Category: Tech

Everything about technology but product reviews

Why LinkedIn is a better fit for me than Orkut

This article compares LinkedIn and Orkut. These are websites that allow you to enter in people you know and then those people can enter in people they know which then lets you perform searches over your entire social 'network'. Orkut focuses primarily on personal relationships while LinkedIn focuses on business relationship. I prefer LinkedIn because it provides tools and features that are extremely useful to me both in hiring people and in putting myself in a better position to be hired if I should need a new job.
Continue reading Why LinkedIn is a better fit for me than
Orkut

My Experiences Installing Mandrake 9.2

It took one day and a few hassles but upgrading from Mandrake 9.1 to Mandrake 9.2 was worth all the effort. Better yet, the amount of effort required was much less then with either Windows (3.1, 95, 98, 2000) or other versions of Linux I have installed. Not only was I able to get access to the latest versions of Open Office, Mozilla & GnuCash but the new Mandrake 9.2 fonts are just outstanding. The increased clarity makes reading the screen pleasurable and by letting me keep my windows smaller it gives me extra screen real estate. It's like getting a new monitor.

Continue reading My Experiences Installing Mandrake 9.2

The Book I Want To Read on Web Services

I just finished reviewing a chapter in an upcoming computer science textbook on Web Services. The authors made a heroic effort to give the reader a solid grounding in Web Services including HTTP, SOAP, WSDL, BPEL, WS-TX, WS-CO, UDDI, etc. all in 60 or so pages. In terms of information density, the result was the book equivalent of depleted uranium. To make matters worse many of the specifications they were describing had already changed since the time they wrote the chapter and will surely change even more before they publish. Which got me to thinking about the book I would want to read about Web Services Protocols.
Continue reading The Book I Want To Read on Web Services

Microsoft and Office Innovation

As a thought for today, Microsoft really hasn't managed to add too many compelling features to their Office monopoly. On top of that their recent licensing games have certainly given customers an incentive to look elsewhere. At the high end Office is still the best there is. But the majority of customers are no where near the high end. There are now a large number of competing products, many available for free, that more than meet the needs of most users.

The irony is that Microsoft has traditionally killed off the majority players in any market they enter by offering a product that is 'good enough', lower priced and leverages their existing monopolies. The Office alternatives has reached the point where they have certainly nailed two out of three. Given Office's institutionalized refusal to base any of its code on Windows it doesn't even get to take advantage of the Window's monopoly beyond getting first access to new versions and getting its bugs fixed NOW. But given how mature the office market is and given the generally low rate of new features it isn't clear how much of an advantage that really is.

In the short term none of this matters, no body gets fired for buying Microsoft. But in the medium and long term the change of events spells interesting times. How long until some CIO becomes a hero by slashing the company's Office budget to near zero and the pattern is set? I realize training and support is a bigger issue but anyone who knows Office will be comfortable with the alternatives and I suspect the support for the free/low cost choices are generally better then anything Microsoft can offer.

Microsoft's lack of innovation and pricing games is exactly the sort of help its competitors need.



Debian… oy.

This weekend I installed Debian on an old computer. Most of the install was just blindly pressing return but I ran into a problem getting X to start. It turned out that in the configuration they suggested uses a frame buffer setting that doesn't work on my machine but it took me an hour or two to figure this out. I am still setting Debian up with features I was able to trivially get on Windows. For example, I have a utility called Tardis on windows that uses NTP to set my clock. Of course I could get the NTP client for Debian. It was just a matter of doing a quick package search and then using apt-get. But I then had to go find a list of suitable NTP sites and type them in manually. Tardis came pre-configured. Tardis also shows me that it is working but I can't be sure the NTP deamon is running (I haven't checked) and even if it is I'm not 100% sure that the KDE clock is listening to it. I'm sure everything works just fine but there is no pro-active indication of this. (BTW, I like KDE a lot more than Gnome) No, none of its a big deal, it's just that everything is a little harder on Debian. Still, with crossover I'm hoping to become Windows free within the next few weeks. I need the crossover plugin to get things like Quicktime and Crossover office to get access to Quicken. I'm happy to pay the software fees required to get Crossover. Other then that there seem to be reasonable alternatives for everything else. I will keep y'all informed.

Yaron's Rules of Standardization

Most of my job for the last few years has been working on standards. In that time I noticed a fairly obvious pattern for technologies that tend to make successful standards, they meet three criteria:

Yaron's rules of standardization:

  • The technology must be very old and very well understood
    • 20 years is a good rule of thumb
  • Everyone must implement the technologies in essentially the same way
    • A good rule of thumb is, how hard would it be to build a bi-directional proxy between the various players implementation of the technology?
  • Standardizing the technology must provide greater advantage to the software writing community then keeping the technology incompatible
    • Even in the open source world standards can fail if there isn't enough advantage in it, just look at the fights over RSS.

Darik's Boot and Nuke & Mandrake 9.1

I have an old computer I finally want to get rid of and I need to destroy all the data on the drives before sending it to be recycled. There is a simple GPL solution called Darik's Boot and Nuke which does the trick. But the directions on the Website don't quite work for Mandrake 9.1. Here were the changes I had to make:

To copy the image to a floppy I had to use dd if=dban-*.img of=/dev/fd0. [1] I had trouble with /dev/random generating enough bits. The easiest workaround is to substitute /dev/urandom for /dev/random. But there is a more paranoid choice. [2]

[1] In most system apparently /dev/floppy points to the floppy device but in Mandrake 9.1 it points to a directory with the floppy's contents. /dev/fd0 is what Mandrake uses to point at the actual floppy image, it's a link to /dev/floppy/0.

[2] /dev/urandom uses a crypto grade hash to extend the amount of random data collected by /dev/random. This should be fine but I'm really paranoid so instead what I do is first run dd if=/dev/random of=MySeedFile bs=512 count=20. This will give me more than the 512 bytes required by DBAN. You will have to play with the mouse a bit to generate enough random data to fill the request. Once the dd returns then I execute dd if=MySeedFile of=MySeedFile2 bs=512 count=1 and then delete MySeedFile.