I have repaired a couple of UPS. Every time it just required that I replace the batteries.
I regularly see cheap =$20 UPS at the local Silicon Valley swaps where usually the only thing wrong is that they need new batteries.
I picked up a Tripplite Omnipro 1400VA UPS for $25 and a few months later picked up 5 batteries for $20. I also picked up a small Deltec UPS for $5 and a new battery for $12.
As for the Tripplite, it appeared totally dead when plugged in with the dead batteries that came with it.
I have a Philips DVX 8000 home theater PC which has been awesome. It looks like a piece of home theater equipment from the knobs to the flourescent display and the various buttons on the front. It supports multiple inputs and outputs, has 5.1 audio processing, and a very good line-doubler for all video sources. It is also a DVD player, has a fully programmable remote and keyboard, and runs Windows 95 (ugh). Fortunately almost everything still works when Windows has crashed.
It is powered by a lowly 233MHZ MMX Pentium CPU. Of course, all of the MPEG decoding for DVD and audio processing is handled by dedicated hardware so the CPU isn't involved for most operations and in fact for the most part it sits idle.
As for the PC part, I would be happier if it DIDN'T have the PC part. Trying to use it as a PC on a 32" monitor isn't all that great, since the monitor is only capable of 800x600.
I have yet to see another unit with similar capabilities. What I like about the line doubling support is that for DVD it is all done in the digital domain, without any analog between the MPEG decoder to the RGB output. The line doubling also looks great with video from other sources, without the usual line doubling artifacts caused by bob-and-weave line doubling.
As for the graphics card in the unit? It uses a Trident graphics chip which is fed video by a Genesis line doubler chip. A good video card is not necessarily a good graphics card since the two do very different operations.
Of course had you gone into the control center, click "Look & Feel" and "Launch Feedback" you can easily disable this. In addition there are numerous other look and feel settings to make KDE look and feel quite different than Windows.
I recently upgraded 3 different computers from SuSE 7.3 to 8.0. I had a number of issues with the original 7.3 but those were fixed with later patches (like the updated kernel).
SuSE 8.0, while mostly stable, has a number of annoying bugs.
1. sudo relies on the environment variables. I need sudo support to start and stop various services under/etc/rc.d and that doesn't work because/usr/sbin is not in the sudo path. If the user's path is being inherited this could cause security holes.
2. The upgrade on one system failed because 8.0 remapped one of the SCSI devices causing a failed mount of fstab. I had to go back and search the logs to figure out what went wrong. When problems occur I shouldn't have to go to the logs to find out what happened. It should have popped up a dialog or something. As it was, the mount was for my USB zip drive. If I could have told YAST to just ignore it I wouldn't have had to start over.
I also found that yast failed to properly upgrade everything. For example, on all of my systems gpg stopped working properly. I had to manually reinstall the rpm to fix this.
I also wish yast2 were more extensive. The firewall configuration could be improved, and many more modules are needed for configuring things like a DNS server (for my internal home LAN), an imap/pop mail server, a news server, and samba. Hardware configuration needs to add support for installing a CDRW drive.
I also found it a real PITA to get my CDRW working again after upgrading to 8.0. In addition the KDE tool I used before for burning CDs keeps crashing whenever I try and configure it.
I found that the video for Linux support is working much better than it did in earlier releases (although the 2.4.16 kernel upgrade for 7.3 was stable as well).
I also like KDE 3.0, which I am also running on Solaris at work.
During the initial upgrade I missed being able to do detailed selection of packages to install in the categorized way it was in 7.3.
Over all I am satisfied with SuSE 8.0, but I think it is a.0 release and needs some more polishing. If you don't need the bleeding edge support, stay with 7.3 and wait for 8.1.
I have an Epson Stylus Color 800 printer and am surprised at the high cost of replacement ink cartridges. There's no printhead in the cartridge or a chip. I have started buying non-OEM print cartridges and have found that often the quality is at least as good and in some cases much better than the original Epson cartridges. Not to mention that the alternatives are often a fraction the price.
I remember King's Quest I and bought it when it first came out. The very first version was for the IBM PCjr to show off its graphics and sound capability (it could do 320x200 with 16 colors! and had 3 channel sound+noise support) Remember at this time most PCs with color could only do 320x200 4 colors (CGA). What was amazing was that they could fit the entire game on a 360K floppy! The graphics were actually rendered at 160x200 so they could store the coordinates in a single byte to save space. As I recall, when entering a new screen the game would draw outlines of everything and fill them in.
Sierra also came out with the PC version which looked like crap by comparison (4 dithered colors in CGA).
I also remember the later Sierra games where you could get them into interesting modes for script debugging and so forth.
After I upgraded to an 8 MHz 8088 XT with EGA with a hard drive I wrote a program to get around the copy protection of all of the Sierra games so I didn't have to keep inserting the floppies every time I wanted to play. It would intercept the disk requests and simulate floppy accesses for the copy protection. Things have sure changed since then.
I have heard many arguments that the US adopting the Kyoto protocol for the reduction of greenhouse gasses is too expensive, unsafe, or inconvienent. For example, BP has already met the conditions of Kyoto in 4 years and in fact is now saving money because of it. DuPont has made even better progress, reducing their contribution of greenhouse gasses by 50%!
Cutting greenhouse gasses is not necessarily that difficult. For example, last year I added more insulation to my home. I saw a 25% reduction in natural gas because of this, and with various other improvements I have made over the last few years I am sure my home consumes far less energy than it did in 1991. The net result of these improvements to my house is that I spend less money on heating and lighting and in only a few years all of the changes will more than pay for themselves. Not only that, but with the added insulation my house is more comfortable.
Everyone seems to think that increasing the milage of cars is the most important step. While it is important and I believe easily doable with todays technology, many other areas are even easier.
How many of your homes have old furnaces and sub-standard insulation?
Perhapse if we had to pay the true cost of energy things would change. Here in California where we are stuck with outragiously high electricity costs (my bill is over $0.20/kwh) and very high gas prices, many people have taken advantage of methods to reduce energy usage. The state has helped as well by offering rebates. For example, it is now not unusual to buy a 100 watt equivelent compact flourescent light bulb for less than $1.
One doesn't have to be a rabid environmentalist to see the benefits from reducing greenhouse gasses. It also makes sense in the pocket book.
Actually the Hammer series adds 8 more general purpose registers and more SSE registers. Read up on X86-64, there's more to it than just going 64-bit. For example, there's better support for relocatable code (i.e. shared libraries).
I've been investigating this chip for embedded use and it looks like a great solution. It basically has just about everything one would need on a single chip... serial ports, Gig-E/Ethernet MACs, DDR ram controller, PCI controller, and more. It also has a Hypertransport interface where two chips can be connected back-to-back with a 12Gbps connection between them.
Broadcom also supports Hard-Hat Linux on this chip.
One solution to the problem would be to establish randomly distributed honey pot computers which act as if they're infected by one of the various script-kiddie trojans. Log everything that happens to those computers, but do not allow those computers to actually perform DoS attacks (the script-kiddie probably won't know the difference).
After collecting evidence, the perpetrator should be fined and prosecuted. It would likely cost nothing to the tax payers since it could fund itself from the fines imposed on the perpetrators. If it's just a kid, then hold the parents responsible.
I agree that it's apathy. A couple of years ago I was running a Perl script that emulated a PC infected with Back Orifice. The script would simulate everything the user would normally do and log everything.
I once caught a user attempting to launch massive DoS attacks against other users by sending ping storms. Unknown to the attacker, nothing actually was going out. I notified the attacker's ISP (Quest) and the target's ISP. The target ISP was very thankful and said that they had noticed very high bandwidth directed to that user. The originating ISP could care less and refused to do anything, even after several more attempts by the same user.
I became frustrated and sent all the information on to the FBI.
Sadly I can no longer find the simulator script on the net (it was called Boobie).
While China may want to get into the software industry they have a major impediment. In India, all of the educated Indians speak fluent english. In China, this is not the case. In my group where I work most of the team is Indian. The only non-Indian people are myself, a Chinese fellow, and my boss (who's Canadian). For the most part there are no problems working with the Indian people (once I got used to their accent). However, the Chinese guy is another story. It is often very painful to try and explain basic concepts to him such that he understands what I am saying. I spent half an hour just getting him to understand that a parameter to my API was limited to 20 bits and to place the upper bits elsewhere. I felt like that father in Monty Python's Holy Grail telling the guards not to let his son leave the room.
He is an extreme example. I have worked with other Chinese software developers who have better English skills, yet the language issue always ends up becomming a problem.
One of the problems is that English is completely different from Chinese with absolutely nothing in common. Europe doesn't suffer this problem since all of the languages have many similarities in sentence structure, pronunciation, spelling, tense, character sets, and so on. Also, in Europe the schools have been teaching English for some time and there is no problem getting fluent English speaking teachers.
I imagine that China has a lot of difficulty getting fluent English teachers over there, especially with all of the opportunities in the Western world without the corruption or restrictions.
I've worked on computers running the German version of Windows NT. By the text I can still figure out what things are and how they relate to the English equivelent. It was not too difficult for me to reconfigure the networking on the boxes, even though I am not a Windows NT andministration expert). Also, if I had any problems, I could ask anyone since all of the Germans spoke fluent English. When I sat in front of a Chinese version once, I was totally lost.
Another major problem is China's lack of respect for IP. India has a fairly good reputation and many large companies have offices over there (i.e. Cisco, Nortel, etc.). China, OTH, has a very bad reputation. You can go anywhere and buy expensive pieces of software for only $1. Or for that matter, you can get a DVD of the latest movie within days after it's released into theaters long before the official releases come out.
Most companies are scared to death of their IP getting out. On this front the Chinese have a well deserved *very* bad reputation.
No company in their right mind would farm anything out to China.
While CUPS makes life easier, I still have a lot of problems with Linux. I run SuSE Linux 7.3 and have an Epson Stylus Color 800 printer connected to a networked print server (Netgear PS104). While Linux talks to the print server just fine, no matter what I try Linux takes 2 sheets of paper per page (yes, I have it configured for Letter and tried the inkjet letter settings). What is worse, however, is I get random horizontal lines on the page. Note that the output is going through Ghostscript.
Now, with the same setup, print server, etc. I have no problems when printing from OS/2 or Windows. Furthermore I do not get the horizontal lines (so it's not a cabling problem or a problem with the print server).
Now the Epson is a fairly common printer with well-documented control codes. I guess the only way to print properly in Linux is with a Postscript printer.
When I first joined @Home in Silicon Valley in the late 90's there was no upstream cap. I ran a TTCP benchmark between two cable modems as a test and saw over 640KBps (over 5Mbps). Just think, I could transfer all the memory I'd ever need in a second!
In addition, there was no policy about running servers and security was very lax. Network Neighborhood would show all of your clueless neighbor's computers hard drives.
Those were the days. Then they added a 1.5Mbps upstream cap, and I was still quite happy. Then they added the 128Kbps cap and from then on it sucked. Send a big email and forget browsing. FTP??? We're not supposed to upload anything.
Things have been very bad since AT&T took over. Last night at around midnight the ping latency to www.yahoo.com was between 500-1500 ms! Granted, no packets were lost, but my 28.8 modem would have been significantly faster than the cable modem.
I'll see what happens when Comcast takes over, but I'm very seriously considering DSL at this point. AT&T just plain sucks. Their cable service also sucks. They keep replacing good channels with infomercials.
Years ago in the early '90s I worked at a company called GRiD Systems where Jeff Hawkins got started with handwriting recognition. I believe he originally developed the technology as a graduate student and then licensed or sold it to GRiD, who made hand-held computers using his algorithms.
Handwriting recognition is extremely difficult for computers (hell, I often can't read my own handwriting). While his algorithm was good, it was not perfect. Trying to recognize the difference, for example between an "a" or a "d", or an "r" or an "n" is very difficult. The only reason humans can read most handwriting is that we can understand the context of what is written.
Jeff realized that there was no way to easily reduce the error rate (which was very high for some letters). Instead of trying to develop a better algorithm (which I believe would be next to impossible given the computing resources at the time) he figured it would be better to change how people write letters to make it easier for the recognition engine.
Jeff tried to sell the idea of Graffiti to GRiD, but they decided they were not interested (by this time GRiD was part of Tandy corporation, hence its later downfall).
So rather than give up, Jeff left and founded Palm computing.
The beauty of his algorithm was that it worked fine, even on an 8088, whereas the competing algorithms from companies like Microsoft needed far more powerful processors.
Also, while at GRiD we worked with a Casio device called a "Zoomer" that had many similarities to the Palm. It had a PCMCIA slot, used a V20 CPU and ran DOS. On top of DOS it ran Geoworks with Jeff's handwriting recognition algorithm. While it was a really cool device (it had digital audio, IR, a serial port, and lasted 40 hours on a set of batteries) it was a bit too thick to fit into a shirt pocket. Also, the user interface was more mouse oriented than pen oriented (although Geoworks was quite cool).
Now all of this happened in the early 90's. I imagine that this was well before Xerox filed for a patent. I also know that Jeff was the originator of the algorithm and Graffiti and not Xerox.
I am a software engineer writing networking code that runs in embedded environments. No one in their right mind would choose Windows XP for embedded development for a number of reasons.
1. Embedded devices use a wide variety of microprocessors, memory architectures, and hardware, often with custom boot code. At least in the hard-core networking industry, X86 is not a big player since RISC processors tend to be a better fit (almost all networking protocols are big endian, for example). As far as I know MS is X86 only (not counting Intel's Unobtainium).
Also, many embedded products are based around non-Intel processors to cut costs and power requirements. There are a wide variety of RISC processors out there with varying levels of integration.
In the networking area I'm in (dealing with network processors from companies like Agere, Broadcom, Intel, MMC, Vitesse, etc.) most of the parts come with support for Linux and VxWorks. I do not see much support for Windows.
2. Embedded developers often need very low level access to the operating system. I doubt very much that MS will make available the source code to their OS. Where I work, we have the full source code to VxWorks ($$$, but worth it), which our product is based. This allows us great flexibility in terms of adding features or fixing bugs (what, bugs in commercial software???). If we had to wait for Wind River (or Microsoft for that matter) to fix a bug we'd never get anything done.
3. Linux is becomming very popular in the embedded area in part due to its open source and the licensing issues in many instances. Linux also has a proven track record of having been ported to a wide variety of hardware, from IBM mainframes to the Sega Dreamcast and other platforms. XP has no such track record.
4. Many embedded environments do not want features like IE, media player, or even a GUI. In the networking products I've worked on the only front end is either through a serial port or a telnet session.
Embedded developers are not your typical programmers. It's one thing to write an app in a nice GUI front-end and be able to step through the code in a debugger (like one can do with user-space apps). In the embedded world it's often the case that everything is running at the kernel layer.
As someone who works in the embedded networking area, we laugh at Microsoft. One reason embedded Linux is making inroads is that it supports a wide variety of processors and architectures. Windows XP will not run on Power PC, Mips, or a variety of other platforms. Also, it is much easier to write embedded code for Linux just because it is open source.
VxWorks is also popular (but it's very $$$ with full source) because it can run from a very small footprint. The last project I worked on had a flash footprint of around 1.6MB and ran quite well with 8MB of RAM (this was a L3 switch with a lot of additional proprietary software).
Many embedded areas don't care about wiz-bang user interfaces or multimedia. Also, with embedded Linux there are some nice alternatives. There is an embedded version of QT and KDE that do not require X Windows that includes a full-featured web browser that better follows the standards than IE.
Also, getting close to the hardware is much easier in Linux than Windows. Writing kernel loadable modules is a trivial task for Linux, and setting up communications between user and kernel drivers is also trivial (through ioctls or even the proc filesystem).
Linux also comes with a wealth of sample code from which to base a driver or application. With Windows you get whatever comes with the DDK, whereas with Linux you get the source to every non-proprietary driver available.
Embedded developers also like to have the full source code to everything. When something goes wrong, we don't have time to wait for a 3rd party to fix a bug. Can you imagine waiting for Microsoft to fix a bug that only affects a few people?
I know this isn't exactly related, to the topic, but I was wondering about what good tape backup software exists for Linux. Currently I am using kbackup and find it difficult to use. While tar is simple for backing up a directory, it isn't good for doing a full system backup when multiple tapes are required.
Are there any good open-source GUI-based tape backup programs for Linux? I really miss BackAgain/2 for OS/2 when using Linux.
As a former @Home (now AT&T) broadband user, since my email address changed I am no longer receiving the 30+ daily spams I got with @Home. I had my former @Home address for over 5 years and early on I wasn't as careful as I should have been about protecting it. I now use multiple email addresses, where I use an alternate address for services or postings which have the potential to be picked up by spammers.
I suspect that it's only a matter of time until my new email address becomes another toilet for the spammers to piss in.
I wired my older house with tons of CAT5E and coax (use RG6 for coax, BTW). I ran 1 to 2 lines from each room to a central closet where I also installed an AC outlet. In the closet I have a firewall router, cable modem, fast Ethernet switch, and a UPS ($5 at a swap).
This has worked out quite well. With this setup I can control the cable and network of each room from one central location.
This was an older home. It was fairly easy to do since I have a crawl space below the house and an attic above.
If there are places that will be difficult to reach later, put in some tough nylon string to feed wires or cable through later. You will be glad you did. Also, it never hurts to run extra cat5e, as you never know when you'll need it. Cat5e also makes excellent phone cable.
At the company I work for we've had very little trouble in terms of viruses and worms. Then again, they banned Outlook long ago. Whenever someone does manage to get hit, IT sends out a broadcast email warning everyone and telling everyone that this user was using banned software. A guy a couple cubes away ran Outlook and got hit with one of the worms. IT was there within minutes and immediately uninstalled Outlook, pulled the network connection, and left a nice note for the guy when he returned the next day.
It's too bad other companies don't wake up and just fix the problem to begin with... ban Outlook.
It's too bad that it can't use AC to refuel itself
by cracking H2O back into hydrogen to refill its fuel tanks when the AC is on. Now this would be cool.
This is exactly what the product I'm working on at Net.com does, but it's targeted at DSL (although it could work for Ethernet as well). The box can shape traffic independently going to each subscriber and the bandwidth can be changed on the fly via a web portal. Not only that, but it can shape the traffic coming from different sources at different rates. For example, it is possible to place a video server in the CO and a customer could order a movie and the bandwidth pipe between the video server and the subscriber would be opened up accordingly (say 4Mbps, while standard Internet (i.e. web) traffic would remain as it was (say, 512Kbps).
This is the way we see things going, with ISPs differentiating based on service. The more bandwidth you want, the more you pay. With a box like ours you can choose the bandwidth you want when you want and be charged according to the bandwidth you order.
You want to download a big MPEG file or a bunch of MP3's but your 384Kbps is too slow? Go to the portal and crank it up to 1Mbps and when done put it back to 384K.
Slashdot Mirror
I have repaired a couple of UPS. Every time it
just required that I replace the batteries.
I regularly see cheap =$20 UPS at the local Silicon Valley swaps where usually the only thing wrong is that they need new batteries.
I picked up a Tripplite Omnipro 1400VA UPS for $25 and a few months later picked up 5 batteries for $20. I also picked up a small Deltec UPS for $5 and a new battery for $12.
As for the Tripplite, it appeared totally dead when plugged in with the dead batteries that came with it.
I have a Philips DVX 8000 home theater PC which has been awesome. It looks like a piece of home theater equipment from the knobs to the flourescent display and the various buttons on the front. It supports multiple inputs and outputs, has 5.1 audio processing, and a very good line-doubler for all video sources. It is also a DVD player, has a fully programmable remote and keyboard, and runs Windows 95 (ugh). Fortunately almost everything still works when Windows has crashed.
It is powered by a lowly 233MHZ MMX Pentium CPU. Of course, all of the MPEG decoding for DVD and audio processing is handled by dedicated hardware so the CPU isn't involved for most operations and in fact for the most part it sits idle.
As for the PC part, I would be happier if it DIDN'T have the PC part. Trying to use it as a PC on a 32" monitor isn't all that great, since the monitor is only capable of 800x600.
I have yet to see another unit with similar capabilities. What I like about the line doubling support is that for DVD it is all done in the digital domain, without any analog between the MPEG decoder to the RGB output. The line doubling also looks great with video from other sources, without the usual line doubling artifacts caused by bob-and-weave line doubling.
As for the graphics card in the unit? It uses a Trident graphics chip which is fed video by a Genesis line doubler chip. A good video card is not necessarily a good graphics card since the two do very different operations.
With the 3Gb limit, how will we ever be able to keep up with Microsoft's security patches? This could be a national security threat.
Of course had you gone into the control center, click "Look & Feel" and "Launch Feedback" you can easily disable this. In addition there are numerous other look and feel settings to make KDE look and feel quite different than Windows.
I recently upgraded 3 different computers from SuSE 7.3 to 8.0. I had a number of issues with the original 7.3 but those were fixed with later patches (like the updated kernel).
/etc/rc.d and that doesn't work because /usr/sbin is not in the sudo path. If the user's path is being inherited this could cause security holes.
.0 release and needs some more polishing. If you don't need the bleeding edge support, stay with 7.3 and wait for 8.1.
SuSE 8.0, while mostly stable, has a number of annoying bugs.
1. sudo relies on the environment variables. I need sudo support to start and stop various services under
2. The upgrade on one system failed because 8.0 remapped one of the SCSI devices causing a failed mount of fstab. I had to go back and search the logs to figure out what went wrong. When problems occur I shouldn't have to go to the logs to find out what happened. It should have popped up a dialog or something. As it was, the mount was for my USB zip drive. If I could have told YAST to just ignore it I wouldn't have had to start over.
I also found that yast failed to properly upgrade everything. For example, on all of my systems gpg stopped working properly. I had to manually reinstall the rpm to fix this.
I also wish yast2 were more extensive. The firewall configuration could be improved, and many more modules are needed for configuring things like a DNS server (for my internal home LAN), an imap/pop mail server, a news server, and samba. Hardware configuration needs to add support for installing a CDRW drive.
I also found it a real PITA to get my CDRW working again after upgrading to 8.0. In addition the KDE tool I used before for burning CDs keeps crashing whenever I try and configure it.
I found that the video for Linux support is working much better than it did in earlier releases (although the 2.4.16 kernel upgrade for 7.3 was stable as well).
I also like KDE 3.0, which I am also running on Solaris at work.
During the initial upgrade I missed being able to do detailed selection of packages to install in the categorized way it was in 7.3.
Over all I am satisfied with SuSE 8.0, but I think it is a
I have an Epson Stylus Color 800 printer and am surprised at the high cost of replacement ink cartridges. There's no printhead in the cartridge or a chip. I have started buying non-OEM print cartridges and have found that often the quality is at least as good and in some cases much better than the original Epson cartridges. Not to mention that the alternatives are often a fraction the price.
I remember King's Quest I and bought it when it first came out. The very first version was for the IBM PCjr to show off its graphics and sound capability (it could do 320x200 with 16 colors! and had 3 channel sound+noise support) Remember at this time most PCs with color could only do 320x200 4 colors (CGA). What was amazing was that they could fit the entire game on a 360K floppy! The graphics were actually rendered at 160x200 so they could store the coordinates in a single byte to save space. As I recall, when entering a new screen the game would draw outlines of everything and fill them in.
Sierra also came out with the PC version which looked like crap by comparison (4 dithered colors in CGA).
I also remember the later Sierra games where you could get them into interesting modes for script debugging and so forth.
After I upgraded to an 8 MHz 8088 XT with EGA with a hard drive I wrote a program to get around the copy protection of all of the Sierra games so I didn't have to keep inserting the floppies every time I wanted to play. It would intercept the disk requests and simulate floppy accesses for the copy protection. Things have sure changed since then.
I have heard many arguments that the US adopting the Kyoto protocol for the reduction of greenhouse gasses is too expensive, unsafe, or inconvienent. For example, BP has already met the conditions of Kyoto in 4 years and in fact is now saving money because of it. DuPont has made even better progress, reducing their contribution of greenhouse gasses by 50%!
Cutting greenhouse gasses is not necessarily that difficult. For example, last year I added more insulation to my home. I saw a 25% reduction in natural gas because of this, and with various other improvements I have made over the last few years I am sure my home consumes far less energy than it did in 1991. The net result of these improvements to my house is that I spend less money on heating and lighting and in only a few years all of the changes will more than pay for themselves. Not only that, but with the added insulation my house is more comfortable.
Everyone seems to think that increasing the milage of cars is the most important step. While it is important and I believe easily doable with todays technology, many other areas are even easier.
How many of your homes have old furnaces and sub-standard insulation?
Perhapse if we had to pay the true cost of energy things would change. Here in California where we are stuck with outragiously high electricity costs (my bill is over $0.20/kwh) and very high gas prices, many people have taken advantage of methods to reduce energy usage. The state has helped as well by offering rebates. For example, it is now not unusual to buy a 100 watt equivelent compact flourescent light bulb for less than $1.
One doesn't have to be a rabid environmentalist to see the benefits from reducing greenhouse gasses. It also makes sense in the pocket book.
Actually the Hammer series adds 8 more general purpose registers and more SSE registers. Read up on X86-64, there's more to it than just going 64-bit. For example, there's better support for relocatable code (i.e. shared libraries).
I've been investigating this chip for embedded use and it looks like a great solution. It basically has just about everything one would need on a single chip... serial ports, Gig-E/Ethernet MACs, DDR ram controller, PCI controller, and more. It also has a Hypertransport interface where two chips can be connected back-to-back with a 12Gbps connection between them.
Broadcom also supports Hard-Hat Linux on this chip.
-Aaron
It's too bad they don't have a processor like this MIPS-based processor. 600MHz is pretty slow, even for MIPS.
Apparently the above processor is becomming popular for areas other than networking, its intended market.
-Aaron
One solution to the problem would be to establish randomly distributed honey pot computers which act as if they're infected by one of the various script-kiddie trojans. Log everything that happens to those computers, but do not allow those computers to actually perform DoS attacks (the script-kiddie probably won't know the difference).
After collecting evidence, the perpetrator should be fined and prosecuted. It would likely cost nothing to the tax payers since it could fund itself from the fines imposed on the perpetrators. If it's just a kid, then hold the parents responsible.
I agree that it's apathy. A couple of years ago I was running a Perl script that emulated a PC infected with Back Orifice. The script would simulate everything the user would normally do and log everything.
I once caught a user attempting to launch massive DoS attacks against other users by sending ping storms. Unknown to the attacker, nothing actually was going out. I notified the attacker's ISP (Quest) and the target's ISP. The target ISP was very thankful and said that they had noticed very high bandwidth directed to that user. The originating ISP could care less and refused to do anything, even after several more attempts by the same user.
I became frustrated and sent all the information on to the FBI.
Sadly I can no longer find the simulator script on the net (it was called Boobie).
While China may want to get into the software industry they have a major impediment. In India, all of the educated Indians speak fluent english. In China, this is not the case. In my group where I work most of the team is Indian. The only non-Indian people are myself, a Chinese fellow, and my boss (who's Canadian). For the most part there are no problems working with the Indian people (once I got used to their accent). However, the Chinese guy is another story. It is often very painful to try and explain basic concepts to him such that he understands what I am saying. I spent half an hour just getting him to understand that a parameter to my API was limited to 20 bits and to place the upper bits elsewhere. I felt like that father in Monty Python's Holy Grail telling the guards not to let his son leave the room.
He is an extreme example. I have worked with other Chinese software developers who have better English skills, yet the language issue always ends up becomming a problem.
One of the problems is that English is completely different from Chinese with absolutely nothing in common. Europe doesn't suffer this problem since all of the languages have many similarities in sentence structure, pronunciation, spelling, tense, character sets, and so on. Also, in Europe the schools have been teaching English for some time and there is no problem getting fluent English speaking teachers.
I imagine that China has a lot of difficulty getting fluent English teachers over there, especially with all of the opportunities in the Western world without the corruption or restrictions.
I've worked on computers running the German version of Windows NT. By the text I can still figure out what things are and how they relate to the English equivelent. It was not too difficult for me to reconfigure the networking on the boxes, even though I am not a Windows NT andministration expert). Also, if I had any problems, I could ask anyone since all of the Germans spoke fluent English. When I sat in front of a Chinese version once, I was totally lost.
Another major problem is China's lack of respect for IP. India has a fairly good reputation and many large companies have offices over there (i.e. Cisco, Nortel, etc.). China, OTH, has a very bad reputation. You can go anywhere and buy expensive pieces of software for only $1. Or for that matter, you can get a DVD of the latest movie within days after it's released into theaters long before the official releases come out.
Most companies are scared to death of their IP getting out. On this front the Chinese have a well deserved *very* bad reputation.
No company in their right mind would farm anything out to China.
While CUPS makes life easier, I still have a lot of problems with Linux. I run SuSE Linux 7.3 and have an Epson Stylus Color 800 printer connected to a networked print server (Netgear PS104). While Linux talks to the print server just fine, no matter what I try Linux takes 2 sheets of paper per page (yes, I have it configured for Letter and tried the inkjet letter settings). What is worse, however, is I get random horizontal lines on the page. Note that the output is going through Ghostscript.
Now, with the same setup, print server, etc. I have no problems when printing from OS/2 or Windows. Furthermore I do not get the horizontal lines (so it's not a cabling problem or a problem with the print server).
Now the Epson is a fairly common printer with well-documented control codes. I guess the only way to print properly in Linux is with a Postscript printer.
When I first joined @Home in Silicon Valley in the late 90's there was no upstream cap. I ran a TTCP benchmark between two cable modems as a test and saw over 640KBps (over 5Mbps). Just think, I could transfer all the memory I'd ever need in a second!
In addition, there was no policy about running servers and security was very lax. Network Neighborhood would show all of your clueless neighbor's computers hard drives.
Those were the days. Then they added a 1.5Mbps upstream cap, and I was still quite happy. Then they added the 128Kbps cap and from then on it sucked. Send a big email and forget browsing. FTP??? We're not supposed to upload anything.
Things have been very bad since AT&T took over. Last night at around midnight the ping latency to www.yahoo.com was between 500-1500 ms! Granted, no packets were lost, but my 28.8 modem would have been significantly faster than the cable modem.
I'll see what happens when Comcast takes over, but I'm very seriously considering DSL at this point. AT&T just plain sucks. Their cable service also sucks. They keep replacing good channels with infomercials.
-Aaron
Years ago in the early '90s I worked at a company called GRiD Systems where Jeff Hawkins got started with handwriting recognition. I believe he originally developed the technology as a graduate student and then licensed or sold it to GRiD, who made hand-held computers using his algorithms.
Handwriting recognition is extremely difficult for computers (hell, I often can't read my own handwriting). While his algorithm was good, it was not perfect. Trying to recognize the difference, for example between an "a" or a "d", or an "r" or an "n" is very difficult. The only reason humans can read most handwriting is that we can understand the context of what is written.
Jeff realized that there was no way to easily reduce the error rate (which was very high for some letters). Instead of trying to develop a better algorithm (which I believe would be next to impossible given the computing resources at the time) he figured it would be better to change how people write letters to make it easier for the recognition engine.
Jeff tried to sell the idea of Graffiti to GRiD, but they decided they were not interested (by this time GRiD was part of Tandy corporation, hence its later downfall).
So rather than give up, Jeff left and founded Palm computing.
The beauty of his algorithm was that it worked fine, even on an 8088, whereas the competing algorithms from companies like Microsoft needed far more powerful processors.
Also, while at GRiD we worked with a Casio device called a "Zoomer" that had many similarities to the Palm. It had a PCMCIA slot, used a V20 CPU and ran DOS. On top of DOS it ran Geoworks with Jeff's handwriting recognition algorithm. While it was a really cool device (it had digital audio, IR, a serial port, and lasted 40 hours on a set of batteries) it was a bit too thick to fit into a shirt pocket. Also, the user interface was more mouse oriented than pen oriented (although Geoworks was quite cool).
Now all of this happened in the early 90's. I imagine that this was well before Xerox filed for a patent. I also know that Jeff was the originator of the algorithm and Graffiti and not Xerox.
-Aaron
I am a software engineer writing networking code that runs in embedded environments. No one in their right mind would choose Windows XP for embedded development for a number of reasons.
1. Embedded devices use a wide variety of microprocessors, memory architectures, and hardware, often with custom boot code. At least in the hard-core networking industry, X86 is not a big player since RISC processors tend to be a better fit (almost all networking protocols are big endian, for example). As far as I know MS is X86 only (not counting Intel's Unobtainium).
Also, many embedded products are based around non-Intel processors to cut costs and power requirements. There are a wide variety of RISC processors out there with varying levels of integration.
In the networking area I'm in (dealing with network processors from companies like Agere, Broadcom, Intel, MMC, Vitesse, etc.) most of the parts come with support for Linux and VxWorks. I do not see much support for Windows.
2. Embedded developers often need very low level access to the operating system. I doubt very much that MS will make available the source code to their OS. Where I work, we have the full source code to VxWorks ($$$, but worth it), which our product is based. This allows us great flexibility in terms of adding features or fixing bugs (what, bugs in commercial software???). If we had to wait for Wind River (or Microsoft for that matter) to fix a bug we'd never get anything done.
3. Linux is becomming very popular in the embedded area in part due to its open source and the licensing issues in many instances. Linux also has a proven track record of having been ported to a wide variety of hardware, from IBM mainframes to the Sega Dreamcast and other platforms. XP has no such track record.
4. Many embedded environments do not want features like IE, media player, or even a GUI. In the networking products I've worked on the only front end is either through a serial port or a telnet session.
Embedded developers are not your typical programmers. It's one thing to write an app in a nice GUI front-end and be able to step through the code in a debugger (like one can do with user-space apps). In the embedded world it's often the case that everything is running at the kernel layer.
-Aaron
As someone who works in the embedded networking area, we laugh at Microsoft. One reason embedded Linux is making inroads is that it supports a wide variety of processors and architectures. Windows XP will not run on Power PC, Mips, or a variety of other platforms. Also, it is much easier to write embedded code for Linux just because it is open source.
VxWorks is also popular (but it's very $$$ with full source) because it can run from a very small footprint. The last project I worked on had a flash footprint of around 1.6MB and ran quite well with 8MB of RAM (this was a L3 switch with a lot of additional proprietary software).
Many embedded areas don't care about wiz-bang user interfaces or multimedia. Also, with embedded Linux there are some nice alternatives. There is an embedded version of QT and KDE that do not require X Windows that includes a full-featured web browser that better follows the standards than IE.
Also, getting close to the hardware is much easier in Linux than Windows. Writing kernel loadable modules is a trivial task for Linux, and setting up communications between user and kernel drivers is also trivial (through ioctls or even the proc filesystem).
Linux also comes with a wealth of sample code from which to base a driver or application. With Windows you get whatever comes with the DDK, whereas with Linux you get the source to every non-proprietary driver available.
Embedded developers also like to have the full source code to everything. When something goes wrong, we don't have time to wait for a 3rd party to fix a bug. Can you imagine waiting for Microsoft to fix a bug that only affects a few people?
-Aaron
I know this isn't exactly related, to the topic, but I was wondering about what good tape backup software exists for Linux. Currently I am using kbackup and find it difficult to use. While tar is simple for backing up a directory, it isn't good for doing a full system backup when multiple tapes are required.
Are there any good open-source GUI-based tape backup programs for Linux? I really miss BackAgain/2 for OS/2 when using Linux.
As a former @Home (now AT&T) broadband user, since my email address changed I am no longer receiving the 30+ daily spams I got with @Home. I had my former @Home address for over 5 years and early on I wasn't as careful as I should have been about protecting it. I now use multiple email addresses, where I use an alternate address for services or postings which have the potential to be picked up by spammers.
I suspect that it's only a matter of time until my new email address becomes another toilet for the spammers to piss in.
I wired my older house with tons of CAT5E and coax (use RG6 for coax, BTW). I ran 1 to 2 lines from each room to a central closet where I also installed an AC outlet. In the closet I have a firewall router, cable modem, fast Ethernet switch, and a UPS ($5 at a swap).
This has worked out quite well. With this setup I can control the cable and network of each room from one central location.
This was an older home. It was fairly easy to do since I have a crawl space below the house and an attic above.
If there are places that will be difficult to reach later, put in some tough nylon string to feed wires or cable through later. You will be glad you did. Also, it never hurts to run extra cat5e, as you never know when you'll need it. Cat5e also makes excellent phone cable.
-Aaron
At the company I work for we've had very little trouble in terms of viruses and worms. Then again, they banned Outlook long ago. Whenever someone does manage to get hit, IT sends out a broadcast email warning everyone and telling everyone that this user was using banned software. A guy a couple cubes away ran Outlook and got hit with one of the worms. IT was there within minutes and immediately uninstalled Outlook, pulled the network connection, and left a nice note for the guy when he returned the next day.
It's too bad other companies don't wake up and just fix the problem to begin with... ban Outlook.
It's too bad that it can't use AC to refuel itself
by cracking H2O back into hydrogen to refill its fuel tanks when the AC is on. Now this would be cool.
This is exactly what the product I'm working on at Net.com does, but it's targeted at DSL (although it could work for Ethernet as well). The box can shape traffic independently going to each subscriber and the bandwidth can be changed on the fly via a web portal. Not only that, but it can shape the traffic coming from different sources at different rates. For example, it is possible to place a video server in the CO and a customer could order a movie and the bandwidth pipe between the video server and the subscriber would be opened up accordingly (say 4Mbps, while standard Internet (i.e. web) traffic would remain as it was (say, 512Kbps).
This is the way we see things going, with ISPs differentiating based on service. The more bandwidth you want, the more you pay. With a box like ours you can choose the bandwidth you want when you want and be charged according to the bandwidth you order.
You want to download a big MPEG file or a bunch of MP3's but your 384Kbps is too slow? Go to the portal and crank it up to 1Mbps and when done put it back to 384K.
-Aaron