This is exactly why my preference list for distributions is:
1 - Slackware (unaltered packages, tgz wrapping, no "packaging" to speak of) 2 - Redhat (Fedora) 3 - Everybody else
Redhat slides in at number 2 because they have to security conscious. Fedora squeaks in with Redhat. Strange part is my trust of Slackware (the "smallest" distributor) is on par with Redhat (the "biggest"). I trust Patrick to not muck it up; he wouldn't have the time!
mplayer (google it). Both mplayer and mencoder are available for Linux, and Windows.
Any format that can be played can be converted. And they do support FLV format. Now, how to download the FLV? Firefox. Download the "Video Download Helper" add-on, and tell it to save the FLV.
Easy peasy (especially on Linux, not so much on Windows, but easy enough).
Code that won't work generally breaks because Linux 2.0 used BSD style ptys, and Linux 2.6 uses System V style ptys. No great issue, but you do have to do a little work...
Beyond that? I think everything else in 2.0 is still ok (but I may have missed a little something).
However, open() was never "deprecated" and yet carried forward. The original design was strong enough to accommodate the changes over the years. POSIX is, of course, your friend.
But we speak of Windows. I really believe that Microsoft is in an "API hell" now. Partly, it's an issue of being a closed source operating system. A dumb idea, because applications can only go by documentation, and by the actual API behaviour. To give an example (simplified, so not be in the Windows domain), p = malloc(0); p = realloc(p, 100); is supposed to work.
But say the application was written before the behavior of malloc(0) (and thus the behavior of the subsequent realloc) was not really documented. Did this work? Didn't it? Before it was documented, the only way was "try and see". If applications then relied on that behaviour, what should be done if the API is updated? If the old (undocumented) behavior does not fit into the new model, do you force the application vendor to change? In the closed software world, both the API implementation and the usage are likely closed source. The application vendor may not even be in business anymore! So the API is forced into an accommodation. Maybe for just one application. This leads to cruft.
Of course, anal programmers would have done:
p = (n == 0) ? NULL : malloc(n); p = (p == NULL) ? malloc(100) : realloc(p, 100);
if that behavior had not been mentioned. But, there is the drive to write the most efficient and readable code! Especially for the Windows API where calls can have LOTS of parameters, each of which would need suitable checking.
After trying out "anti-SPAM" solutions, I have reverted to no such protection on my domains. I do not allow relaying, but I also do not filter ANY messages.
If a user gets thousands of SPAM... well, its the nature of the beast.
After all, why would I compound theft of bandwidth with theft of compute resources to even LOOK at it? If the SPAM producers kill email, well, they kill the golden goose. After all, if no one reads it, it WILL go the way of USENET.
RIP SMTP, your time is near. Either that, or the spammers will reign in.
Since my domains do not produce SPAM; not even the tiniest morsel, I do not feel it is my problem (and if any of my users produces SPAM, I will immediately cut them off -- even as it is being produced). As it is, all outbound mail is relayed by my ISP, and they only allow 10 username/domains that are NOT theirs (new policy).
Frankly, I have given up giving a shit about SPAM, or anything having to do with it. If too much bandwidth is consumed, I will simply stop using email.
Yes, it will like your Hauppage card. With prodding.
On the other hand, YOU WILL NOT LIKE IT. Because you couldn't get it working on other distributions, you won't get it working with Slackware, either. As an "added bonus", Slackware has no GUI tools. No user manager, service manager, or graphical network configuration. Everything is configured with a text editor. Slackware doesn't even really like the SysV init system (although, it begrudgingly supports it), preferring the old-style BSD init instead.
Slackware is probably the only distribution (may gentoo, but I haven't looked) where it is pretty common to use init level 3, and startx when needed...
On the plus side of Slackware -- if you DO get your remote working with it, you will know how to get it working on any other distribution. Bad news is: you will probably end up bypassing all that "GUI goodness" and automated configuration to get the job done.
Slack philosophy is to use unaltered kernels, and unaltered packages, along with building wrappers. No custom patches. Easy to understand. The packaging tool is simply a tgz file, along with a catalog. Dependencies? For wimps. It is normal to have kernel sources on the machine. If you don't like "configure/make/install", don't use Slack. If you can't use vi or emacs, don't use Slack.
No. Let go marketroid (them's Darl words): Linux is an independent implementation of an Operating System supporting POSIX compliant applications, the FSF GNU tool chain, Xorg GUI and higher level systems such as GNOME and KDE.
The only commonality that Linux and Unix have in common is 3 letters of the name and the application support. This application support demands certain features of the Operating System (such as the fork() primitive) and certain file system semantics. Linux does support certain binary level conventions for compatibility (such as signal values, and binary UFS file system), but most of these are deprecated (UFS), or forced by header file source compatibility.
In no way is Linux a COPY of Unix. Not only is there no copyright infringement, there is no obvious "design infringement" either. If there were, it would be directed at the FSF for GNU (ls, cp, rm, bash, etc.). And at POSIX for specifying the API (IEEE 1003.1) *and* the command tools (IEEE 1003.2)
If a client doesn't have a clear idea, I help them by working through a process. I will help develop a "Requirements Specification". We need to be able to isolate what we will build.
This needs to be known, because we need to know the problem we are addressing. After this is settled, we work on a "Design Specification". That will tell us how we will go about building that thing.
Simply jumping in and coding is, well, wrong. Without understanding where you want to go, how will you know what direction to go in?
Of course GUIs can actually promote this. If all applications look the same, you can produce an application that looks like all the others. What it does may be of little concern to you (and, given your methodology, it probably isn't). You can then fill in "functionality" fed from the client. This will, of course, model around the GUI (and not the actual problem domain). There will be little choice of unifying concept. There will be a lot of "duplicated" code, which will make you think that a "reuse" policy would be a good idea.
I would suggest that you supply a higher level of service to your client. Supply a real design service. This will benefit your coding as well -- you will know what you need to code.
I have two laser printers at home - a colour model, and a b&w multifunction.
The b&w multifunction is an HP 3015. Printing? fast enough (15ppm). Resolution? ok (600x600, 1200x1200 w/ RET). Toner? Uses a cartridge, and the entire thing (including the drum) is replaced. Each cartridge runs $56 to $75, and they are available through Costco. You get around 2000 pages (more in "economy" mode) per cartridge, which is around 4 cents a page (or less; not including the price of paper). When printing, doesn't smell obnoxiously of ozone. Printing (b&w), scanning (colour), photocopying, faxing. Can email scans to users on your network. Reasonably quiet. USB 2 and parallel connections (I use the USB 2).
I have been using the 3015 for a few years now, with no problems.
Main complaints:
The 3015 doesn't have a flatbed scanner (only feed driven). Paper capacity is limited (both in and out). Limited paper handling. No automatic double-sided printing.
Recommended for home/soho light duty usage. (My colour unit has a straight feed path (able to handle card stock), and I have a stand alone flatbed scanner. Generally, I use the 3015 for 90% of my soho needs).
"It also eliminates the need to seize a computer itself, which typically involves disconnecting from a network, turning off the power and potentially losing data. Instead, the investigator can scan for evidence on site."
Which implies that the tool can extract data from a RUNNING MACHINE. The issue at hand is: Do we now trust Windows cryptography?
"These are things that we invest substantial resources in, but not from the perspective of selling to make money," Smith said in an interview.
Now there is not much in the article -- just enough to eliminate Windows encryption as viable. Indeed, I would also not use TruCrypt either; we have to assume that it is also comprimised by simply being on the platform. After all, it is a closed source platform, and one of the key Microsoft players has just let the cat out of the bag.
Now, the tool may end up being "nothing"; maybe just ntfs recovery on a USB key. But... the trust is gone. The proper response is "I am sorry. There is no way to access the data without the passphrase. No recovery is currently possible.". The security issue is then reduced to key management.
The surprise with this announcement is that the CSP (cryptographic service providers) in Windows XP have been FIPS 140 certified (at least to level 1, and possibly level 2 for some things), which means that any "back-doors" must be very clever indeed. I would even venture that the new Microsoft Crack Box doesn't attack the CSP, but somehow bypasses it. In other words, best results probably if the system ISN'T powered off... But that's just speculation on my part.
As a PS: I always thought (but I could be wrong) that FIPS 140 certification ALSO meant best practices with key control in memory, including scrubbing after using. I would now look for weakness in that area, if I were a "black-hat".
Of course, your trust quotient may be much higher than mine.
An example: Say I modify GNU C to support a new platform. Say, an embedded platform. I can sell this for a high price -- say $50,000. I give the compiler and source on the sale.
The question is: does this SINGLE sale destroy the market? No, it doesn't. If I can support it, and it works, why would the company that purchased it give it away? That would allow a competitor (to THEM) to start up at a lower cost.
Of course, the purchaser could decide to enter the business of compilers... in which case they would compete with ME. But then, why purchase in the first place? (I have more expertise in the area).
Now on to a real example: I used to produce prepress proofing software. Typical sale would be $30,000 (in the '80s). The software required a "dongle" for security. If the customer simply requested it, they would be supplied with a "no dongle" version that could be simply copied. The typesetting industry is intensely competitive -- and so NO ONE ever requested that version. They were more afraid that the software would be copied, providing a competitor with the same advantage WITHOUT paying for it.
Not "open source", but you do get the idea. You CAN charge for open source software. This will never work for a commodity program, but can work in vertical markets.
I agree with you. I was responding to the post which stated that protocols were somehow under one of the intellectual properties. I wanted to drive the point that the documentation was the only thing that could be considered of value; that a protocol couldn't be in one of the possible categories. (Well, actually, it could be a trade secret, as I pointed out, but that would defeat the purpose completely).
The "protocol" used with the flat-head screw driver is a slit. There are two parts - the screwdriver (which can be patented) and the screw (which can be patented). The slit (protocol)?
As you pointed out, a coin can be used instead of the screwdriver. And the receptor may be something other than a screw (say, a snap-tab).
How can a protocol be a trade secret? After all, the machine using it is spewing it! How can a protocol even be copyrighted? After all, it isn't a static expression; that would be useless.
Which leaves patents -- but a protocol is neither a machine or a process! Which means (arguably) that it can't be patented, either.
Which leaves documentation.
But let's think about that "licensing" again... What is being licensed? Compare to music -- the "protocol" in written music is NOT the music, it's the staff, and individual marks that COULD represent notes (if the protocol is followed). There is no "trade secret", "copyright" or "patent" in that protocol -- it would be ludicrous and self-defeating. After all, if no one can learn that protocol, the music will be dead. In the same way, type fonts cannot be copyrighted, and are not a trade secret! (the "protocol" for reading).
The solution? CutePuppies.exe is not executable. End of discussion.
If you want to actually execute it, you have to:
1 - save it to disk 2 - change its permissions 3 - then (and only then) execute it.
It is preferable to force a command line session (terminal window) for step 2, with a "difficult" sequence. Say.. chmod +x CutePuppies.exe. And it should show up on the desktop either...
No "is this allowed?" dialog. No "please enter your password" dialog. Just.. don't.. execute.. it.
I would even go so far as to force a manual base64 or uu decode in there.
It's the integration. Market a computer to someone who doesn't know what a program is, and allow that person to run programs. Without having to consult with anyone. So you are right; some basic training is in order.
But the layout and features of Windows needs to come under examination as well. WHY does double clocking.jpg.exe work? Especially if part of the information is hidden? Who thought that would EVER be a sensible idea? I used to try to allay peoples fears years ago, by saying "No, you cannot get malware from email -- the thought is ludicrous". And then we get the integration that lets it happen easily.
Unix (Linux) isn't a "magic bullet". And, indeed, applications have been dragged into that "tight integration" role -- email must talk to the calendar, and multimedia, and random web pages must be allowed to run code, without sand-boxing! I swear, it makes me want to become a luddite. My kids are insisting that it is "ok" to run random games from the internet now.
At least I keep tabs of all traffic, incoming/outgoing. If it spikes, I find out why. Pull the plug. It's happened a few times.
"Where do you want to go today?" ads. Probably as socially responsible as marketing 400kmph racing cars to people who barely know how to drive. And when they crash, just engaging in hand-waving "They should have known".
But the government will look to have their needs met. Instead of the "one-size-fits-all" solution, customized solutions can be proposed.
The "Linux toys": one is a media extender, fixed software with TV/remote control. Another is a a router "kit" -- pick the functions and build it yourself. The last is a general purpose laptop (openoffice, ssh). They don't look the same, configure the same, or have ANYTHING to do with each other -- except that they are all based on commodity Linux.
Could be based on something else, but the OS is only resource control and some APIs. Commodity. And, further, the OS becomes the base. For example, the router uses Linux. But it isn't an x86! And that fact turns out not to be relevant.
It could have been Linux, BSD, or something else. Just happens that Linux won.
What it does is lock in a platform, and everything that isn't IDENTICAL to that platform gets turfed. Want a recent example? One in which Microsoft get caught in the their OWN trap?
Microsoft decides to include an "autorun" feature. Considered a VERY BAD IDEA, but it becomes a checkmark. Microsoft then introduces VISTA, and "modifies" the autorun feature. Is is simply removed? No, the feature remains. It now opens a dialog, asking if the autorun should be done.
This is EXACTLY the same as the user opening the application on the inserted media. In neither case is the autorun, well... autorun. Even Microsoft cannot retract the autorun feature, even though it is no longer autorun. It is now bloat. Created by platform lock-in.
The effect of removing Microsoft from the software equation would be a broad examination of HOW things should be done. Microsoft believes in a "one size fits everything" (ok, 8 sizes). Since source was not provided, it was impossible to tune installations for purpose.
Within the Linux ecosystem, the Linux running on a DSM-320 is NOT AT ALL like the Linux running on my laptop. It is ALSO different from the Linux running on my router. There can be individual modifications made to support specific uses and platforms. The Linux running on the eeepc is not the same flavor as the Linux running on my T43.
We will see broader individualized deployments. A government department could (reasonably) have their own "distribution" customized to their needs. There would be an ecosystem of people who supply that service (similar to the legion of Microsoft supporters, including the "MVP"s).
This has happened with Solaris and Linux is the server space, and has happened with Microsoft in the "desktop" space. You would now see "Linux Desktop Specialist" (or whatever the position would be called). And once the customizations start rolling, more users would see the benefit of having the environment completely customizable.
The T2000 (for example) has one die clocked at 1.2 or 1.4 Ghz. On that die are 8 processor cores. Each of these has 4 CMT threads (sort of what Intel used to call hyper-threading). 32 "virtual" cpus, 2U form factor. $6,995 base
The M-series, lets take the M5000 is built by Fujitsu. That has 8 processing elements, each clocked at 2.1 Ghz, with each dual-core. 16 "virtual" cpus. $47,000 base. 10U form factor.
The 5220 has 8 cores, 8 CMT per core, for 64 virtual cpus, 1.4 Ghz. $10,995 base. 2U,
The issue here is that all of this threading doesn't help "straight line" programs. The Power6 does much better (at 5 Ghz)! For "straight line" code, the best machine here (of SUNs offerings) is the M5000 -- machines (more or less) selected randomly from SUNs catalog. Only 16 virtual CPUs, but boy does it cook! (and, its 5 times the size, and 5 times the price).
The best "laptop" I've ever used was the Radioshack M100. Could take notes ALL DAY (maybe even all week). Upload the notes, and continue. The size of a clipboard, and a reasonable keyboard (for typing).
Now? *If* it plays movies (DVD, or other), it should be able to play at least 2 full movies (at least 4 hour battery life, although the M100 lasted 20 to 40 hours!). I should be able to pull it out and type on it (capture notes) without waiting minutes for it to "boot". It should be dead quiet for use in meetings. It should be (almost) indestructible. It should offer telnet/ssh connectivity (bonus if it supports X). It should be able to use standard batteries of some kind (AA?), or a common DC input (12V? 6V? but with a wide tolerance). It should support USB ports for additional storage. It should have integrated WIFI and RJ45 network plug. It should NOT be larger than 8.5x11x1. The keyboard should have full (typist) travel.
Does this product exist? I don't know. My current laptop (Thinkpad T43) occasionally goes "super loud" (its fan kicks in), and even blows papers off of my desk. It is too warm to use comfortably on my lap. The battery only lasts 1.5 hours (not quite a movie). The keyboard doesn't have enough travel, but it does run Linux (and therefore telnet, ssh and X). It also takes up to a minute to wake up sometimes. I use it, but I am sure not happy with it.
I don't think my "ideal" laptop exists yet -- but I have to look into the new ultra-small units (not for the size; I think that the keyboards will be too small). I am also very interested in the Apple Air.
I have tried to purchase "business class" service from my ISP. Have been turned down every time (3 times and counting). I try every 6 months to a year.
No, I don't want petabytes of storage, or zetabytes of bandwidth. I use between 15GB and 100GB of "bandwidth" per month. Storage? zilch.
Now, I have 60GB per month with an overage charge of $2/GB. As to SPEED... As long as its reasonable, I frankly don't care. What is reasonable? 6 Mbps to allow download Standard Definition TV at "real time" rates. Of course, I don't get that:( (my ISP *claims* I get 6 Mbps, and delivers 0.5 to 6, averaging 3).
I can download any audio recording (movies, books, etc.) that was fixed prior to... 1912, I believe. WWII hasn't yet entered the public domain, culturally. Not the Vietnam conflict, Not the racial marches. Not the moon landing.
It's all under copyright; nothing is public domain.
Give it all up. Go wear a straw hat, or, if you are female, do the "flapper" thing. Until its pushed back further. I'm ok with this (personally). But then, I think Beethoven's 9th is the best piece of music ever written. I can pass on the rest of "culture" until I can own it and contribute (make use of it).
I am a person, not just a "consumer" or an "eyeball".
Since copyright law is SO SKEWED, I feel quite content ignoring it. When I was a kid, we didn't have VHS, DVDs. I went to the movies two (or three tops) times a month (less if there were no new releases). Never saw a movie twice (had better things to do with my money). Wouldn't tolerate advertisements EXCEPT trailers in the movie.
Had three TV channels, and records. Bought the occasional 45 and LP.
The media companies were happy.
Now -- I still go to movies. Pop and Popcorn the same price as the tickets, or more. Have to put up with Card ads, Soft drink ads, Still have trailers. Then, the movie comes up for rental/DVD purchase. I pay again. There are toy promotions, etc. I have in excess of 100 channels on the TV.
The media companies have much more opportunity to market and make money (I now pay for movies TWICE routinely, if I see them in the theater).
However, there is a push to control what I can further do with this deluge that is pushing in, trying to become culture. I can't even use Mickey yet! What's the answer? I figure that there is SO MUCH media being pumped (radio, TV, satellite radio, wireless internet all beaming stuff to where I am, 24/7) that it will cause some health issues. And yet, the motto is "Look, but don't touch". I would argue that it is almost IMPOSSIBLE to not break copyright. "We control your culture".
Ok, I am probably insane; just wish I had some answers. "They" make it very difficult.
"Older movies" are on film. Not recorded on video (HD or otherwise). So, a quality HD encoding will make a difference. For home use, HD (HD DVD or Blue Ray) is about as good as it gets. On a 42" screen, I can't tell the difference from film.
Slashdot Mirror
This is exactly why my preference list for distributions is:
1 - Slackware (unaltered packages, tgz wrapping, no "packaging" to speak of)
2 - Redhat (Fedora)
3 - Everybody else
Redhat slides in at number 2 because they have to security conscious. Fedora squeaks in with Redhat.
Strange part is my trust of Slackware (the "smallest" distributor) is on par with Redhat (the "biggest"). I trust Patrick to not muck it up; he wouldn't have the time!
Toonol
mplayer (google it). Both mplayer and mencoder are available for Linux, and Windows.
Any format that can be played can be converted. And they do support FLV format. Now, how to download the FLV? Firefox. Download the "Video Download Helper" add-on, and tell it to save the FLV.
Easy peasy (especially on Linux, not so much on Windows, but easy enough).
Generally, yes.
Code that won't work generally breaks because Linux 2.0 used BSD style ptys, and Linux 2.6 uses System V style ptys. No great issue, but you do have to do a little work...
Beyond that? I think everything else in 2.0 is still ok (but I may have missed a little something).
However, open() was never "deprecated" and yet carried forward. The original design was strong enough to accommodate the changes over the years. POSIX is, of course, your friend.
But we speak of Windows. I really believe that Microsoft is in an "API hell" now. Partly, it's an issue of being a closed source operating system. A dumb idea, because applications can only go by documentation, and by the actual API behaviour. To give an example (simplified, so not be in the Windows domain), p = malloc(0); p = realloc(p, 100); is supposed to work.
But say the application was written before the behavior of malloc(0) (and thus the behavior of the subsequent realloc) was not really documented. Did this work? Didn't it? Before it was documented, the only way was "try and see". If applications then relied on that behaviour, what should be done if the API is updated? If the old (undocumented) behavior does not fit into the new model, do you force the application vendor to change? In the closed software world, both the API implementation and the usage are likely closed source. The application vendor may not even be in business anymore! So the API is forced into an accommodation. Maybe for just one application. This leads to cruft.
Of course, anal programmers would have done:
p = (n == 0) ? NULL : malloc(n);
p = (p == NULL) ? malloc(100) : realloc(p, 100);
if that behavior had not been mentioned. But, there is the drive to write the most efficient and readable code! Especially for the Windows API where calls can have LOTS of parameters, each of which would need suitable checking.
After trying out "anti-SPAM" solutions, I have reverted to no such protection on my domains. I do not allow
relaying, but I also do not filter ANY messages.
If a user gets thousands of SPAM... well, its the nature of the beast.
After all, why would I compound theft of bandwidth with theft of compute resources to even LOOK at it? If the SPAM producers kill email, well, they kill the golden goose. After all, if no one reads it, it WILL go the way of USENET.
RIP SMTP, your time is near. Either that, or the spammers will reign in.
Since my domains do not produce SPAM; not even the tiniest morsel, I do not feel it is my problem (and if any of my users produces SPAM, I will immediately cut them off -- even as it is being produced). As it is, all outbound mail is relayed by my ISP, and they only allow 10 username/domains that are NOT theirs (new policy).
Frankly, I have given up giving a shit about SPAM, or anything having to do with it. If too much bandwidth is consumed, I will simply stop using email.
RIP SMTP
Yes, it will like your Hauppage card. With prodding.
On the other hand, YOU WILL NOT LIKE IT. Because you couldn't get it working on other distributions, you won't get it working with Slackware, either. As an "added bonus", Slackware has no GUI tools. No user manager, service manager, or graphical network configuration. Everything is configured with a text editor. Slackware doesn't even really like the SysV init system (although, it begrudgingly supports it), preferring the old-style BSD init instead.
Slackware is probably the only distribution (may gentoo, but I haven't looked) where it is pretty common to use init level 3, and startx when needed...
On the plus side of Slackware -- if you DO get your remote working with it, you will know how to get it working on any other distribution. Bad news is: you will probably end up bypassing all that "GUI goodness" and automated configuration to get the job done.
Slack philosophy is to use unaltered kernels, and unaltered packages, along with building wrappers. No custom patches. Easy to understand. The packaging tool is simply a tgz file, along with a catalog. Dependencies? For wimps. It is normal to have kernel sources on the machine. If you don't like "configure/make/install", don't use Slack. If you can't use vi or emacs, don't use Slack.
Good luck with your quest.
A COPY?
No. Let go marketroid (them's Darl words): Linux is an independent implementation of an Operating System supporting POSIX compliant applications, the FSF GNU tool chain, Xorg GUI and higher level systems such as GNOME and KDE.
The only commonality that Linux and Unix have in common is 3 letters of the name and the application support. This application support demands certain features of the Operating System (such as the fork() primitive) and certain file system semantics. Linux does support certain binary level conventions for compatibility (such as signal values, and binary UFS file system), but most of these are deprecated (UFS), or forced by header file source compatibility.
In no way is Linux a COPY of Unix. Not only is there no copyright infringement, there is no obvious "design infringement" either. If there were, it would be directed at the FSF for GNU (ls, cp, rm, bash, etc.). And at POSIX for specifying the API (IEEE 1003.1) *and* the command tools (IEEE 1003.2)
If a client doesn't have a clear idea, I help them by working through a process. I will help develop a "Requirements Specification". We need to be able to isolate what we will build.
This needs to be known, because we need to know the problem we are addressing. After this is settled, we work on a "Design Specification". That will tell us how we will go about building that thing.
Simply jumping in and coding is, well, wrong. Without understanding where you want to go, how will you know what direction to go in?
Of course GUIs can actually promote this. If all applications look the same, you can produce an application that looks like all the others. What it does may be of little concern to you (and, given your methodology, it probably isn't). You can then fill in "functionality" fed from the client. This will, of course, model around the GUI (and not the actual problem domain). There will be little choice of unifying concept. There will be a lot of "duplicated" code, which will make you think that a "reuse" policy would be a good idea.
I would suggest that you supply a higher level of service to your client. Supply a real design service. This will benefit your coding as well -- you will know what you need to code.
I have two laser printers at home - a colour model, and a b&w multifunction.
The b&w multifunction is an HP 3015. Printing? fast enough (15ppm). Resolution? ok (600x600, 1200x1200 w/ RET). Toner? Uses a cartridge, and the entire thing (including the drum) is replaced. Each cartridge runs $56 to $75, and they are available through Costco. You get around 2000 pages (more in "economy" mode) per cartridge, which is around 4 cents a page (or less; not including the price of paper). When printing, doesn't smell obnoxiously of ozone. Printing (b&w), scanning (colour), photocopying, faxing. Can email scans to users on your network. Reasonably quiet. USB 2 and parallel connections (I use the USB 2).
I have been using the 3015 for a few years now, with no problems.
Main complaints:
The 3015 doesn't have a flatbed scanner (only feed driven).
Paper capacity is limited (both in and out).
Limited paper handling.
No automatic double-sided printing.
Recommended for home/soho light duty usage. (My colour unit has a straight feed path (able to handle card stock), and I have a stand alone flatbed scanner. Generally, I use the 3015 for 90% of my soho needs).
The fine article states:
"It also eliminates the need to seize a computer itself, which typically involves disconnecting from a network, turning off the power and potentially losing data. Instead, the investigator can scan for evidence on site."
Which implies that the tool can extract data from a RUNNING MACHINE. The issue at hand is: Do we now trust Windows cryptography?
"These are things that we invest substantial resources in, but not from the perspective of selling to make money," Smith said in an interview.
Now there is not much in the article -- just enough to eliminate Windows encryption as viable. Indeed, I would also not use TruCrypt either; we have to assume that it is also comprimised by simply being on the platform. After all, it is a closed source platform, and one of the key Microsoft players has just let the cat out of the bag.
Now, the tool may end up being "nothing"; maybe just ntfs recovery on a USB key. But... the trust is gone. The proper response is "I am sorry. There is no way to access the data without the passphrase. No recovery is currently possible.". The security issue is then reduced to key management.
The surprise with this announcement is that the CSP (cryptographic service providers) in Windows XP have been FIPS 140 certified (at least to level 1, and possibly level 2 for some things), which means that any "back-doors" must be very clever indeed. I would even venture that the new Microsoft Crack Box doesn't attack the CSP, but somehow bypasses it. In other words, best results probably if the system ISN'T powered off... But that's just speculation on my part.
As a PS: I always thought (but I could be wrong) that FIPS 140 certification ALSO meant best practices with key control in memory, including scrubbing after using. I would now look for weakness in that area, if I were a "black-hat".
Of course, your trust quotient may be much higher than mine.
If you disagree with Microsoft's EULA, you cannot use the software.
If you disagree with the GPL, you can use the software.
I guess it does make a difference. But probably not the difference you were thinking about.
Depends on the software.
An example: Say I modify GNU C to support a new platform. Say, an embedded platform. I can sell this for a high price -- say $50,000. I give the compiler and source on the sale.
The question is: does this SINGLE sale destroy the market? No, it doesn't. If I can support it, and it works, why would the company that purchased it give it away? That would allow a competitor (to THEM) to start up at a lower cost.
Of course, the purchaser could decide to enter the business of compilers... in which case they would compete with ME. But then, why purchase in the first place? (I have more expertise in the area).
Now on to a real example: I used to produce prepress proofing software. Typical sale would be $30,000 (in the '80s). The software required a "dongle" for security. If the customer simply requested it, they would be supplied with a "no dongle" version that could be simply copied. The typesetting industry is intensely competitive -- and so NO ONE ever requested that version. They were more afraid that the software would be copied, providing a competitor with the same advantage WITHOUT paying for it.
Not "open source", but you do get the idea. You CAN charge for open source software. This will never work for a commodity program, but can work in vertical markets.
I agree with you. I was responding to the post which stated that protocols were somehow under one of the intellectual properties. I wanted to drive the point that the documentation was the only thing that could be considered of value; that a protocol couldn't be in one of the possible categories. (Well, actually, it could be a trade secret, as I pointed out, but that would defeat the purpose completely).
The "protocol" used with the flat-head screw driver is a slit. There are two parts - the screwdriver (which can be patented) and the screw (which can be patented). The slit (protocol)?
As you pointed out, a coin can be used instead of the screwdriver. And the receptor may be something other than a screw (say, a snap-tab).
Is the slit itself patentable?
How can a protocol be a trade secret? After all, the machine using it is spewing it! How can a protocol even be copyrighted? After all, it isn't a static expression; that would be useless.
Which leaves patents -- but a protocol is neither a machine or a process! Which means (arguably) that it can't be patented, either.
Which leaves documentation.
But let's think about that "licensing" again... What is being licensed? Compare to music -- the "protocol" in written music is NOT the music, it's the staff, and individual marks that COULD represent notes (if the protocol is followed). There is no "trade secret", "copyright" or "patent" in that protocol -- it would be ludicrous and self-defeating. After all, if no one can learn that protocol, the music will be dead. In the same way, type fonts cannot be copyrighted, and are not a trade secret! (the "protocol" for reading).
The solution? CutePuppies.exe is not executable. End of discussion.
If you want to actually execute it, you have to:
1 - save it to disk
2 - change its permissions
3 - then (and only then) execute it.
It is preferable to force a command line session (terminal window) for step 2, with a "difficult" sequence. Say.. chmod +x CutePuppies.exe. And it should show up on the desktop either...
No "is this allowed?" dialog. No "please enter your password" dialog. Just.. don't.. execute.. it.
I would even go so far as to force a manual base64 or uu decode in there.
Get off my lawn, you damn kids!
It's the integration. Market a computer to someone who doesn't know what a program is, and allow that person to run programs. Without having to consult with anyone. So you are right; some basic training is in order.
.jpg.exe work? Especially if part of the information is hidden? Who thought that would EVER be a sensible idea? I used to try to allay peoples fears years ago, by saying "No, you cannot get malware from email -- the thought is ludicrous". And then we get the integration that lets it happen easily.
But the layout and features of Windows needs to come under examination as well. WHY does double clocking
Unix (Linux) isn't a "magic bullet". And, indeed, applications have been dragged into that "tight integration" role -- email must talk to the calendar, and multimedia, and random web pages must be allowed to run code, without sand-boxing! I swear, it makes me want to become a luddite. My kids are insisting that it is "ok" to run random games from the internet now.
At least I keep tabs of all traffic, incoming/outgoing. If it spikes, I find out why. Pull the plug. It's happened a few times.
"Where do you want to go today?" ads. Probably as socially responsible as marketing 400kmph racing cars to people who barely know how to drive. And when they crash, just engaging in hand-waving "They should have known".
No, I don't have the answer.
But the government will look to have their needs met. Instead of the "one-size-fits-all" solution, customized solutions can be proposed.
The "Linux toys": one is a media extender, fixed software with TV/remote control. Another is a a router "kit" -- pick the functions and build it yourself. The last is a general purpose laptop (openoffice, ssh). They don't look the same, configure the same, or have ANYTHING to do with each other -- except that they are all based on commodity Linux.
Could be based on something else, but the OS is only resource control and some APIs. Commodity. And, further, the OS becomes the base. For example, the router uses Linux. But it isn't an x86! And that fact turns out not to be relevant.
It could have been Linux, BSD, or something else. Just happens that Linux won.
No.
What it does is lock in a platform, and everything that isn't IDENTICAL to that platform gets turfed. Want a recent example? One in which Microsoft get caught in the their OWN trap?
Microsoft decides to include an "autorun" feature. Considered a VERY BAD IDEA, but it becomes a checkmark. Microsoft then introduces VISTA, and "modifies" the autorun feature. Is is simply removed? No, the feature remains. It now opens a dialog, asking if the autorun should be done.
This is EXACTLY the same as the user opening the application on the inserted media. In neither case is the autorun, well... autorun. Even Microsoft cannot retract the autorun feature, even though it is no longer autorun. It is now bloat. Created by platform lock-in.
The effect of removing Microsoft from the software equation would be a broad examination of HOW things should be done. Microsoft believes in a "one size fits everything" (ok, 8 sizes). Since source was not provided, it was impossible to tune installations for purpose.
Within the Linux ecosystem, the Linux running on a DSM-320 is NOT AT ALL like the Linux running on my laptop. It is ALSO different from the Linux running on my router. There can be individual modifications made to support specific uses and platforms. The Linux running on the eeepc is not the same flavor as the Linux running on my T43.
We will see broader individualized deployments. A government department could (reasonably) have their own "distribution" customized to their needs. There would be an ecosystem of people who supply that service (similar to the legion of Microsoft supporters, including the "MVP"s).
This has happened with Solaris and Linux is the server space, and has happened with Microsoft in the "desktop" space. You would now see "Linux Desktop Specialist" (or whatever the position would be called). And once the customizations start rolling, more users would see the benefit of having the environment completely customizable.
Um... they are NOT all processing elements
The T2000 (for example) has one die clocked at 1.2 or 1.4 Ghz. On that die are 8 processor cores. Each of these has 4 CMT threads (sort of what Intel used to call hyper-threading). 32 "virtual" cpus, 2U form factor. $6,995 base
The M-series, lets take the M5000 is built by Fujitsu. That has 8 processing elements, each clocked at 2.1 Ghz, with each dual-core. 16 "virtual" cpus. $47,000 base. 10U form factor.
The 5220 has 8 cores, 8 CMT per core, for 64 virtual cpus, 1.4 Ghz. $10,995 base. 2U,
The issue here is that all of this threading doesn't help "straight line" programs. The Power6 does much better (at 5 Ghz)! For "straight line" code, the best machine here (of SUNs offerings) is the M5000 -- machines (more or less) selected randomly from SUNs catalog. Only 16 virtual CPUs, but boy does it cook! (and, its 5 times the size, and 5 times the price).
The best "laptop" I've ever used was the Radioshack M100. Could take notes ALL DAY (maybe even all week). Upload the notes, and continue. The size of a clipboard, and a reasonable keyboard (for typing).
Now? *If* it plays movies (DVD, or other), it should be able to play at least 2 full movies (at least 4 hour battery life, although the M100 lasted 20 to 40 hours!). I should be able to pull it out and type on it (capture notes) without waiting minutes for it to "boot". It should be dead quiet for use in meetings. It should be (almost) indestructible. It should offer telnet/ssh connectivity (bonus if it supports X). It should be able to use standard batteries of some kind (AA?), or a common DC input (12V? 6V? but with a wide tolerance). It should support USB ports for additional storage. It should have integrated WIFI and RJ45 network plug. It should NOT be larger than 8.5x11x1. The keyboard should have full (typist) travel.
Does this product exist? I don't know. My current laptop (Thinkpad T43) occasionally goes "super loud" (its fan kicks in), and even blows papers off of my desk. It is too warm to use comfortably on my lap. The battery only lasts 1.5 hours (not quite a movie). The keyboard doesn't have enough travel, but it does run Linux (and therefore telnet, ssh and X). It also takes up to a minute to wake up sometimes. I use it, but I am sure not happy with it.
I don't think my "ideal" laptop exists yet -- but I have to look into the new ultra-small units (not for the size; I think that the keyboards will be too small). I am also very interested in the Apple Air.
Very witty. Made me spew coffee! Thanks.
I have tried to purchase "business class" service from my ISP. Have been turned down every time (3 times and counting). I try every 6 months to a year.
:( (my ISP *claims* I get 6 Mbps, and delivers 0.5 to 6, averaging 3).
No, I don't want petabytes of storage, or zetabytes of bandwidth. I use between 15GB and 100GB of "bandwidth" per month. Storage? zilch.
Now, I have 60GB per month with an overage charge of $2/GB. As to SPEED... As long as its reasonable, I frankly don't care. What is reasonable? 6 Mbps to allow download Standard Definition TV at "real time" rates. Of course, I don't get that
http://www.washingtonpost.com/wp-dyn/content/article/2006/10/01/AR2006100101090.html
Well... subway cars are heavier than tires, so we won't have anything else to worry about. Really, this time its ok.
Let's put this into perspective.
I can download any audio recording (movies, books, etc.) that was fixed prior to... 1912, I believe. WWII hasn't yet entered the public domain, culturally. Not the Vietnam conflict, Not the racial marches. Not the moon landing.
It's all under copyright; nothing is public domain.
Give it all up. Go wear a straw hat, or, if you are female, do the "flapper" thing. Until its pushed back further. I'm ok with this (personally). But then, I think Beethoven's 9th is the best piece of music ever written. I can pass on the rest of "culture" until I can own it and contribute (make use of it).
I am a person, not just a "consumer" or an "eyeball".
Since copyright law is SO SKEWED, I feel quite content ignoring it. When I was a kid, we didn't have VHS, DVDs. I went to the movies two (or three tops) times a month (less if there were no new releases). Never saw a movie twice (had better things to do with my money). Wouldn't tolerate advertisements EXCEPT trailers in the movie.
Had three TV channels, and records. Bought the occasional 45 and LP.
The media companies were happy.
Now -- I still go to movies. Pop and Popcorn the same price as the tickets, or more. Have to put up with Card ads, Soft drink ads, Still have trailers. Then, the movie comes up for rental/DVD purchase. I pay again. There are toy promotions, etc. I have in excess of 100 channels on the TV.
The media companies have much more opportunity to market and make money (I now pay for movies TWICE routinely, if I see them in the theater).
However, there is a push to control what I can further do with this deluge that is pushing in, trying to become culture. I can't even use Mickey yet! What's the answer? I figure that there is SO MUCH media being pumped (radio, TV, satellite radio, wireless internet all beaming stuff to where I am, 24/7) that it will cause some health issues. And yet, the motto is "Look, but don't touch". I would argue that it is almost IMPOSSIBLE to not break copyright. "We control your culture".
Ok, I am probably insane; just wish I had some answers. "They" make it very difficult.
"Older movies" are on film. Not recorded on video (HD or otherwise). So, a quality HD encoding will make a difference. For home use, HD (HD DVD or Blue Ray) is about as good as it gets. On a 42" screen, I can't tell the difference from film.