SMF instead of systemd? ZFS instead of BTRFS, meaning you can boot from ZFS a mirrored pool along with the other features? A platform that is outside the Red Hat/Ubuntu corporate circles? Multiarch support (i86 and x86_64) that just kinda works instead of the current Linux solutions?
How about you try this as "what could this become?" rather than "why does this look like everything else I've tried?"? How about looking at what Illumos and friends could be ported to? Your mileage, and consequently exposure to new ideas, may vary
Have them select an episode of Mythbusters and then analyze their scientific method. Have them state the hypothesis then compare their approach to that taken on the show. Have the student analyze the results, talk about variables and risk management. Then have them discuss some next steps for the experiment, and discuss practical applications of the findings. Mythbusters is awesome laying out a question and applying the scientific method, but due to time, they don't always pursue a question far as possible. I think they provide a good stepping stone to scientific research, and encourage critical thinking.
Although an incredibly dry beginning, reaching into information theory may be the most rewarding path. A neat thing about information theory is that it isn't all discrete sets and number crunching. You could start with an oscilloscope and just show what data looks like moving around. Such as a serial port--very easy to visualize and a stepping stone to any fancier stream. Also with a scope you can show AM and FM with different phase/amplitude modulations. Which leads right up to a ham radio license. I suspect she'll get just as caught up in error correcting/detecting codes, such as Hamming codes and Reed-Solomon codes, where some redundancy can make all the difference in receiving data correctly. And how much noise can a transmission tolerate? (Clue: CRC) Ethernet 10-base-T might be a nice case study of all of the above, and how carrier sense and collision detection was necessary. On top of error correcting codes you can introduce her to compression, covering both dictionary (LZA) and frequency (Huffman) approaches. Anyone that has dial-up (v.42bis and friends) is employing all of the above. After compression, introduce cryptography--being in some ways the opposite of error correction and compression. And there are a wealth of examples for cryptography and information theory, both full mechanical and computer programs! I think once you know where to look, this is a great opportunity!
Running Man by Richard Bachman (Steven King). Not like the movie at all--no mass market appeal. But the development of the main character and ending left me dazed for days.
New laws aren't really necessary, as the existing ones can be used to force additional improvements in this area. The manufacturers of the tubes are already held to FDA 820 for design and development, 21 CFR 820, of which section 820.30 (g) calls out risk analysis during validation. All the FDA has to do is indicate that the risk analysis includes foreseeable misuse which includes incorrect tubing connections, and all manufacturers will have to be able to demonstrate that their device is incompatible with a (justifiable) subset of existing similar tubes.
The FDA is already aware of the adverse events (21 CFR 820.198), and may even see trends. Some of these cases are the result of overstressed individuals, for which better staffing is the best answer. But for some of these cases I'll borrow from Forrest Gump, "How do you fix stupid?"
Moreover, would you want the government backing a specific industry standard for tubing? I can foresee many ways in which that system could be abused. The FDA is hoping someone else steps up and makes it easy for the different manufacturers to cooperate on this.
I use OpenSolaris on two systems at home, my desktop and a server (print and squid).
OpenSolaris provides a choice of userland (utilities) by setting the order of your path. So for those concerned about "incompatibilities" or "taint", you can choose your own userland, associated trappings, and level of standards conformance.
OS upgrades, or other significant package installs, are a no risk endeavor since the zfs root clones mean that you can boot to or rollback to a previous (working) instance if things go wrong. Snapshots from package installs and time-slider also provide a sense of security that you can "get back to that thing you just lost".
The package system deliberately has a concept of OS release, and ties the packages you install to that version. This reduces some of the confusion and problems that come from choosing packages implicitly based on a URL. Moreover, it also has means for a third-party vendor to host their own package repo and integrate it into the overall package scheme with user accounts, as appropriate.
The version you can download has all of the same features as that used in server rooms; so you can see how Sun met their customer requests for enterprise features compared to other distributions. Not just iSCSI and zones, but also MPIO among others are there for review.
The introduction of features to Open/Solaris feels more mature and predictable. Linux leads the bleeding edge with ideas and issues, whereas Solaris is a few steps behind, with solutions that feel more thought out.
If you'd like to meet OpenSolaris half-way, try Nexenta: http://nexenta.org/ which is the OpenSolaris kernel with a GNU (Debian) userland.
* Warlock in Spite of Himself by Christopher Stasheff This is a very interesting bridge between science and fantasy, while spending equal time on both. He continued on with this main character for several more books, and then started a spin-off series for the main character's son.
* The Stainless Steel Rat by Harry Harrison This is a delightful sci-fi series that provides comedy and fun for younger readers, but more food for thought when re-read later.
* Jhereg (Vladimir Taltos) books by Steven Brust More fantasy than sci-fi, but a delightful series with lots of character development that stays good for any age reader.
I hope this helps fill out your summer list, and many to come. I know it has added some to mine!
It probably helps greatly that UGS has released Parasolid for Linux. I can't find the press release describing this, but the link for supported platforms says enough:
I am in agreement. This is a great opportunity for M$ to loose ground to free software.
The possiblity of companies being able to take the same source and build it on Linux/*BSD as well as Windows means that much more of a market for them, at nearly no extra cost. Perhaps they might have to polish off certain calls, perhaps it might not be as smooth, but the audience/target market is that much larger.
I am excited by this chance! As WINE gets better and better, it will allow more and more companies te chance to realse programs for multiple platforms! I eagerly look forward to the day when software pacakges at Wal-Mart have a sticker that says
Built to run on Windows 98/ME/2000/XP/WINE
This will be an elegant triumph of open source over M$ monopoly! (not trying to make a flame war, but I see this as the easiest way to break the monopoly.)
To me, a humble student who really likes linux, this looks like they have a plan of some sort.. And one that will leave people gasping in surprise if it works...
Consider:
They have the infrastructure to build and produce LOTS of these. The plan might be to get people to hook onto successive buys like cell phones, but that's still less of a market gimmik than M$ everything...
They have the marketing base to promote this out of obscurity. Not necessarily to challange the top players, but enough to make this a house hold question: Media Top or XBox or PSX2??
They have taken a completely different tack on profit for this. They expect to make some revenue off the hardware and none off the software.
They already have some big name players (RealNetworks and Loki) helping to make things move along, and have asked the Linux community in general to help out.
I think this is a carefully crafted move to get a foothold in the home appliance market. They probably looked at it as a gamble, but the upside of winning is Killer User Base for them.
Further, I think this is a great opportunity for Linux to become a common household name. A large, international company has just planned to offer cheap, useable, Linux boxes to the masses for interactive tv and games... and then how much of a hop to the desktop and/or workplace??
*duck first fireball** not to start a flame war on linux's place in the world, but to prod those who feel a great need to evangalize to put out code for this product instead..
This is a seperate shameless plug, but once again might prove usefull to you...
<disclaimer> I am currently co-op-ing at the parent company, http://www.ugs.com, so add salt to taste.
</disclaimer>
You might consider IMAN. It is a multi-platform, version and project control system for large projects. I know it will be binary safe because most of what it was built to handle (CAD part files) are binary. It has a graphical interface, and sounds good from what I've seen. It also has a web portal for off-site workers or remote companion businesses. It also stands a chance of being overkill for your particular problem...
Rather than play games with identification, switching users, and arbitrary collections of capabilities/permissions, what of a bind-proxy?
How about modifying the bind code itself, in the system library, to authenticate a request to bind a port. Then a normal, non-special binary could call the routine, and ask for a priviliged port. The library, like PAM, would evaluate the request and decide to grant or deny.
And for those that ask about trojans, I don't think it would take much to have some other level of verification that the binary image requesting is the one you want running. Since this is calling from the kernel, it would have access to the exact file location, and image in memory. For example, it could compare a MD5 hash of the in memory binary with some secured list of allowed (port,program,hash) information.
You have marked yourself too stupid to use a computer.
Shannon's work provided the groundwork for programs you use all the time, whether you know it or not!
Data compression Huffman's compression algorithm reads like an appositive to Shannon's work.
Error Correction Like your data to arrive in one piece? No corruption?? Thank Shannon for providing the groundwork for efficient insights to error correction.
Encryption Without Shannon's take on information theory, encryption would be the same hunt-and-peck game that was played during WW2.
Have a little respect for something you couldn't do.
Can someone fill me in on where the matter goes?? i.e., is this 'Honey, I shrunk the kids' or something better??
What I remember of physics implies that it should be impossible to make any of those coins any more dense. Perhaps more brittle, and a little smaller, but not that much...
Your media distribution comment is an interesting one, because I've already seen the hooks in Win2K. For device drivers and downloads off the 'Net, the OS will ask you if you want to install this, if you trust the source, and if you always want to trust this source. (It makes me laugh whenever it asks if I want to always trust MS!).
The hooks are in place for them to do that kind of lockdown for the OS itself. So, I presume it could be possible for them to request hardware vendors to cooperate...
Even as apparent and possible as this plan sounds, I don't see it happening. As with the Cue Cat, people expect to do what they wish with hardware. If the hardware complains, they will fuss to the vendors..
And, Hooray for the mass quantities of legacy and slightly out of date hardware available!!!
I recall on OS/2 Warp (which I don't have installed right now, or I'd go home and check first), a similar functionality.
You could select some objects from the desktop (all items are objects and have methods and properties) and from the context menu choose Pick. What this did was allow the collection to be "dropped" on multiple items. In effect, if you could choose the proper items first, and have the proper items to drop on, you could "drop" them repeatedly. Like a pipe.
I recall that to buld programs to do this, you had to buy the SOM related products from IBM, which poor college students can't afford. So I did'nt have much of a chance to look into more details.
My personal goal is to find a job that allows me to attend all the conferences. What I do in the mean time I haven't worked out yet.. Prefereably systems/network development...
Slashdot Mirror
Looks like they've reinvented identd, with all the same features and flaws.
SMF instead of systemd?
ZFS instead of BTRFS, meaning you can boot from ZFS a mirrored pool along with the other features?
A platform that is outside the Red Hat/Ubuntu corporate circles?
Multiarch support (i86 and x86_64) that just kinda works instead of the current Linux solutions?
How about you try this as "what could this become?" rather than "why does this look like everything else I've tried?"?
How about looking at what Illumos and friends could be ported to?
Your mileage, and consequently exposure to new ideas, may vary
Have them select an episode of Mythbusters and then analyze their scientific method. Have them state the hypothesis then compare their approach to that taken on the show. Have the student analyze the results, talk about variables and risk management. Then have them discuss some next steps for the experiment, and discuss practical applications of the findings. Mythbusters is awesome laying out a question and applying the scientific method, but due to time, they don't always pursue a question far as possible. I think they provide a good stepping stone to scientific research, and encourage critical thinking.
Although an incredibly dry beginning, reaching into information theory may be the most rewarding path. A neat thing about information theory is that it isn't all discrete sets and number crunching. You could start with an oscilloscope and just show what data looks like moving around. Such as a serial port--very easy to visualize and a stepping stone to any fancier stream. Also with a scope you can show AM and FM with different phase/amplitude modulations. Which leads right up to a ham radio license. I suspect she'll get just as caught up in error correcting/detecting codes, such as Hamming codes and Reed-Solomon codes, where some redundancy can make all the difference in receiving data correctly. And how much noise can a transmission tolerate? (Clue: CRC) Ethernet 10-base-T might be a nice case study of all of the above, and how carrier sense and collision detection was necessary. On top of error correcting codes you can introduce her to compression, covering both dictionary (LZA) and frequency (Huffman) approaches. Anyone that has dial-up (v.42bis and friends) is employing all of the above. After compression, introduce cryptography--being in some ways the opposite of error correction and compression. And there are a wealth of examples for cryptography and information theory, both full mechanical and computer programs! I think once you know where to look, this is a great opportunity!
Running Man by Richard Bachman (Steven King). Not like the movie at all--no mass market appeal. But the development of the main character and ending left me dazed for days.
New laws aren't really necessary, as the existing ones can be used to force additional improvements in this area. The manufacturers of the tubes are already held to FDA 820 for design and development, 21 CFR 820, of which section 820.30 (g) calls out risk analysis during validation. All the FDA has to do is indicate that the risk analysis includes foreseeable misuse which includes incorrect tubing connections, and all manufacturers will have to be able to demonstrate that their device is incompatible with a (justifiable) subset of existing similar tubes.
The FDA is already aware of the adverse events (21 CFR 820.198), and may even see trends. Some of these cases are the result of overstressed individuals, for which better staffing is the best answer. But for some of these cases I'll borrow from Forrest Gump, "How do you fix stupid?"
Moreover, would you want the government backing a specific industry standard for tubing? I can foresee many ways in which that system could be abused. The FDA is hoping someone else steps up and makes it easy for the different manufacturers to cooperate on this.
I use OpenSolaris on two systems at home, my desktop and a server (print and squid).
OpenSolaris provides a choice of userland (utilities) by setting the order of your path. So for those concerned about "incompatibilities" or "taint", you can choose your own userland, associated trappings, and level of standards conformance.
OS upgrades, or other significant package installs, are a no risk endeavor since the zfs root clones mean that you can boot to or rollback to a previous (working) instance if things go wrong. Snapshots from package installs and time-slider also provide a sense of security that you can "get back to that thing you just lost".
The package system deliberately has a concept of OS release, and ties the packages you install to that version. This reduces some of the confusion and problems that come from choosing packages implicitly based on a URL. Moreover, it also has means for a third-party vendor to host their own package repo and integrate it into the overall package scheme with user accounts, as appropriate.
The version you can download has all of the same features as that used in server rooms; so you can see how Sun met their customer requests for enterprise features compared to other distributions. Not just iSCSI and zones, but also MPIO among others are there for review.
The introduction of features to Open/Solaris feels more mature and predictable. Linux leads the bleeding edge with ideas and issues, whereas Solaris is a few steps behind, with solutions that feel more thought out.
If you'd like to meet OpenSolaris half-way, try Nexenta: http://nexenta.org/ which is the OpenSolaris kernel with a GNU (Debian) userland.
Also off the beaten path:
* Warlock in Spite of Himself by Christopher Stasheff
This is a very interesting bridge between science and fantasy, while spending equal time on both. He continued on with this main character for several more books, and then started a spin-off series for the main character's son.
* The Stainless Steel Rat by Harry Harrison
This is a delightful sci-fi series that provides comedy and fun for younger readers, but more food for thought when re-read later.
* Jhereg (Vladimir Taltos) books by Steven Brust
More fantasy than sci-fi, but a delightful series with lots of character development that stays good for any age reader.
I hope this helps fill out your summer list, and many to come. I know it has added some to mine!
Cogline
http://www.eds.com/products/plm/parasolid/portfoli o/supported.shtml
I'm eager to see what other high end CAD/CAM (UGS, CATIA) programs get ported as a response to this...
The possiblity of companies being able to take the same source and build it on Linux/*BSD as well as Windows means that much more of a market for them, at nearly no extra cost. Perhaps they might have to polish off certain calls, perhaps it might not be as smooth, but the audience/target market is that much larger.
I am excited by this chance! As WINE gets better and better, it will allow more and more companies te chance to realse programs for multiple platforms! I eagerly look forward to the day when software pacakges at Wal-Mart have a sticker that says
This will be an elegant triumph of open source over M$ monopoly! (not trying to make a flame war, but I see this as the easiest way to break the monopoly.)
Consider:
- They have the infrastructure to build and produce LOTS of these. The plan might be to get people to hook onto successive buys like cell phones, but that's still less of a market gimmik than M$ everything...
- They have the marketing base to promote this out of obscurity. Not necessarily to challange the top players, but enough to make this a house hold question: Media Top or XBox or PSX2??
- They have taken a completely different tack on profit for this. They expect to make some revenue off the hardware and none off the software.
- They already have some big name players (RealNetworks and Loki) helping to make things move along, and have asked the Linux community in general to help out.
I think this is a carefully crafted move to get a foothold in the home appliance market. They probably looked at it as a gamble, but the upside of winning is Killer User Base for them.Further, I think this is a great opportunity for Linux to become a common household name. A large, international company has just planned to offer cheap, useable, Linux boxes to the masses for interactive tv and games... and then how much of a hop to the desktop and/or workplace??
*duck first fireball** not to start a flame war on linux's place in the world, but to prod those who feel a great need to evangalize to put out code for this product instead..
my $0.02 and SCl... ;-)
<disclaimer>
I am currently co-op-ing at the parent company, http://www.ugs.com, so add salt to taste.
</disclaimer>
You might consider IMAN. It is a multi-platform, version and project control system for large projects. I know it will be binary safe because most of what it was built to handle (CAD part files) are binary. It has a graphical interface, and sounds good from what I've seen. It also has a web portal for off-site workers or remote companion businesses. It also stands a chance of being overkill for your particular problem...
Best of luck!
look at: this location.
don't know that it will help, but it does exist.
How about modifying the bind code itself, in the system library, to authenticate a request to bind a port. Then a normal, non-special binary could call the routine, and ask for a priviliged port. The library, like PAM, would evaluate the request and decide to grant or deny.
And for those that ask about trojans, I don't think it would take much to have some other level of verification that the binary image requesting is the one you want running. Since this is calling from the kernel, it would have access to the exact file location, and image in memory. For example, it could compare a MD5 hash of the in memory binary with some secured list of allowed (port,program,hash) information.
Is this possible, or am I off my rocker??
Shannon's work provided the groundwork for programs you use all the time, whether you know it or not!
Data compression
Huffman's compression algorithm reads like an appositive to Shannon's work.
Error Correction
Like your data to arrive in one piece? No corruption?? Thank Shannon for providing the groundwork for efficient insights to error correction.
Encryption
Without Shannon's take on information theory, encryption would be the same hunt-and-peck game that was played during WW2.
Have a little respect for something you couldn't do.
What I remember of physics implies that it should be impossible to make any of those coins any more dense. Perhaps more brittle, and a little smaller, but not that much...
The hooks are in place for them to do that kind of lockdown for the OS itself. So, I presume it could be possible for them to request hardware vendors to cooperate...
Even as apparent and possible as this plan sounds, I don't see it happening. As with the Cue Cat, people expect to do what they wish with hardware. If the hardware complains, they will fuss to the vendors..
And, Hooray for the mass quantities of legacy and slightly out of date hardware available!!!
You could select some objects from the desktop (all items are objects and have methods and properties) and from the context menu choose Pick. What this did was allow the collection to be "dropped" on multiple items. In effect, if you could choose the proper items first, and have the proper items to drop on, you could "drop" them repeatedly. Like a pipe.
I recall that to buld programs to do this, you had to buy the SOM related products from IBM, which poor college students can't afford. So I did'nt have much of a chance to look into more details.
My personal goal is to find a job that allows me to attend all the conferences. What I do in the mean time I haven't worked out yet.. Prefereably systems/network development...