Win XP crashes solved by disabling Nvidia Driver service

I finally solved a long-standing problem with my Windows XP laptop.

My CPU would slowly grow to about 15% even when idle.  Then a window would pop-up telling me I had 60 seconds to reboot because an RPC service had crashed.  This occurred about once a week.  Often enough to be annoying, but not often enough to make me dig to find the problem.  The crashed "RPC service" was not very instructive, because it's a service used by lots of programs.  So clearly some software running on my computer was "abusing" this service.

At first I thought it may have been due to the Intel PROSet/Wireless software running on my computer that was doing something bad with RPC, it turns out the problem was with Nvidia driver software for my Nvidia graphics card.  The problems were on my Asus A8JS laptop, under XP Pro SP2 & SP3, with an Nvidia GeForce Go 7700 graphics card, and Nvidia Forceware 163.44 drivers, with DirectX 9.0c.

I found the solution here:
http://forum.notebookreview.com/showthread.php?p=3076277#post3076277

I had to disable the Nvidia Driver service (see detailed instructions below).  After that, I've not had any of the symptoms.

This describes the exact symptoms I was seeing (including the 1 minute to reboot):
http://www.notebookforums.com/thread212236.html

Apparently, you can shutdown the Nvidia Driver helper without much consequences. There may be some multi-monitor stuff that doesn't work, but you can always start it manually, or re-enable it to start automatically.
http://www.glaryutilities.com/processlibrary/process/nvsvc32.exe.html

Read on for detailed instructions…

Read More …

Return of the sneakernet?

Ages ago (pre-1990), most computers weren't connected to networks. You had to use sneakernet to transfer files — put them on a floppy disk, and walk to the other computer (in your sneakers).

The rise of networking killed off sneakernet, but maybe only temporarily. Rasmus Fleischer writes in The Future of Copyright:

The capacity of portable storage devices is increasing exponentially, much faster than Internet bandwidth, according to a principle known as "Kryder's Law". The information in our pockets yesterday was measured in megabytes, today in gigabytes, tomorrow in terabytes and in a few years probably in petabytes (an incredible amount of data). Within 10-15 years a cheap pocket-size media player will probably be able to store all recorded music that has ever been released — ready for direct copying to another person's device.

In other words: The sneakernet will come back if needed. "I believe this is a 'wild card' that most people in the music industry are not seeing at all," writes Swedish filesharing researcher Daniel Johansson. "When music fans can say, 'I have all the music from 1950-2010, do you want a copy?' — what kind of business models will be viable in such a reality?"

Regardless of the mistakes the recording industry is making, I suspect the sneakernet will come back, at least for transferring large amounts of content.

This reminds me of this quote I first heard in the mid-90's: "Never underestimate the bandwidth of a station wagon filled with CDs."

Martian headsets, and why standards are hard

Joel Spolsky writes about the real-world difficulty of standards in "Martian Headsets". The article was spurred by Joel's observations on the Microsoft IE8 team's rock-and-a-hard-place predicament. The article uses a clever example of Martian MP3 players and headphones to show why standards start out easy to implement but end up nearly impossible despite everyone's well-meaning efforts.

Joel also tosses in an interesting anecdote about the Bible as a "standard":

If you've ever visited the ultra-orthodox Jewish communities of Jerusalem, all of whom agree in complete and utter adherence to every iota of Jewish law, you will discover that despite general agreement on what constitutes kosher food, that you will not find a rabbi from one ultra-orthodox community who is willing to eat at the home of a rabbi from a different ultra-orthodox community. And the web designers are discovering what the Jews of Mea Shearim have known for decades: just because you all agree to follow one book [of standards] doesn't ensure compatibility, because the laws are so complex and complicated and convoluted that it's almost impossible to understand them all well enough to avoid traps and landmines.

Standards are a great goal, of course, but before you become a standards fanatic you have to understand that due to the failings of human beings, standards are sometimes misinterpreted, sometimes confusing and even ambiguous.

After reading the article, I feel sorry for the IE8 team. And it's just a matter of time before Firefox, Opera, and all the other browsers feel the pain, too…

[via "The Flamewar of the Century" at john-ahrens.com]

50+ Tricks to Get to a Real Person

To get a live person when calling a support number, I use tricks like pressing zero even if it's not given as a choice, or doing nothing to make the system think that I don't have touch tone phone.

50+ Hacks and Tips to Get to a Real Person at Any Corporation in 10 Seconds or Less has some clever ones I've never thought of before, like:

  • Choose the Spanish option, which often has a shorter wait time, and you'll probably be connected with a bilingual person.
  • Press everything. By pressing multiple numbers, you can trick systems into thinking you're on a rotary phone or that you're crazy. Either way, you're in.

The article also has links to some good resources, like 800-numbers.net and Gethuman.com.

Using anonymous cellular data to indentify traffic

In-car GPS vendors are starting to integrate real-time traffic data into their route finding/planning — rather than taking the shortest route, the GSP tries to take traffic into account.  In the US, many cities, counties, & states now collect traffic data, and display this data on their own websites.  But for GPS companies, hooking into all these various sources is initally expensive, and then continues to be expensive to maintain.  And where the city/state/county does not collect this data, there is simply no source of traffic information.

TomTom is trying out an interesting new source of traffic data — partner with a cell phone company (Vodaphone) to get anonymous phone usage statistics.  If the phone company sees an unusually large number of cell phones near freeways, that's a good indicator of unusually bad traffic.  By partnering with Vodaphone, TomTom can get these anonymous stats from Vodaphone to make better route suggestions.  TomTom is calling this service "High Definition Traffic (HD Traffic)".

This is intially a trial in the Netherlands only.  If it works well, they will try more Eurpoean countries, and then try to partner with US cell phone companies.  If this works, it gives TomTom fewer different sources to integrate with — only once per cell phone company, rather than once per city/county/state.  It will also work in areas where the city/county/state does not collect real-time traffic data, or else does not make that data available.

TomTom & TeleAtlas "get it" for community-provided map information

Web-sites like Amazon and Wikipedia have been leveraging the value of community-provide content for years. With all the GPS devices on the road, it seems like map data providers could find a way to collect community-provided map updates from these users.

It looks like TomTom & TeleAtlas "get it" around community-provided map data. TomTom's recently announced "Map Share" system and their proposed acquisition of TeleAtlas will be a powerful combination.

Read More …