The Dichotomous Trekkie 2.0

A simple screenshot for now:

Taken on Liz this evening.

Since I've not managed to post anything in about three weeks, I'll start with another minor computer type post -- some browser benchmarks. They weren't conducted the most rigorously, but it definitely gets the idea across. Test machine was Athena, the "Weighted" benchmark is Sunspider/V8 (so it approximately squares the mean of the difference), and you can download Chrome here.

OTOOS is a really good book, and I must say, Chapter 3 in particular has some great imagery. But, to the task at hand! Its well-known that various antiviruses (antivirii?) have different memory footprints, different scan speeds, different effectivness, etc ... but which is the best? Well, faithful readers, I present to you the somewhat perplexing graph below.

Here, the red line is under Windows Live OneCare, the blue line with no antivirus installed, the yellow line with Nod32, and the green line with VIPRE antivirus. That's right -- it ran *faster* under VIPRE than with no antivirus. My testing methodology was to do a reboot, wait five minutes, and run two benchmarks and use the best one. The benchmark was run with PerformanceTest 6.1 64-bit under High priority, changed from Windows Task Manager. I'm not sure why the NoAV score was so low -- I'm almost certain I remembered to change the priority, but that might be the culprit. In any case, the three antivirus lines speak for themselves, though -- VIPRE wins by a long shot.

So, Google Chrome is out, and I must say, its a very attractive browser.

It also doubles FF3's Sunspider javascript benchmark speed, nearly triples Opera 9.52's, and nearly quadruples IE8's. Very slim browser with an installation routine that is under 500 kb and the whole install takes well under a minute, and installs cleanly -- doesn't ask for admin privledges.

On a technical basis, it has uses a tweaked WebKit renderer (after Safari) but doesn't do Safari's horrible proprietary font-smoothing; copies IE8's tab process isolation (thus, if one tab crashes only that tab crashes, not the browser); has a startup page very similar to Opera's speed dial (but lacking Opera's wand feature and speed-dial shortcuts); largely matches Vista's UI conventions in Vista (big plus); renders SVGs like Opera (see if this will render in your browser, and see what happens when you zoom).

Gotta say, I really like it. We'll see if it becomes browser of choice over Opera for speed, or if Opera will end up winning with its extra keyboard and mouse shortcuts.

So, I actually will post the interesting sci-tech tidbit in a bit (I've been behind in reading papers this week -- the camp was a blast, but it ate up time like no tomorrow), but I thought I'd post a breifish update on why developers hate IE. I think it is best encapsulated in these two photos:

IE 8 rendering http://beta.revealedsingularity.net

Opera on the same

As is evident, Opera and Firefox and Safari will all render it one way, and IE another. To be perfectly fair, the utter mutilation in the image above is IE8; IE7 just does a hack job on places that images are supposed to blend into backgrounds. Unfortunately, this is something I'll have to look up to fix.

I'll try to post the promised science tidbit sometime in the next few days, along with a rundown of Adapt or Die. Till then, by popular demand from the students, here's a link to the Evolution Game (best played with more than four players).

Interesting. It would appear that inline CSS editing may not be doable after the fact in LifeType, and that re-editing the source code in Opera 9.52 is buggy. Curious ... very curious ... C'est la vie, still a damn sight better than Wordpress.  I've also tried importing the blog into Facebook ... let's see if I'll regret this.  If anyone comments, please try to comment on the original blog -- I rather rarely check FB anymore!

Wow, I didn't think I hadn't updated in a month. I've not even updated LJ much. I suppose I've actually been a bit busy. The macro site is getting along quite nicely. I'm a fair way into it, but certainly not done. Much of my work has been directed at building the collapsing, (soon to be) pop-outable phylogenetic tree and navigation, as well as the live-preveiwing contribution page. With this site, I've gotten a lot more up-to-date in my knowledge of CSS and its failing across browsers, particularly in implementation of psuedo-elements, floats, and CSS3.

Meanwhile, my work for LHS has been coming to a head. I start teaching Adapt or Die next week, which has been intense to gather all this information, make sure I'm up-to-date, and marshal it down into what a 4^th or 6^th grader can understand. Of course, much comes in the way of actual live critters, and some in the way of games, etc. Here, intrepid reader, I give you a tiny preview: an evolution game that will, over the course of six disasters, give you a feel for how evolution works. Some weird things get put in, some "good" things get taken out, and all in all, you get something that works better than its competition but is by no means perfect. Rob, Alyssa, and I tested this, and it seems that about 3/4 of the population has gone extinct every 3 rounds or so, which can lead to population bottlenecking (sound familiar?) if too small a playgroup is used. I reccommend at least six. Also along the LHS lines, I provide you with a graphical tree of life!

Its admittedly weak in places, and doesn't show all kinds of interesting diversity points I feel its most notable in Carnivora, Afrotheria, Metatheria, and Perissodactyla, but I'm a tetrapod man. I know it lacks in Dinosauria, but only so much one can put in with images. Also lacking is inverts (using the paraphyletic definition here), rather obscenely actually. Still, impressive looking.  Catch: most CGI images are screens from BBC's "Walking with" series.  I think this is still fair use, but just a noteworthy caveat.  The other images are either my own or Wiki images from the appropriate articles, with a slew of CC variants.

Now, I will try to update at least once a week, and include an intersting science or technology tidbit each time. So, let's go ahead and start high-rolling. Science has a rather interesting tidbit about snake fangs this week, in which analysis shows that that advanced (Caenophidian) snakes have their various fang morphologies derived from a rear-fanged ancestor. The nifty part? Selective expression of the "sonic hedgehog" gene (shh) relating to the anterior development of the maxilla meant the fangs moved effectivley forward, giving the "front-fanged" appearance of Viperids & Elapids, but these front fangs are actually the rear fangs -- and this single change was utilized twice! Cool stuff.

Now, lets see if I actually follow through with these updates ...

Having just returned from a trip to Hawai'i, I would very strongly recommend it to anyone who is thinking about going -- O'ahu in particular is a beautiful island, and the ocean is beautiful. I've uploaded many photos to my Flickr page, should anyone be curious.

Tomorrow I return to Berkeley via Amtrak. Thankfully, there are power cords -- I fully intend on working more on sauropods and doing some course outlining for the LHS Adapt or Die camp. Later this month, I will talk with Jim, Kevin, and Gibor about recommendations; and, around August, if things go well, I will talk to Erica about recommendations as well.

What will be irritating about the next few days is lack of internet -- its not getting activated until the 12^th, which, needless to say, is painfully long from now. I suspect much of my days will be spent setting up the new place, but if I have down time, I will see about working on the stellar models from the UGAstro room in Campbell. I have a few colortable ideas, and in addition I will probably rewrite some of the interface when I get it done in my head. No use having so many programs to call and save files when you can do it better internal to one program or by passage of variables. Relying on saving of structures almost obviates their use.

It really is strking how internet dependant everything is nowadays -- my work with Kepler is strictly network based, and I rely on the internet to keep copies and synchronized versions of my biomechanics work across multiple computers. No internet will be hard, indeed.

As I've been working on the sauropod paper (aside from feeling a bit neglectful of Kepler), I've definitely felt the wish to "feature creep". For those unfamiliar with the term, its one I borrowed from computer program development; its the wish to include ever more "new" stuff into a program that you're writing, often to the detriment of timeliness. Aside from finding ever more species to want to add to my analysis, I find that I am having a difficult time finding the line where I say "enough -- I've explained this to my satisfaction." There are some legitimate holes that are left over from its "essay" form, such as comparisons on different ways of getting the taper constant, ways of extrapolating neck length and accuracy, and even I can't remember what I did to find the trunk length.

I also was a bit lax about my citing in the initial essay (oh, the usefullness of BibTeX -- I wish I knew it sooner. Kudos to Kit for helping me out there), I also really want it to look nice.

So, with its current state of revisions, it is being looked at by Sara W, Sarah W, Sarah K, and Andrew. Hopefully, it will be ready for submission by the end of June, July at the latest. Sarah (K) brought up a good point though -- where to submit it? Its hardly Nature or Science quality, but that still leaves:

1) Journal of Paleontology
2) Paleontology
3) Historical Biology

As the prime candidates at the moment. Hm. I'm leaning towards Historical Biology, but we'll see.

Hopefully this will be out of my system by the time Hawaii is over -- then I can do some solid astronomy. Helps that that's the one I get paid for, too.

So, as my housemate Kit very properly asked: why am I starting this blog?

 Its true!  I have (two) defunct Blogger addresses, and I regularly stick to my Livejournal -- so why this one now?

 Well, this is an OSS expandable platform with by own domain name, which is nice.  But, more to the point, I think this is where my sciency posts will live.  Posts with more quality input than discussion of girls, homework, the latest movie ... if this blog starts to become substantial, I may end up migrating over from LJ, but we shall see about that.  This is to be used for something else.

 And, well, I need to work out the kinks still.  I am still not sure if timestamps are working right, and I'm wrestling with remote Atom posting.

So we'll see.