Microsoft Live Search is now… uhm… live. Naturally, I’ve tried it out – and lost interest promptly after the first query.
Microsoft is clearly trying very hard to replicate Google success. The problem is, part of the Google appeal is clean and usable UI. Unfortunately, Microsoft Search folks still don’t get it. Branding and replication of [rather garish] desktop UI elements pretty much ruin the experience.
To substantiate and quantify my qualms, here is a list of things that made me think this round of improvements is still years behind Google in terms of both UI and usability:
I understand that Microsoft will eventually fix these bugs and will become at least as usable as Google. Still, this is pretty sad: the company that defines the UI on 98% of the desktops in the world is incapable of replicating the usability of a white webpage with a bunch of search results on it! This might explain why people have to curse through changed UI paradigms every time a new version of Windows or Office is released.
Forget MIMO. Donate your Linksys speed booster to charity. Now you can boost you wireless network speed dramatically using the scientifically proven technology.
“..WiFi Speed Spray™ can overcome the effects of pollution, increase fidelity, and provide you with the fastest wireless data transfer possible. Compatible with ALL 802.XXx standards!”
I hope it is patent pending.
Your actions are (choose one):
Correct choice: 3.
Recently, I have finally decided to upgrade our home server. My wife’s old Dell Dimension 8200 that has been in this role for a while would move to a summer house as a main PC and I would get a new server.
After shopping around, I finally ended up with the following do-it-yourself configuration:
Using ASUS Pundit as barebone seemed like a good idea. It is small, elegant, inexpensive, and it was said to have a good Linux support. The entire configuration cost $400+, which is significantly lower, compared to the Dell and HP prices for similar systems.
Much to my surprise, the configuration and installation was not quite as seamless as I expected it to be. At this point of time I usually take for granted the fact that Linux installs and runs seamlessly on most of hardware available. Installing Linux on my new server turned out to be more challenging and finding the solution was not a metter of a simple Google search.
I was a bit scared after reading a couple reports about the supposedly loud noise emitted by the CPU fan. The fan turned out to be quite loud indeed but only when running at top speed. Setting the BIOS option for Q-FAN on make the fan go quiet. I am yet to see the amount of number crunching I have to run on the machine to get the fan loud again.
Wrong VCORE temperature
Turns out that Pundit’s default BIOS revision (1007) does not play nice with Celeron D CPUs, which results in a rather ominous BIOS halt with a message about a hardware problem. The problem reported is wrong VCORE temperature read by BIOS. The solution is to update the BIOS to revision 1010.
The Pundit’s manual is very detailed on the BIOS update procedure and it clearly explains each step of it: how to copy the update file to the floppy disk and how to tell the AFLASH program (or to the BIOS update routine) what the filename is. There is only one tiny problem: the system does not have a floppy drive!
The solution, graciously provided by Asus support, is to use the Asus Update program supplied on the CD (and not documented anywhere). Unfortunately, in order to use the program, one has to install Windows first, but the inconvenience is minor, especially when compared to other options, like trying to create a CD-ROM bootable into MS-DOS and posing as A: drive.
Disk geometry problems
It turns out, SuSE 9.3 installer has a bug when dealing with SIS-based chipsets (like the one in Pundit). As a result, the installer kept using wrong geometry information for my hard drive causing dozens of scary IDE errors and telling me that my hard drive is actually only half the size of what I thought it was (interestingly, none of the other Linux distributions, including Knoppix, Fedora Core 3, and even SuSE 9.1 exhibited this problem). The solution is to pass the option insmod=ide-generic as a boot parameter when starting the installer. See also SuSE Support Page.
One thing that still doesn’t work is reboots and poweroffs. At the point where the PC is to be rebooted of turned off, the system locks up. I have tried various combinations of apm and acpi options but with no success so far. I would appreciate if someone suggests a way to deal with it.
Overall, Asus Pundit is a nice little box ideal for a small home server or an HTPC. Once I got the problems above resolved, I was good to go. and my new server keeps on humming in the corner. So far I am pretty happy with it and if it keeps that way, I will be soon buying another one for a home HTPC.
Smartphone Thoughts is running a rather lengthy discussion about whether the development tools for Windows Mobile 5.0 are not going to be free. The Slashdot crowd has immediately jumped onto the issue as usual blaming Microsoft of a vile conspiracy.
The core of the issue is the fact that Microsoft is trying to unify all of its development tools underneath the Visual Studio 2005 umbrella. Since Visual Studio is not freely distributed (unlike the Embedded Visual Tools development system), the argument goes, there will no longer be any free development for the Windows Mobile platform.
I find the argument flawed for a number of reasons. If we look at the trend, there are actually many more free development tools available from Microsoft compared to the past. In fact, looking specifically at CE development tools: in order to develop for Windows CE some 3-4 years ago one had:
Nowadays, one just has to spend 15 minutes downloading eVC++ 4 for free. Same goes for Windows-oriented development. Same for dozens of other SDKs. There are ways to develop for each and every Microsoft platform using free tools from Microsoft.
What is going to happen actually, in my opinion, will be similar to the situation with development tools for .NET framework:
Now on a personal note. I think, I am willing to pay $299 if this would give me a single tool that would provide coverage for all Pocket PC and Windows Mobile target platforms without the sheer madness of having to install eVC++ 3, eVC++ 4, VS 6, and VS.NET 2003 (let alone a half dozen of platform SDKs). This alone is well worth $299. 😉
Телегадюшник под названием RTVi не перестает умилять. Во вчерашнем выпуске новостей – опрос мнения телезрителей: “Одобряете ли вы выбор кардиналов в отношении нового Папы?” (я не шучу)
Как оказалось, 63 процента виртуального Брайтон-Бич решения кардиналов не одобряют…
I discovered Steve Pavlina’s weblog by accident about one year ago and have been an avid reader of his articles ever since. I have been constantly striving to incorporate some of his tips into my own “workflow.”
Steve has recently published an article titled “30 Days to Success.” His idea is very simple: it is pretty difficult to commit oneself to drop a bad habit or to acquire a good one with no time horizon. Instead, he suggests having a “free trial” period for 30 days, which is much more achievable. After 30 days, the thinking goes, the new habit becomes a second nature, so it would be much easier to stick to it.
I am going to take Steve up for his challenge. For the next 30 days, I will go to bed no later than 12am and I will also work out no less than 3 times a week. I have been trying to stick to these rules since time immemorial and I never managed to last more than a couple of days. Let’s see how it will pan out now.
What will you succeed at in the next 30 days?
“Сионизм явление повседневное в жизни Новосибирской области”. Ну разве не прелесть? Я думал, такие мастодонты уже повывелись – ан нет, есть Frauen в русских селеньях.
Оттуда же: “Чиновники-расхитители – золотой фонд сионизма.” – это хоть сейчас на плакат. Интересно, взялся бы кто-нибудь в каком-нибудь Cafepress сделать пару десятков футболок с данным лозунгом? Я бы точно купил.
Когда-нибудь инопланетяне, роясь на пепелище планеты Земля, проанализируют электронную почту, дабы составить коллективный образ Человека. Получившийся персонаж будет выглядеть следующим образом: жирный, похотливый, лысеющий козел, с крошечным членом, депрессиями, долгами, постоянно покупающий дома в рассрочку и страхующий автомобили…