This is a great thing, but the problem I'm really seeing here is that there was nothing in the contract that suggested UI research be done. The truth is, the UIs of these programs they're going to be using (including the back-ends) are really pretty horrific.
KMail, etc. really need some interface overhauls, and Kolab could really benefit from a central "component-type" (see Horde) administration interface. What would make more sense? Having to use a bunch of different open terminals, web admin tools, and configuration scripts? Or a single management interface that would let administrators install Kolab and manage Kolab as Kolab. Not as Cyrus+Sendmail+ProFTPd+OpenLDAP+etc.
Well, maybe for you, but not for me. The only problem is that I need a very customizable setup. That's why a single box like this wouldn't cut it for me. Here's some of what I have:
Linux + Windows + Mac + ViewSonic V50HRTV -> 4-channel stereo mixer -> 5.1 speakers with Dolby Digital, Dolby Surround, Dolby Pro Logic.
(Console systems + television + CD player) -> ViewSonic V50HRTV + Linux + Windows + FreeBSD/Solaris -> 4-port KVM + Mac -> side-by-side 21" CRTs.
That said, I have to wonder if Bill Gates (for once) was right when he suggested several years ago that "media convergence" isn't really a thing that people want. People want to compute on computers, watch TV on a television, and watch movies in a movie theater. Converging the three into the single PC -- or the PC breakout box hooked up to a PC -- is nifty and very George Jetson-like (and who can forget his boy Elroy spiralling down from the old man's hoverbug in a mini-hoverbug of his very own?) -- but it seems that technology (in this case and others [palladium and MS's MediaPC's especially) is thinking too far ahead for its own good.
"I've never gone behind Mr. Gates' back before but Congress's copyright-expansionist views, er, conflict with my...choice of lifestyle. All I can do is give you one name: Richard Stallman. Find him, and you'll find your answer."
If the four buildings just need to be networked together, and you _don't_ need Internet access (a lot of other posts seem to think this is for Internet access, but neatan only mentioned the Internet for use with VPN), then you should be able to simply use the copper wiring to create your own little 4-building "DSL-LAN".
This was posted on/. a while ago: I, Cringely's Roll Your Own DSL. He gives you some basic direction for grabbing extra copper pairs from the telcos and plunking modems on each end. I'm sure there's more to it than what he describes, but that should give you 2Mbps symmetric bandwidth between your four buildings. The Lariat guys might be able to give you more help.
In a recent issue of Wired (the one with Moby on the cover), Moby himself stated that his new album, 18, is very similar to his previous work Play. He also stated that the "management" was concerned that having such a similar album to Play would hurt sales. While I do own Play and enjoy it very much, I personally prefer Moby's older works, which is one reason I have not purchased 18. Perhaps the reason 18 is having slower sales is because it is so similar to Play; with Electronica, I believe people are less likely to purchase similar material (a lot of Electronica can sound the same).
I'm betting that the.pro TLD will become pretty worthless just because the people approving assignment of.pro domains won't be able to verify/authenticate the legitimacy or professional competence and ability of the applicants. In the end, there will be a lot of.pro sites which don't really have respected or competent professionals. Anyone with a degree or certificate will be granted a domain, much like how commercial companies are granted.net and.org.
Anyway, how do you find a lawyer, accountant, or doctor? I don't think you just search the Internet and pick the first few which return with a.pro domain. The reason 1-800-DENTIST is successful is because they are promising high quality dentists, and they make money by fulfilling that promise. I doubt.pro is really prepared or willing to do that. Someone check their business model.
Inkwell is probably the most significant addition any software developer could provide as an operating system addition. This has the potential to be the most important change in user interfaces since the GUI.
Why? Because an extremely large percentage of the world's population does not use the Roman alphabet. Think about how revolutionary it would be for Chinese, Japanese, Korean, Arabic, etc. people to be able to interface with computers in their native languages. All the klunky hacks for using keyboards to type in one of 65,000 different characters are no longer necessary.
(This is actually something I was just thinking about playing Ragnarok Online where more than half the people don't speak English natively.)
So we just have to see if Inkwell is made available for the foreign language features of Mac OS X. Hopefully it will, because so far Apple's been very good about supporting many different languages simultaneously in Mac OS X.
A great book on the BeOS file system, its design and the issues involved, is Practical File System Design with the Be File System by Dominic Giampaolo. This was the first book I read about file systems and is a great discussion on the types of decisions the BeOS folks made to address the specific users and OS functionality they were looking for. It's not too heavy and is really a casual read in comparison to most other Computer Science books.
How about Group Organizer by Chronos Software? Not only does it provide the typical PIM functions, but it is specifically designed for developers/consultants and contains features like Gantt charts.
I used an older version of Personal Organizer (it used to be called Consultant) and it was great.
If you want some real experience working on software development, try asking your professors. After all, many of them are working on large software projects and could always use your slave labor. The resources and projects are there. It sounds like you just haven't taken the time to look in your own backyard.
If you really want open-source, then professors are still a good person to ask because they are probably working on open-source software projects funded by the NSF or some other government agency (well, at least those at public institutions probably are, since the private ones may have deals with industry). I've been working on Open Mash, a very large open-source project, for two years now because my faculty advisor is the guy heading that project.
I think OTcl and TclCL does everything on that list, at least to some extent. I'm mostly dealing with the OTcl part here.
While I agree with many of the points made by other people about how the language isn't as important the problem, it's quite possible that the boss knows that if a language can be found to do all that, it will make the implementation a lot easier. Either that or he/she's got pointy-hair genes.
Anyway, down the list:
IDE - wish + text editor, e.g. emacs. This isn't your "traditional" IDE, but it does pretty much everything an IDE needs to. Easy to debug because you're in an interactive shell which does everything right in front of you. Easy to write files and then execute/debug them immediately. But no point-and-click creation of code.
GUI Design - wish. Type it, see it immediately. Decent layout management.
Event Handling - Tcl/Tk event handling. Pretty basic. You can simply "bind" events to functions.
Advanced Error Handling - this is your job, but the feature is there. e.g. catch.
Advanced Object Oriented Design, Multiple Inheritance, Abstract Classes - oTcl gives this to you on a Tcl level. Abstract classes aren't specifically provided, but you could enforce "abstract" classes via the constructor/init code.
Garbage Collection - There's no automatic garbage collection, but it does reference counting.
Overloading - yep.
Cross-platform - Tcl/Tk runs just about everywhere.
I fail to see how this little rectangular box is cuter than an iMac the poster has seen. Apparently, said poster gets excited over tool boxes and tiny file cabinets as well.
I wonder how this is going to affect system administration at the University of California, Berkeley. We use Solaris 8 x86 machines in the undergraduate labs, along with a SunRay "cluster" running off a SparcStation.
My feeling is that we use Solaris because it's something of a standard as far as Internet communications goes. It doesn't provide all these extra libraries or APIs that you might find on Linux, BSD, or Windows while still complying completely with all the standards. I could easily be wrong here since I've never had access to HP-UX, AIX, etc. But if you coded network stuff for Solaris, I was able to get it to compile and run on Linux and BSD. Not so the other way around.
I also have a Solaris 8 x86 box at home (previously Solaris 7 x86) for the sole purpose of development at UC Berkeley. My home box is also used as a test machine for Solaris x86 compatibility of the Open Mash project I work on.
If x86 support is removed for Solaris 9, then those Solaris 8 x86 boxes in the undergraduate labs will eventually have to be replaced by a non-Solaris OS, as Sun's application development and support leaves Solaris 8 behind. At this point, that would mean Linux or BSD. Hard to say which, since there are currently both BSD and Linux machines running for undergraduate use, but probably BSD. It also means I'm either going to have to get a SparcStation or be stuck with Solaris 8. And I doubt I'm getting a SparcStation.
Sun may need to rethink the role Solaris x86 plays from a long-term viewpoint, instead of a bottom-line viewpoint.
Whenever a student in CS150 here at UC Berkeley launches IE while logged into a lab computer, MSN's home page comes up. Of course, we immediately go to the CS150 home page, but the MSN page does pop up first. If you add that up for a few months, that's quite a few MSN page views that shouldn't really count.
And then, we had a Microsoft guy come to the last day of class to give out prizes to groups with the best projects. Hm...
When was the last time you read about a buffer exploit in Lasso or the TCP/IP stack resulting in an serious compromise in Mac OS 9?
Oh, right about the time of this message [fi.upm.es] to bugtraq...
I stand corrected:) Although I would argue this isn't as serious a compromise (it's still serious) since you can't do anything more than read the data.
Sounds like Mac OS 9 is the best protection against this now. (+5 informative)
WTF?!
Slashdot moderation system (-5 overrated).
Mac OS 9 (and previous Mac OS operating systems) is the only operating system I know of that can provide a great deal of Internet-related services (web server, routing, database access, etc.) while not providing an attacker with much to work with. When was the last time you read about a buffer exploit in Lasso or the TCP/IP stack resulting in an serious compromise in Mac OS 9?
I guess I should've been more clear. I meant locking them down as in cutting off external access to data.
Also, I'm pretty sure making their web site unavailable is simply one part of the judge's sweeping order to cut off Internet access to all of their systems. What I saw indicated cutting off everything, not just their web server.
I would think that not having access to the Internet from your workplace would have a great effect on your ability to do function, for the majority of large businesses and organizations.
Lots of information is available at the Indian Trust: Cobell v. Norton web site. Press releases plus offical court documents.
Of particular interest is this document, which more fully explains why the judge ordered all Internet access to the Department of Interior. Apparently, court investigators were able to break in and modify lots of important information without any response from the DoI.
Seems like this sets a legal precendence for locking down an entire business, organization, or corporation involved in a legal situation. If it can be demonstrated that it would be possible for an outside entity to modify data crucial to the proceeding of the case (such data would be subpeonaed), the judge can order all external access to that data cut off.
Since simply running a some Microsoft software makes it possible for a large number of outside entities to modify such data without difficulty, and to know that doing so is possible without having to figure it out, I could see this becoming a problem for businesses and organizations that run said Microsoft software.
However, it also means that lax UNIX administrators could have their systems' access cut off if court investigators demonstrate that they are able to get in. Sounds like Mac OS 9 is the best protection against this now.
The Apple iPod does not have any "anti-piracy" features built in. Specifically, it states something to the effect of "Piracy is a social issue, not a technological one" on the packaging. You can use it with Windows (via Mediafour's XPod) or Macs, and probably soon with Linux. If you use the iPod as a portable drive on a Mac, you can simply copy files back and forth at will. But if you use iTunes to sync, the MP3 files are invisible. Some information on this is available here, and here's a simple utility to access the invisible MP3 files.
Also, the iPod supports a variety of encodings. It should support up to 256Kbps (or is it 320Kbps?), variable bit rate, joint stereo or normal stereo, because that's what iTunes supports. The 1000 songs it advertises is for 160Kbps songs.
In other news, Sony has announced a new Television Family License which allows all members of a family or household (up to 5 individuals) to watch the same television, without violating the Sony Home Electronics License Agreement.
I've heard that this attempt at humor described above was actually true. Apparently, the movie industry was originally opposed to movie rentals. The reason? They wouldn't be able to control how many people would be watching the movie displayed on a home television set. In their eyes, they were losing the cost of a ticket every time an additional person watched the movie.
A lot of people are providing excellent links to free scientific software resources, but one of the key points mentioned was the lack of a high speed Internet connection. FlashBoltzman also specifically said they were looking at Debian because of the applications bundled with it.
Based on this, I think what would help the most would be hard copies of the scientific software people have mentioned. I would also recommend looking at SuSE's boxed distribution, because it contains 7 CDs or 1 DVD worth of software. Spending a few hundred U.S. dollars to get a box into every organization is probably much cheaper than the amount they would have to spend on their slow Internet connections to download several GB of data to each organization.
However, those scientific and research packages mentioned aren't going to be part of any distribution. FlashBoltzman can post the resources listed here to a web page, but maybe he should routinely grab software and then each month or quarter burn a few CDs or DVDs to send to Africa.
What Microsoft is really proposing is simply to have Bill donate a lot of money to his existing Bill & Melinda Gates Foundation? If you look at the grant breakdown, you'll see that $1.6 Billion has already been "granted" to Education. Increasing that to $2.7 Billion over five years makes an excellent tax write-off, although I'm not too certain that it wouldn't have reached $2.7 Billion in five years on its own anyway.
BG: I know! I'm giving away money each year to education anyway; let's tell them that I'm going to do it to settle this. Mwahahaahaha!
It's also worth noting that $160 Million goes towards what is essentially an MSCE-primer school, and then $38 Million goes towards paying those MSCE-primer students to support to new computers. And 200,000 reconditioned computers and laptops? In other words, they are simply redirecting what would otherwise be either landfill or freely donated anyway. I don't understand the $90 Million in teacher training either, unless it is not how to use computers, but how to make use of computers in an educational environment. Wasn't Windows XP supposed to be as easy to use as a Mac? They copied everything else, why not ease of use? (Microsoft doesn't have R&D, only D.) And yeah, $900 Million in software probably has a real cost to Microsoft of $1 Million. People need to know that the cost of duplicating software is nil.
For Mash, we have a C++ base that provides all of the lower-level multicast multimedia functionality and we use Tcl/Tk to create the applications. The C++ code lets us implement the low-level functionality (such as JPEG decoding, RTP transport, etc.) efficiently, but using Tcl/Tk to interface with the C++ makes it very easy to develop applications, test new ideas, and not worry too much about cross-platform UI.
Of course, you can't take a normal Tcl/Tk binaries and just arbitrarily call compiled C code, so we do require custom tclsh and wish binaries which have the C++ objects built into it. Our custom binary lets you define a C++ class as a subclass of TclClass. Then, we use otcl and tclcl within tclsh and wish to refer to those C++ classes as Tcl objects.
Christopher mentions three domains which have bandwidth costs of $10,000/mo. Seems to me like this is a good time to start mirrors of www.limewire.com and www.limewire.org. Not sure how router.limewire.com is used (I haven't read up on the Gnutella network's behavior), but couldn't that be distributed as well?
This is a great thing, but the problem I'm really seeing here is that there was nothing in the contract that suggested UI research be done. The truth is, the UIs of these programs they're going to be using (including the back-ends) are really pretty horrific.
KMail, etc. really need some interface overhauls, and Kolab could really benefit from a central "component-type" (see Horde) administration interface. What would make more sense? Having to use a bunch of different open terminals, web admin tools, and configuration scripts? Or a single management interface that would let administrators install Kolab and manage Kolab as Kolab. Not as Cyrus+Sendmail+ProFTPd+OpenLDAP+etc.
Well, maybe for you, but not for me. The only problem is that I need a very customizable setup. That's why a single box like this wouldn't cut it for me. Here's some of what I have:
Linux + Windows + Mac + ViewSonic V50HRTV -> 4-channel stereo mixer -> 5.1 speakers with Dolby Digital, Dolby Surround, Dolby Pro Logic.
(Console systems + television + CD player) -> ViewSonic V50HRTV + Linux + Windows + FreeBSD/Solaris -> 4-port KVM + Mac -> side-by-side 21" CRTs.
That said, I have to wonder if Bill Gates (for once) was right when he suggested several years ago that "media convergence" isn't really a thing that people want. People want to compute on computers, watch TV on a television, and watch movies in a movie theater. Converging the three into the single PC -- or the PC breakout box hooked up to a PC -- is nifty and very George Jetson-like (and who can forget his boy Elroy spiralling down from the old man's hoverbug in a mini-hoverbug of his very own?) -- but it seems that technology (in this case and others [palladium and MS's MediaPC's especially) is thinking too far ahead for its own good.
"I've never gone behind Mr. Gates' back before but Congress's copyright-expansionist views, er, conflict with my...choice of lifestyle. All I can do is give you one name: Richard Stallman. Find him, and you'll find your answer."
Haven't you ever seen John Cage on Ally McBeal? He enjoys a fresh bowl.
"Unclean bowls trouble me."
"Sometimes people leave in haste forgetting to flush, other times they leave residual reminents. I like a fresh bowl."
If the four buildings just need to be networked together, and you _don't_ need Internet access (a lot of other posts seem to think this is for Internet access, but neatan only mentioned the Internet for use with VPN), then you should be able to simply use the copper wiring to create your own little 4-building "DSL-LAN".
/. a while ago: I, Cringely's Roll Your Own DSL. He gives you some basic direction for grabbing extra copper pairs from the telcos and plunking modems on each end. I'm sure there's more to it than what he describes, but that should give you 2Mbps symmetric bandwidth between your four buildings. The Lariat guys might be able to give you more help.
This was posted on
In a recent issue of Wired (the one with Moby on the cover), Moby himself stated that his new album, 18, is very similar to his previous work Play. He also stated that the "management" was concerned that having such a similar album to Play would hurt sales. While I do own Play and enjoy it very much, I personally prefer Moby's older works, which is one reason I have not purchased 18. Perhaps the reason 18 is having slower sales is because it is so similar to Play; with Electronica, I believe people are less likely to purchase similar material (a lot of Electronica can sound the same).
I'm betting that the .pro TLD will become pretty worthless just because the people approving assignment of .pro domains won't be able to verify/authenticate the legitimacy or professional competence and ability of the applicants. In the end, there will be a lot of .pro sites which don't really have respected or competent professionals. Anyone with a degree or certificate will be granted a domain, much like how commercial companies are granted .net and .org.
.pro domain. The reason 1-800-DENTIST is successful is because they are promising high quality dentists, and they make money by fulfilling that promise. I doubt .pro is really prepared or willing to do that. Someone check their business model.
Anyway, how do you find a lawyer, accountant, or doctor? I don't think you just search the Internet and pick the first few which return with a
Inkwell is probably the most significant addition any software developer could provide as an operating system addition. This has the potential to be the most important change in user interfaces since the GUI.
Why? Because an extremely large percentage of the world's population does not use the Roman alphabet. Think about how revolutionary it would be for Chinese, Japanese, Korean, Arabic, etc. people to be able to interface with computers in their native languages. All the klunky hacks for using keyboards to type in one of 65,000 different characters are no longer necessary.
(This is actually something I was just thinking about playing Ragnarok Online where more than half the people don't speak English natively.)
So we just have to see if Inkwell is made available for the foreign language features of Mac OS X. Hopefully it will, because so far Apple's been very good about supporting many different languages simultaneously in Mac OS X.
A great book on the BeOS file system, its design and the issues involved, is Practical File System Design with the Be File System by Dominic Giampaolo. This was the first book I read about file systems and is a great discussion on the types of decisions the BeOS folks made to address the specific users and OS functionality they were looking for. It's not too heavy and is really a casual read in comparison to most other Computer Science books.
How about Group Organizer by Chronos Software? Not only does it provide the typical PIM functions, but it is specifically designed for developers/consultants and contains features like Gantt charts.
I used an older version of Personal Organizer (it used to be called Consultant) and it was great.
If you want some real experience working on software development, try asking your professors. After all, many of them are working on large software projects and could always use your slave labor. The resources and projects are there. It sounds like you just haven't taken the time to look in your own backyard.
If you really want open-source, then professors are still a good person to ask because they are probably working on open-source software projects funded by the NSF or some other government agency (well, at least those at public institutions probably are, since the private ones may have deals with industry). I've been working on Open Mash, a very large open-source project, for two years now because my faculty advisor is the guy heading that project.
I think OTcl and TclCL does everything on that list, at least to some extent. I'm mostly dealing with the OTcl part here.
While I agree with many of the points made by other people about how the language isn't as important the problem, it's quite possible that the boss knows that if a language can be found to do all that, it will make the implementation a lot easier. Either that or he/she's got pointy-hair genes.
Anyway, down the list:
IDE - wish + text editor, e.g. emacs. This isn't your "traditional" IDE, but it does pretty much everything an IDE needs to. Easy to debug because you're in an interactive shell which does everything right in front of you. Easy to write files and then execute/debug them immediately. But no point-and-click creation of code.
GUI Design - wish. Type it, see it immediately. Decent layout management.
Event Handling - Tcl/Tk event handling. Pretty basic. You can simply "bind" events to functions.
Advanced Error Handling - this is your job, but the feature is there. e.g. catch.
Advanced Object Oriented Design, Multiple Inheritance, Abstract Classes - oTcl gives this to you on a Tcl level. Abstract classes aren't specifically provided, but you could enforce "abstract" classes via the constructor/init code.
Garbage Collection - There's no automatic garbage collection, but it does reference counting.
Overloading - yep.
Cross-platform - Tcl/Tk runs just about everywhere.
I fail to see how this little rectangular box is cuter than an iMac the poster has seen. Apparently, said poster gets excited over tool boxes and tiny file cabinets as well.
I wonder how this is going to affect system administration at the University of California, Berkeley. We use Solaris 8 x86 machines in the undergraduate labs, along with a SunRay "cluster" running off a SparcStation.
My feeling is that we use Solaris because it's something of a standard as far as Internet communications goes. It doesn't provide all these extra libraries or APIs that you might find on Linux, BSD, or Windows while still complying completely with all the standards. I could easily be wrong here since I've never had access to HP-UX, AIX, etc. But if you coded network stuff for Solaris, I was able to get it to compile and run on Linux and BSD. Not so the other way around.
I also have a Solaris 8 x86 box at home (previously Solaris 7 x86) for the sole purpose of development at UC Berkeley. My home box is also used as a test machine for Solaris x86 compatibility of the Open Mash project I work on.
If x86 support is removed for Solaris 9, then those Solaris 8 x86 boxes in the undergraduate labs will eventually have to be replaced by a non-Solaris OS, as Sun's application development and support leaves Solaris 8 behind. At this point, that would mean Linux or BSD. Hard to say which, since there are currently both BSD and Linux machines running for undergraduate use, but probably BSD. It also means I'm either going to have to get a SparcStation or be stuck with Solaris 8. And I doubt I'm getting a SparcStation.
Sun may need to rethink the role Solaris x86 plays from a long-term viewpoint, instead of a bottom-line viewpoint.
Whenever a student in CS150 here at UC Berkeley launches IE while logged into a lab computer, MSN's home page comes up. Of course, we immediately go to the CS150 home page, but the MSN page does pop up first. If you add that up for a few months, that's quite a few MSN page views that shouldn't really count.
And then, we had a Microsoft guy come to the last day of class to give out prizes to groups with the best projects. Hm...
Mac OS 9 (and previous Mac OS operating systems) is the only operating system I know of that can provide a great deal of Internet-related services (web server, routing, database access, etc.) while not providing an attacker with much to work with. When was the last time you read about a buffer exploit in Lasso or the TCP/IP stack resulting in an serious compromise in Mac OS 9?
I guess I should've been more clear. I meant locking them down as in cutting off external access to data.
Also, I'm pretty sure making their web site unavailable is simply one part of the judge's sweeping order to cut off Internet access to all of their systems. What I saw indicated cutting off everything, not just their web server.
I would think that not having access to the Internet from your workplace would have a great effect on your ability to do function, for the majority of large businesses and organizations.
Lots of information is available at the Indian Trust: Cobell v. Norton web site. Press releases plus offical court documents.
Of particular interest is this document, which more fully explains why the judge ordered all Internet access to the Department of Interior. Apparently, court investigators were able to break in and modify lots of important information without any response from the DoI.
Seems like this sets a legal precendence for locking down an entire business, organization, or corporation involved in a legal situation. If it can be demonstrated that it would be possible for an outside entity to modify data crucial to the proceeding of the case (such data would be subpeonaed), the judge can order all external access to that data cut off.
Since simply running a some Microsoft software makes it possible for a large number of outside entities to modify such data without difficulty, and to know that doing so is possible without having to figure it out, I could see this becoming a problem for businesses and organizations that run said Microsoft software.
However, it also means that lax UNIX administrators could have their systems' access cut off if court investigators demonstrate that they are able to get in. Sounds like Mac OS 9 is the best protection against this now.
The Apple iPod does not have any "anti-piracy" features built in. Specifically, it states something to the effect of "Piracy is a social issue, not a technological one" on the packaging. You can use it with Windows (via Mediafour's XPod) or Macs, and probably soon with Linux. If you use the iPod as a portable drive on a Mac, you can simply copy files back and forth at will. But if you use iTunes to sync, the MP3 files are invisible. Some information on this is available here, and here's a simple utility to access the invisible MP3 files.
Also, the iPod supports a variety of encodings. It should support up to 256Kbps (or is it 320Kbps?), variable bit rate, joint stereo or normal stereo, because that's what iTunes supports. The 1000 songs it advertises is for 160Kbps songs.
I've heard that this attempt at humor described above was actually true. Apparently, the movie industry was originally opposed to movie rentals. The reason? They wouldn't be able to control how many people would be watching the movie displayed on a home television set. In their eyes, they were losing the cost of a ticket every time an additional person watched the movie.
A lot of people are providing excellent links to free scientific software resources, but one of the key points mentioned was the lack of a high speed Internet connection. FlashBoltzman also specifically said they were looking at Debian because of the applications bundled with it.
Based on this, I think what would help the most would be hard copies of the scientific software people have mentioned. I would also recommend looking at SuSE's boxed distribution, because it contains 7 CDs or 1 DVD worth of software. Spending a few hundred U.S. dollars to get a box into every organization is probably much cheaper than the amount they would have to spend on their slow Internet connections to download several GB of data to each organization.
However, those scientific and research packages mentioned aren't going to be part of any distribution. FlashBoltzman can post the resources listed here to a web page, but maybe he should routinely grab software and then each month or quarter burn a few CDs or DVDs to send to Africa.
What Microsoft is really proposing is simply to have Bill donate a lot of money to his existing Bill & Melinda Gates Foundation? If you look at the grant breakdown, you'll see that $1.6 Billion has already been "granted" to Education. Increasing that to $2.7 Billion over five years makes an excellent tax write-off, although I'm not too certain that it wouldn't have reached $2.7 Billion in five years on its own anyway.
BG: I know! I'm giving away money each year to education anyway; let's tell them that I'm going to do it to settle this. Mwahahaahaha!
It's also worth noting that $160 Million goes towards what is essentially an MSCE-primer school, and then $38 Million goes towards paying those MSCE-primer students to support to new computers. And 200,000 reconditioned computers and laptops? In other words, they are simply redirecting what would otherwise be either landfill or freely donated anyway. I don't understand the $90 Million in teacher training either, unless it is not how to use computers, but how to make use of computers in an educational environment. Wasn't Windows XP supposed to be as easy to use as a Mac? They copied everything else, why not ease of use? (Microsoft doesn't have R&D, only D.) And yeah, $900 Million in software probably has a real cost to Microsoft of $1 Million. People need to know that the cost of duplicating software is nil.
For Mash, we have a C++ base that provides all of the lower-level multicast multimedia functionality and we use Tcl/Tk to create the applications. The C++ code lets us implement the low-level functionality (such as JPEG decoding, RTP transport, etc.) efficiently, but using Tcl/Tk to interface with the C++ makes it very easy to develop applications, test new ideas, and not worry too much about cross-platform UI.
Of course, you can't take a normal Tcl/Tk binaries and just arbitrarily call compiled C code, so we do require custom tclsh and wish binaries which have the C++ objects built into it. Our custom binary lets you define a C++ class as a subclass of TclClass. Then, we use otcl and tclcl within tclsh and wish to refer to those C++ classes as Tcl objects.
Christopher mentions three domains which have bandwidth costs of $10,000/mo. Seems to me like this is a good time to start mirrors of www.limewire.com and www.limewire.org. Not sure how router.limewire.com is used (I haven't read up on the Gnutella network's behavior), but couldn't that be distributed as well?