trivial details that fill my life. I'd been using VoodooPad for these sort of things, but, as mentioned above, it doesn't travel with me so I needed something web based. I've grown fairly comfortable with Wiki -style editing, and I definitely love being able to create new pages basically "at the flick of the wrist" (by joining wiki-words), so I started to think: Why not just configure some proper WikiWikiWeb software? I already have AwkiAwki installed to serve my FAQ pages, but it's not exactly feature-ful. I tried PurpleWiki as well, but had some problems setting it up (adding Perl modules on OS X usually involves invoking dark forces.) MoinMoin is powerful enough to have served the Atom project, and it was dead simple to set up at work (praise Jebus for the FreeBSD ports system), where I'm evaluating it as a possible internal tech-support mechanism, so I decided to try it here. Frankly, the installation was a pain in the ass (mostly my fault), but I got it working.
Anyway, I get full text searching and an index and stuff "for free." I can see myself using it as an idea scratchpad for long blog entries, for the book about absolutely nothing I may write someday, and whatever else.
I've restricted it by IP address for now (Google, world, and dog don't need my grocery list), so I can reach it from home, the office, and I figure any other place I might need to have access from in the future is just a SSH session away.
You don't have to tell me that normal people don't do this. Well, duh... Proudly without a life since at least 1985...
Looks like Ford is going to be doing something with Linux desktops (via the Reg and /.). I am actually drooling (drooling, I say!) when I think about the all the Ford-dependent Windows-based automotive suppliers in the Detroit area who are going to need some *nix-savvy consulting as this happens. cha-ching!
Yay, a new toy.
Geekier than normal content follows. Mom, you should skip to the next entry.
Looks like a number of corporations have chipped in to ensure a secure base of funding for projects under the Mozilla banner, under the leadership of Mitch Kapor. This is great news, since there has been plenty of doubt about AOL's commitment to Mozilla ever since they crawled under the desk.
edit: I wrote the above paragraph before I read that "AOL has cut or will cut the remaining team working on Mozilla in a mass firing and are dismantling what was left of Netscape". Ugh. My heart goes out to those folks, who worked their butts off for the usual corporate reward.
Mike Pinkerton: "Tonight I pour one for myself, and one for my homies."
Chuq Von Rospach has an excellent suggestion for aggregator developers here. A couple of aggregator developers applaud the suggestion in his comments, so maybe we'll start to see these sorts of smarts in a software generation or so.
I collect RSS aggregators like some folks collect stamps. I (obviously) don't do this for practical reasons. Realistically, no one really needs an aggregator (though if you're an avid reader of weblogs they're pretty much essential for keeping things manageable.) Anyway, I've got a ton of them.
An extended ramble on tools, aggregation, and people who you should probably be tired of by now.
There’s got to be more to life than compile-and-go.