Catch up on stories from the past week (and beyond) at the Slashdot story archive


Forgot your password?

Submission + - Intel Announces 'Smaller than a Penny' SSD (

Tech.Luver writes: "On Dec 14 Intel announced the Intel Z-P140 PATA Solid State Drive (SSD), an ultra-small — smaller than a penny, weighing less than a drop of water — complete storage solution for mobile digital entertainment, and embedded applications, offering low-power, high performance, and durability, which is also 400x smaller than a 1.8-inch hard-drive. Right Capacity 2, 4, 8, and 16GB capacities are enough to support operating system storage, applications, data, and media storage, meeting mainstream density requirements for most computing markets. ( )"

Submission + - Brain grown in jar pilots F-22 sim ( 5

NJ Hewitt writes: "Florida scientists have grown a brain in a petri dish and taught it to pilot an F-22 jet simulator." The brain, with neurons connected to 60 electrodes, at first had no ability to pilot the fighter jet, but slowly learned and can now reliably navigate through even hurricane-force winds in the simulator.

Submission + - Linux Networking Cookbook

stoolpigeon writes: "[Editors: I've read the slashdot guidelines and tried to adhere to them as best I could. What follows is my review of a brand new O'Reilly book, the Linux Networking Cookbook. My email is I can be reached at 321-695-4295 and I'm usually logged into AIM during the week as ronpeckjr. I would never ask this on a regular submission, but if this review is rejected, due to my writing, could I get a short note letting me know that and possibly what I could have done to improve it? I would really appreciate that.]

The Linux Networking Cookbook is Schroder's companion to her earlier book, The Linux Cookbook. As the title suggests, this is a set of networking 'recipes'. The scope is wide, but the recipes are concise and to the point. Schroder wastes little time getting to the focus of each section, making this an excellent reference guide for any of the technologies that are covered. For the reader interested in a deep, long running discussion of how and why things work the way they do, this is not what they want. The book is perfect though for the reader looking for examples and thorough instructions on getting things installed and running.

The back cover of the Linux Networking Cookbook says, "This wide-ranging recipe collection covers everything you need to know as a Linux network administrator, whether you're new to the job or have years of experience." I'm on the new to the job side of the spectrum when it comes to networking experience. My guess is that for the experienced administrator, there may not be a lot of depth. But the breadth of the information is such that they are very likely to find something new here. The book assumes a basic familiarity with administering Linux. Instructions are detailed but the reader will need to know how to navigate the file system, edit files, create user and group, change permissions and other similar tasks.

Like many other cookbooks, my questions were "Will I understand the recipes?", "Are these recipes within my skill level?" and "Are these things that I want to make?". Schroder has done an excellent job making everything in this book extremely clear and understandable. There is enough explanation to get the reader started, not so much that it feels slow or like she was padding for length. Anyone with even slight exposure to the command line in Linux should be able to dig right in, follow the instructions and enjoy the satisfaction of seeing these recipes work. It often felt to me like I had taken a few hours of research on Google, cut out the useless and outdated, cut out the excess verbiage, and was left with a distilled set of advice, examples and references for further reading. I really see this book as being strongest as a time saver and a great platform for learning. That answers my first two questions with an emphatic yes. Now all that is left is the question of, "Are these things that I want to make?" Here, really the best answer is to read through the table of contents. I'm going to comment on what stood out, but there are just too many subtopics covered to mention them all. So it would be worthwhile for any who might be interested in this book to take the time to read through them.

The only drawback to the book is that to really get a lot out of it, one is going to need access to some equipment. For many chapters some PCs are enough. For other chapters, purchasing hardware will probably be necessary. While this keeps the book from being perfect for everyone, I would say that it also means that the reader is going to get a solid understanding of the topics rather than one that is only theoretical. This is a strength of the book in my opinion, but it is good that the buyer is aware of this before they purchase.

Each chapter follows the same format. They begin with a brief overview of the technology and concepts for that chapter. Chapter 1, "Introduction to Linux Networking" contains only this overview, and is the shortest chapter of the book. The overview includes not only the primary concepts but what hardware will be required to work through the recipes and any software that may be required as well. The overview is followed by a series of subtopics, each presented with a problem, solution, discussion of the solution and a list of other resources. The solutions are given with instructions applicable to the Fedora and Debian distributions. I would assume that this makes the book immediately useful for the majority of linux users, as their distribution will probably be very similar to one of those two.

The second chapter is Building a Linux Gateway on a Single-Board Computer. This chapter is somewhat unique in that the recipes are all written with the goal of installing Pyramid Linux on a Soekris 4521 board. The recipes also require a CF card (or microdrive), power supply and null-modem cable. These hardware requirements mean that working through this chapter requires spending more than an insignificant amount. The Soekris board runs about $150 and then there will be the smaller costs of the other equipment. For the reader unwilling to spend the money, this chapter is not much use. On the other hand, someone who may be interested in learning this kind of embedded work and finding out more about network devices will find this to be an excellent chapter. I remember spending more on my first Mindstorms kit to start learning about embedded programming.

The third chapter is Building a Linux Firewall. This chapter has recipes to build an iptables firewall from scratch. The problems and solutions in this chapter are excellent and cover a wide ranging number of situations. Many distributions have gui tools for managing iptables, but I know I've run into problems with these applications more than once. Schroder lays out how to get around such difficulties and deal directly with iptables for everything from getting multiple SSH host keys past NAT to logging. The references to outside resources are also extremely useful as in the other chapters.

Chapter four is Building a Linux Wireless Access Point. This chapter builds on the previous two chapters. These recipes, combined with the previous recipes, and once again pointed at a devoice running pyramid linux, will have the reader building a very capable wireless access point. These three chapters work together as a very nice unit that could be seen a little costly for the hobbiest or as incredibly inexpensive for the network administrator. I think that they offer very attractive options to the shop with a smaller budget but a reluctance to settle on less capable hardware.

Chapter five takes a turn and is Building a VoIP Server with Asterisk. This chapter can be done with a few pcs, and hardware that allow for putting sound in and getting sound out. I found the chapter to be a little difficult to follow but I have absolutely no experience with telecommunications. I think that anyone else in my position may need to do some other supplementary reading (Schroder points out plenty) and an allowance for plenty of time to expirement and learn. Like the other chapters, the coverage is wide, and given enough time readers could have a very capable system built with the guidance given in this chapter.

Chapter six is Routing with Linux. Unlike chapters two and four, the recipes in this chapter are written assuming Debian or Fedora as opposed to Pyramid Linux. Schroder starts with calculating subnets and moves on quickly to static and dynamic routing.

Chapters seven, eight, nine and ten are all about connecting remotely. Seven is Secure Remote Administration with SSH. The basics are covered as well as hardening SSH, tunneling X Windows securely and even sshfs. I've personally spent a great deal of time tracking down little bits of information for many of these, and here they are all collected together in a very easy to read and use format. Chapter eight is Using Cross-Platform Graphical Desktops. This chapter covers rdesktop, FreeNX and VNC. The majority of the chapter focuses on FreeNX and VNC. The VNC portions give some very nice recipes for working securely and in a mixed environment that includes windows machines. I work in just such an environment, and I look forward to being able to reference this book on those occasions when I need to connect to one of our Windows servers. It doesn't happen too often, and that makes a reliable reference that much more valuable. Chapter nine is Building Secure Cross-Platform Virtual Private Networks with OpenVPN. The recipes begin with instructions on setting up a lab to work with OpenVPN and test things without interfering with the rest of a network. This is a quick chapter and lays out setting up the server and connecting with clients. Chapter ten is similar to nine but has recipes to build a Linux PPTP VPN server.

Chapters eleven and twelve move things back inside the LAN. They are Single Sign-on with Samba for Mixed Linux/Windows LANs and Centralized Network Directory with OpenLDAP. The recipes include migrating away from and into Windows networking solutions. I can see the value here not only for network administrators, but also Linux system administrators who may find themselves needing to integrate into a predominately Windows environment. I know there are often questions on Slashdot about getting Linux in the door. Well, for those who succeed, these chapters could come in very handy.

Chapters thirteen and fourteen are about network monitoring using Nagios and MRTG. The recipes cover a wide number of monitoring options and could really get an IT shop on well on their way from finding out about outages from their users, to being truly on top of their network.

Chapter fifteen is a quick treatment of IPv6. Chapter 16 covers network installs. Chapters seventeen and eighteen cover administration via serial console directly and over a modem. These tried and true methods are probably less relied upon than in the past, but they are covered well and it speaks to the thorough coverage of the book. Chapter nineteen closes the book with a host of generic recipes centered around troubleshooting network issues. This is a solid chapter, not just a quick troubleshooting grid tacked on to the end of the book. There are three appendices. The first is a list of recommended resources that is dominated by O'Reilly titles, though there are others. The second is a glossary and the third is a very useful kernel building reference.

The index is decent. It is not exceptional, but it is not bad either. This is somewhat alleviated by the fact that the book comes with access to it on Safari for 45 days. There is also a web site with all of the examples available for download. The author's website is also a good launch point for related articles and information."

Submission + - Role of endogenous retroviruses in human evolution

mhackarbie writes: The current edition of the New Yorker magazine has a fascinating story about endogenous retroviruses in the genomes of humans and other species. Although researchers have known about such non-functional retroviral 'fossils' in the human genome for some time, the large amount of recent genomic data underscores just how pervasive they are, in a compelling tale that involves humans, their primate cousins, and a variety of viral invaders. Some researchers are even bringing back non-functional viral remnants from the dead by fixing their broken genes.

Submission + - Microcontroller for the hobbyist? 5

TomTheGeek writes: "I'm a programmer that's done some assembly language before and would like to start programming microcontrollers. I've heard about the BASIC Stamps from Parallax, the PIC series from Microchip, the MAKE Controller Kit, and the AVR series from Atmel but they seem to be focused on a development board that is too expensive to dedicate to a single project. Having an expensive development board is fine but I want the microcontroller to be cheap (<$10) enough that I don't have to disassemble my previous project in order to start a new one. I'll be doing the programming in Ubuntu so compatible development tools and drivers are required."

Submission + - Thirty Meter Telescope Moving Ahead

Hugh Pickens writes: "California Institute of Technology and the University of California have announced they have received a $200 million commitment over nine years toward the further development and construction of the Thirty-Meter Telescope (TMT). The core of the TMT Observatory will be a wide-field, alt-az Ritchey-Chretien telescope with a 492 segment, 30 meter diameter primary mirror, a fully active secondary mirror and an articulated tertiary mirror. TMT will be the first ground-based astronomy telescope designed with adaptive optics as an integral system element. The adoptive optics will sense atmospheric turbulence in real-time, correct the optical beam of the telescope to remove its effect, and enable true diffraction-limited imaging on the ground. For many astronomical observations, this is equivalent to observing above the Earth's atmosphere at a fraction of a cost of a space-based observatory. Relative to the Hubble Space Telescope, TMT will have 144 times the collecting area and more than a factor of 10 better spatial resolution at near-infrared and longer wavelengths."
The Internet

Submission + - Is Comcast at it again? 3

dreamchaser writes: "I'm a guitarist and frequently record my music to MP3 via a digital mixer/recorder. Yesterday I tried to share a few of my songs with a friend who happens to have Comcast Cable. We were using ICQ at the time and after I shot my friend a picture or two, I tried sending her a song. The transfer fizzled out at about 200k. Tried it a few more times, no luck. Tried another (large) picture and it worked. Scratching my head, I renamed the MP3's to BIN and they whoosed right through the old Internet tube like they should have in the first place.

In light of the previous news about Comcast throttling P2P apps, it now appears that merely trying to exchange a file of a 'bad' type (MP3 in this case) gets your transfer throttled. I do not have Comcast anymore, having given it up for Verizon's FIOS. My question for Slashdot is can those of you who DO have Comcast test this and see if it consistently like is all over? It's beyond ridiculous that one cannot send legal content to a friend via a direct IM connection wihtout having the filenames (and who knows what else) sniffed out and the transfer killed. Is anyone else seeing this? What can we do about it other than raise awareness?"

Submission + - Cryptography experts sounding alarms

netbuzz writes: "First we learn from Bruce Schneier that the NSA may have left itself a secret back door in an officially sanctioned cryptographic random-number generator that would allow the good guys to easily decipher encrypted messages sent between bad guys (not to mention anyone else). Now Adi Shamir is warning that a math error unknown to a chip maker but discovered by a bad guy could lead to serious consequences, too. Remember the Intel blunder of 1996?"

Submission + - Possible Development in Primate Cloning (

explosivejared writes: "Researchers from Oregon allege that they have successfully derived embryonic stem cells (ESCs) from a blastocyst created through somatic cell nuclear transfer. If confirmed, this would be a world first and could potentially lead to the creation of human ESCs through cloning.

While details of the new research — led by Shoukhrat Mitalipov and performed at the Oregon National Primate Research Center — will be unavailable until the work is published in the journal Nature, his past work may provide some indication of what to expect. (Given the fraud of Hwang Woo-Suk, who came out of nowhere, one can't be too careful.)

If this bears any fruit this will be sure to stir a frenzied debate over the ethics of cloning all over again."

The Internet

Submission + - 'When the patient is a Googler'

netbuzz writes: "A New York orthopedist's essay carrying that headline sets out to make the point that patients who rely more on their Web browsers than the expertise of their doctors are likely to be ill served by both. It's a valid enough point, but the doctor in this case so savages the patient he puts forth as Exhibit A — a 40-something Mom with an unruly 3-year-old — that he's likely to send even more skeptics of modern medicine scurrying for the Internet."
Portables (Apple)

Submission + - Ipod's success baffles Sony Japan (

An anonymous reader writes: Sony, large, powerful and technological wizards cant work out why apples ipod is so popular? Duh These quotes from a recent article here in Japan explain why Sony is baffled. "We researched the iPod and fixed what [we considered] was difficult to use about it," said Takashi Kinouchi, head of Sony's product planning department". "it is not easy to lure customers used to Apple's software over to Sony products". Source: Sony trying to bite into Apple's iPod market share, Shigeki Kurokawa / Yomiuri Shimbun Staff Writer, November 5, 2007. Daily Yomiuri

Submission + - The first image taken with an ultra low field MRI (

KentuckyFC writes: "MRI machines are about to get smaller, much smaller. Most of their bulk is taken up by the huge superconducting magnets required to generate fields of about a Tesla. Now a team at the Los Alamos National Lab in New Mexico has built a machine that can produce images using a field of only a few microTesla. So giant superconducting magnets aren't necessary, a development that has the potential to make MRI machines much smaller, perhaps even suitcase sized. Today, the team has posted sections of the first 3D brain image taken with the device (abstract, pdf)."

Submission + - Pseudoscience winning weblog award ( 1

OneMHz writes: A pseudoscience climate change denialist blog is currently winning the science weblog award. Why? Because conservative websites are encouraging people to go vote for it, whether they read it or not. Their strongest competitor, Bad Astronomy, needs help. I encourage people to go read it, then vote for it. They've embedded the science poll in the page, so you can vote there, or here. It's a sad day when pseudoscience wins over real sience...

Slashdot Top Deals

Two percent of zero is almost nothing.