Good news. This will be nice to have in software, a 4096-byte sector. As others have said, this will exactly match the cluster size used by most modern filesystems, and the page size used by the Intel x86 architecture. This happy coincidence will mean that operating systems can just do a 1-to-1 read/write, and not need to waste time blocking/deblocking.
Isn't this already done in drive hardware? I thought, that in order to save physical space on the disc surface by reducing inter-sector overhead, the drive already internally uses much larger sectors than the current 512-byte standard. The disc controller just takes care of this automatically, doing the blocking/deblocking transparently. It reads/writes the larger sectors just fine, but just serves 512 bytes at a time to the computer, buffering the rest.
From a software point of view, 4096-byte sectors will be nice. I hope they take the time to get in a new partition table format! Drop the obsolete CHS fields, as they've been maxed out for a long time now. As for the LBA fields, widen them to 64 bits, so that there's plenty of room for the future. With 512-byte sectors and the current 32-bit LBA fields, the maximum is 4G sectors, 2TB. With RAID becoming popular these days, this limit is easily reachable! Going to 4096-byte sectors will push this limit back to 16TB, a good thing, but widening the fields to 64 bits will really push this limit out of sight. 72ZB. Maybe that will even be enough for Google?:)
Wow, I'm amazed. I have been out of the graphics card market for some time, and thought AGP cards had lost their value, as everything seems to be PCI-E these days.
I thought I had a collector's item, the fastest AGP video card ever made. No more.
"Back when I had money" I bought an Asus V9999 Ultra Deluxe. Exorbitant price. We're talking multiple arms and multiple legs here. Still, hard to find at that price. I Froogled it and was amazed to see that it is still holding its value! Wow. Very surprising. Maybe now's the time to sell it on Ebay?
My card smoked, though: a GeForce 6800 Ultra, 256MB, AGP 8X, and factory overclocked just a little. Rock-solid stable, once I got over a few initial tweaks having to deal with flaky Windows drivers.
As others have said, though, it would be kind of pointless to upgrade. The AGP bandwidth is maxed out. The CPU/motherboard is the bottleneck, since AGP systems are typically older systems. The newest chips all seem to prefer PCI-E motherboards. When I was playing games on my AGP system, the CPU was the bottleneck in almost all cases, even graphics-intensive games like Doom 3!
I've played several MMORPG's, and WoW is unfortunately the worst. The reason is that it uses TCP, instead of UDP like the others (Everquest, CoH, and so on).
With TCP, one dropped packet, and your connection lags out until the packet can be regenerated -- typically on the order of several seconds. More than one dropped packet every minute or so, and your game becomes unplayable. Have fun playing on a wireless connection or oversubscribed residential cable/DSL line!:)
I would love it if Blizzard would put some effort into making their network connection reliable under load. IMO, it is one of the few things Everquest got right. Due to the game's age, it was designed for people playing on analog modems, and as a result, its connection dropout-tolerance is quite robust.
[SCTP would be the best of both worlds, but unfortunately, it is still a pipe dream, and I'll save that rant for another day.]
I can often tell my line has hiccuped when my partner shouts "Oh crap" as his WoW screen freezes, even though my Web browsing and Internet radio playing continues on just fine for me:)
Don't forget, the DVD series of the show is already out now, in the UK (Region 2, PAL). Comes in a monstrously big TARDIS-shaped box that opens diagonally, kind of awkward, but clever. If it takes up too much space on your shelf, you can store the discs in those black plastic DVD cases that AOL spam-mailed out a while ago:)
Use your favorite multiregion DVD player to view it, or rip it first. No need to wait until later in the year.
I BitTorrented the shows as they came out, but bought the DVD because I wanted to support the show (and get the episodes in better quality). Still find it ironic that they say it's "coming soon" to the USA, but is already sitting here on my table....
I can see this being very useful to have around with you in the car, competing with satellite radio, and potentially offering more channels. Many car kits already include external speakers.
Sound quality over a cellphone may be rough, but this is made up for by the ability to have the music follow you as you get out of your car:)
Since Cingular themselves are offering this service, it would be absurd of them to charge for minutes/KB that it uses (I wouldn't put it past them, though). Cingular could offer an exemption from normal airtime billing for all minutes/KB used by this radio service. I'm assuming that this will be billed instead as a separate per-month flat charge, on the monthly bill.
Commercials could also subsidize the charges completely. If commercials could make enough money for Cingular to offset the cost of the airtime, then this competes with terrestrial radio as well (not just satellite radio)!
If Cingular also/instead gives the choice of normal minutes/KB billing for airtime used by this radio service, I could see this being a useful way to burn up excess minutes. I have 4000 minutes rolled over that I'll never use, Cingular expires minutes after 12 months, and I'm at 10 months now!
I'm hoping they have a wide selection of channels to choose from. I'd love to see more affiliates than just Music Choice (their selection of channels, already on most cable TV services, is pretty lame).
An acquaintance of mine's startup company tried to do this about 5 years ago, but failed.
The idea was to do basically what IBM has announced: automated spidering of websites, in an effort to automatically gauge public opinion about a given topic.
Intended uses were: corporate image monitoring, public relations, marketing, early warning about negative public opinion, politics, etc.
The distinctive thing is that it used natural language processing, in an effort to learn the tone and emotion behind the writings it discovered. The company was founded by a linguist.
This use of language is what distinguished it from being just a generic search engine or aggregator.
Unfortunately, the company failed. I don't think the startup had enough connections to reach their target market of large corporations. There also weren't enough blogs and other easily discoverable opinionated content online back then, as blogging was not yet mainstream.
Nowadays, IBM will have better luck on both counts, of course!
And unfortunately, I now forget the name of the startup, otherwise I'd dredge up some information and try to post it here....
Re:They aren't as dangerous as before
on
Ma Bell is Back
·
· Score: 4, Insightful
There's a reason for why they charge $50 to click "OK".
The $50-or-so price is cast in stone, as a tariffed rate!
Back about 15 years ago, when the price was merely $33 for flipping a switch (no fancy "OK" buttons to click here), a family friend of ours got a phone line activated.
Turns out, the wires were too ratty/old to hold voice service: static, buzzing, dropped calls, and the like.
The phone company came out, and ran over ONE MILE of new wiring, including telephone poles, through a forest, just to reach his house!
This was in a small little rural town, as you might have guessed. No way would he have been able to pay the true market rate for the labor/equipment to install the phone line, which I guess would have cost at least $10,000.00 if he had hired a crew to do it privately. "Universal service" at work!
This is why you're paying $50 for them to hit a button: the cost to you, and essentially everybody else, was $0.05 for 15 seconds of a call-center employee's time. It's just these rare exceptions, that bring the average subsidized rate up to $50 or so.
And, no, the phone company will NOT give him DSL service today, nor install a second voice line. I wonder why?:)
It sucks to lose a name that 99.999% of players have no problem with, when you have the bad luck of running into a GM who's had a bad day and is feeling vindictive. This happened to me in EQ1: during the middle of a raid, a GM appeared and made one of our GUILD LEADERS change their player name!! Much confusion, and scramble to update friends and other lists, ensued.
(The GM's name was Rybarin, on The Nameless server. BTW, I no longer play EQ1 or have an active account on any Sony multiplayer game.)
I see 3 technical solutions to this problem:
1) Behind the scenes, the servers should keep track of players by ID numbers, not names. ID numbers are permanent, unique, and never change. When a player is renamed for any reason, the ID number would remain unchanged. All features in the game, such as friends, ignore, membership, etc. would still work, and automatically pick up the new name! (If there's anybody at Blizzard who's ever used a database, this is called a "primary key".)
2) Names that are forcefully changed by GM's should be added to a permanent list of blacklisted names. After a name has been banned by a GM, no future players could be created under that name, so newbies and griefers wouldn't be able to immediately reuse the name (essentially the same as IRC nick-stealing when a server rejoins after a split).
3) Players could have a means of reporting a vindictive/unfair GM, perhaps by voting. If a GM gets enough votes from unique paid-for ACCOUNTS (not players), this triggers a flag somewhere and that GM gets looked at more closely by their superiors (server admins, higher GM's, whatever). This could make for interesting in-game politics:)
1. Install VMware on Linux. 2. Install some random older version of Windows into VMware. 3. Take checkpoint within VMware. 4. Insert copy-protected CD in VMware, let it install its DRM crap. 5. Play copy-protected CD in VMware. 6. Divert audio output using patched Linux audio driver, and save to disk. 7. Eject CD when complete, restore checkpoint in VMware to completely remove all trace of DRM crap. Nice! As you said, it only takes 1 copy to be released, so DRM of any sort is pointless really....
Notice that when you use MSIE on Windows, it shows you the true URL of the site you are downloading from. In the download box, it will show you the URL it's downloading from, and you can see Mozilla's choice of mirrors around the world.
With Firefox, however, you don't get to see this by default. It just shows the basename of the file you are downloading, not the full URL containing the hostname and directory path. By right-clicking on the progress bar in the Downloads popup window, and choosing Properties, you can then view the true URL, but many users don't know about this.
If the user has turned on the "Ask me where to save every file" option, the popup file-chooser window also unfortunately does not show the true URL. It would be an ideal place to show it in this window, as there seems to be plenty of room there.
Right now, I have to download the file multiple times, open the Properties to make sure I'm getting a different mirror, and then diff the files to make sure they're the same, before I can consider them trustworthy enough to install.
By itself, this is just a nitpick, but it turns into a nasty bug when combined with other things:
1) The user not being able to easily see the true originating URL of a file, before making the download decision
2) Mozilla's decision to use a huge variety of seemingly random sites as mirrors, some more questionable than others
3) Mozilla's decision to not have any way whatsoever of verifying the integrity of the download, such as a cryptographic signature
Put all three together, and it's virus time!
Microsoft: Smug Mode.
With the large numbers of mirrors Mozilla uses, spread throughout the world, the odds of someone sneaking malware in there (either by ignorance, hacking, or a good old-fashioned bribe) is quite high.
The solution probably lies in a plugin. If there's not already a plugin to let the user plainly see the true URL and verify where files are coming from, it should be made (I wish I knew how). The plugin should also have some cryptographic method of verifying a downloaded file, and Mozilla should sign all releases with a strong key. It's just basic common sense, and I'm shocked Mozilla hasn't done this already.
I remember playing the original SimCity for the Amiga, in high school, around the time of the first Gulf War. A bad floppy caused the game world to load up one time absolutely blank, with all empty terrain (not even grass). We happily saved that map as "Baghdad":)
MPI is already very good at converting data between the various computers involved in a parallel MPI program.
There's almost an absurd number of datatype declaration, conversion, etc. functions in MPI. If you properly set up MPI_Datatype types to hold your data, then the MPI library will be able to handle it all internally. Then, when sending and receiving messages, it will automatically do conversions as needed (between big-endian and little-endian machines, and so on).
So the problem isn't one of sending/receiving data between machines of differing architecture. The problem is writing this data to a file, and then reading it in again at a later date, possibly on a different machine. This is a harder problem.
The MPI I/O extensions (part of MPI-2) tried to address this somewhat. There is a file format "external32" in the spec, that was supposed to be universal, with a standard encoding for all datatypes, and so on. However, evidently it was never implemented fully, as I haven't been able to find it.
I take it you've aleady read section 7.5 in MPIv2. If you haven't, now's the time!
Unfortunately, I know of no other MPI I/O implementations, other than ROMIO, that can simply be plugged into an existing MPI stack. You might want to ask around at the new project OpenMPI, a new-from-the-ground-up MPI implementation that is currently in development. I'd be curious to learn the level of MPI I/O support that they claim!
Assuming you are stuck with a MPI stack that only supports the "native" representation, the problem you face becomes one of data representation in general. As you know, there's bajillions of different ways of storing floating-point numbers, and if you write them to disk, the files will be only valid for exactly that CPU.
As a last resort, a brute-force solution is to write the numbers as human-readable text, and then parse them in again accoringly. It's a waste of file space, but there's no ambiguity in the datatype representation, and it is very tolerant of floating point differences between machines.
-1.2345234523452345 2.345634563456365e+13 -3.2121212121e-24 And so on.
This shouldn't be much of a hotspot in your code, since ideally it would only be done at start, stop, and checkpoint time. Also, if you need paralellism, and don't care about wasted file space or future precision improvements, you could use a fixed-length string for each number (with much padding), thus allowing you to read your numbers random-access instead of sequential.
BOINC has built-in caching (like SetiQueue, but built-in to the client). You have to enable it on the website, which is rather odd, and hope your client picks up the new preferences during its next resync. BOINC will then download a lot of data, and keep your CPU's busy even during downtime. My CPU's are still running old SETI units that BOINC has cached!
And, there are other BOINC projects you can choose, besides SETI.
Or, you can save a watt or two. I checked, and my best computer uses 110 watts when idle, and 160 watts when running SETI! That is a significant difference.
Personally, I think funds should be spent on longer hours for libraries, before getting the latest computer toys.
In San Jose, California, we have a new downtown library that's hooked up like you wouldn't believe. It's not open enough hours for the public to truly use it well, though. Fortunately, the library is jointly owned by the nearby college, and the college funds additional hours during the school year. Extended hours at the library are quite convenient because most downtown parking in San Jose becomes free after 6pm!
Unfortunately with government projects it's often easier to get money for new construction/projects instead of maintenance. New toys are sexy, and sexiness gets votes.
If the funding for your library is with strings attached, and those strings have to be spent on new computer technology, I suggest these:
* Free Wi-Fi everywhere in the library and as far into the surrounding areas as your access points can reach, if you don't already have this.
* CD-burning kiosks that burn CD's full of public domain books, from the Gutenberg Project and other sources.
* Similarly, DVD-burning kiosks that burn DVD's full of public domain videos/movies, from the Prelinger Archives and other sources.
* Book-on-demand printing presses for public domain books, something like this!
This is a real shame, and a waste. What was the school district thinking?
1000 laptops worth $500-ish and sold for $50 = $450,000 that the school district just threw away. Over here in California, our school districts could sure use that money!
I wonder what the public cost was, for the extra law enforcement required at this riot? When the injured people sue?
I hope all those reading about this in the Henrico County papers get a lesson in economics, supply, and demand -- at least people there will then learn something from their school district!
My guess is that someone had no clue what the laptops were really worth (heads should roll, in this case), or there was some strange bureaucratic rule capping the price they could charge, or most likely, there was "interesting" accounting going on with regards to depreciation and tax writeoffs (they shouldn't have to do this, since they're a nonprofit school district!).
If they were determined to sell at a price that is ridiculously below market like this, then at a minimum, they should have done nothing on that day except hand out tickets, with the drawing to take place once the crowd disperses (perhaps the next day).
In this day and age, there's no way people should sell desirable items below market price unless the customer base is strictly limited somehow (lottery drawing, prequalification, etc.), due to the ease of "scalpers" flipping them on Ebay!
A 2D game that has a plane that loops around clockwise/counterclockwise?
Could be "Combat" for Atari 2600 or "Two Tigers" for arcade.
If the destroyed enemies crash down as kamikaze, and the long-term goal of the game is to tear through a large obstacle (carrier ship, etc.) by crashing enough destroyed enemies into it, then it's definitely "Two Tigers"!
Give it a try on MAME if you get the chance. One of my favorite arcade games. Got fairly good at it, too....
BTW, it's surreal to grow up playing a computer game that's a clone of some arcade machine, not knowing at the time that it's a clone -- then seeing the arcade game in real life, years after the fact. I was blown away to finally see a "Firepower" pinball machine decades later, after playing "Raster Blaster" as a child!
Can't blame me for that one, though, I worked on just the ANA-7711 cards (TOE only, without any additional iSCSI support). At Adaptec, there was an internal division between TOE-only cards and iSCSI cards, and I was not part of the iSCSI group.
The ANA-7711 cards didn't accelerate anything past layer 4, so if you wanted encryption and iSCSI, the CPU still had to handle that.
We tested and solved many data corruption issues, and interactions with TCP stacks on various other operating systems. To my knowledge, no ANA-7711 cards ever shipped with data corruption problems.
Unfortunately, just this year they laid off my last friend I knew who was still working there. It's a real shame what has happened to Adaptec over the last few years. There were, and may still be, some really good engineers working there....
Accelerating Ethernet in hardware, while remaining 100% compatible with the standard protocols on the wire, isn't all that new. Just over 2 years ago, I worked on a TOE (TCP offload engine) card at Adaptec.
It was a complete TCP stack in hardware (with the exception of startup/teardown, which still was intentionally done in software, for purposes of security/accounting).
Once the TCP connection was established, the packets were completely handled in hardware, and the resulting TCP payload data was DMA'ed directly to the application's memory when a read request was made. Same thing in the other direction, for a write request. Very fast!
I'm not sure of the exact numbers but we reduced CPU utilization to around 10%-20% of what it was under a non-accelerated card, and were able to saturate the wire in both directions using only a 1.0Ghz CPU. This is something that was difficult to do, given the common rule of thumb that you need 1Mhz of CPU speed to handle every 1Mbit of data on the wire.
To make a long story short, it didn't sell, and I (among many others) was laid off.
The reason was mostly about price/performance: who would pay that much for just a gigabit ethernet card? The money that was spent on a TOE-accelerated network card would be better spent on a faster CPU in general, or a more specialized interconnect such as InfiniBand.
When 10Gb Ethernet becomes a reality, we will once again need TOE-accelerated network cards (since there are no 10GHz CPU's today, as we seem to have hit a wall at around 4Ghz). I'd keep my eye on Chelsio: of the Ethernet TOE vendors still standing, they seem to have a good product.
BTW, did you know that 10Gb Ethernet is basically "InfiniBand lite"? Take InfiniBand, drop the special upper-layer protocols so that it's just raw packets on the wire, treat that with the same semantics as Ethernet, and you have 10GbE. I can predict that Ethernet and InfiniBand will conceptually merge, sometime in the future. Maybe Ethernet will become a subset of InfiniBand, like SATA is a subset of SAS....
About seamlessly combining raster and vector image editing: It's nothing new.
I miss SuperPaint, by Silicon Beach Software, for the old black-and-white Macintosh.
It had the *perfect* combination of raster and vector painting. It also had a good balance between photo-editing features (aka Photoshop) and tools for creating a new image from scratch (aka DeluxePaint).
At the time it was written, there were no color Macintoshes, but there was still some functions in the original QuickDraw API to deal with color (mostly to support multicolored ribbons in Apple ImageWriter printers). Icons for colors in SuperPaint just had words like "Orange", "Blue", etc. that would just show up black on the screen. The amazing thing is, when the first color Macintosh II came out, I tried out SuperPaint on it, and it WORKED!! It was so well written that everything on screen was now in color, even though the developers had no way of testing it before! Amazing....
SuperPaint was eventually bought out by Aldus, and after that, it just wasn't the same. Adobe eventually bought Aldus, and SuperPaint faded into obscurity. (What is with Adobe? The same thing happened in the Windows world just a few years ago, when we lost Cool Edit, essentially the only high-quality audio editing program that was within the budget of the average home user.)
I use and prefer Lite-On drives. They are cheap and standards-compliant. The same can't be said for Plextor (although they have improved in recent years).
What really put me over the fence for Lite-On is the freeware (closed source) program KProbe. This works only with Lite-On drives, and is only for Windows (unfortunately). Even though it is closed source, it is a free download (unlike Plextor's commercial utility).
The program KProbe seems to perform similar functions to what these Plextor utilities do: show the true low-level bit error rates of the disc, and allow certain drive settings to be overridden as desired (minimum and maximum speeds, DVD+R booktype, DVD region, and so on).
With the ability for end users to get at this information, it becomes possible to make informed decisions when buying blank DVD media.
If only these low-level functions were standardized, *sigh* -- it would be great if mainstream DVD and CD-burning software could use them to check disc reliability, and the OS could even inform the user if a disc were about to fail.
With more and more people storing home movies and other keepsakes on DVD these days, this will become more important. The ability to see this low-level info has saved me from losing data several times due to cheap media: beware "Great Quality" from Fry's!
Slashdot Mirror
Good news. This will be nice to have in software, a 4096-byte sector. As others have said, this will exactly match the cluster size used by most modern filesystems, and the page size used by the Intel x86 architecture. This happy coincidence will mean that operating systems can just do a 1-to-1 read/write, and not need to waste time blocking/deblocking.
:)
Isn't this already done in drive hardware? I thought, that in order to save physical space on the disc surface by reducing inter-sector overhead, the drive already internally uses much larger sectors than the current 512-byte standard. The disc controller just takes care of this automatically, doing the blocking/deblocking transparently. It reads/writes the larger sectors just fine, but just serves 512 bytes at a time to the computer, buffering the rest.
From a software point of view, 4096-byte sectors will be nice. I hope they take the time to get in a new partition table format! Drop the obsolete CHS fields, as they've been maxed out for a long time now. As for the LBA fields, widen them to 64 bits, so that there's plenty of room for the future. With 512-byte sectors and the current 32-bit LBA fields, the maximum is 4G sectors, 2TB. With RAID becoming popular these days, this limit is easily reachable! Going to 4096-byte sectors will push this limit back to 16TB, a good thing, but widening the fields to 64 bits will really push this limit out of sight. 72ZB. Maybe that will even be enough for Google?
Wow, I'm amazed. I have been out of the graphics card market for some time, and thought AGP cards had lost their value, as everything seems to be PCI-E these days.
:)
I thought I had a collector's item, the fastest AGP video card ever made. No more.
"Back when I had money" I bought an Asus V9999 Ultra Deluxe. Exorbitant price. We're talking multiple arms and multiple legs here. Still, hard to find at that price. I Froogled it and was amazed to see that it is still holding its value! Wow. Very surprising. Maybe now's the time to sell it on Ebay?
My card smoked, though: a GeForce 6800 Ultra, 256MB, AGP 8X, and factory overclocked just a little. Rock-solid stable, once I got over a few initial tweaks having to deal with flaky Windows drivers.
As others have said, though, it would be kind of pointless to upgrade. The AGP bandwidth is maxed out. The CPU/motherboard is the bottleneck, since AGP systems are typically older systems. The newest chips all seem to prefer PCI-E motherboards. When I was playing games on my AGP system, the CPU was the bottleneck in almost all cases, even graphics-intensive games like Doom 3!
Make offer if interested
I've played several MMORPG's, and WoW is unfortunately the worst. The reason is that it uses TCP, instead of UDP like the others (Everquest, CoH, and so on).
:)
:)
With TCP, one dropped packet, and your connection lags out until the packet can be regenerated -- typically on the order of several seconds. More than one dropped packet every minute or so, and your game becomes unplayable. Have fun playing on a wireless connection or oversubscribed residential cable/DSL line!
I would love it if Blizzard would put some effort into making their network connection reliable under load. IMO, it is one of the few things Everquest got right. Due to the game's age, it was designed for people playing on analog modems, and as a result, its connection dropout-tolerance is quite robust.
[SCTP would be the best of both worlds, but unfortunately, it is still a pipe dream, and I'll save that rant for another day.]
I can often tell my line has hiccuped when my partner shouts "Oh crap" as his WoW screen freezes, even though my Web browsing and Internet radio playing continues on just fine for me
Don't forget, the DVD series of the show is already out now, in the UK (Region 2, PAL). Comes in a monstrously big TARDIS-shaped box that opens diagonally, kind of awkward, but clever. If it takes up too much space on your shelf, you can store the discs in those black plastic DVD cases that AOL spam-mailed out a while ago :)
Use your favorite multiregion DVD player to view it, or rip it first. No need to wait until later in the year.
I BitTorrented the shows as they came out, but bought the DVD because I wanted to support the show (and get the episodes in better quality). Still find it ironic that they say it's "coming soon" to the USA, but is already sitting here on my table....
I'm getting deja vu here....
:)
Did anybody else read Ray Bradbury's short story, "The Rocket", many years ago?
It's the final story in his compilation novel _The_Illustrated_Man_.
I won't give away any spoilers, but after reading this story, I feel like I don't need to watch this TV show at all.
I can see this being very useful to have around with you in the car, competing with satellite radio, and potentially offering more channels. Many car kits already include external speakers.
:)
Sound quality over a cellphone may be rough, but this is made up for by the ability to have the music follow you as you get out of your car
Since Cingular themselves are offering this service, it would be absurd of them to charge for minutes/KB that it uses (I wouldn't put it past them, though). Cingular could offer an exemption from normal airtime billing for all minutes/KB used by this radio service. I'm assuming that this will be billed instead as a separate per-month flat charge, on the monthly bill.
Commercials could also subsidize the charges completely. If commercials could make enough money for Cingular to offset the cost of the airtime, then this competes with terrestrial radio as well (not just satellite radio)!
If Cingular also/instead gives the choice of normal minutes/KB billing for airtime used by this radio service, I could see this being a useful way to burn up excess minutes. I have 4000 minutes rolled over that I'll never use, Cingular expires minutes after 12 months, and I'm at 10 months now!
I'm hoping they have a wide selection of channels to choose from. I'd love to see more affiliates than just Music Choice (their selection of channels, already on most cable TV services, is pretty lame).
An acquaintance of mine's startup company tried to do this about 5 years ago, but failed.
The idea was to do basically what IBM has announced: automated spidering of websites, in an effort to automatically gauge public opinion about a given topic.
Intended uses were: corporate image monitoring, public relations, marketing, early warning about negative public opinion, politics, etc.
The distinctive thing is that it used natural language processing, in an effort to learn the tone and emotion behind the writings it discovered. The company was founded by a linguist.
This use of language is what distinguished it from being just a generic search engine or aggregator.
Unfortunately, the company failed. I don't think the startup had enough connections to reach their target market of large corporations. There also weren't enough blogs and other easily discoverable opinionated content online back then, as blogging was not yet mainstream.
Nowadays, IBM will have better luck on both counts, of course!
And unfortunately, I now forget the name of the startup, otherwise I'd dredge up some information and try to post it here....
There's a reason for why they charge $50 to click "OK".
:)
The $50-or-so price is cast in stone, as a tariffed rate!
Back about 15 years ago, when the price was merely $33 for flipping a switch (no fancy "OK" buttons to click here), a family friend of ours got a phone line activated.
Turns out, the wires were too ratty/old to hold voice service: static, buzzing, dropped calls, and the like.
The phone company came out, and ran over ONE MILE of new wiring, including telephone poles, through a forest, just to reach his house!
This was in a small little rural town, as you might have guessed. No way would he have been able to pay the true market rate for the labor/equipment to install the phone line, which I guess would have cost at least $10,000.00 if he had hired a crew to do it privately. "Universal service" at work!
This is why you're paying $50 for them to hit a button: the cost to you, and essentially everybody else, was $0.05 for 15 seconds of a call-center employee's time. It's just these rare exceptions, that bring the average subsidized rate up to $50 or so.
And, no, the phone company will NOT give him DSL service today, nor install a second voice line. I wonder why?
It sucks to lose a name that 99.999% of players have no problem with, when you have the bad luck of running into a GM who's had a bad day and is feeling vindictive. This happened to me in EQ1: during the middle of a raid, a GM appeared and made one of our GUILD LEADERS change their player name!! Much confusion, and scramble to update friends and other lists, ensued.
:)
(The GM's name was Rybarin, on The Nameless server. BTW, I no longer play EQ1 or have an active account on any Sony multiplayer game.)
I see 3 technical solutions to this problem:
1) Behind the scenes, the servers should keep track of players by ID numbers, not names. ID numbers are permanent, unique, and never change. When a player is renamed for any reason, the ID number would remain unchanged. All features in the game, such as friends, ignore, membership, etc. would still work, and automatically pick up the new name! (If there's anybody at Blizzard who's ever used a database, this is called a "primary key".)
2) Names that are forcefully changed by GM's should be added to a permanent list of blacklisted names. After a name has been banned by a GM, no future players could be created under that name, so newbies and griefers wouldn't be able to immediately reuse the name (essentially the same as IRC nick-stealing when a server rejoins after a split).
3) Players could have a means of reporting a vindictive/unfair GM, perhaps by voting. If a GM gets enough votes from unique paid-for ACCOUNTS (not players), this triggers a flag somewhere and that GM gets looked at more closely by their superiors (server admins, higher GM's, whatever). This could make for interesting in-game politics
Krellan
Or....
1. Install VMware on Linux. 2. Install some random older version of Windows into VMware. 3. Take checkpoint within VMware. 4. Insert copy-protected CD in VMware, let it install its DRM crap. 5. Play copy-protected CD in VMware. 6. Divert audio output using patched Linux audio driver, and save to disk. 7. Eject CD when complete, restore checkpoint in VMware to completely remove all trace of DRM crap. Nice! As you said, it only takes 1 copy to be released, so DRM of any sort is pointless really....
This has been a worry of mine for some time.
Notice that when you use MSIE on Windows, it shows you the true URL of the site you are downloading from. In the download box, it will show you the URL it's downloading from, and you can see Mozilla's choice of mirrors around the world.
With Firefox, however, you don't get to see this by default. It just shows the basename of the file you are downloading, not the full URL containing the hostname and directory path. By right-clicking on the progress bar in the Downloads popup window, and choosing Properties, you can then view the true URL, but many users don't know about this.
If the user has turned on the "Ask me where to save every file" option, the popup file-chooser window also unfortunately does not show the true URL. It would be an ideal place to show it in this window, as there seems to be plenty of room there.
Right now, I have to download the file multiple times, open the Properties to make sure I'm getting a different mirror, and then diff the files to make sure they're the same, before I can consider them trustworthy enough to install.
By itself, this is just a nitpick, but it turns into a nasty bug when combined with other things:
1) The user not being able to easily see the true originating URL of a file, before making the download decision
2) Mozilla's decision to use a huge variety of seemingly random sites as mirrors, some more questionable than others
3) Mozilla's decision to not have any way whatsoever of verifying the integrity of the download, such as a cryptographic signature
Put all three together, and it's virus time!
Microsoft: Smug Mode.
With the large numbers of mirrors Mozilla uses, spread throughout the world, the odds of someone sneaking malware in there (either by ignorance, hacking, or a good old-fashioned bribe) is quite high.
The solution probably lies in a plugin. If there's not already a plugin to let the user plainly see the true URL and verify where files are coming from, it should be made (I wish I knew how). The plugin should also have some cryptographic method of verifying a downloaded file, and Mozilla should sign all releases with a strong key. It's just basic common sense, and I'm shocked Mozilla hasn't done this already.
Thanks for posting this! I had heard of a third levee breach, but wasn't aware of its location.
Here are Google Maps links for two levee breaches.
1 7275,-90.121467&spn=0.002356,0.002879&t=e&hl=en
7 2056,-90.023303&spn=0.009430,0.011517&t=e&hl=en
http://maps.google.com/maps?q=new+orleans&ll=30.0
http://maps.google.com/maps?q=new+orleans&ll=29.9
Once more into the... never mind!
That's cute....
:)
I remember playing the original SimCity for the Amiga, in high school, around the time of the first Gulf War. A bad floppy caused the game world to load up one time absolutely blank, with all empty terrain (not even grass). We happily saved that map as "Baghdad"
Kind of morbid, but all in fun.
The Salvation Army is anti-gay. That is unfortunate.
t m
1 61
i d=354
http://www.commondreams.org/headlines01/1113-06.h
http://outrage.nabumedia.com/pressrelease.asp?ID=
http://www.glaad.org/media/np_archive_detail.php?
Please keep this in mind when you consider donating to them. Thank you.
Josh
MPI is already very good at converting data between the various computers involved in a parallel MPI program.
There's almost an absurd number of datatype declaration, conversion, etc. functions in MPI. If you properly set up MPI_Datatype types to hold your data, then the MPI library will be able to handle it all internally. Then, when sending and receiving messages, it will automatically do conversions as needed (between big-endian and little-endian machines, and so on).
So the problem isn't one of sending/receiving data between machines of differing architecture. The problem is writing this data to a file, and then reading it in again at a later date, possibly on a different machine. This is a harder problem.
The MPI I/O extensions (part of MPI-2) tried to address this somewhat. There is a file format "external32" in the spec, that was supposed to be universal, with a standard encoding for all datatypes, and so on. However, evidently it was never implemented fully, as I haven't been able to find it.
I take it you've aleady read section 7.5 in MPIv2. If you haven't, now's the time!
Unfortunately, I know of no other MPI I/O implementations, other than ROMIO, that can simply be plugged into an existing MPI stack. You might want to ask around at the new project OpenMPI, a new-from-the-ground-up MPI implementation that is currently in development. I'd be curious to learn the level of MPI I/O support that they claim!
Assuming you are stuck with a MPI stack that only supports the "native" representation, the problem you face becomes one of data representation in general. As you know, there's bajillions of different ways of storing floating-point numbers, and if you write them to disk, the files will be only valid for exactly that CPU.
As a last resort, a brute-force solution is to write the numbers as human-readable text, and then parse them in again accoringly. It's a waste of file space, but there's no ambiguity in the datatype representation, and it is very tolerant of floating point differences between machines.
-1.2345234523452345
2.345634563456365e+13
-3.2121212121e-24
And so on.
This shouldn't be much of a hotspot in your code, since ideally it would only be done at start, stop, and checkpoint time. Also, if you need paralellism, and don't care about wasted file space or future precision improvements, you could use a fixed-length string for each number (with much padding), thus allowing you to read your numbers random-access instead of sequential.
Hope this helps!
Josh
BOINC has built-in caching (like SetiQueue, but built-in to the client). You have to enable it on the website, which is rather odd, and hope your client picks up the new preferences during its next resync. BOINC will then download a lot of data, and keep your CPU's busy even during downtime. My CPU's are still running old SETI units that BOINC has cached!
And, there are other BOINC projects you can choose, besides SETI.
Or, you can save a watt or two. I checked, and my best computer uses 110 watts when idle, and 160 watts when running SETI! That is a significant difference.
Personally, I think funds should be spent on longer hours for libraries, before getting the latest computer toys.
In San Jose, California, we have a new downtown library that's hooked up like you wouldn't believe. It's not open enough hours for the public to truly use it well, though. Fortunately, the library is jointly owned by the nearby college, and the college funds additional hours during the school year. Extended hours at the library are quite convenient because most downtown parking in San Jose becomes free after 6pm!
Unfortunately with government projects it's often easier to get money for new construction/projects instead of maintenance. New toys are sexy, and sexiness gets votes.
If the funding for your library is with strings attached, and those strings have to be spent on new computer technology, I suggest these:
* Free Wi-Fi everywhere in the library and as far into the surrounding areas as your access points can reach, if you don't already have this.
* CD-burning kiosks that burn CD's full of public domain books, from the Gutenberg Project and other sources.
* Similarly, DVD-burning kiosks that burn DVD's full of public domain videos/movies, from the Prelinger Archives and other sources.
* Book-on-demand printing presses for public domain books, something like this!
Good luck with your funding!
This is a real shame, and a waste. What was the school district thinking?
1000 laptops worth $500-ish and sold for $50 = $450,000 that the school district just threw away. Over here in California, our school districts could sure use that money!
I wonder what the public cost was, for the extra law enforcement required at this riot? When the injured people sue?
I hope all those reading about this in the Henrico County papers get a lesson in economics, supply, and demand -- at least people there will then learn something from their school district!
My guess is that someone had no clue what the laptops were really worth (heads should roll, in this case), or there was some strange bureaucratic rule capping the price they could charge, or most likely, there was "interesting" accounting going on with regards to depreciation and tax writeoffs (they shouldn't have to do this, since they're a nonprofit school district!).
If they were determined to sell at a price that is ridiculously below market like this, then at a minimum, they should have done nothing on that day except hand out tickets, with the drawing to take place once the crowd disperses (perhaps the next day).
In this day and age, there's no way people should sell desirable items below market price unless the customer base is strictly limited somehow (lottery drawing, prequalification, etc.), due to the ease of "scalpers" flipping them on Ebay!
This is very shameful and a real waste....
A 2D game that has a plane that loops around clockwise/counterclockwise?
Could be "Combat" for Atari 2600 or "Two Tigers" for arcade.
If the destroyed enemies crash down as kamikaze, and the long-term goal of the game is to tear through a large obstacle (carrier ship, etc.) by crashing enough destroyed enemies into it, then it's definitely "Two Tigers"!
Give it a try on MAME if you get the chance. One of my favorite arcade games. Got fairly good at it, too....
BTW, it's surreal to grow up playing a computer game that's a clone of some arcade machine, not knowing at the time that it's a clone -- then seeing the arcade game in real life, years after the fact. I was blown away to finally see a "Firepower" pinball machine decades later, after playing "Raster Blaster" as a child!
Wow, that's nasty, and an embarassment.
Can't blame me for that one, though, I worked on just the ANA-7711 cards (TOE only, without any additional iSCSI support). At Adaptec, there was an internal division between TOE-only cards and iSCSI cards, and I was not part of the iSCSI group.
The ANA-7711 cards didn't accelerate anything past layer 4, so if you wanted encryption and iSCSI, the CPU still had to handle that.
We tested and solved many data corruption issues, and interactions with TCP stacks on various other operating systems. To my knowledge, no ANA-7711 cards ever shipped with data corruption problems.
Unfortunately, just this year they laid off my last friend I knew who was still working there. It's a real shame what has happened to Adaptec over the last few years. There were, and may still be, some really good engineers working there....
Accelerating Ethernet in hardware, while remaining 100% compatible with the standard protocols on the wire, isn't all that new. Just over 2 years ago, I worked on a TOE (TCP offload engine) card at Adaptec.
l ymatrix.html?cat=%2fTechnology%2fNAC+Cards%2fNAC+C ards
http://www.adaptec.com/worldwide/product/prodfami
It was a complete TCP stack in hardware (with the exception of startup/teardown, which still was intentionally done in software, for purposes of security/accounting).
Once the TCP connection was established, the packets were completely handled in hardware, and the resulting TCP payload data was DMA'ed directly to the application's memory when a read request was made. Same thing in the other direction, for a write request. Very fast!
I'm not sure of the exact numbers but we reduced CPU utilization to around 10%-20% of what it was under a non-accelerated card, and were able to saturate the wire in both directions using only a 1.0Ghz CPU. This is something that was difficult to do, given the common rule of thumb that you need 1Mhz of CPU speed to handle every 1Mbit of data on the wire.
To make a long story short, it didn't sell, and I (among many others) was laid off.
The reason was mostly about price/performance: who would pay that much for just a gigabit ethernet card? The money that was spent on a TOE-accelerated network card would be better spent on a faster CPU in general, or a more specialized interconnect such as InfiniBand.
When 10Gb Ethernet becomes a reality, we will once again need TOE-accelerated network cards (since there are no 10GHz CPU's today, as we seem to have hit a wall at around 4Ghz). I'd keep my eye on Chelsio: of the Ethernet TOE vendors still standing, they seem to have a good product.
BTW, did you know that 10Gb Ethernet is basically "InfiniBand lite"? Take InfiniBand, drop the special upper-layer protocols so that it's just raw packets on the wire, treat that with the same semantics as Ethernet, and you have 10GbE. I can predict that Ethernet and InfiniBand will conceptually merge, sometime in the future. Maybe Ethernet will become a subset of InfiniBand, like SATA is a subset of SAS....
About seamlessly combining raster and vector image editing: It's nothing new.
I miss SuperPaint, by Silicon Beach Software, for the old black-and-white Macintosh.
It had the *perfect* combination of raster and vector painting. It also had a good balance between photo-editing features (aka Photoshop) and tools for creating a new image from scratch (aka DeluxePaint).
At the time it was written, there were no color Macintoshes, but there was still some functions in the original QuickDraw API to deal with color (mostly to support multicolored ribbons in Apple ImageWriter printers). Icons for colors in SuperPaint just had words like "Orange", "Blue", etc. that would just show up black on the screen. The amazing thing is, when the first color Macintosh II came out, I tried out SuperPaint on it, and it WORKED!! It was so well written that everything on screen was now in color, even though the developers had no way of testing it before! Amazing....
SuperPaint was eventually bought out by Aldus, and after that, it just wasn't the same. Adobe eventually bought Aldus, and SuperPaint faded into obscurity. (What is with Adobe? The same thing happened in the Windows world just a few years ago, when we lost Cool Edit, essentially the only high-quality audio editing program that was within the budget of the average home user.)
I use and prefer Lite-On drives. They are cheap and standards-compliant. The same can't be said for Plextor (although they have improved in recent years).
What really put me over the fence for Lite-On is the freeware (closed source) program KProbe. This works only with Lite-On drives, and is only for Windows (unfortunately). Even though it is closed source, it is a free download (unlike Plextor's commercial utility).
The program KProbe seems to perform similar functions to what these Plextor utilities do: show the true low-level bit error rates of the disc, and allow certain drive settings to be overridden as desired (minimum and maximum speeds, DVD+R booktype, DVD region, and so on).
With the ability for end users to get at this information, it becomes possible to make informed decisions when buying blank DVD media.
If only these low-level functions were standardized, *sigh* -- it would be great if mainstream DVD and CD-burning software could use them to check disc reliability, and the OS could even inform the user if a disc were about to fail.
With more and more people storing home movies and other keepsakes on DVD these days, this will become more important. The ability to see this low-level info has saved me from losing data several times due to cheap media: beware "Great Quality" from Fry's!
Now, if only KProbe were open source....