0xDECAFBAD

Wine is Getting Good: “Anyone else notice lately how good Wine is getting?“

I haven’t checked Wine out in awhile, but I’m a big fan. I was using it to run Radio UserLand under Linux, about 5 years ago. (Wow, has it been that long?) I’ve kind of assumed that Wine would get Good Enough at some point to run almost everything that matters that was written for Win95, Win98, Win2K, and eventually even WinXP. And, games aside, pretty much everything that matters and that anyone bothers with runs on this range of Windows versions. And, seeing as Windows Vista hasn’t exactly set the world afire, I’m betting that Wine might actually have a chance to catch up well enough to be trouble.

Oh yeah, and CrossOver Mac is based on Wine. And CrossOver Mac supports Team Fortress 2 right now, without purchase of a Windows license. Think about that. So, maybe we shouldn’t put games aside. Really, unless Microsoft can somehow patent the crucial APIs exposed by Windows - if they haven’t already - resistance is futile. There are nerds motiviated and annoyed enough to reverse engineer and assimilate. I’m not a lawyer, though: I suppose by the point Wine actually becomes a threat of any sort, nasty things can be sprung.

Michael McCracken: “Once you get a piece of code to the point where you believe it works - it’s passing its tests - go back over it and edit it. That is, go back and edit it for clarity, flow, and style. Just as if it were an essay.“

This is something I just don’t hear enough of - code as writing, code as communication. While I don’t always make time to polish as much as I’d like, I’m always thinking about things like narrative and flow of a file of source code as well as the aesthetic shape and form of the code. Code is fiction, code is poetry.

Compilation free of syntax and parse errors is baseline competence - and sometimes just good luck. Apparent correct operation is not good enough: Moving on before the code has beauty is a poverty. Ugly code kills motivation and comprehension - which attracts bugs and induces a downward spiral in morale.

Unit tests clear the way for courage and bravery; beauty yields living code worth fighting for.

Update: Oh! And since even I missed this link in the original blog post, you can request to be invited. Not sure if that results in an email, either, though.

Update #2: Also, we’re trying out Vanilla forums as an alternative to the delicious mailing list. If you’re invited to the preview, your login should work there too. We might even launch public forums once del 2 is out.

In case anyone’s forgotten: I work on delicious. Since I’ve had a steady stream of friends and acquaintances asking me about how they can get into the invite, I thought I’d mention this:

You may already have been invited!

Unlike most Web 2.0 sites that open up a preview or invite-only thing with a mysterious sign-up form, we can’t fire off an email when we create an account for someone on our preview. It’s thanks to our privacy policy - although we have the email addresses of most everyone we’ve been adding to the preview, we’re not allowed to send anything other than password recovery emails. Weird, I know, but sometimes organizations try to follow things like that.

So, the problem is that we have many more people invited to the preview than have actually started using it. As briefly mentioned in the blog post announcing the preview, you need to check your “Links for You” to see if you’ve been sent a link to the preview.

But, if your invite has possibly been buried in other links from friends - you might just try signing into the preview with the username and password from your existing delicious account! It might just work.

One bit of hackery I did this weekend was to explore the CodeIgniter PHP framework, which is even less magic yet more glued-together out of the box than the Zend Framework. Also, very much less magic than CakePHP and barely in the same universe as Symfony. I think I may have found my sweet spot of simplicity and comprehensible magic.

CodeIgniter has is a PHP 4 / 5 framework that provides a lightweight Model / View / Controller arrangement with URL routing and a useful but minimal constellation of utility classes and helpers. The view templates are in PHP, with just a bit of pre-processing. The controllers are plain-old PHP classes whose methods are called via simple URL routes.

The model layer of CodeIgniter, however, is the weakest part of the trio - which I actually like a lot. It’s not a magical ORM - instead it’s just a simple superclass from which to derive, mostly there to suggest that a separate model is a good thing to write. The DB abstraction classes just make the job of writing SQL statements easier and don’t pretend that there’s no impedance mismatch between objects and relational tables.

The depth of code in CodeIgniter is shallow enough that I felt I was able to get into developing quickly while still taking side trips through the framework methods I was calling and get at least a glance at what being taken care of for me. There’s always danger in magic, but this stuff doesn’t really get out of hand.

To doodle around with an idea, I threw together a layout decorator class to support the sort of nested templates I’m used to from Symfony. Out of the box, CI views are simple, offering facilities for calling sub-templates and “partials” but not directly providing conventions for such. My first swatches of code add a bit of that.

After playing around with layouts, the next thing I did was to use CI sessions and the PHP OpenID Library to develop both an auth producer and consumer. CodeIgniter doesn’t come with a ready-made user auth system, though it does come with its own implementation of sessions using encrypted browser cookies to replace PHP’s native sessions - another feature I really like.

Kind of odd - but on the same toy app I can sign up for local accounts, sign into other sites with OpenID URLs based on local accounts, and sign into my CI app with external OpenID identities.

The wild bit is that I can even sign in locally with one OpenID URL, then use the newly created local profile to get a second OpenID URL to sign in somewhere else. Seems interesting yet possibly dangerous. I mostly just glued the right parts together and got the Auth_OpenID black-box modules working - now that I have an operational OpenID system, I need to study it further.

The next thing I want to do is work more on profile managment, offer some of the OpenID Simple Registration fields, and look into supporting multiple auth IDs linked to the same user profile.

It might also be fun to do something like “catalogue the sites that a user logs into and automatically construct a homepage.” Overall, I’ve got some vague notional ideas to build a personal status collector and maybe some federated social networking and messaging stuff. But, one hack at a time.

Hopefully, I’ll get this code I’ve got so far into better shape and commit to my SVN repository. Sooner than later, otherwise it’ll never get off my dev box.

Scott Burkett, Stop Twittering and Go Solve a Problem: “Aside from aesthetically looking like something a 12 year old cobbled up, the entire tool seems utterly pointless to me.”

For Scott, Twitter probably is pointless at first glance. And that’s because no one’s talking to him. Really, this should be a FAQ for all social media, to be read by all journalists and commentators: Listening in on random strangers’ conversations is more likely to be boring than not, and that’s what you’re doing if you just go to twitter.com and start clicking around. But, to assume because you’re bored that the service as a whole is boring is, again, to be one of the Blind Men with an Elephant.

Twitter becomes immensely interesting when it turns out that you’ve amassed a group of contacts who tend to run in similar circles as you, because even their off-handed remarks and random burps have a decent chance of surfacing something interesting or entertaining. When it’s good, this sets up a nice ambient chatter like sitting in a coffee shop filled with just your kind of people.

This same “I’m bored, so it’s boring” mistake has popped up again and again - applied to blogging, MMORPGs, social bookmarking, MySpace, what have you. Replace “bored/boring” with “confused/confusing” or “annoyed/annoying” and you’ve got my own early reactions to MySpace.

But, really, it’s not addressed to you if there’s a buzz and you’re not groking it. That’s a clue that you’re probably missing something. Granted, people can be insane, but it really is more likely that you’re missing something - even if it’s just that there’s a particular brand of entertaining insanity in progress to which you’re not attuned.

Update (5/18): I am, of course, only an amateur hoarder. Be sure to check out this comment from professional-grade hoarder Jason Scott (in brief): “I agree with you in a tangential way. Outsourcing hoarding is a good idea, but you have to stay on top of the pros and cons of it, just like real outsourcing. … In other words, the risk you take with outsourcing is having your “value metric” not match any other active hoarding entities. The less exotic your tastes, the less of a problem this is. My tastes are exotic; I suspect yours are too.”

Trying out the OPML Editor again, pasting in an entry composed over at my OPML blog.

Copyright and DRM and litigation aside, it’s recently occurred to me as inevitable that digitized media and cultural artifacts will become increasingly more available and easier to obtain. That in itself is not my big realization. My realization is that I’m starting to feel less and less of a compulsion to hoard and collect things.

For example: I used to carefully stick comic books in mylar bags with backing boards - not for resale value, but because I wanted them in good shape to read again years from now. But nowadays, though I still pay for and buy comics, I see that complete archives of things like The Amazing Spider-Man are available as multi-gigabyte torrents. Someday, you won’t be able to buy Spider-Man anymore - but I bet that torrent will still be out there in some form or another.

Another example: In high school, some friends and I semi-obsessively searched for and collected Nine Inch Nails albums and CD singles that fell into a numbered “Halo” sequence. Never mind that this is probably what Trent Reznor was talking about when he recently wrote “Nothing but a consumer rip-off that I’ve been talked into my whole career.”

Living in a small town with crap for record stores, the Halos seemed so rare at the time, and having gaps in the numbered set was frustrating. And imagine when we discovered that there was a “Halo 00″! Of course, now you can find a complete archive of all the Halos in one big torrent - meticulously labeled, bestowed with cover art, and with lots of seeders.

And that’s barely scraping the surface. You can find almost any music, any obscure old scifi TV show - anything that’s got at least one person interested in scanning / converting / compressing the material and seeding a share. And as long as there remains at least one person interested, the material continues on into the future.

Now, mind you, this is all academic - I’m not encouraging you to seek out and download this stuff, and I’m not saying that I’ve downloaded this stuff. That sort of thing could get you into trouble.

Of course, someday, it might not get you into trouble. And, in fact, someday all this might have become accepted tradition - an inherent human right even! Right now, it’s the norm for quite a number of people. What makes it amazing and inevitable to me is that the trend seems to be toward more and more of this, not less. And eventually, even the self-selecting nerd filters will fall away, everyone will play, and the content will stop skewing toward the techies.

Someone, somewhere, with some old recording or tome or whatever will get ambitious enough some weekend to digitize the thing and share it - and 10 years later, it’ll still be floating around because throughout that time at least one person at any given time felt it was important enough to keep going. Really, all it takes is one person - and odds are, there’ll always be at least one person interested who has the skill and inclination to relay into the future.

And, really, all of this is just yet another corollary of Mark Pilgrim’s “the only long-term effect of copy protection is to ensure that those who defeat it are immortalized.”

So, yeah, I’m starting to feel optimistic enough to think maybe I don’t need to hoard and collect the things that I feel are definitive of me. That maybe it’s okay to expect, down the road, to be able to search for and find that stuff again with ease and lighten up my own personal clutter. Not quite there yet, and I don’t trust it enough to shred my comics and CDs, but I think a turning point has passed.