Rhonabwy

Last night I attended a Google Tech Talk in Seattle – they were doing an overview of Google Website Optimizer. For those who haven’t heard of this, it’s a feature buried farther than most, under the AdWords application functionality. But what it does is pretty damn cool – and available for use for free…

The whole idea is enabling A/B (or multi-variate) testing of websites to see which is “better”. I’m pretty sure it’s bolted into the AdWords because it was originally devised to help folks who were spending $$ with AdWords to get folks to their site. One of the key components of this kind of testing is determining “which is better” – which means you need a goal or target action to occur to determine it. In this case, Google’s taken advantage of their own Analytics engine and uses the conversion as that goal measure.

A super-brief overview of what this thing does – you set up two or more variations of a page that you want to test. The “goal” of the test (how you know it worked) is determined by folks viewing your site getting to a goal page – maybe a thank you for signing up page or the like. Then you start plugging in javascript tidbits that do the tracking and control what the user will see by some nifty javascript document.location.replace() method calls. There is a lot of failsafe setup – if javascript isn’t available or the like – and attention paid to make sure that the testing doesn’t break your site, and that the “choose A vs B (vs any others)” is kept as light and quick as possible. Since it was a Google tech talk, I’d expect that you could find the video of this talk in another few days or so, and if you’re interested in this topic, watch it.
The cool thing that I learned last night (or one of them, the speakers were excellent – part of the engineering team that is currently developing out this tool) is that you don’t need to spend $$ to make use of the tool. You’ve got to “sign up” for AdWords, but that doesn’t mean putting in any billing information or even setting up an adwords campaign. The result is that pretty much anyone can now take advantage of the testing mechanisms – something that Amazon and other “big guys” have been doing for ages. Normally its a hell of a lot of infrastructure and thought into enabling that kind of setup, and even more effort on the statistical side of the analysis.

Of course this kind of testing requires traffic. Google has their reports set up so that it’s clear what’s happening with the results, so depending on how many variations and such you’re providing, you’ll get details on what’s doing better or worse, and with what confidence interval. They suggested running any test for at least a week to take out weekly variations, but you can certainly run the tests longer. If you have a low traffic site, you can reduce the number of variations to get meaningful results a little quicker.

Yeah!!! I’ve got IMAP enabled on my Gmail account. I’m sure I’ll continue to use the web interface during the day, but having more general access to it (and through Apple’s Mail.app) is going to be great!

It’s been an interesting day on the laptop. I picked up Leopard this morning, and after having backed up my laptop last night, I did the install.

I thought about an upgrade install, but I decided against it – maybe good, maybe bad. I stashed my entire home directory onto a USB drive (turns out that was a damn good idea), sync’ed up my data with .Mac, and did a complete wipe/install.

The only significant “Oh shit!” in the process was that .Mac didn’t sync *back* all the stuff I thought it would. I asked it to pull everything from .Mac down to replace bookmarks in Safari, the Address Book entries, Mail accounts, keychains, etc. What I got was Mail accounts and my keychain. Safari ended up with nothing for bookmarks – just wiped it all out, and the Address Book was completely empty. Fortunately between a dump from my desktop computer of those items, and my iPhone, I was able to get them all back into sync without issue. Apple needs some serious love against the .Mac stack as far as I can tell – a 2 year old API that still hasn’t gone out of beta and the periodic issues like I’m seeing – all without errors… it’s a definite weak spot.

So a word to the wise: if you’re following the “clean install” path, export your bookmarks and vcards and back those up explicitly for importing when it’s all back in operation.

I used the Application Support folder in my home directories “Library” to populate out all the goodies for Camino, Firefox, and NetNewsWire. For most of the other applications, I just copied in the Apps and re-registered them. That took care of Acorn, VoodooPad, TextMate, Coda, OmniGraffle, and (ya gotta have it!) Escape Velocity Nova. The other apps (iWork ‘08, the iLife stuff, etc) I pulled in from CD or DVD as they were available, and while that took a while to get done, it all worked very smoothly.

The new developer tools got installed (Of course!!), and I really like what Apple has done when it first kicked into gear – there was a welcome page with a plethora of links, notes, and topics – sort of a “hey, check it out!” advertisement if you will (except WAY more useful than most adverts). It was a nice touch. I didn’t poke around in there too much, but I had to fire everything up at least once to see what popped out. Lots of folks are hearing that Xray was renamed “Instruments” – fine, whatever. The only phrase that really matters to me about it is the critical “It’s DTrace under the hood!”, which hasn’t changed. There’s a world to dig into there, and I’m looking forward to really pushing that about. And the ADC has up a few new articles talking about the advancements in Objective-C 2.0 that are available in Leopard, as well as a nice overview of the graphics updates (which are huge). I spotted them in the XCode “Hi there” window first, but James also tossed up a link on his blog – seeing as how he probably wrote most of that… When it comes down to development tools, of course I had to check:

[akoop:~] heckj$ python
Python 2.5.1 (r251:54863, Oct  5 2007, 21:08:09)
[GCC 4.0.1 (Apple Inc. build 5465)] on darwin
Type "help", "copyright", "credits" or "license" for more information.
>>>

and subversion is built right in!

I’m quickly migrating my muscle memory habits into using Spaces (virtual desktops) instead of my older habit of “hide the window for now”. I’ve got a nice setup and have been moving applications between windows to split everything up. So far, it is working extremely well.

There are a lot of updates associated with Applications to work with Leopard. I’ve spent most of the day solidly slamming my DSL line to get them all down and in place. It took quite a while to synchronize out my IMAP accounts as well – I guess I’ve just got a lot of crap stored up in there.

The final tidbit for my “Upgrading to Leopard” post is that this is the first Mac OS release where things feel a bit slower than before. I’ve been getting damn spoiled by previous releases, and some of it may be that I’m trying to fire up enough applications to eat 64GB of RAM… but seriously, sometimes there are these weird sort of “pauses” in the finder, mail, or whatever that I haven’t quite figured out. I don’t know if I’m overdriving my laptop right now (maybe it’s reindexing, caching, or whatever) – I hope so, but its something I’m definitely going to keep an eye on for a few days to see how it rolls. Not that I’m going back – nope, just deal with it. Shoot – I spend most of my work day waiting for Outlook on a MUCH faster machine than this. I can deal with some slight UI perturbations and quirks. For the most part, the whole OS has a much higher level of polish. Lots of little tweaks and touches – you can really tell that “how it looked” was a critical part of the whole release this time – from the OS to small utility applications. It’s great detail.

I spotted this tool at first from jonathan saggau’s, but I didn’t know what it was. mdmunoz pointed me to the tool – GrandPerspective. Man – this is a tool I’ve been looking for.

I’ve seen the grid-maps used before – mostly market related graphs. It really answers the question of “who’s largest” in a huge collection, giving a great sense of scale and size in a hierarchical structure. Which is perfect for that nagging question of “Where the hell did my disk space go?” Okay – so an 80GB drive just isn’t enough anymore with media. That’s just the facts now.

I freed up a good double-handful of gigabytes already this evening with the tool, just checking around in my home directory. This critter is going to be a fairly constant companion now… at least until I have more disk on my computer than I easily need at any time, which I don’t think will be any time soon.

Leopard’s new “read me that…” voice (the built-in voice synthesis) is just amazing. I spotted a post talking about it at Jonathan Saggau’s blog.

Anyone know what Jonathan was using to get that relative size map of file sizes? I’d love to use something like that…

Google has announced IMAP access to GMail accounts, but it’s apparently not enabled for everyone. I guess they might be doing it in sections or something, but when I followed the directions for setting it up, the “Enable IMAP” portion of those directions doesn’t show up.

I noticed a few other folks reporting the same on my Twitter-feed, so maybe it’s just a “wait for it” feature…

A few weeks ago, I took some surveys and received some Amazon gift certificates in return for my semi-random murmurs and reasonings. I did what any other geek might do – I bought stuff from Amazon with them – geek books. (Ok, not just geek books… just mostly geek books).

One of those geek books is Making Things Talk, by Tom Igoe. I have another book that Tom had a hand in writing: Physical Computing (He maintains a web site on the topic of physical computing as well).

If you’re interested in hooking up something physical to something virtual, get the book. In fact, get both of ‘em. Making Things Talk walks you through all the steps of setting up basic electronic circuits to working with a wireless mesh network to communicate data about. And that’s what it is really focused on (just the like title says…) – communicating. Getting data in and out of other places.

It’s always semi-bothered me that computers really only talk to other computers, or maybe other “computer things”. All the coding that we’re doing are these simple to immensely complex structures – that don’t physically exist. The architecture and construction metaphors are reasonably appropriate and abound – but sometimes there’s not a whole hell of a lot physical to show for extended labors. I still don’t think my grandmother gets what I do… Thank god for the advent of the web (and that she uses it), or I’d never be able to halfway explain it.

But the Igoe books make that link. Making things talk is set up in a very “make” style of book – fairly explicit projects that walk you through the basics of using some “building block” style electronics and chips to make your stuff work. They’ll reference using micro-controllers like Arduino. That’s talking the Electronic Circuits I class I learned in college and pumping it up several levels to basically have something that is damn near plug-and-play for the electrical circuits world. (I will, forever, have the habit of ducking when plugging in my own electrical circuits from that class). The book doesn’t really dwell heavily on the micro-processors – it’s just a means to an end, and that end is a number of different communication setups from serial lines to wireless mesh networking with XBee radios.

The final chapter does a light-weight overview of identifying and finding things – talking about wiring up a GPS chip or reading RFID tags. The book leaves you just wanting to fiddle more and see what’s next in the lego-block world of electronics.

Where the book didn’t talk was “making things happen” – invoking motion or action in the physical world from the virtual. You’ve got to image that once you grok sending signals one way, it’s pretty possible to send them the other way too. That’s really where the other book (Physical Computing) ties in. It talks about a tad lower level – which you kind of need to hit in electronics. There’s not (yet?) plugin and play stepper motors or actuators that I’m aware of. But the Physical Computing book does a damn good job of laying out how and why, including specifics, of making exactly that kind of thing happen. It seems like a re-write of that kind of book – “make” style – would be an excellent successor to Making Things Talk: especially if there’s plug-and-play actuators available that I’m not aware of.

I had a light introduction to the Arduino setup at OSCON this past year – I attended Jon Oxer’s talk/tutorial Hardware / Software Hacking: Joining the Real and the Virtual. I thought we were going to get some Arduino’s ourselves to play with, but it looks like that never really quite panned out… But anyway – it looks like Jon did a tech talk at Google that’s available online if you’re interested in seeing what he put together and how he was doing it.

If you’re interested in playing with an Arduino board – the current boards appear to be running $35 (Maker Store, SparkFun)

The story is (briefly, I suspect) available on Apple’s “hot news” page. (Although I caught the news of it on Twitter from Buzz, Gruber and Dan Pasco).

The text of the announcement reads like we’ll be seeing some sort of signed binaries mechanism for the executables. That’s something that Apple had been starting to push 18 months ago at WWDC 2006, so it’s really no surprise. It will be interesting to see what it takes to sign and deploy code within the iPhone or iPod touch – we’ll apparently be able to deploy to both with this SDK.
I think a large open question will be distribution mechanisms for Applications. AppTapp Installer made the hacks really easy to download and install. I hope they’ll be back with the official SDK and a set of common technologies to make getting applications super-easy.

I’m excited to see the SDK coming available. Now I won’t feel like I’m potentially wasting time developing apps down that path. Of course, I don’t really have a clue of what I’m going to develop – but the potential is very exciting.

Here’s an odd one I ran into tonight.

I had been using curl to debug some http REST code I’ve been working on, and recently it stopped working. I finally tracked it down to a version of curl that I installed as a dependency when I was playing with learning about git. Turns out, the macports version of curl that I installed (@7.17.0_0+darwin_8) simply didn’t want to recognize http://localhost:8080/ as valid, even though it was.

I was running a simple python WSGI server on that port. At first I thought maybe I was binding it funny – but heading to that address with Safari worked fine, I could telnet to localhost port 8080 and get a connect, and lsof was showing me that it was bound to a normal TCP socket. I’m still not sure what exactly was the issue, but when I manually invoked the system’s built-in curl (/usr/bin/curl instead of /opt/local/bin/curl), it finally worked.

To fully resolve this, I uninstalled git-core and curl from MacPorts, but that still seems really strange. I can only guess there’s some weird down-lower-level error that cropped up, but nothing was making it to the surface of the code. It just constantly refused to believe that localhost had port 8080 open. Worked fine for remote hosts.

Tonight we finished Halo 3. My brother in law (Dan) and I finished it up this evening, playing “Heroic”. Of course, now we have to take it down on Legendary…