I've kind of thought that a highspeed RC car could launch, use military technology for tracking, get under the car and attach itself with a bunch of hard drive magents. Either use that as the GPS device (which eliminates the whole impact debate here) or emits a small, localized EMP. Yes, I know that EMP takes a lot of energy, but nothing a couple of really large capacitors could be revved up to handle. Few people are likely to initiate a chase in any vehicle that doesn't have a computer nowadays, so the only weak spot would be diesels.
For the critics who complain about what an RC car would need for resources, they are forgetting that two things are in its favor: the vehicle only needs to run at full speed for about 2 minutes (>2minutes is a failure), and its carrier is already travelling at full speed. Heck, for that matter why not make it a hovercraft and avoid all those pavement anomalies.
In the spirit of what people idealize/. to be, I present the URL to an opensource project called DIY Zoning (that is, Do-It-Yourself for those who live in an apartment). It is a very well designed website with links to best practices and pointers to basic parts and the sourceforge-based software. From the site:
A collection of bits and pieces of knowledge to explain how to put together mostly off-the-shelf inexpensive equipment and make a temperature zoning system out of it, and a software product that allows you to control the hardware.
This is a typical case of working yourself out of a job. Offering to telecommute is like handing your boss a permission slip to outsource you - either to a lesser paid employee or to an overseas subcontractor. If you can do the job from the beach in Bali, so can the Indian working in a sweatshop of Bangalore. At that point, it's just training.
If you want to keep your job, get one where you directly support a customer and must show up or live on-site. The momentum required to replace you is even higher. Make your job something that requires semi-routine, hands-on access to hardware and you keep your job for a really long time. If your customer is always growing, your daily job is to deal with all the new work they have, not keeping existing equipment functioning.
But perhaps you are confusing the hubris of the Perl programmer which states that all Perl programmers are inherently lazy. Yes, but by requesting to telecommute, you are exhibiting laziness and an unwillingness to work with people. The Perl mantra refers to an unwillingness to repeat the same tasks day after day when the computer can be taught to perform the same work.
I don't really have sympathy for people who are too lazy to come to work. Yes, good employers will find ways to accomodate valued employees that have a strong urge to live somewhere else (e.g. a military spouse which gets restationed) in the hope they someday they will be able to come back. Beyond that, telecommuting is an excuse for terrible employers to make their existing employees work even more on their free-time at home.
If Microsoft really wants to kill themselves on this, not only will the/. crowd love it, but so will Sony, and Nintendo. Of course, all those commercial companies that have figured out how to run an actual business on top of MS's OS may not appreciate having a bunch of gaming/entertainment execs that can tug on the ship helm at will.
I'm sure shareholders won't look too lightly on the profit losing division which represents the XBox. Shareholders are likely to examine the idea quality vs. profitability ratio from this division and smell bad news. Microsoft will probably just claim that they are being given access to more areas of the company hoping to foster horizontal integration, but they'll forget that only PHBs like buzz-phrases like that and shareholders will see the fluff for what it is.
Why is this new? Richard Feynman talked about ants long time ago. Even as far back as when he was a kid, as he discusses in his book Surely You're Joking Mr. Feynman (which has the text of the book, and this section, about 1/3 of the way down).
First non-lamer post.
Network Operations Centers use these kinds of setups all the time (except for the audio). This is going to seem like a plug, but I've been researching this for work. One that I've found so far is the MediaWall 2000. You can control it via serial or Ethernet, so the presumption is that you might have a customzed/handheld interface developed.
They also have other, related products.
You say that you want an on-screen display with remote, but most configurations this complex aren't being built for you, they are going to emergency response centers and such where there is a person in the hot seat watching various view points. It needs to be quick, easy, and customized (label the inputs for example and put the most used at the front of the list). This quickly points to an embedded solution. Talk to a sales rep and find out what your choices are. I'm going to end up doing the same in a few weeks.
I'm quite sure this isn't the only vendor for this type of product, but I did search for a while and came up somewhat cold. Talk to a local audio/video specialist for help, although they probably will want a piece of the action.
As for the audio - you and I both know you need a mixing board, even a simplistic 16-channel version. You said that you want picture within a picture, but these really complex solutions might mean you want four pictures on screen and to have each feed come from one of four speakers, presumably in pseudo-surround mode. A mixer could work in two modes - aux1-aux4 feed amps that drive four speakers around you. The main setup could also just drive your fronts, although you'll lose and 6.1 sound you might be running.
In the end, I think you're overdoing it. You probably can't process more than two pictures at a time, much less 16, unless you're watching sports. You want to avoid a "bunch of VCRs" but yet you want 16 channels of individually tunable inputs ala tuners. I don't get it. You won't be able to maintain surround sound at 6.1 and still have multiple audio feeds. Since you want so much versatility, I assume you're one of those rich boys that will gladly plunk out US$3000 every couple of years for a new display. Then it may also be safe to assume you would pay someone US$1000 to design such a system to your specifications. People do this for a living so as the yellow pages in your area for help with Audio-Visual-Consultants.
Other interesting finds:
Zektor audio/video switches. 4 channels, but HD and digital audio supported
Everyone expects that Microsoft would want to shut such a site down. Believe it or not, if the Trusted Computing Machine paradigm is to really take hold, Microsoft is going to have to wait it out. Lots of companies have worked on other tamper-proof technology. If this platform can withstand a very large portion of that attack, then they will have a reputation to be proud of - from a security perspective.
Bruce Schneier reminds us of several attributes in his book Secrets and Lies.
Tamper proof hardware through zeroization techniques (no evidence thus far), but may involve destruction when a critical chip is removed.
Revocation of privilege to participate. If the hypervisor detects trouble, it fails to a safe position.
"Only the key is secret" (and only for so long). Call this a free update CD every XBox360 owner must run after two years from launch - this is a valid application with new Microsoft keys.
Compromise in one section does not compromise the whole unit (defense in depth)
Assume something like the James Bond 007 game save buffer overflow will happen again, and the damage should require everyone to purchase this game to continue running non-standard code - during which time the bug will be patched in the new distribution discs.
Fiercly litigate anyone that builds disc reading/writing technology for the XBox 360, specifically targetting hardware vendors.
Develop a method by which an honest enthusiast can work in a sandbox that does anything. They'll never be entirely happy, but it will keep all but the most zealous enthusiasts at bay. Make this disk cost, oh, say $150 to cover the lost profit, or $40 per year per console. Call this a bizaare toy for the sophisticated adult, and the cost should make it a disincentive to commercial distribution of competitive products.
For the record, I have no interest in playing on a 360, much less compromizing one, but if Microsoft can apply the above principles, then they will have a reputation and platform other non-gaming industries can embrace. Even Sony couldn't buy that with money. I do, however, have my doubts that Microsoft has focused on security robustness because their first and formost motto should be "It's all about the gaming experience." Fail that and the thing dies anyway.
It looks like they didn't even have the name Riya as of August of this year. It was Ojos.
Their main guy even has a blog where you can follow up on their Series B financing from the VCs. If this guy wants to make it big, their US and India teams should get the technology polished and then license it to Google for inclusion in their Google Desktop, with support for external media (e.g. DVDs full of photos).
Later, they could expand it out to search for the same faces in movies. Whoa! Hold on. You all are getting carried away. This is not recognizing everybody that passes by a security camera. The success of their codes depends on you having photographed the same individual multiple times. That's why facial recognition in security camera's is such a bad idea - you normally only get to train such a system on: one good photograph, or lots of terrible photographs. And then people want to pick out the bad guy as one of 5,000 people per camera per day???
If you tie this in to the running email address debate, you'll understand a little deeper. They do in fact keep track of the facial (and other) highlights, but there are too many to compare them to everyone ever scanned. Using an e-mail address significantly narrows the fields down, and then they reused the feature attributes. But otherwise, they statistically never reuse the patterns outside of your photo collection.
Since when was id a fan of DirectX, let alone Direct3D?
(If this sentence doesn't make any sense, you didn't watch the video)
An
article from the time clearly show's id's lead engine developer,
John Carmack in complete opposition to the Direct3D API.
I haven't followed things after that, but let's assume the API hasn't
gotten any simpler, especially with all the new shading specifications.
Did somebody say Easter Egg? Splash screen clearly not giving your boss credit for the work. Better yet, if it was developed on their own time/dime, make a click-through EULA that grants them the right to test the code on a development network, but production implementations must be licensed separately.
Make sure to embed the Easter Egg that gives you extra credit for having written it. Then if they steal it, you could prove that you own it (at just the right moment, in front of official witnesses and lawyers) and they didn't pay for it.
If the U.S. Copyright Office is soliciting ideas on this concept, a method exists to address accessibility for those disabled, and the "lost due to bankruptcy" sides. Two parts address each side.
Part 1: Copyright holders should be required to register each anti-circumvention method with the copyright office in exchange for which a number (which has no legal value) will be assigned to the method. IP firms which specialize in protection methods could register their methods as "base classes" for others. References to patent numbers might also be helpful.
Part 2: The holder then keeps a list available for public viewing which indicates the works and which copy protection methods they have employed. (Example: ebook "How To Be A Dummy," published 8Oct2005, document format: PRC, DRM method: 1,554,776 (Ref 334,665) ). Note that this is not registering your works. Ideally, this list should be notarized, either electronically or physically to prevent post-litigation tampering.
Part 1 takes the good old cryptography rule to heart, "The algorithm is never the secret. The input key is the secret." Two things happen here: bad protection is shamed away from being used, and the court system has a public notice of intent on the behalf of the holder requesting that the DMCA provide anti-circumvention enforcement. Then the U.S. Courts would only need to prove or disprove that the declared method was actually in use and attempting to protect the work in question. If expert commentary were recognized on at least "base class" methods, then prebuilt testimony could be used in courts to make enforcement of base-64 encoding methods an embarassment to the litigant.
Part 2 gives the courts the benefits of documentation regarding protected works. It also lets those who provide support to disabled individuals have a course of action with regard to legal circumvention. So it could, for example, allow someone who converts DRMed ebooks to audiobooks to become a limited, authorized agent to perform such action with nothing more than some paperwork from the copyright holder. If the copyright holder has disappeared (death or otherwise), then the method is publicly known and could be cracked within governed rules.
In short, there exists the potential to not take the teeth out of the DMCA while still executing it a more efficient way. Hopefully the methods described above represent ideas that tilt more power towards the public without actually removing any power of the copyright holder. Yes, many of us believe it should disappear, but the U.S. did sign an international treaty of which the DMCA is only a manifestation. Since the U.S. is a WTO member, it was a willing participant, although it could have fought back a little harder.
Treat it just like any other project that uses a cyclic lifecycle management. I'm supposing you already have your foot in the door, you are just unsure as to how to conduct yourself. At the end of each round, the customer can decide if they like the kind of progress being made and has the option to cancel the contract after each round if they disagree with methods or results. Start small and simple and develop their trust. If they really have security problems, you are best off finding a way to make them want to change rather than just telling them off.
Round 1: Spend one week writing a paper on the intellectual or physical property deemed essential to the company, and then document what measures the company believes they are practicing to protect them. At this point, you should also define your known enemies, be it a competitor or vast amounts of time wasted during virus outbreaks. Don't dwell on anything but the obvious as we all learned in the Six Dumbest Ideas In Computer Security document.
Round 2: Propose a paper exercise approach to physical security, both in the server room and in the cubicle farms. Spend a week and not too much money. This will confirm or deny that declared in Round 1.
Round 3: Address disaster recovery options because arson and other DOS techniques are just as bad for protecting IP as is an electronic attack. This is a check to see if the current protections methods covered this usually underfunded area. Don't forget offsites.
Round 4: Propose, via contractual methods, solutions for closing gaping holes in the protection measures. That is, cover the areas for which no protection is provided, be it physical, procedural, or electronic. Implement if approved and have alternate, albeit less-effective approaches for those rejected due to cost or time.
Round 5: Propose a development area be established to test current and future configurations of electronic equipment for known attack vectors (e.g. new patches on a firewall don't open new ports). [At this stage, your customer has confidence that you know what you're doing, but it took you this long before you really started touching the inside of their network.] You never subject the production network to most scans, except maybe for proper patch deployment. All the exploit attempts happen in the lab.
Round 6: Like every good reader of Bruce Schneier's Secrets and Lies, you now propose methods and procedures for monitoring and reacting to attacks against the core intellectual or physical property documented in Round 1. Depending on your company goals, you can hope to win this one, or you can let them run the service while you move on to another customer.
Tips: If you get lots of resistance at Round 1 telling you that you aren't moving fast enough, beware because you will be the victim of the blame game in Round 6. Don't forget that sometimes the attack vector is physical theft - encrypt core files anywhere they are found, most especially on laptops. Round 1 may have identified Internet access as a risk, so in Round 4, consider using a private, internal network and force all users to use thin-client tools for Internet access - no removable media, highly-enforced group policies, and the ability to quarantine viruses at the door. For that matter, proxy all Internet access and monitor it in Round 6.
I have to admit, the article is quite enlightening. I am most interested in the object and economic systems. Fortunately, Mr. Koster was a little shy on the macro-scale things, instead focusing on software design issues like: usability, maintainability, and scalability.
Persistence means it never goes away was a very pertinent topic and one that I address in much more detail. Problems here include: trusting the system owners to do good backups, proving proper transfer of world objects between owners, and avoiding client-side corruption. The Never trust the client problem has a potential solution in my paper. It's not completely solved, but corruption of in-world objects is at least trivialized.
He touches on in-game community, but has neglected the larger interoperability problem between games and vendors. Too many clients and connection methods. Game persistence lasts only as long as the operating system or game console are in use, which is absurd.
The idea that really tickled me was the economy theorem that was so obvious I missed it. "Players will hate having this drain, but if you do not enforce ongoing expenditures, you will have Monty Haul syndrome, infinite accumulation of wealth, overall rise in the "standard of living" and capabilities of the average player." This methedology really need connected with the Attention is the currency of the future and you will see the solution to the small-time MMORPG player and the 30+hours a week gamer. Us small-timers walk in and get creamed because these full-timers are super heroes.
Perchance, we ought to tie increased skill with increased responsibility, just like in real life. Level 30+ character, you must now lead larger missions. Short term players can now join the group, like showing up at the gym and playing a game of basketball with strangers. A couple of guys are there all the time and are either friendly or they aren't. Not friendly means no squads to go on more complex missions.
Read more on their new CEO Hurd at The Register. This is not the last layoff. I know the industry reports say that Hurd will do great things just like he did for NCR, but I live here in Dayton, OH, where NCR is headquartered.
I don't work for, near, or against NCR in any market, I just live here. My perception has been that Hurd took NCR and focused them on ATM machines, their core business. Lots and lots of other unrelated things were shed, including long-term employees and facilities. Most recently, NCR has turned over maintenance of their world-class headquarters to a local office-real estate company (Miller Valentine). Their ATM sales, by the way, are in competition with the infamous Diebold, Inc. And in that market, Diebold is innovating with ways to keep banks coming back to them. NCR just never seemed to get the hang of it.
Now back to HP and the recent high speed printing invention , and I would have to say we can all expect HP to shed all the unprofitable businesses and focus heavy on the printing. Well, heavier than they already do.
I expect that if HP is in some select-contract, highly-profitable, niche market that they will stay there. NCR has their TeraData database. HP had, during the merger claimed that Compaq's worldwide sales and services forces would allow them to dominate global industries, but I don't think that really every took foot they way they wanted to. So, unless they drum up something other than calculators and home PCs, those segments are likely to get hit hard. Hurd likely won't wait for the home PC market to do something unique because they've had a couple of dozen years to find a niche and haven't. I'd better say goodbye to the calculator segment too before Cringely goes around saying something stupid like "you saw it here first."
So that leaves the saturated inkjet market. Since the DMCA cannot be used, and since we're already paying $3,800US per gallon of ink, increasing profitability will be difficult to do without large customer outbursts. Of course, NCR was so full of waste, those of us in Dayton didn't think they could ever shed all of it and yet they did.
PowerManga, TuxRacer, Frozen Bubbles, SuperTux. Violence (other than squishing the bad guys) is out. Put blood/dismemberment in, and the girls leave. Require large portions of their time, and they'll spend it caring for the family, watching TV/Movies, or playing with real, physical toys (Tinker Toys).
It still doesn't discuss in-house applications that are meant to solve production needs, but will never be sold or given away because the code is: too customized, reflects a particular business model, and/or not well polished. Perl/Qt is so much better than Perl/Tk, but we tolerate Tk because it's free.
Now I don't object to the core developer of applications needing a license, since that is what corporations do with Visual Studio, but scripted languages is kind of fuzzy. I think the best way to solve this is to officially support some of the language wrappers, like Perl and Python and provide some mechanism that determines an elegant way to "lock the interface layout" unless a developer license can be had. For example, only precompiled QtDesigner User Interface files could be added run-time, or something like that.
Then distribute it with ActiveState's ActivePerl or ActivePython. That version would only allow GPLed applications to be built, maybe through a variable like $main::LICENSE="GPL" to be declared, or something like that. You would have to purchase ActivePerl Pro Studio in order to develop commercial applications with the Qt widget set, but again the same "lock it down without a license" could be had.
Official support for scripted languages is the only way to sort out the need for quick, small programs for which C++ is inappropriate.
SOX has a compand operator that lets you tune this stuff and parameters that let you tweak how fast it responds, how hard it attempts to correct the sound, and how fast it "lets go." Put this in line with your CD RIP process after you run a few through a test bed.
Don't forget to specify settings for each channel (normally two).
If you're willing to forego ID3 tags, or can hack them in yourself with mp3info, you can use cdda2wav to do sox processing: cdda2wav -O wav -t 7 -D/dev/cdrom - | sox [sox options] | lame - foo.mp3
(and why more companies are starting to care also)...
Hidden text, unknown OLE links, undo and revision information. Too many things are found floating around a document. Even though PPT isn't part of this equation, Word documents can now have a (relatively compilicated) stylesheet applied against them as part of the "scrubbing" process.
Be it metadata, or routine edits and changes, Word is a dangerous portal into a company's opinions or sensitve government data. What everyone wants is the simple, provable method for knowing only their best foot is placed forward.
While governments play only a minor role in the balance sheets of Microsoft, changes like this solve the only real, outstanding technical hangup governments have with Office (excluding the PPT exclusion).
FutureProof said that Apple is putting watermarking in their music and they are looking for the lack of that watermark in future versions of iTunes (both to stop competitors and most likely identify those who would rip from iTunes and resell it illegally). Nothing has stated that the watermark is an Apple-wide watermark (i.e. distributed to all users) or if it is a per user watermark added on top of the Apple watermark (double water-marked).
Unless this makes your head swim, there is an excellent book that most folks with a bachelor's degree in some field which involved math should be able to read and understand: Information Hiding Techniques - Techniques for Steganography and Digital Watermarking (ISBN 1-580-53035-4), by Katzenbeisse. This and some other related books can been seen at forensics.nl.
Note: I am not affiliated with any of these publishers or authors, but merely read through the above mentioned book and found it appropriate for the topic.
In every distaster, be it a natural disaster like this one, or refuge camps from civil war, the NGOs which run the aid efforts must use some sort of software. The classic problem appears to be connecting families. If I were a relief worker (and I've never been one), I think the best software would provide:
Short video clip saying their name, village name, and names of parents and children (if any).
Local dialects spoken, unicode records of their names (I imagine most refuges are illiterate and workers must guess at their names), and any relatives whose names they can spell.
Also, GIS coordinates of their original home, if available, would be helpful. This really could be as simple as finding a village name and recording that, or if they gave you an address in a city.
Still frame, mug shot that facial recognition software could compare against photos provided by family members, if they become available. (hint: use facial recognition where false alarms don't get people arrested, but rather are welcomed)
Some sort of indexing system that could operate over low bandwidth, like what might be available over HAM radio. Treat the file on a refugee like a BitTorrent link and as one person at another site gets the records from other sites, cached copies start appearing around the mesh of camps speeding up the search process as time drags on.
Except for the video capture, don't have extensive hardware and software requirements (read high cost per terminal)
Give each person an RFID bracelet so that refuge workers can find people quickly. If the camp gets raided, people can tear off their bracelets, or if they get relocated, they just need to check in and their demographics are copied to the local camp.
Let a relative looking for the lost open a case and now it becomes a question of data mining and the application of existing tools.
Relatives looking for lost people could indicate the approximate location, the language the person(s) speak, and letter sequences known to be in the person's name]
People who lived there that could effectively communicate with aid workers could assist in correctly spelling names and addresses, making the need for translations to occur only once.
Okay. I know I'm dreaming, but all this stuff can be done with real databases that support blobs, and torrent links aren't that hard to index. Drop facial recognition into a central facility (say the NGO headquarters) and they can issue recommendations for people to hook up. Heck, make it a Knoppix-like live-CD where the local HD is for cache and data acquisition, and building a reliable workstation is a piece of cake - distribute CDs and replace broken hardware quickly and efficiently.
Have any NGOs really looked into starting open-source projects to do these kinds of things or do they already have adequate tools of their own? Anybody have any insight? (they're all probably in the Pacific right now)
I say open-source because NGOs are not in competition for anything except money, and sometimes not even then. Given a uniform software base, they could work together and participate much more uniformly and thus speed the disaster relief efforts all that much more. Add the cost of running open-source and the myriad of commercial vendors looking for a piece of the action (not all will be as generous as ESRI is for now) will be numerous. Open source is the only way to keep the cost down, and the NGO could still pay someone to develop this software, but agree to keep the work in the open.
Re:Releasing code?
on
Wish Cancelled
·
· Score: 2, Interesting
If the vendor had planned it, from the outset, to be able to live past their ability or desire to support the product, then perhaps the world could live on. Per my
Virtual World Bylaws page, this is one of the primary rules which must be followed if an author's dream world is to be realized. I don't propose a solution, only that without solving this fundamental, doom is certain. The additional point above is well made by mentioning that the graphics engine is a big deal.
1. The world must be able to exist when the vendor loses interest and shuts down the hardware.
Peer-to-peer networking is an excellent example of letting a good thing keep running even when somebody wants it to come down. The same resiliency should be applied to virtual worlds. Distribute the servers that manage the virtual areas or worlds and localized hardware problems only mean a degredation in service.
Likewise, folks who are open with their underlying engine will find their technology replicated so that the world continues without them. Find a balance with openness and intellectual property or risk having your worlds be meaningless six months after creation. Note the difference between a graphics engine and an object interaction/scene description engine.
Vendors around the world will testify that customer service is the hardest part of MMOGs. No vendor wants to keep a customer support team paid and trained for 300 total players. If the technology is open, then those 300 players can play when they want and they will be expert enough to keep it running.
But what about artwork and the licensing that goes with it? Yes, that is a troubling sticky point. So what I recommend is that at the moment the vendor decides it's time to abandon, they exercise a clause they signed with all the graphic artists that says all royalty payments stop when the support section for that world closes. The artwork should then be released under something like the Creative Commons License (http://creativecommons.org/). This ties the company's profit line to the royalty distribution, if one exists at all, and as long as the venture is profitable to someone, the artists involved should be compensated.
A professor friend of mine developed just such an algorithm that he believes will allow robust splitting, privacy, redundancy, and resliency to malicious attack. He just never connected the idea with P2P for distribution. I'm a moron for having not done the association for him.
Schneier might call it snake oil. I don't think so. If there are holes, it's because they haven't been vetted by cryptographers. The professor is open to formal critiques and papers. He's opened his source for those who would use it for free (as in beer) or analysis. He's a business professor with a keen interest in software and is leaning towards open source, if only he can find the right license.
For those in the know, this is Dr. Doug Lowry at Franciscan University in Steubenville, OH. Tell him he got his name on Slashdot. -ct
An excerpt from my Virtual World Bylaws (rules to live by). This exact scenario fits the bylaw perfectly.
The world must be able to exist when the vendor loses interest and shuts down the hardware.
Peer-to-peer networking is an excellent example of letting a good thing keep running even when somebody wants it to come down. The same resiliency should be applied to virtual worlds. Distribute the servers that manage the virtual areas or worlds and localized hardware problems only mean a degredation in service.
Likewise, folks who are open with their underlying engine will find their technology replicated so that the world continues without them. Find a balance with openness and intellectual property or risk having your worlds be meaningless six months after creation. Note the difference between a graphics engine and an object interaction/scene description engine.
Vendors around the world will testify that customer service is the hardest part of MMOGs. No vendor wants to keep a customer support team paid and trained for 300 total players. If the technology is open, then those 300 players can play when they want and they will be expert enough to keep it running.
But what about artwork and the licensing that goes with it? Yes, that is a troubling sticky point. So what I recommend is that at the moment the vendor decides it's time to abandon, they exercise a clause they signed with all the graphic artists that says all royalty payments stop when the support section for that world closes. The artwork should then be released under something like the Creative Commons License. This ties the company's profit line to the royalty distribution, if one exists at all, and as long as the venture is profitable to someone, the artists involved should be compensated.
Tcl and Postscript both offer lists and lists within lists, without having to do anything fancy. Perl does have this, to an extent, but only on static lists. Anything dynamic adds a few lines of code and confuses things pretty easily. Of course, all of this came from Lisp, but I learned car and cdr and that was more than enough for me.
In short, lists only work when recursion is right at the tip of your fingertips in your language. If XSLT-2.0 works on tree fragments, we will have lists there as well.
Slashdot Mirror
I've kind of thought that a highspeed RC car could launch, use military technology for tracking, get under the car and attach itself with a bunch of hard drive magents. Either use that as the GPS device (which eliminates the whole impact debate here) or emits a small, localized EMP. Yes, I know that EMP takes a lot of energy, but nothing a couple of really large capacitors could be revved up to handle. Few people are likely to initiate a chase in any vehicle that doesn't have a computer nowadays, so the only weak spot would be diesels.
For the critics who complain about what an RC car would need for resources, they are forgetting that two things are in its favor: the vehicle only needs to run at full speed for about 2 minutes (>2minutes is a failure), and its carrier is already travelling at full speed. Heck, for that matter why not make it a hovercraft and avoid all those pavement anomalies.
P.S. This counts as prior art. :-)
In the spirit of what people idealize /. to be, I present the URL to an opensource project called DIY Zoning (that is, Do-It-Yourself for those who live in an apartment). It is a very well designed website with links to best practices and pointers to basic parts and the sourceforge-based software. From the site:
This is a typical case of working yourself out of a job. Offering to telecommute is like handing your boss a permission slip to outsource you - either to a lesser paid employee or to an overseas subcontractor. If you can do the job from the beach in Bali, so can the Indian working in a sweatshop of Bangalore. At that point, it's just training.
If you want to keep your job, get one where you directly support a customer and must show up or live on-site. The momentum required to replace you is even higher. Make your job something that requires semi-routine, hands-on access to hardware and you keep your job for a really long time. If your customer is always growing, your daily job is to deal with all the new work they have, not keeping existing equipment functioning.
But perhaps you are confusing the hubris of the Perl programmer which states that all Perl programmers are inherently lazy. Yes, but by requesting to telecommute, you are exhibiting laziness and an unwillingness to work with people. The Perl mantra refers to an unwillingness to repeat the same tasks day after day when the computer can be taught to perform the same work.
I don't really have sympathy for people who are too lazy to come to work. Yes, good employers will find ways to accomodate valued employees that have a strong urge to live somewhere else (e.g. a military spouse which gets restationed) in the hope they someday they will be able to come back. Beyond that, telecommuting is an excuse for terrible employers to make their existing employees work even more on their free-time at home.
If Microsoft really wants to kill themselves on this, not only will the /. crowd love it, but so will Sony, and Nintendo. Of course, all those commercial companies that have figured out how to run an actual business on top of MS's OS may not appreciate having a bunch of gaming/entertainment execs that can tug on the ship helm at will.
I'm sure shareholders won't look too lightly on the profit losing division which represents the XBox. Shareholders are likely to examine the idea quality vs. profitability ratio from this division and smell bad news. Microsoft will probably just claim that they are being given access to more areas of the company hoping to foster horizontal integration, but they'll forget that only PHBs like buzz-phrases like that and shareholders will see the fluff for what it is.
Why is this new? Richard Feynman talked about ants long time ago. Even as far back as when he was a kid, as he discusses in his book Surely You're Joking Mr. Feynman (which has the text of the book, and this section, about 1/3 of the way down). First non-lamer post.
Network Operations Centers use these kinds of setups all the time (except for the audio). This is going to seem like a plug, but I've been researching this for work. One that I've found so far is the MediaWall 2000. You can control it via serial or Ethernet, so the presumption is that you might have a customzed/handheld interface developed. They also have other, related products.
You say that you want an on-screen display with remote, but most configurations this complex aren't being built for you, they are going to emergency response centers and such where there is a person in the hot seat watching various view points. It needs to be quick, easy, and customized (label the inputs for example and put the most used at the front of the list). This quickly points to an embedded solution. Talk to a sales rep and find out what your choices are. I'm going to end up doing the same in a few weeks.
I'm quite sure this isn't the only vendor for this type of product, but I did search for a while and came up somewhat cold. Talk to a local audio/video specialist for help, although they probably will want a piece of the action.
As for the audio - you and I both know you need a mixing board, even a simplistic 16-channel version. You said that you want picture within a picture, but these really complex solutions might mean you want four pictures on screen and to have each feed come from one of four speakers, presumably in pseudo-surround mode. A mixer could work in two modes - aux1-aux4 feed amps that drive four speakers around you. The main setup could also just drive your fronts, although you'll lose and 6.1 sound you might be running.
In the end, I think you're overdoing it. You probably can't process more than two pictures at a time, much less 16, unless you're watching sports. You want to avoid a "bunch of VCRs" but yet you want 16 channels of individually tunable inputs ala tuners. I don't get it. You won't be able to maintain surround sound at 6.1 and still have multiple audio feeds. Since you want so much versatility, I assume you're one of those rich boys that will gladly plunk out US$3000 every couple of years for a new display. Then it may also be safe to assume you would pay someone US$1000 to design such a system to your specifications. People do this for a living so as the yellow pages in your area for help with Audio-Visual-Consultants.
Other interesting finds:
Everyone expects that Microsoft would want to shut such a site down. Believe it or not, if the Trusted Computing Machine paradigm is to really take hold, Microsoft is going to have to wait it out. Lots of companies have worked on other tamper-proof technology. If this platform can withstand a very large portion of that attack, then they will have a reputation to be proud of - from a security perspective.
Bruce Schneier reminds us of several attributes in his book Secrets and Lies.
For the record, I have no interest in playing on a 360, much less compromizing one, but if Microsoft can apply the above principles, then they will have a reputation and platform other non-gaming industries can embrace. Even Sony couldn't buy that with money. I do, however, have my doubts that Microsoft has focused on security robustness because their first and formost motto should be "It's all about the gaming experience." Fail that and the thing dies anyway.
It looks like they didn't even have the name Riya as of August of this year. It was Ojos. Their main guy even has a blog where you can follow up on their Series B financing from the VCs. If this guy wants to make it big, their US and India teams should get the technology polished and then license it to Google for inclusion in their Google Desktop, with support for external media (e.g. DVDs full of photos).
Later, they could expand it out to search for the same faces in movies. Whoa! Hold on. You all are getting carried away. This is not recognizing everybody that passes by a security camera. The success of their codes depends on you having photographed the same individual multiple times. That's why facial recognition in security camera's is such a bad idea - you normally only get to train such a system on: one good photograph, or lots of terrible photographs. And then people want to pick out the bad guy as one of 5,000 people per camera per day???
If you tie this in to the running email address debate, you'll understand a little deeper. They do in fact keep track of the facial (and other) highlights, but there are too many to compare them to everyone ever scanned. Using an e-mail address significantly narrows the fields down, and then they reused the feature attributes. But otherwise, they statistically never reuse the patterns outside of your photo collection.
Since when was id a fan of DirectX, let alone Direct3D? (If this sentence doesn't make any sense, you didn't watch the video) An article from the time clearly show's id's lead engine developer, John Carmack in complete opposition to the Direct3D API. I haven't followed things after that, but let's assume the API hasn't gotten any simpler, especially with all the new shading specifications.
Make sure to embed the Easter Egg that gives you extra credit for having written it. Then if they steal it, you could prove that you own it (at just the right moment, in front of official witnesses and lawyers) and they didn't pay for it.
If the U.S. Copyright Office is soliciting ideas on this concept, a method exists to address accessibility for those disabled, and the "lost due to bankruptcy" sides. Two parts address each side.
Part 1: Copyright holders should be required to register each anti-circumvention method with the copyright office in exchange for which a number (which has no legal value) will be assigned to the method. IP firms which specialize in protection methods could register their methods as "base classes" for others. References to patent numbers might also be helpful.
Part 2: The holder then keeps a list available for public viewing which indicates the works and which copy protection methods they have employed. (Example: ebook "How To Be A Dummy," published 8Oct2005, document format: PRC, DRM method: 1,554,776 (Ref 334,665) ). Note that this is not registering your works. Ideally, this list should be notarized, either electronically or physically to prevent post-litigation tampering.
Part 1 takes the good old cryptography rule to heart, "The algorithm is never the secret. The input key is the secret." Two things happen here: bad protection is shamed away from being used, and the court system has a public notice of intent on the behalf of the holder requesting that the DMCA provide anti-circumvention enforcement. Then the U.S. Courts would only need to prove or disprove that the declared method was actually in use and attempting to protect the work in question. If expert commentary were recognized on at least "base class" methods, then prebuilt testimony could be used in courts to make enforcement of base-64 encoding methods an embarassment to the litigant.
Part 2 gives the courts the benefits of documentation regarding protected works. It also lets those who provide support to disabled individuals have a course of action with regard to legal circumvention. So it could, for example, allow someone who converts DRMed ebooks to audiobooks to become a limited, authorized agent to perform such action with nothing more than some paperwork from the copyright holder. If the copyright holder has disappeared (death or otherwise), then the method is publicly known and could be cracked within governed rules.
In short, there exists the potential to not take the teeth out of the DMCA while still executing it a more efficient way. Hopefully the methods described above represent ideas that tilt more power towards the public without actually removing any power of the copyright holder. Yes, many of us believe it should disappear, but the U.S. did sign an international treaty of which the DMCA is only a manifestation. Since the U.S. is a WTO member, it was a willing participant, although it could have fought back a little harder.
Treat it just like any other project that uses a cyclic lifecycle management. I'm supposing you already have your foot in the door, you are just unsure as to how to conduct yourself. At the end of each round, the customer can decide if they like the kind of progress being made and has the option to cancel the contract after each round if they disagree with methods or results. Start small and simple and develop their trust. If they really have security problems, you are best off finding a way to make them want to change rather than just telling them off.
Round 1: Spend one week writing a paper on the intellectual or physical property deemed essential to the company, and then document what measures the company believes they are practicing to protect them. At this point, you should also define your known enemies, be it a competitor or vast amounts of time wasted during virus outbreaks. Don't dwell on anything but the obvious as we all learned in the Six Dumbest Ideas In Computer Security document.
Round 2: Propose a paper exercise approach to physical security, both in the server room and in the cubicle farms. Spend a week and not too much money. This will confirm or deny that declared in Round 1.
Round 3: Address disaster recovery options because arson and other DOS techniques are just as bad for protecting IP as is an electronic attack. This is a check to see if the current protections methods covered this usually underfunded area. Don't forget offsites.
Round 4: Propose, via contractual methods, solutions for closing gaping holes in the protection measures. That is, cover the areas for which no protection is provided, be it physical, procedural, or electronic. Implement if approved and have alternate, albeit less-effective approaches for those rejected due to cost or time.
Round 5: Propose a development area be established to test current and future configurations of electronic equipment for known attack vectors (e.g. new patches on a firewall don't open new ports). [At this stage, your customer has confidence that you know what you're doing, but it took you this long before you really started touching the inside of their network.] You never subject the production network to most scans, except maybe for proper patch deployment. All the exploit attempts happen in the lab.
Round 6: Like every good reader of Bruce Schneier's Secrets and Lies , you now propose methods and procedures for monitoring and reacting to attacks against the core intellectual or physical property documented in Round 1. Depending on your company goals, you can hope to win this one, or you can let them run the service while you move on to another customer.
Tips: If you get lots of resistance at Round 1 telling you that you aren't moving fast enough, beware because you will be the victim of the blame game in Round 6. Don't forget that sometimes the attack vector is physical theft - encrypt core files anywhere they are found, most especially on laptops. Round 1 may have identified Internet access as a risk, so in Round 4, consider using a private, internal network and force all users to use thin-client tools for Internet access - no removable media, highly-enforced group policies, and the ability to quarantine viruses at the door. For that matter, proxy all Internet access and monitor it in Round 6.
I have to admit, the article is quite enlightening. I am most interested in the object and economic systems. Fortunately, Mr. Koster was a little shy on the macro-scale things, instead focusing on software design issues like: usability, maintainability, and scalability.
My own thoughts on the matter are entitled Virtual World Bylaws and are my own.
Persistence means it never goes away was a very pertinent topic and one that I address in much more detail. Problems here include: trusting the system owners to do good backups, proving proper transfer of world objects between owners, and avoiding client-side corruption. The Never trust the client problem has a potential solution in my paper. It's not completely solved, but corruption of in-world objects is at least trivialized.
He touches on in-game community, but has neglected the larger interoperability problem between games and vendors. Too many clients and connection methods. Game persistence lasts only as long as the operating system or game console are in use, which is absurd.
The idea that really tickled me was the economy theorem that was so obvious I missed it. "Players will hate having this drain, but if you do not enforce ongoing expenditures, you will have Monty Haul syndrome, infinite accumulation of wealth, overall rise in the "standard of living" and capabilities of the average player." This methedology really need connected with the Attention is the currency of the future and you will see the solution to the small-time MMORPG player and the 30+hours a week gamer. Us small-timers walk in and get creamed because these full-timers are super heroes.
Perchance, we ought to tie increased skill with increased responsibility, just like in real life. Level 30+ character, you must now lead larger missions. Short term players can now join the group, like showing up at the gym and playing a game of basketball with strangers. A couple of guys are there all the time and are either friendly or they aren't. Not friendly means no squads to go on more complex missions.
I don't work for, near, or against NCR in any market, I just live here. My perception has been that Hurd took NCR and focused them on ATM machines, their core business. Lots and lots of other unrelated things were shed, including long-term employees and facilities. Most recently, NCR has turned over maintenance of their world-class headquarters to a local office-real estate company (Miller Valentine). Their ATM sales, by the way, are in competition with the infamous Diebold, Inc. And in that market, Diebold is innovating with ways to keep banks coming back to them. NCR just never seemed to get the hang of it.
Now back to HP and the recent high speed printing invention , and I would have to say we can all expect HP to shed all the unprofitable businesses and focus heavy on the printing. Well, heavier than they already do.
I expect that if HP is in some select-contract, highly-profitable, niche market that they will stay there. NCR has their TeraData database. HP had, during the merger claimed that Compaq's worldwide sales and services forces would allow them to dominate global industries, but I don't think that really every took foot they way they wanted to. So, unless they drum up something other than calculators and home PCs, those segments are likely to get hit hard. Hurd likely won't wait for the home PC market to do something unique because they've had a couple of dozen years to find a niche and haven't. I'd better say goodbye to the calculator segment too before Cringely goes around saying something stupid like "you saw it here first."
So that leaves the saturated inkjet market. Since the DMCA cannot be used, and since we're already paying $3,800US per gallon of ink, increasing profitability will be difficult to do without large customer outbursts. Of course, NCR was so full of waste, those of us in Dayton didn't think they could ever shed all of it and yet they did.
PowerManga, TuxRacer, Frozen Bubbles, SuperTux. Violence (other than squishing the bad guys) is out. Put blood/dismemberment in, and the girls leave. Require large portions of their time, and they'll spend it caring for the family, watching TV/Movies, or playing with real, physical toys (Tinker Toys).
The legal jargon surrounding the use of scripting languages and Qt is still kind of vague. Buried very deep in their FAQ section is the question Can I develop commercial applications with PerlQt or PyQt or other Qt wrappers? which isn't that helpful.
It still doesn't discuss in-house applications that are meant to solve production needs, but will never be sold or given away because the code is: too customized, reflects a particular business model, and/or not well polished. Perl/Qt is so much better than Perl/Tk, but we tolerate Tk because it's free.
Now I don't object to the core developer of applications needing a license, since that is what corporations do with Visual Studio, but scripted languages is kind of fuzzy. I think the best way to solve this is to officially support some of the language wrappers, like Perl and Python and provide some mechanism that determines an elegant way to "lock the interface layout" unless a developer license can be had. For example, only precompiled QtDesigner User Interface files could be added run-time, or something like that.
Then distribute it with ActiveState's ActivePerl or ActivePython. That version would only allow GPLed applications to be built, maybe through a variable like $main::LICENSE="GPL" to be declared, or something like that. You would have to purchase ActivePerl Pro Studio in order to develop commercial applications with the Qt widget set, but again the same "lock it down without a license" could be had.
Official support for scripted languages is the only way to sort out the need for quick, small programs for which C++ is inappropriate.
Serial Number 76604206 at http://www.uspto.gov/ (direct links aren't so easy to get there)
SOX has a compand operator that lets you tune this stuff and parameters that let you tweak how fast it responds, how hard it attempts to correct the sound, and how fast it "lets go." Put this in line with your CD RIP process after you run a few through a test bed.
Don't forget to specify settings for each channel (normally two).
If you're willing to forego ID3 tags, or can hack them in yourself with mp3info, you can use cdda2wav to do sox processing: cdda2wav -O wav -t 7 -D /dev/cdrom - | sox [sox options] | lame - foo.mp3
(and why more companies are starting to care also)...
Hidden text, unknown OLE links, undo and revision information. Too many things are found floating around a document. Even though PPT isn't part of this equation, Word documents can now have a (relatively compilicated) stylesheet applied against them as part of the "scrubbing" process.
Be it metadata, or routine edits and changes, Word is a dangerous portal into a company's opinions or sensitve government data. What everyone wants is the simple, provable method for knowing only their best foot is placed forward.
While governments play only a minor role in the balance sheets of Microsoft, changes like this solve the only real, outstanding technical hangup governments have with Office (excluding the PPT exclusion).
FutureProof said that Apple is putting watermarking in their music and they are looking for the lack of that watermark in future versions of iTunes (both to stop competitors and most likely identify those who would rip from iTunes and resell it illegally). Nothing has stated that the watermark is an Apple-wide watermark (i.e. distributed to all users) or if it is a per user watermark added on top of the Apple watermark (double water-marked).
Unless this makes your head swim, there is an excellent book that most folks with a bachelor's degree in some field which involved math should be able to read and understand: Information Hiding Techniques - Techniques for Steganography and Digital Watermarking (ISBN 1-580-53035-4), by Katzenbeisse. This and some other related books can been seen at forensics.nl.
Note: I am not affiliated with any of these publishers or authors, but merely read through the above mentioned book and found it appropriate for the topic.
In every distaster, be it a natural disaster like this one, or refuge camps from civil war, the NGOs which run the aid efforts must use some sort of software. The classic problem appears to be connecting families. If I were a relief worker (and I've never been one), I think the best software would provide:
Okay. I know I'm dreaming, but all this stuff can be done with real databases that support blobs, and torrent links aren't that hard to index. Drop facial recognition into a central facility (say the NGO headquarters) and they can issue recommendations for people to hook up. Heck, make it a Knoppix-like live-CD where the local HD is for cache and data acquisition, and building a reliable workstation is a piece of cake - distribute CDs and replace broken hardware quickly and efficiently.
Have any NGOs really looked into starting open-source projects to do these kinds of things or do they already have adequate tools of their own? Anybody have any insight? (they're all probably in the Pacific right now)
I say open-source because NGOs are not in competition for anything except money, and sometimes not even then. Given a uniform software base, they could work together and participate much more uniformly and thus speed the disaster relief efforts all that much more. Add the cost of running open-source and the myriad of commercial vendors looking for a piece of the action (not all will be as generous as ESRI is for now) will be numerous. Open source is the only way to keep the cost down, and the NGO could still pay someone to develop this software, but agree to keep the work in the open.
A professor friend of mine developed just such an algorithm that he believes will allow robust splitting, privacy, redundancy, and resliency to malicious attack. He just never connected the idea with P2P for distribution. I'm a moron for having not done the association for him.
The software is called Pryvit and the site is at www.privit.net ( cached ). A detailed description sits at their site.
Schneier might call it snake oil. I don't think so. If there are holes, it's because they haven't been vetted by cryptographers. The professor is open to formal critiques and papers. He's opened his source for those who would use it for free (as in beer) or analysis. He's a business professor with a keen interest in software and is leaning towards open source, if only he can find the right license.
For those in the know, this is Dr. Doug Lowry at Franciscan University in Steubenville, OH. Tell him he got his name on Slashdot. -ct
An excerpt from my Virtual World Bylaws (rules to live by). This exact scenario fits the bylaw perfectly.
The world must be able to exist when the vendor loses interest and shuts down the hardware.
Peer-to-peer networking is an excellent example of letting a good thing keep running even when somebody wants it to come down. The same resiliency should be applied to virtual worlds. Distribute the servers that manage the virtual areas or worlds and localized hardware problems only mean a degredation in service.
Likewise, folks who are open with their underlying engine will find their technology replicated so that the world continues without them. Find a balance with openness and intellectual property or risk having your worlds be meaningless six months after creation. Note the difference between a graphics engine and an object interaction/scene description engine.
Vendors around the world will testify that customer service is the hardest part of MMOGs. No vendor wants to keep a customer support team paid and trained for 300 total players. If the technology is open, then those 300 players can play when they want and they will be expert enough to keep it running.
But what about artwork and the licensing that goes with it? Yes, that is a troubling sticky point. So what I recommend is that at the moment the vendor decides it's time to abandon, they exercise a clause they signed with all the graphic artists that says all royalty payments stop when the support section for that world closes. The artwork should then be released under something like the Creative Commons License. This ties the company's profit line to the royalty distribution, if one exists at all, and as long as the venture is profitable to someone, the artists involved should be compensated.
Tcl and Postscript both offer lists and lists within lists, without having to do anything fancy. Perl does have this, to an extent, but only on static lists. Anything dynamic adds a few lines of code and confuses things pretty easily. Of course, all of this came from Lisp, but I learned car and cdr and that was more than enough for me.
In short, lists only work when recursion is right at the tip of your fingertips in your language. If XSLT-2.0 works on tree fragments, we will have lists there as well.