I'm just a programmer who enjoys cool code and cool ideas. I'd likely start with FVision and add SIFT. I haven't yet found any details on the algorithm they're using to do 'reverse raytracing'. Anyone else has some pointers?
This is not limited to static scenes as one comment says. It could be used to reconstruct moving objects just as well, with a bit more software.
You could very accurately construct physical models of criminals from security tapes.
You could also construct an accurate model of how they walk. Since every person has a unique walk, that would be more difficult to disguise than physical appearance alone.
You could discover identifying details of the cars they drive, like a small dent in the fender.
This would be perfect for eBay, you could send them a short film of the object you're selling and they would post a 3D model of the item.
This heralds the end of both motion capture and the existing hours long '3D scanning' of clay models used in films like LoTR. Instead of requiring a mechanical stylus to touch every point of a model, you just film it.
Once the software has the ability to turn multiple 2D viewpoints into a single 3D image, this will be the perfect replacement for VR gloves as well. You could have a cameras on either shoulder and your hands would be your 3D mice. That sounds like a nicely intuitive interface.
Moving companies could find this useful. They could film the objects, the moving truck, and in return get an optimal packing order. You could also film the stairway up to an apartment and the software could figure out how to get through any of the particularly tight spots, if it's possible.
This would be good for the sort of augmented reality that washington.edu has researched. When the software can regognize the separate parts in a machine it can display directions for disassembly on a heads up display.
Oh, I can think of lots more uses, but better to get hold of the code and try to implement some of the random ideas above.
IceHouse! Lots of Games in one!
on
Fun Tabletop Games?
·
· Score: 3, Informative
IceHouse is awesome! It's a game construction set and has rules for lots of games, and you can make up your own. It's perfect for the creative geeky type, for the easily bored type, check it out! Looney Labs has lots of other neat games too, Fluxx, Aquarius, many more.
Check out Lava at Xilinx, Lava at Chalmers, Hawk, the Hardware Design and Synthesis section of Haskell Application Papers on readscheme.org.
The links above lead to programs that are used by companies like Xilinx and Intel to help designers build better chips with existing technology. There are more interesting hardware approaches being investigated with Haskell. Two that come to mind immediately are quantum computing and dataflow-based simulations more related to the Lustre and Lucid languages. Though I do know of some unfinished research in the dataflow/hardware design area, I can't find any published papers at the moment.
One day I'll get around to buying a PCI card with a FPGA and use Haskell to turn it into a reprogrammable coprocessor. So many cool things to learn, so little time...
First of all, bubble packets let you get around blocked incoming ports.
Second of all, how is this different from disallowing any incoming or outgoing connections?
Some ISPs don't allow any incoming connections, some don't allow port 25 outgoing. Blocking port 25 outgoing would be easy to fix, only block it for microsoft mailer agents. (try this on your spam filter).
Why don't ISPs upgrade to IPv6 so users have the most useful technology available to them?
ISPs don't upgrade to IPv6 because they maximize profit, not user services.
As for IPv6 being the 'most useful', it would mean every user could have a public static IP, and run their own services.
Every user could use true multicast (BitTorrent is fake multicast) meaning webcasts, online games, and many other applications would use a tiny fraction of the bandwidth they use now.
I suspect both general purpose and scientific computing can benefit from nested data parallelism.
The Nepal Project at the University of New South Wales concentrates on Multiple Program Multiple Data (MPMD). In a nutshell, any problem that can be specified as array operations can be flattened, unrolled, and automatically parallelized. This is not the holy grail of general purpose transparent parallelization of purely functional programs, but instead nested data parallelism. This extends research done in data parallel languages such as Nesl, Sisal, and really nifty algorithm shape research done in FISh.
This is the best approach to transparent parallelism that I've seen yet. Anyone know anything better?
declarative parallelism will always scale better than threads or whatever else
micro-optimizations will always be faster than declaractive parallelism
Manual parallelism won't scale well from one core to sixty-four cores, but will be faster in static situations like running on one Cell CPU in the PS3 where the configuration is known at design time of the app.
This is the same trade-off as manual memory allocation versus garbage collection. Garbage collection is easier and more automatic than manual memory control in C, but if you put enough effort in a C program will be more efficient than a GC-using program.
So the essence of the answer is that declaractive parallelism gives you development speed, but manual parallelism gives you execution speed. You choose what you want.
I have a two CPU machine right now, and I'm very much looking forward to the rumored SMP version of Haskell that uses Software Transactional Memory. That's gonna rock!
This is a standard paradigm in the Haskell world.
You write a new language that fits the problem domain. These are called Domain Specific Languages
For example, Simon Peyton-Jones wrote a combinator library to describe financial contracts and used it to describe the collapse of Enron. (With fascinating conclusions!) Paul Hudak has written Dance and Haskore. Dance is a language that describes dance choreography, with a handy OpenGL viewer. Haskore is a music scoring language where code looks like:
> cMajScale = Tempo 2 > (line [c 4 en [], d 4 en [], e 4 en [], f 4 en [], > g 4 en [], a 4 en [], b 4 en [], c 5 en []])
Languages, spoken or programming, or any other means of expression is most efficient when it fits the problem domain.
If this sort of thing interests you, Lambda The Ultimate is a good forum to learn more.
I like Emacs Code Browser, it's fast, featureful, and it deals with a bunch of different languages, see the screenshots
Of course, if you don't use emacs, it won't be nearly as handy.
Zope Page Templates (ZPTs) are much easier to deal with than JSPs (for example).
ZPTs put all their programmatic stuff inside tag attributes. That means that any HTML or XML understanding tool will be fine with ZPT, and you'll still get scripting power.
Here's a short example where a new <tr> is created for each user in the list user_list: <table>
<tr tal:repeat="user user_list">
<td tal:content="user/name">Name</td>
<td tal:content="user/phone">Phone</td>
</tr> </table>
Thi is really nice because our economist project leader can happily write this sort of stuff without a week-long course. And I can write Python scripts that he can easily call from inside ZPTs.
Mark, do you have any suggestions for getting Open Source accepted and used by governments and schools, both local and national?
I heard you speak at EuroPython 2004, and you said that coordinated schools were good schools, but that well-funded schools were not necessarily good schools.
Do you have advice on using open source or other resources to bring that coordination to the schools, governments or other civil organizations around us?
The difference is that you need a 'seed crystal' to get started.
If the company funds enough of the drivers to get them working decently and into mainline Linux, they'll have cards that will be used and loved for years.
If they incrementally improve the hardware design so that improving the drivers increases the performance on newer fully open hardware, they'll have a dynasty of completely open cards.
I very much want OpenGL on my SMP desktop, and I tend to use strange setups/configs that nVidia just can't handle. I would be happy to fix the problems in the nVidia code, if only I could.
If TechSource gives me a fully open and documented platform with decent hardware 3D support for less than 300 euro, and even the most basic of OpenGL drivers, I will buy it, because I can improve it and fix it myself.
Group Flow is a different thing, but it's even more powerful than single flow.
I got the power when it comes to creative and unusual ideas, but I get easily distracted into investigating all those ideas. When I work with a good partner, they point out the flaws that are obvious to them and things move a lot faster.
My experiences with PairProgramming were euphoric, each time my partner and I produced far more high quality code in a few hours than I usually produce by myself in days.
Anyway, the point of 'Agile Programming' is to be agile, you wiggle around till you find what works for you and the rest of your team. If you try pair programming and it doesn't work for you, don't do it.
One important part of eXtreme Programming is that qualified people make decisions, programmers make tech decisions, and managers support the programmers who are actually doing the work. That means pair programming or whatever should not be mandated.
The idea of XP is to form a tight and smooth team, like a jazz band, where everyone can jam together. Whether you like Acid Jazz or prefer something closer to Blues, it's the choice of the team, and every team is unique.
I own a company with my fiancee, we've been working from home for a few years.
I really enjoy it. I can sing loudly, jump up and down and scream at my code, or hack obsessively for 36 hours in a row.
We have clients in four countries and the variety keeps me interested and continually improving.
Our overhead is much much lower because we live in a very small (low rent, inexpensive food, etc) town in Sweden and our clients are mostly located in the high population density areas of Europe.
I suspect that the ultimate in flexible working is where companies range in size from one to eight people, with four being the best size.
I believe a good company works like a good jazz band, you're always improvising the next solution, so it's best to have team (band) members you know and trust.
I suspect that a company of one hundred people would be more efficient as many separate partner companies of two to six. 'Corporate reorganization' would still happen, but in that case it would show up in the changing business relations between those many units.
Think of Conway's Life applied to business.
I believe this would dramatically improve the agility of businesses.
Why should you use Windows for *anything* just because you want Windows for games?
I run Linux, and I think it's much better in every way. I'm happy to purchase native linux binary games, but I am not happy to purchase WineX or games that integrate Transgaming's WineX code to run Windows binaries on Linux.
For older games, I can just use Wine, and it doesn't cost me anything. For newer games, why would I pay money to be a second-class citizen?
Dual (or triple or...) platform development has some great advantages for the extra effort involved.
In summary, I want to fund *Linux* gaming development.
When do we get real pluggable hardware so I can buy a nice tri-band send/receive unit that plugs into my laptop/palmtop/handset and lets me have coverage from multiple vendors if I feel like it?
Linux already supports OBEX and much more.
Just give me a 'cell-phone modem' so I can have actual control over the hardware I purchase.
Specifically, I have a 6310i with tri-band and bluetooth, but the crappy black on white display and flaky software makes it effectively a voice-only phone.
I wish I could swap out the display and run Linux on it so I could write my own apps that use the Linux IrDA code.
IPv6 has both multicast and QoS that would help with this, and distributed hash systems like Chord, Bamboo, Pastry, and Tapestry could give you a system that doesn't have servers, but instead works like a low-bandwidth peer to peer MMORPG.
Creative and worthwhile ideas do exist, but who has the money to take a year off work to code them up?
Simpler solution:
What's really missing is accountability, notice earlier comments that say "9 out of 10 spams are untraceable." I'm writing an email server that works as detailed below.
In short, Only send 80 char max notifications, and make the sender keep the email on their own server for the receiver to go get by himself.
The problem with spam is that once it's in the system, it's totally trusted, and the system bears the cost of transport and storage.
If you shift the cost to the sender, spam won't be economically viable.
If spammers have to hold spam on their own servers, the servers will quickly be found out and blacklisted.
The greatest benefit is that real geeks like us will shutdown or blacklist spam server before grandma and joe q. public do their weekly email check.
Q: What about Spammed Notifications?
A: will still be an improvement over full spam emails, and takes a lot less time to download.
Q: Will mailing list servers require lots of extra space?
A: not if you consider them mailing list archives as well.
Q: How does this work for the average user that has an account with an ISP?
A: You send your email to your ISP via SMTP, just as always. Your email remains there on the server, and the server sends a notification to the final destination. The final destination then chooses when it wants to pick up the mail from the ISP's server.
As for receiving email, your client will need to pick up from many different POP3 servers, rather than just picking up from one as now.
User Stories:
A Spammer registers an account with an ISP, and sends lots of Spam.
Result: That spam remains on the server until the spammer uses up their storage quota and flags the sysadmin (who should immediately kill the account and any non-picked up spam)
Or the public blacklists list the user@host once the first few spams have been picked up, and that user@host is not accepted by clients that check blacklists.
A spammer sets up their own server, and sends lots of Spam.
Result: the server is listed in the public blacklists, and is not accepted by clients that check blacklists.
A spammer tries to forge an email sender.
Result: your client can't pick up an email from a server that doesn't exist.
Oh, it's very popular.
Check out the skilled optometrists at contactjuggling.org
Michael Moschen is the most famous contact juggler I know. You've probably seen him in the movie The Labyrinth, or maybe his PBS special.
I dunno is this particularly on-topic though.
I've been working from home for four years, I need both sufficient exercise and sufficient socialization to keep myself balanced. When I'm balanced, I can work well.
I've found that contact juggling, unicycling, and regular old juggling are both exercise and easily lead to socialization.
If that's not your style, make a point of finding new socialization opportunities whenever possible.
Slashdot Mirror
I'd like to see the Haskell sources for comparison.
http://minorgems.sf.net/Haskell.hs doesn't exist, though the the C++ and OCaml code are there.
I'm just a programmer who enjoys cool code and cool ideas. I'd likely start with FVision and add SIFT. I haven't yet found any details on the algorithm they're using to do 'reverse raytracing'. Anyone else has some pointers?
Two time-separated frames in a video and sufficiently intelligent software is a stereoscopic camera.
This is not limited to static scenes as one comment says. It could be used to reconstruct moving objects just as well, with a bit more software.
You could very accurately construct physical models of criminals from security tapes.
You could also construct an accurate model of how they walk. Since every person has a unique walk, that would be more difficult to disguise than physical appearance alone.
You could discover identifying details of the cars they drive, like a small dent in the fender.
This would be perfect for eBay, you could send them a short film of the object you're selling and they would post a 3D model of the item.
This heralds the end of both motion capture and the existing hours long '3D scanning' of clay models used in films like LoTR. Instead of requiring a mechanical stylus to touch every point of a model, you just film it.
Once the software has the ability to turn multiple 2D viewpoints into a single 3D image, this will be the perfect replacement for VR gloves as well. You could have a cameras on either shoulder and your hands would be your 3D mice. That sounds like a nicely intuitive interface.
Moving companies could find this useful. They could film the objects, the moving truck, and in return get an optimal packing order. You could also film the stairway up to an apartment and the software could figure out how to get through any of the particularly tight spots, if it's possible.
This would be good for the sort of augmented reality that washington.edu has researched. When the software can regognize the separate parts in a machine it can display directions for disassembly on a heads up display.
Oh, I can think of lots more uses, but better to get hold of the code and try to implement some of the random ideas above.
IceHouse is awesome! It's a game construction set and has rules for lots of games, and you can make up your own. It's perfect for the creative geeky type, for the easily bored type, check it out!
Looney Labs has lots of other neat games too, Fluxx, Aquarius, many more.
Check out Lava at Xilinx, Lava at Chalmers, Hawk, the Hardware Design and Synthesis section of Haskell Application Papers on readscheme.org.
The links above lead to programs that are used by companies like Xilinx and Intel to help designers build better chips with existing technology. There are more interesting hardware approaches being investigated with Haskell. Two that come to mind immediately are quantum computing and dataflow-based simulations more related to the Lustre and Lucid languages. Though I do know of some unfinished research in the dataflow/hardware design area, I can't find any published papers at the moment.
One day I'll get around to buying a PCI card with a FPGA and use Haskell to turn it into a reprogrammable coprocessor. So many cool things to learn, so little time...
First of all, bubble packets let you get around blocked incoming ports.
Second of all, how is this different from disallowing any incoming or outgoing connections?
Some ISPs don't allow any incoming connections, some don't allow port 25 outgoing.
Blocking port 25 outgoing would be easy to fix, only block it for microsoft mailer agents. (try this on your spam filter).
Why don't ISPs upgrade to IPv6 so users have the most useful technology available to them?
ISPs don't upgrade to IPv6 because they maximize profit, not user services.
As for IPv6 being the 'most useful', it would mean every user could have a public static IP, and run their own services.
Every user could use true multicast (BitTorrent is fake multicast) meaning webcasts, online games, and many other applications would use a tiny fraction of the bandwidth they use now.
I suspect both general purpose and scientific computing can benefit from nested data parallelism.
The Nepal Project at the University of New South Wales concentrates on Multiple Program Multiple Data (MPMD). In a nutshell, any problem that can be specified as array operations can be flattened, unrolled, and automatically parallelized. This is not the holy grail of general purpose transparent parallelization of purely functional programs, but instead nested data parallelism. This extends research done in data parallel languages such as Nesl, Sisal, and really nifty algorithm shape research done in FISh.
This is the best approach to transparent parallelism that I've seen yet. Anyone know anything better?
- declarative parallelism will always scale better than threads or whatever else
- micro-optimizations will always be faster than declaractive parallelism
Manual parallelism won't scale well from one core to sixty-four cores, but will be faster in static situations like running on one Cell CPU in the PS3 where the configuration is known at design time of the app.This is the same trade-off as manual memory allocation versus garbage collection. Garbage collection is easier and more automatic than manual memory control in C, but if you put enough effort in a C program will be more efficient than a GC-using program.
So the essence of the answer is that declaractive parallelism gives you development speed, but manual parallelism gives you execution speed. You choose what you want.
I have a two CPU machine right now, and I'm very much looking forward to the rumored SMP version of Haskell that uses Software Transactional Memory. That's gonna rock!
In essence, IRC is an old protocol, it needs help.
SILC is unicode-only, encrypt-only, allows MIME messages, uses a hybrid topology to be netsplit resistant, etc.
SILC isn't fully p2p as one poster suggested, but it's a lot better than IRC.
For example, Simon Peyton-Jones wrote a combinator library to describe financial contracts and used it to describe the collapse of Enron. (With fascinating conclusions!)
Paul Hudak has written Dance and Haskore. Dance is a language that describes dance choreography, with a handy OpenGL viewer. Haskore is a music scoring language where code looks like:Languages, spoken or programming, or any other means of expression is most efficient when it fits the problem domain.
If this sort of thing interests you, Lambda The Ultimate is a good forum to learn more.
I like Emacs Code Browser, it's fast, featureful, and it deals with a bunch of different languages, see the screenshots
Of course, if you don't use emacs, it won't be nearly as handy.
The #zope irc channel on the freenode IRC network is very helpful if you're polite.
I've been there off and on the last two or three years. Lots of helpful and knowledgable people.
Zope Page Templates (ZPTs) are much easier to deal with than JSPs (for example).
ZPTs put all their programmatic stuff inside tag attributes. That means that any HTML or XML understanding tool will be fine with ZPT, and you'll still get scripting power.
Here's a short example where a new <tr> is created for each user in the list user_list:
<table>
<tr tal:repeat="user user_list">
<td tal:content="user/name">Name</td>
<td tal:content="user/phone">Phone</td>
</tr>
</table>
Thi is really nice because our economist project leader can happily write this sort of stuff without a week-long course. And I can write Python scripts that he can easily call from inside ZPTs.
Mark, do you have any suggestions for getting Open Source accepted and used by governments and schools, both local and national?
I heard you speak at EuroPython 2004, and you said that coordinated schools were good schools, but that well-funded schools were not necessarily good schools.
Do you have advice on using open source or other resources to bring that coordination to the schools, governments or other civil organizations around us?
The difference is that you need a 'seed crystal' to get started.
If the company funds enough of the drivers to get them working decently and into mainline Linux, they'll have cards that will be used and loved for years.
If they incrementally improve the hardware design so that improving the drivers increases the performance on newer fully open hardware, they'll have a dynasty of completely open cards.
I very much want OpenGL on my SMP desktop, and I tend to use strange setups/configs that nVidia just can't handle. I would be happy to fix the problems in the nVidia code, if only I could.
If TechSource gives me a fully open and documented platform with decent hardware 3D support for less than 300 euro, and even the most basic of OpenGL drivers, I will buy it, because I can improve it and fix it myself.
Group Flow is a different thing, but it's even more powerful than single flow.
I got the power when it comes to creative and unusual ideas, but I get easily distracted into investigating all those ideas. When I work with a good partner, they point out the flaws that are obvious to them and things move a lot faster.
My experiences with PairProgramming were euphoric, each time my partner and I produced far more high quality code in a few hours than I usually produce by myself in days.
Anyway, the point of 'Agile Programming' is to be agile, you wiggle around till you find what works for you and the rest of your team. If you try pair programming and it doesn't work for you, don't do it.
One important part of eXtreme Programming is that qualified people make decisions, programmers make tech decisions, and managers support the programmers who are actually doing the work.
That means pair programming or whatever should not be mandated.
The idea of XP is to form a tight and smooth team, like a jazz band, where everyone can jam together. Whether you like Acid Jazz or prefer something closer to Blues, it's the choice of the team, and every team is unique.
I own a company with my fiancee, we've been working from home for a few years.
I really enjoy it. I can sing loudly, jump up and down and scream at my code, or hack obsessively for 36 hours in a row.
We have clients in four countries and the variety keeps me interested and continually improving.
Our overhead is much much lower because we live in a very small (low rent, inexpensive food, etc) town in Sweden and our clients are mostly located in the high population density areas of Europe.
I suspect that the ultimate in flexible working is where companies range in size from one to eight people, with four being the best size.
I believe a good company works like a good jazz band, you're always improvising the next solution, so it's best to have team (band) members you know and trust.
I suspect that a company of one hundred people would be more efficient as many separate partner companies of two to six. 'Corporate reorganization' would still happen, but in that case it would show up in the changing business relations between those many units.
Think of Conway's Life applied to business.
I believe this would dramatically improve the agility of businesses.
Any feedback on these ideas?
Why should you use Windows for *anything* just because you want Windows for games?
...) platform development has some great advantages for the extra effort involved.
I run Linux, and I think it's much better in every way. I'm happy to purchase native linux binary games, but I am not happy to purchase WineX or games that integrate Transgaming's WineX code to run Windows binaries on Linux.
For older games, I can just use Wine, and it doesn't cost me anything. For newer games, why would I pay money to be a second-class citizen?
Dual (or triple or
In summary, I want to fund *Linux* gaming development.
When do we get real pluggable hardware so I can buy a nice tri-band send/receive unit that plugs into my laptop/palmtop/handset and lets me have coverage from multiple vendors if I feel like it?
Linux already supports OBEX and much more.
Just give me a 'cell-phone modem' so I can have actual control over the hardware I purchase.
Specifically, I have a 6310i with tri-band and bluetooth, but the crappy black on white display and flaky software makes it effectively a voice-only phone.
I wish I could swap out the display and run Linux on it so I could write my own apps that use the Linux IrDA code.
IPv6 has both multicast and QoS that would help with this, and distributed hash systems like Chord, Bamboo, Pastry, and Tapestry could give you a system that doesn't have servers, but instead works like a low-bandwidth peer to peer MMORPG.
Creative and worthwhile ideas do exist, but who has the money to take a year off work to code them up?
On a slightly related subject, is there an inexpensive (10,000 USD or so) way to get decent VR on a linux desktop?
I'd like to have stereo motion-tracked goggles at least, and one or two gloves would be nice also.
Since it's easy enough to run two or more graphics cards at the same time, I'd like to find a headset that takes two DVI inputs.
Simpler solution:
What's really missing is accountability, notice earlier comments that say "9 out of 10 spams are untraceable." I'm writing an email server that works as detailed below.
In short, Only send 80 char max notifications, and make the sender keep the email on their own server for the receiver to go get by himself.
The problem with spam is that once it's in the system, it's totally trusted, and the system bears the cost of transport and storage.
If you shift the cost to the sender, spam won't be economically viable.
If spammers have to hold spam on their own servers, the servers will quickly be found out and blacklisted.
The greatest benefit is that real geeks like us will shutdown or blacklist spam server before grandma and joe q. public do their weekly email check.
Q: What about Spammed Notifications?
A: will still be an improvement over full spam emails, and takes a lot less time to download.
Q: Will mailing list servers require lots of extra space?
A: not if you consider them mailing list archives as well.
Q: How does this work for the average user that has an account with an ISP?
A: You send your email to your ISP via SMTP, just as always. Your email remains there on the server, and the server sends a notification to the final destination. The final destination then chooses when it wants to pick up the mail from the ISP's server.
As for receiving email, your client will need to pick up from many different POP3 servers, rather than just picking up from one as now.
User Stories:
A Spammer registers an account with an ISP, and sends lots of Spam.
Result: That spam remains on the server until the spammer uses up their storage quota and flags the sysadmin (who should immediately kill the account and any non-picked up spam)
Or the public blacklists list the user@host once the first few spams have been picked up, and that user@host is not accepted by clients that check blacklists.
A spammer sets up their own server, and sends lots of Spam.
Result: the server is listed in the public blacklists, and is not accepted by clients that check blacklists.
A spammer tries to forge an email sender.
Result: your client can't pick up an email from a server that doesn't exist.
Oh, it's very popular.
Check out the skilled optometrists at contactjuggling.org
Michael Moschen is the most famous contact juggler I know. You've probably seen him in the movie The Labyrinth, or maybe his PBS special.
I dunno is this particularly on-topic though.
Seriously, I bought myself a unicycle.
I've been working from home for four years, I need both sufficient exercise and sufficient socialization to keep myself balanced. When I'm balanced, I can work well.
I've found that contact juggling, unicycling, and regular old juggling are both exercise and easily lead to socialization.
If that's not your style, make a point of finding new socialization opportunities whenever possible.