According to this post (courtesy of digg.com), Dell has initially selected 3 models of desktop PCs to be configurable with Linux. While some might complain that this is a small number to begin with, I think it’s simply a sign that they’re testing the waters. These machines span a decent portion of Dells range of desktop offerings, and those who were seeking desktops (and not laptops) should be just fine. Linux on laptops isn’t perfect anyways. Heck, even the systems shipped pre-installed with Linux from System 76 that are laptops have issues in this area (See Darter comments here).Now, all of this is well and good, but I’m going to make another prediction regarding what’s going to happen next Thursday after the announcement of these systems. I think they’ll be pretty well received on the whole, and dell will probably sell at least as many machines as they’re estimating (~1% of all sales, 20,000 units/yr). My prediction is that, at least initially, these machines will not be as cheap as Windows Vista Home Basic equipped machines of the same hardware configuration, and that there will be an initial uproar over this. My reasoning is as follows: Dell bundles lots of crap with their machines, and gets discounted versions of Windows from Microsoft. At the end of the day, your basic copy of Windows and maybe some of the hardware is likely subsidized by the pre-installed junkware trial software that comes with each machine.Dell, I’m sure, has either contracted with Canonical or set up in-house staff to support this new software configuration, and that’s going to cost money. That cost will be passed along to the customer. There’s nothing wrong with this, but I expect that the cheapest way to get Linux on a Dell (unsupported) will be to buy a Vista Home Basic PC and then wipe and install Windows.In both the long and short term, it will be interesting to follow the support profile provided for these boxes. It should be very interesting to see what portion of the purchasers are enthusiasts, and what portion buy one so that they can have a stable, virus-free PC for accessing the web. Since most Linux support, so far, has been through a sales and support staff with more training than the average Windows support personnel, I wonder where Dell’s Linux support folks will fall. Will they know Linux, or use a flowchart.Only time will tell.
D5 2F 00 B7 B4 B7 C1 92 ED 7A E8 E0 15 F9 E2 C1
Generated courtesy of this page.If, by chance, you don’t follow check the background links on the linked page. This is kinda the crux of the whole thing.
Tagged with: web
Posted in Uncategorized
According to this post up at TUAW (and most of the remainder of the online tech community), Apple will be offering up DRM-free EMI tracks beginning in May. Upgrades will be offered for previously purchased tracks, and newly purchased individual tracks will be priced at $1.29 (full albums in DRM-free state remain $9.99). DRM laden tracks will still be available for the same price for those concerned that Apple might be pulling a little bait and switch. As an added bonus the tracks will be at “twice the quality” of DRM loaded tracks. Hopefully that means encoding at twice the bitrate or around 256kbits AAC which is nothing to sneeze at.I guess that means that Jobs wasn’t full of it after all when he wrote his piece that appeared on the Apple website. I wonder if this agenda was there from the get-go, or whether this has evolved now that the music industry is entangled in Apple’s web.
I’ve not written in a bit about the trials and tribulations of getting an electrophys rig up and running. As an update, I believe that I’ve located the major problem in the system. The ringers solution being used was made by a previous grad student, and had been sitting in the fridge for some time with parafilm over it. Well, it turns out the parafilm wasn’t sealing things in so well, and the solution was supersaturated to the point of precipitating solid. I didn’t realize this was way off until looking up ringers solutions and finding that not not that much in the way of salts go into the stock, and therefore precipitation shouldn’t be happening. This was making the electrode pulling almost impossible (I’m guessing the ones I pulled before will be rediculously high impedance with the new solution), and led me to try innumerable solutions. While some might consider all that time wasted, I really don’t look at it that way.In graduate school, I believe, one needs to develop a bit of an immunity towards repeated obstacles. Keeping focused on finding solutions, and realizing that much time will be spent getting past obstacles is the only way to survive. If you don’t expect all that, or come to expect all that, I’d imagine that things could be quite miserable. I think that’s also probably why graduate school isn’t for everyone. If one is already frustrated after undergrad, you’re probably not going to pick up steam spending another several years working on things. If, however, you’re not, and you keep in mind Rule #1 (most of your time will be spent trying to fix things), it can be pretty awesome. You basically get to pick out your own project and work on what you like without worrying about getting product out the door by next quarter, regardless of flaws still present. I’m not saying here that one’s research ends up being perfection in the end, I would say instead that things don’t generally get swept under the carpet. They definitely do when immediate profit is the goal.
Well, today puts things again in the realm of micropipette prep, and again things are not coming out precisely as expected. I’ve found that even though I thought previously that backfilling was not taking place in the pipettes that were finely tipped, apparently it is. The pipettes do not seal off on the end, even with a decent amount of elongation, it’s simply that filling them is difficult. The backfill is not visible after letting the electrodes sit for a period of time (10-30 mins), but becomes quite apparent after attempting to fill and being treated with a nice bubble surrounded by fluid towards the tip, and towards the main portion of the pipette. When I have this figured out, I’m going to write a little guide, and perhaps post it on the Wiki I’m thinking of starting, so that at least the other google empowered folks who are attempting to do the same thing won’t have to go through the same steps I did over a number of days to get things working.That said, I’m more and more convinced that this is the sole problem being encountered. Low impedence electrodes with fluid continuity do not have the awful noise issues experienced before, and perhaps as soon as this technique is down, recording will be no problem.
What I thought initially would be hard, proved to be exceptionally easy then exceptionally impossible. I’d heard from others long before I’d gotten into the business of pulling my own electrodes that this would be a road fraught with dismay and frustration, or at least fraught with cursing and many discarded electrodes that did not work out. When I started in on this, I got our lab’s shared laser puller up and running. This part wasn’t hard, and merely required reading the directions. Then after less than half a day of pulling electrodes, I was getting ones that were around 30-40 Megohm. What, I found soon after, however, was that that was 30-40 Megohm with a tiny bit of air still holding out near the tip, and that filling the very tip of the electrode by capillary action is necessary to get a proper electrode. As an aside on this matter, if you have crazy noise issues that creep into your signal and you’re sure it isn’t clogged: you may have a bubble in there somewhere. It seems however, that getting this capillary action going becomes exponentially harder and slower as the tip diameter closes down. Currently, this is showing itself in the sense that I can only get 4-10 Megohm micropipettes that actually sip from the KCL I’m placing them in.I’m writing this, and posting it, both for my own reference, and for the fact that I couldn’t really easily find anything on methodology for pulling electrodes. Perhaps I should start a wiki for such things? It would not have to be limited in scope only to that particular topic, but rather be a site covering lab technique and the various ins and outs. Personally, this rather appeals to me for the reason that a lot of the manual lab work requires what seems to be almost a black art, and either someone in your lab can show you this art, to some extent, or you just have to spend enough time banging into things that you figure it out on your own. I will have to think about this…
So, I’ve recently been training up on doing electrophysiology in the lab, and I thought I’d blog about the experiences a bit (mainly technical details) as I go along since I’ve encountered a number of issues along the way, and some I’ve resolved so far, whereas others still need some work to get going.This is basically both for personal reference, as well as for anyone who happens to be interested in these details. One of my standard solutions when looking for information about things that don’t seem to be working correctly is through googling, and I’ve really not been able to find any good information on all this stuff.
Looks like Apple has rolled out a new update for AirPort that’s recommended for all Intel mac users (find it here). Now, the first thing on everyone’s mind is: does this have anything to do with 802.11n? There’s no detailed description of the changes other than the standard stuff Apple puts with updates it doesn’t want to divulge all the changes on. That said, I’ve been breaking out the good old command line tool “strings” on recent AirPort drivers as they’ve been coming out, and this is no exception. If you download the update, and then extract the following file:
And then run strings on the AirPortAtheros driver file within Contents/MacOS, in the following manner:
strings AirPortAtheros.kext/Contents/MacOS/AirPortAtheros | grep 802.11
You will see the following:
Wireless Network Adapter (802.11 a/b/g/n)Wireless Network Adapter (802.11 a/b/g)
What does this mean, precisely? Hard to say until someone gives things a shot on an 802.11n network!
The latest Apple EFI update adds additional support for bootable DOS mode utilities! Some disk utilities which work on other BIOS based Intel PCs now work, A20 mode still not supported. Unsure whether A20 is required for this, but memtest86 is still broken. Some might ask why use that instead of the Mac memtest application? Well, memtest86 doesn’t require booting a full OS X kernel, and so more of the machine’s memory can be checked.
| digg story
So, I just noticed today that Apple’s Grapher application provides a quick and easy way to format LaTeX equations. I’d been using it recently to copy and paste equations into Pages while I brushed up on using LaTeX equations for taking notes in class when I just noticed this fabulous little feature tonight. Damn cool, since sometimes it can be faster than typing the LaTeX to format an equation, plus it provides a nice GUI. The only downside that I can see at the moment is that it doesn’t look like the exported code can be pasted back to continue GUI editing of an equation.