Most implementations of "micro-managed" memory use the allocate/free model. Programmers are very careful to allocate what they need, and free it when done.
But... Allocation is usually very cheap. You have a big hunk of memory, and a "high water" mark. If the new allocation fits, just take it and advance the mark. Free is not so cheap. Blocks need to be coallesced (sp?).
GC approach is to give the memory (same low cost as allocate), and simply NOT FREE the memory. When we run out of memory is the time that the "free" operation needs to be done. And only then. Also, only "live" memory needs to be dealt with. Everything else is "dead" (garbage). So, the "free" overhead tends to be less.
Consider the following pattern:
a = allocate b = allocate c = allocate d = allocate
free c free a free d free b
In a GC system, non of the free calls is done. a, b, c, d go out of scope or die, and when GC is needed, the dead memory is contiguous. No fancy "joining" needed.
To achieve this is a "careful" non-GC system, the programmer must introduce object caches, and must know about the global behaviour of her program. Typically not the case.
So, in general, GC actually beats "careful micro-managed memory".
"Resource Acquisition is Initialization" idiom... Can easily lead to deadlocks. Pre-declaration of resources is arguably superior for your RTOS. Other approaches are available as well.
For about a year now, I have tried a new anti-spam approach. Previously, tried black listing, white listing, and etc.
Problem is that the spam keeps coming, and sucks *my* bandwidth.
So I thought about it... And here's what I do. I use a hotmail account. With spam protection. Set to DELETE spam. I use a script (gotmail) to read the hotmail and transfer to my "real" email. Hotmail does a pretty good job on anti-spam, *and* I don't have to bother about the bandwidth.
Once we have a relationship established, you get my "real" email.
And I'm sticking to it. Works for now -- I've only had 100ish pieces of spam in the past year. Note that Hotmail seems to have improved the anti-spam features in the last six months (there was a sharp drop in spam).
If I subscribe to a list, I *only* use the hotmail address. Bet your bippy. I don't *trust* you yet, and I don't know where that email address is going. If the list gets bounced -- I don't see it, and, frankly, I don't care that much. Better than getting hammered on my server using my bandwidth.
My sympathies though... good luck dealing with Yahoo, Hotmail, (__fill_in_the_blank___).
Yes - spammers putting random words in is an effort to avoid being automatically judged to be spam based on content. The Habeas poetry is a pre-set sequence that can bypass the same filter (if the filter is aware of it). The idea is that it is *illegal* for the spammer to put that particular piece of poetry into an email.
Of course, the idea is slightly daft, because I get "spams" that offer illegal products (in my jurisdiction). *If* the spammer could be found, they could be shut down -- but they typically don't wait around that long (eg. Nigerian scam, insurance sales, and other).
Why shouldn't a Nigerian scammer put the poetry in? If *actually* caught and charged, I am sure that the additional offense will be simply plead out anyway...
In the embedded space, Linux has the "powerhouse" Motorola/Metrowerks/Lineo that supplies the infrastructure needed.
I use Redhat 9 Intel cross to MIPS embedded core using a modified Metro kernel (Linux 2.4). Target is embedded into TV sets. Win/CE doesn't have all that much traction in this market. It does have traction with PDA style devices, because it runs "standard" applications (by some measure of the word "standard" -- the apps have "Word" (tm) and "Excel" (tm) in the titles).
Other than that; Linux seems much more dominant. There are other OSs in this space, though, even some under the OSS banner. Its just more painful to move a "big" application (like a TV capture, SD TV playback, HD TV playback, PVR, etc). Using Linux, the work can be done in an "ordinary" environment first, and then moved to the embedded device.
With Win/CE, this can *also* happen, except that Win/CE is very restricted (compared to, say, Win/XP). With client side libraries, I ended up with a bunch of #ifdef WINCE sections that shouldn't have been needed. As this was at the lowest app levels, completely separate QA cycles are needed. So, as a developer, I prefer the same kernel embedded and developement. (Different processors, same software). It's smoother. Win/CE is just too much of a different OS (WINE is closer to XP AFAIR).
As to Metro penetration -- I can't really tell you, but it includes a lot of the Japanese and Koreans.
You could do it without Metro, but they do make it easier (and can offer bring-up services, etc. for the smaller).
"Microsoft will win...here is a complete solution..."
It doesn't matter. Perhaps Miguel is trying to "beat Microsoft". If it works, bully for him. Torvalds understands it -- the destruction of Microsoft will be (if it happens) an unintended consequence.
If Longhorn is that good -- it will be used. But lets be reasonable. Linux is a hobbiest OS. And it works really REALLY well. Some companies are deploying it; leveraging its strengths. It will go into cell phones, TV sets, home routers, and internet appliances. Because it is GOOD and CHEAP. It will go into satellite boxes, PDAs, and personal entertainment devices. Because it is GOOD and CHEAP.
It runs on servers, and some desktops. Because... (yup), it is GOOD and CHEAP.
"Pay MS and design the hardware to run MS technology". Not in the embedded world. Why pay a per-unit royalty if you don't have to? And, if the device can run MS technology, it can run Linux. And, Linux supports other non-MS technologies. Like CHEAP embedded MIPS processors. Like IBM mainframes. Like 68K processors. Like SPARC processors. Which have never been supported "MS technology". As a hardware vendor this gives more choice.
Still, the main thrust is the hobby. The "scratch an itch" movement.
You are right -- Linux doesn't solve problems -- it is an enabler (being OSS) that lets people solve their own problems. Which, I believe, is a much more powerful model. Unless a vendor nails MY problems on the head. I can get this by contracting custom software, but it is expensive. Under the OSS model, I can solve my problems, leveraging all other "solved problems".
Interesting that this leads to running OSS software in places that are surprising. Like Windows! [Ob eg - Cross compiling applications for Palm OS 4 under Windows... *most* use GCC].
It also leads to some pretty solutions that work very well. As another Ob Eg. compare and contrast MS IDE against Redhat Source Navigator for modifying large C/C++ projects.
Maybe MS will "win" (defined as making money -- I do own MS stock). I hope so.
But Linux doesn't compete with MS. Redhat does. Novell does (which is why Miguel is tracking this.NET stuff, I guess).
1 - Run WordStar - "yaze" and the "Commercial CP/M Archive". And...
WordStar (up to version 4) stored your files in a very simple format. To convert to ASCII, just strip the 8th bit of each character. Use 'tr' for this..
Off the top of my head:
# tr \\000-\\377 \\000-\\177\\000-\\177 file.txt
should do it (but I haven't botherd to test it... read the man page for tr). So, this is a simple conversion. Let me know if it works, or if you need more help with it (fred_weigel hotmail.com).
2 - Mag tape, audio. If you can play it, and the speed is off, it is a simple matter to modify the player. Or, digitize it at the wrong speed, and adjust the result.
3 - Betamax. Ebay would be your friend. This would be the most serious problem; convert those tapes before its too late... There are service shops that will do this for you (Betamax, PAL, etc to NTSC VHS). I had one convert 8mm to VHS...
4 - Slides. Sure, scan them in... Kodak PictureMaker (available at WalMart) has a slide scanner option.
Let's begin with this. I know its second on your list -- but why not.
The ruling in MAI v. Peak is that loading a program into RAM constitutes making a "copy" under Copyright law. Here (in Canada), this is NOT a violation of copyright, but (apparently) in the US, it is.
So, if you do not have a valid license (specific copyright), you may not execute programs.
Peak used systems at their premises, that had software licensed to MAI customers to access service logs. Basically, it was a violation of copyright to boot the MAI computer!
When downloading, I am not using MY computer to make the copy -- I argue that the reproduction happens on the computer hosting the material (Peak could have had the customer boot the OS, and this would have been legal!). The fault (in our case) lies with the uploader or hosting computer.
Peak knew about the licensing -- they in fact had a valid MAI license, and provided a service to MAI customers. I don't know that copyright assignment of the material. I am NOT responsible for every publisher and tracking copyright of such material. I listen to the radio, watch TV, read books, and download from the internet. In all of these cases I presume proper copyright, or that the copyright holders are pursuing the guilty.
Peak knew the customer had a license, was aware of the license, and booted the OS without a valid license anyway.
And that's the difference.
Enough of MAI v. Peak. (and, ps. it was a BAD decision -- you people in the US should be ashamed of that one).
'Downloading copyrighted music unauthorizedly is illegal '
Awkward, and wrong. Downloading is not illegal. Accessing computer resources without authorization is illegal (in many jursidictions). But downloading? No.
Uploading and sharing copyrighted material without "authorization" -- yes, illegal (in many jurisdictions -- there are exceptions).
Downloading... as much as people tell you it is "illegal", it is not (at least in the US and Canada). It may be "immoral", or "wrong", but illegal? No.
"FAT" systems store the allocation table as a singly linked list on disc. Two copies, in case one gets mangled (but they are adjacent, which is not good). The directory is a list of names, and starting indexes into the allocation table. This makes random access bad, because you have to keep traversing the singly linked list to find blocks. (DOS of early vintage).
CP/M also uses an allocation table, but it is not stored on disk. Instead, a file is broken into "extents". Each extent has a directory entry, and a fixed number of pointers to disk clusters. A single file will have more than one directory entry, if it contains multiple extents.
CP/M built the allocation map for the disc when the disc is "mounted" (used the first time). It does this by reading the directory, and marking blocks that are in use by files.
With FAT, you can have "cross linked" files. The singly linked lists representing the data blocks point into each other. With CP/M you can have multiple directory entries refer to the same data blocks as well. CP/M allows "sparse" files, which FAT doesn't. CP/M has better random access (two levels of index), although many programs pre-built access lists for DOS to improve random file performance (I did that for one application).
CP/M limited file names to emulate PDP-10, FAT limited file names to emulate CP/M -- it's a push.
And, finally, the patent is NOT on FAT, it is on the long file name extensions introduced with Windows 95.
A "FAT" system was in place with Microsoft Disc Basic (AFAIR), Zilog also used a singly linked list block map in their Z80 development platform. I am sure that there is plenty more "prior art" for FAT.
The idea of stored a hash long name into a fixed length directory in multiple pieces, using keys and checksums -- that is what is being contested.
GNU GPL states exactely that -- that NOTHING has compelled you to agree with the GPL.
If you didn't, then you are allowed to use the software.
But, and the GPL is clear on this point, NOTHING else gives you permission to redistribute.
Certainly Copyright restricts you. The GPL specifically allows this, under some conditions. Since you don't have this right, the GPL grants it to you.
No, you don't have to agree to the GPL. No, you don't have to even READ the GPL. But, if you don't, your rights are governed by the prevailing law (in this caase, Copyright). You don't have that, and the punishment can be quite severe.
What is the "value" associated with an infringement? Depends. If it is a product like a security router, and DEPENDS on the GPL software... could be as high as MULTIPLES of all profit.
Now, I don't know if you're actually going to read this: so let's call your reply a sucessful troll...
Tom's article *did* mention Mozilla, Evolution, Xine (maybe MPlayer), GAIM... *and* what they are used for.
You know, the application mix that a hobby user would find useful.
With everything installed, the user can then begin branching out into other applications. This policy (install everything) makes installation of other things easier.
You want frustration? Lead the user into RPM dependency hell... Until the user KNOWS, Tom's advice is the best. Remember: the user will have to be reasonably self-sufficient.
Also, the applications are organized by function (in all the distributions I am familiar with - Redhat 9, Fedora, Mandrake). Also, Redhat 9 (at least) has a CD full of documentation (how to administer your system). I even still refer to it.
And, when the new Linux user becomes familiar with the "new" way of doing things, there are other choices she can make -- which office suite, mail agent, mailer, desktop, etc. to use. But, she will be making these decisions based on some real experience.
If this is a CRIMINAL matter, you are very VERY wrong. Nothing to do with "...beyond a reason doubt." either.
And, just for your information, the US (I assume you are in that jursdication), does allow acts if there is no statement preventing said act. And that's in your constitution.
Not so in Canada, but I sure hope that AC has an agreement in place with the ex-employee. Without a mention of web site usage, they are pretty much fucked. Of course, this could be a last-gasp attempt at increasing AC stock price (what is it now? 1.10 CDN or so?) at WestJets expense.
Now, the ex-employee in question may or may not be a "geek". I'll leave that question alone. But if *I* were given marketing data, updated for five years, for my use *without* a rider restricting that use, I would sure use it. And, if sued, take it to the limit. $5 Million and any profits? Why, the counter-suit would be for the whole fucking company.
And that's why I think this is a very stupid move by the AC CEO. (and I fully expect that he expects a bail-out, and to keep his job).
I believe that there is a Ximian connector for Exchange, but I don't know for sure -- never seen it. I have a Windows 2000 box at work for that reason (and that one alone). Microsoft Exchange...
As to Visio -- you can view Visio inserts in Word docs, but as to editing? I don't know, and have never seen such a thing on Linux or Solaris. It may exist -- but I don't know.
As to VPN -- CISCO has a VPN client for Linux. Others? Don't know.
And, for me, it doesn't matter. Since I need W2K for Exchange, its there. If its "Windows only", I run it there. If not, it runs on my Linux box.
For my document preparation needs, I use LyX, OpenOffice.org, XFig, and some other tools. A lot of them don't go back to the MS world easily, either. Probably "reverse-lockin" -- the MS tools are considered "good enough", and anyone running MS who ACTUALLY wants to edit this stuff can run a Linux box, use VMware, or Cygwin.
Anyway, todays computers are plenty quick enough, and RAM is cheap enough that a VMware solution may be in order. You would still pay Microsoft for the OS (to run the Exchange client).
But, honestly, there isn't anything really wrong with W2K. It does work. I know more about Linux/Solaris though, and prefer those environments. So, my penalty is 2 general purpose boxes on my desk (or money to VMware and more RAM).
So, why are you stressing over not being able to use Linux anyway?
Out of order execution takes the processor to a particular level of performance. Unfortunately, (and especially with the X86 IA), we run out of steam rather quickly, and the processor blocks waiting on registers or memory. The idea behind HT is that the processor's execution elements can then be reassigned to something else waiting in cache.
Of course, this means we need a big fat cache, and something else to execute. Could be another thread or process, but the important thing is that the second job be independent.
This can increase the utilization of the processor's compute elements.
So, yes, the "builds with multiple jobs running at the same time" test makes sense.
I would like to see a benchmark with CPU stalls and utilization summarized at the end. Can't do it myself, because I am far too cheap to replace my current system (and yes, it is an MP box - dual 200Mhz PPRO - and it still does quite nicely).
Anyway, it does look the the Intel took a hit in this benchmark; too bad for them. I looked over the methodology -- and it looked reasonable given the scope of the project.
I just want to point out that your meathead will always keep your (A)I contained and, by design, prevent learning beyond the meatheads initial design.
Where is the fundamental flaw?
Oh, and programs that are introspective can be written, and they can modify themselves, without the original programmer being involved.
I personally don't *want* my computer to simulate a human "Gee, Ratboy, I don't feel like looking up that information right now...". And the worst would be the "sullen, adolescent" version. "..." (that was a lot of silence in answer to my request), or "Why did you crash?" "I don't know. You are ALWAYS picking on me.".
ratboy (IAAPAKTTT - "I am a programmer and know the techical terms")
So fuck off, use your Windows, AIX, cha-cha-cha, and leave us ALONE!
I don't understand, and have never understood, what the big deal about "Linux adoption" is. Happens to be a fine kernel -- and I certainly don't mind if people use it, even make money from it. My contributions were/are never with commercial intent -- simply "this is good, I like it, maybe you will like it too".
So let Dan Farber preach to business. It isn't my job to tear down any "walls to adoption". Really, as a hobby programmer, I don't care.
Now, many companies DO see the point (to them) of using this "open source" stuff. And many don't. Personally, I work for a company that DOES use linux in an embedded role -- works better for me.
Sure, any hobby will attract "Rabid, frothing... zealots". Its passion man! Enjoy it. And it makes not one whit of difference.
And that is why OS vendors are essentially doomed. We do this stuff for fun, and we don't care. Can't really leave it alone. It becomes very hard to compete, given that the cost of distribution is almost 0. (Imagine, someone who can lovingly hand-build cars, and then can turn around and give the final product away for no cost! And, can produce AS MUCH product as any big auto maker! That is the "competition").
If the big software mills COULD produce much higher quality goods at reasonable cost, we wouldn't even be HAVING this conversation.
If (for example) Solaris or Windows really were hyper-secure and had a record of no breakins, AND didn't crash, then, sure, I (you, we) would be a fool to not use them in a business context. And there would be no discussion (NO, NONE, NADA) of "Linux". (or OpenBSD, NetBSD, etc.)
But they aren't. Tough luck for the OS vendors, but the "hobbyist" system is just as good (arguably better in many respects). So, OS vendors, suck it up.
OS vendors have a few choices:
1 - Fight
Sure, give us a better product. I am waiting for "Longhorn". Might be worth it (see above).
2 - Switch
Novel with Netware, IBM with Linux. Phase out your un-competitive products, and back Linux. Why not? Sell something other than an OS (services, hardware, support...)
3 - Leave
Go into a different business. OS vending won't cut it.
Adapt or die.
Note that the question is never "Why should I switch to Linux from...". Whoever it is will know why and when. Because, it doesn't matter to me, because I don't make dime 1 on this. This is my HOBBY. I do this for fun. I also write stuff that very few people read, rant, play very bad guitar, and ski -- for the same reason. There are people who write better than I do -- and I buy and read their books. There are people who rant better than I do, and I elect them. There are people who play better guitar, and I see them in concerts (and sometimes even buy CDs). There are people who ski better than I do, and I watch "World Cup Downhill" events on TV.
So, ask yourself, WHY isn't AIX, Solaris, Windows superior enough to Linux, OpenBSD, NetBSD so we wouldn't have this talk?
'cause you know what?
With the current state of affairs, *even* with my "fuck you" attitude, OS vendors are basically screwed.
Yup, PS is commonly used on Video Servers. Even over unreliable transports. EG. PS over RTP/UDP. Over wireless. With transcoding (sometimes).
Whereas TS should be used in those cases. Makes error recovery easier. Oh, well, live and learn.
As to 720p being superior to 1080i -- honestly, I don't know. I did say "etc". I would imagine sports (action) on 720p is superior, but I just work with canned streams. And the products support all formats. At home, my set is basically left at 1080i -- why? -- because I honestly can't tell the difference, and no one else in my family can either, especially because the highest resolution input we have is a non-progressive scan Sony DVD player. In the sense that I do drivers for HDTV, I guess I am a "professional" of sorts; my opinion is -- I don't think it matters.
MPEG-2 IBBP GOP sequence? Not given. depends on the encoder. To save on broadcast space, there is a push for "...BBBBBBBBBB..." sequences. Of course, that leads to artifacts on transmission errors (which wouldn't be an issue for PS streams, so why not?). Of course, errors are a fact of life with sat, etc. But, the GOP is variable (depends on the encoder and the program).
Interesting - I just want to expand on your "fleshy" point:
Modern TVs use a colour gamut designed to improved flesh tones, yet they have a smaller gamut than the original colour specification. In other words, an antique original colour TV is able to represent MORE colours than your current set.. It just won't do hiqh quality pr0n as well.
As to "near-perfect" signal quality and picture integrity... I would argue that "digital" mpeg encoding reduces quality. The mpeg encode of course relies on "picture integrity" (actually, no, everything is bundled up into 188 byte packets, with the assumption that there will be lossage, and no retransmission).
As to resolution - 480i has been "good enough". Indeed, DVDs are 480i/p as well. Generally, few complaints.
1080i (etc.) HD formats. ARE a major step. Roughly, an order of magnitude improvement. But, for many, 480i/p is "good enough" (please note that HD has 6 times the datarate of a current DVD - and DVD *is not* an HDTV format. The only source of HD will be broadcast (possibly cable or sat..). And, you won't actually be able to *record* an HD signal using normal consumer gear).
And, I find that 480i/p is good enough for me. I do have a largish set, and still don't really have the urge for HDTV. If we had "super-DVD" out there, with 1080i format movies, THEN I would for it. But, I honestly don't care for broadcast formats.
"Bearing in mind the fact that it is impossible to switch to Linux at home for a number of reasons and also that because of the business I need to be online 24/7/365 what the Slashdot community suggest"
So you need Windows. Which is ok -- put Linux on another box, and secure it. I just bought a Compaq with 128MB of memory, 20GB or so hard drive, 400Mhz processor for 100$ CDN (80$ US or so). Used.
Something like that would make a good firewall for you.
Alternately, home routers also have reasonably firewalling. My SMC Barricade (gasp, yet, I know that a REAL geek wouldn't use one) offers the ability to drop in-bound traffic, and only allow certain ports through. This can provide you 80% of what you need (it does for me). Staying on top of patches can bring you the rest of the way. Just don't enable the "DMZ" feature!
As you mentioned, you have external hosting -- which means that you don't have to allow incoming HTTP, or SMTP. If you don't need to administer externally (and since you use Windows, you *probably* don't), you don't need port 22. So, close off ALL inbound connections. Just leaves you with FTP as an issue -- some router boxes will accomodate, or you can learn to love the PASV command (and, AFAIK, MS browser FTP does that automagically).
If you AREN'T using a small home router, GET ONE. They are even cheaper (I have seen brand new units selling here for $20 CDN, approx. $15 US).
Don't forget a good backup plan, just in case you get rooted (or other disaster strikes).
Still, buying a cheap box or two is reasonable. One for a "real" firewall, and another for SAMBA, and other internal services (DNS).
Actually, they are. May not be expert with computers, but you should allow for expertise in the problem area.
"burden on programmers to take responsibility of policing bad behavior"
No. If this is done, then functionality that the programmer has not explicitly thought of is lost. To give you an example -- I needed to introduce a LOT of Postscript to an Apple LaserWriter (the original, with only a megabyte of memory). The code could NOT fit into memory -- yet I was able to tell the interpreter to do a "stupid" thing -- execute the serial port. Saved the day, and it worked fine. I don't think that the programmer had this in mind at all...
As another example: If you allow "printing" to a file, DON'T check to make utterly sure its a file -- it may be a pipe to another program. I guess you could check, at the loss of an enormous amount of flexibility.
Also, users should be encouraged to have "multiple" views of data. An example would be equations. One view is as a typeset equation. Working with that representation is *extremly* tedious. A better model for editing would be a formula view. Or, say, pictures. "Circle, 1 cm diameter". vs. something hand-drawn (and at what resolution?).
Symbolic reasoning is a higher level of abstraction than concrete reasoning. Most of the output produced by the user will be concrete (it will be a tangible, in the sense that it can be experienced). Just don't make the mistake of forcing the user to work in the concrete domain all the time.
I run a DNS server for my little LAN. Internally, machines are added to a ".lan" domain, because they are NAT'd anyway. Can't be accessed from outside, so they don't get the external domain name. Traffic inbound goes to the correct machines.
In the past, I had a guarantee that ".lan" would not be a valid TLD. It can't be a country code, and it's not a standard TLD.
So... now the assurance is going to go away. The price of progress, I guess. I do hope that some TLDs remain reserved for internal use (same as the unroutable IP addresses.)
Of course we need only one such private TLD (I sure hope it's ".lan" because that's easiest for *me*).
But note that the "somebody" who DOWNLOADED the song was NOT "gone after".
The party doing the publishing was brought down -- not the people who used the result of the publication.
It IS legal to download ROMs. May or may not be legal to copy them, and provide them on electronic sites (yes, legal, if copyright has been granted, else no).
I truly would like to see the fireworks if a DOWNLOADER was prosecuted. It would open a nasty can of worms.
And that leaves (1) emulator and (2) the legality of making a copy to run on the emulator.
An emulator is perfectly legal. There may be patent encumberance though... and this is what Nintendo seems to be claiming.
If the emulator is legal, it may be considered a NCESSARY step to copy a ROM to run it. That is allowed in my jurisdiction (Canada) and, I beleive US as well. As well, backup copy can be made.
And there is no difference between material provided on-line (downloaded) and an actual ROM.
YOU CAN'T GET INTO TROUBLE FOR DOWNLOADING. Sorry for the harshness, but it seems that many would have you believe otherwise. There are, of course, exceptions to this rule. In Canada, it is illegal to download child porn, and certain other categories of material. Please note that this has nothing to do with copyright.
YOU CAN GET INTO TROUBLE FOR COPYING OR "UPLOADING". Big time.
If someone places music on-line, there are two possibilities -- the first is that the person has permission to do so (copyright), the second is that the person does not.
If the person *has* permission, then you have the right to download and burn. The thing that you are copying *is* the original.
If the person does NOT have permission, they have broken copyright. Not you... As an example -- if there is a legal music download service, and they have a single illegal track (no copyright granted), the people who download that track are not liable. Only the provider. Big time.
Now -- if you download music from sites that are not located in Canada, it becomes murkier. Are the download sites even part of the Berne Copyright agreement? If so, the Canadian personal copying provision still holds. It is, of course, impossible to tell if the source is legal or not. One presumes that sources that DO NOT have copyright have been or will be shut down by actions of the copyright holder.
All CDs are "copies" to begin with. There is no such thing, effectively, as an "original" (there is the original performance, of course, and there is a part of the Canadian Copyright Act that deals with that).
If the user is ALSO responsible, we have much larger problems. Example -- a teacher copies some pages out of a textbook. The students are later fined for using the material. You would have to verify the copyright of every item that you buy or are given.
Reading this way -- leach away, my Canadian friends!
Slashdot Mirror
Micro-managed memory vs. GC Efficiency
GC wins. And here is why...
Most implementations of "micro-managed" memory use the allocate/free model. Programmers are very careful to allocate what they need, and free it when done.
But... Allocation is usually very cheap. You have a big hunk of memory, and a "high water" mark. If the new allocation fits, just take it and advance the mark. Free is not so cheap. Blocks need to be coallesced (sp?).
GC approach is to give the memory (same low cost as allocate), and simply NOT FREE the memory. When we run out of memory is the time that the "free" operation needs to be done. And only then. Also, only "live" memory needs to be dealt with. Everything else is "dead" (garbage). So, the "free" overhead tends to be less.
Consider the following pattern:
a = allocate
b = allocate
c = allocate
d = allocate
free c
free a
free d
free b
In a GC system, non of the free calls is done. a, b, c, d go out of scope or die, and when GC is needed, the dead memory is contiguous. No fancy "joining" needed.
To achieve this is a "careful" non-GC system, the programmer must introduce object caches, and must know about the global behaviour of her program. Typically not the case.
So, in general, GC actually beats "careful micro-managed memory".
"Resource Acquisition is Initialization" idiom... Can easily lead to deadlocks. Pre-declaration of resources is arguably superior for your RTOS. Other approaches are available as well.
Ratboy
For about a year now, I have tried a new anti-spam approach. Previously, tried black listing, white listing, and etc.
Problem is that the spam keeps coming, and sucks *my* bandwidth.
So I thought about it... And here's what I do. I use a hotmail account. With spam protection. Set to DELETE spam. I use a script (gotmail) to read the hotmail and transfer to my "real" email. Hotmail does a pretty good job on anti-spam, *and* I don't have to bother about the bandwidth.
Once we have a relationship established, you get my "real" email.
And I'm sticking to it. Works for now -- I've only had 100ish pieces of spam in the past year. Note that Hotmail seems to have improved the anti-spam features in the last six months (there was a sharp drop in spam).
If I subscribe to a list, I *only* use the hotmail address. Bet your bippy. I don't *trust* you yet, and I don't know where that email address is going. If the list gets bounced -- I don't see it, and, frankly, I don't care that much. Better than getting hammered on my server using my bandwidth.
My sympathies though... good luck dealing with Yahoo, Hotmail, (__fill_in_the_blank___).
Ratboy.
TykeClone
Yes - spammers putting random words in is an effort to avoid being automatically judged to be spam based on content. The Habeas poetry is a pre-set sequence that can bypass the same filter (if the filter is aware of it). The idea is that it is *illegal* for the spammer to put that particular piece of poetry into an email.
Of course, the idea is slightly daft, because I get "spams" that offer illegal products (in my jurisdiction). *If* the spammer could be found, they could be shut down -- but they typically don't wait around that long (eg. Nigerian scam, insurance sales, and other).
Why shouldn't a Nigerian scammer put the poetry in? If *actually* caught and charged, I am sure that the additional offense will be simply plead out anyway...
Ratboy.
In the embedded space, Linux has the "powerhouse" Motorola/Metrowerks/Lineo that supplies the infrastructure needed.
I use Redhat 9 Intel cross to MIPS embedded core using a modified Metro kernel (Linux 2.4). Target is embedded into TV sets. Win/CE doesn't have all that much traction in this market. It does have traction with PDA style devices, because it runs "standard" applications (by some measure of the word "standard" -- the apps have "Word" (tm) and "Excel" (tm) in the titles).
Other than that; Linux seems much more dominant. There are other OSs in this space, though, even some under the OSS banner. Its just more painful to move a "big" application (like a TV capture, SD TV playback, HD TV playback, PVR, etc). Using Linux, the work can be done in an "ordinary" environment first, and then moved to the embedded device.
With Win/CE, this can *also* happen, except that Win/CE is very restricted (compared to, say, Win/XP). With client side libraries, I ended up with a bunch of #ifdef WINCE sections that shouldn't have been needed. As this was at the lowest app levels, completely separate QA cycles are needed. So, as a developer, I prefer the same kernel embedded and developement. (Different processors, same software). It's smoother. Win/CE is just too much of a different OS (WINE is closer to XP AFAIR).
As to Metro penetration -- I can't really tell you, but it includes a lot of the Japanese and Koreans.
You could do it without Metro, but they do make it easier (and can offer bring-up services, etc. for the smaller).
Nothing like putting your TV set on the 'net!
L8R
Ratboy
"Microsoft will win...here is a complete solution..."
.NET stuff, I guess).
It doesn't matter. Perhaps Miguel is trying to "beat Microsoft". If it works, bully for him. Torvalds understands it -- the destruction of Microsoft will be (if it happens) an unintended consequence.
If Longhorn is that good -- it will be used. But lets be reasonable. Linux is a hobbiest OS. And it works really REALLY well. Some companies are deploying it; leveraging its strengths. It will go into cell phones, TV sets, home routers, and internet appliances. Because it is GOOD and CHEAP. It will go into satellite boxes, PDAs, and personal entertainment devices. Because it is GOOD and CHEAP.
It runs on servers, and some desktops. Because... (yup), it is GOOD and CHEAP.
"Pay MS and design the hardware to run MS technology". Not in the embedded world. Why pay a per-unit royalty if you don't have to? And, if the device can run MS technology, it can run Linux. And, Linux supports other non-MS technologies. Like CHEAP embedded MIPS processors. Like IBM mainframes. Like 68K processors. Like SPARC processors. Which have never been supported "MS technology". As a hardware vendor this gives more choice.
Still, the main thrust is the hobby. The "scratch an itch" movement.
You are right -- Linux doesn't solve problems -- it is an enabler (being OSS) that lets people solve their own problems. Which, I believe, is a much more powerful model. Unless a vendor nails MY problems on the head. I can get this by contracting custom software, but it is expensive. Under the OSS model, I can solve my problems, leveraging all other "solved problems".
Interesting that this leads to running OSS software in places that are surprising. Like Windows! [Ob eg - Cross compiling applications for Palm OS 4 under Windows... *most* use GCC].
It also leads to some pretty solutions that work very well. As another Ob Eg. compare and contrast MS IDE against Redhat Source Navigator for modifying large C/C++ projects.
Maybe MS will "win" (defined as making money -- I do own MS stock). I hope so.
But Linux doesn't compete with MS. Redhat does. Novell does (which is why Miguel is tracking this
Ratboy.
Fixes:
1 - Run WordStar - "yaze" and the "Commercial CP/M Archive". And...
WordStar (up to version 4) stored your files in a very simple format. To convert to ASCII, just strip the 8th bit of each character. Use 'tr' for this..
Off the top of my head:
# tr \\000-\\377 \\000-\\177\\000-\\177 file.txt
should do it (but I haven't botherd to test it... read the man page for tr). So, this is a simple conversion. Let me know if it works, or if you need more help with it (fred_weigel hotmail.com).
2 - Mag tape, audio. If you can play it, and the speed is off, it is a simple matter to modify the player. Or, digitize it at the wrong speed, and adjust the result.
3 - Betamax. Ebay would be your friend. This would be the most serious problem; convert those tapes before its too late... There are service shops that will do this for you (Betamax, PAL, etc to NTSC VHS). I had one convert 8mm to VHS...
4 - Slides. Sure, scan them in... Kodak PictureMaker (available at WalMart) has a slide scanner option.
Ratboy
MAI v. Peak
Let's begin with this. I know its second on your list -- but why not.
The ruling in MAI v. Peak is that loading a program into RAM constitutes making a "copy" under Copyright law. Here (in Canada), this is NOT a violation of copyright, but (apparently) in the US, it is.
So, if you do not have a valid license (specific copyright), you may not execute programs.
Peak used systems at their premises, that had software licensed to MAI customers to access service logs. Basically, it was a violation of copyright to boot the MAI computer!
When downloading, I am not using MY computer to make the copy -- I argue that the reproduction happens on the computer hosting the material (Peak could have had the customer boot the OS, and this would have been legal!). The fault (in our case) lies with the uploader or hosting computer.
Peak knew about the licensing -- they in fact had a valid MAI license, and provided a service to MAI customers. I don't know that copyright assignment of the material. I am NOT responsible for every publisher and tracking copyright of such material. I listen to the radio, watch TV, read books, and download from the internet. In all of these cases I presume proper copyright, or that the copyright holders are pursuing the guilty.
Peak knew the customer had a license, was aware of the license, and booted the OS without a valid license anyway.
And that's the difference.
Enough of MAI v. Peak. (and, ps. it was a BAD decision -- you people in the US should be ashamed of that one).
Ratboy.
'Downloading copyrighted music unauthorizedly is illegal '
Awkward, and wrong. Downloading is not illegal. Accessing computer resources without authorization is illegal (in many jursidictions). But downloading? No.
Uploading and sharing copyrighted material without "authorization" -- yes, illegal (in many jurisdictions -- there are exceptions).
Downloading... as much as people tell you it is "illegal", it is not (at least in the US and Canada). It may be "immoral", or "wrong", but illegal? No.
And yes, we do read Copyright law.
Ratboy.
FAT12 and extension to CP/M file system?
No.
"FAT" systems store the allocation table as a singly linked list on disc. Two copies, in case one gets mangled (but they are adjacent, which is not good). The directory is a list of names, and starting indexes into the allocation table. This makes random access bad, because you have to keep traversing the singly linked list to find blocks. (DOS of early vintage).
CP/M also uses an allocation table, but it is not stored on disk. Instead, a file is broken into "extents". Each extent has a directory entry, and a fixed number of pointers to disk clusters. A single file will have more than one directory entry, if it contains multiple extents.
CP/M built the allocation map for the disc when the disc is "mounted" (used the first time). It does this by reading the directory, and marking blocks that are in use by files.
With FAT, you can have "cross linked" files. The singly linked lists representing the data blocks point into each other. With CP/M you can have multiple directory entries refer to the same data blocks as well. CP/M allows "sparse" files, which FAT doesn't. CP/M has better random access (two levels of index), although many programs pre-built access lists for DOS to improve random file performance (I did that for one application).
CP/M limited file names to emulate PDP-10, FAT limited file names to emulate CP/M -- it's a push.
And, finally, the patent is NOT on FAT, it is on the long file name extensions introduced with Windows 95.
A "FAT" system was in place with Microsoft Disc Basic (AFAIR), Zilog also used a singly linked list block map in their Z80 development platform. I am sure that there is plenty more "prior art" for FAT.
The idea of stored a hash long name into a fixed length directory in multiple pieces, using keys and checksums -- that is what is being contested.
Ratboy
And, firewood, you are absolutely correct.
GNU GPL states exactely that -- that NOTHING has compelled you to agree with the GPL.
If you didn't, then you are allowed to use the software.
But, and the GPL is clear on this point, NOTHING else gives you permission to redistribute.
Certainly Copyright restricts you. The GPL specifically allows this, under some conditions. Since you don't have this right, the GPL grants it to you.
No, you don't have to agree to the GPL. No, you don't have to even READ the GPL. But, if you don't, your rights are governed by the prevailing law (in this caase, Copyright). You don't have that, and the punishment can be quite severe.
What is the "value" associated with an infringement? Depends. If it is a product like a security router, and DEPENDS on the GPL software... could be as high as MULTIPLES of all profit.
Ratboy.
AC:
Now, I don't know if you're actually going to read this: so let's call your reply a sucessful troll...
Tom's article *did* mention Mozilla, Evolution, Xine (maybe MPlayer), GAIM... *and* what they are used for.
You know, the application mix that a hobby user would find useful.
With everything installed, the user can then begin branching out into other applications. This policy (install everything) makes installation of other things easier.
You want frustration? Lead the user into RPM dependency hell... Until the user KNOWS, Tom's advice is the best. Remember: the user will have to be reasonably self-sufficient.
Also, the applications are organized by function (in all the distributions I am familiar with - Redhat 9, Fedora, Mandrake). Also, Redhat 9 (at least) has a CD full of documentation (how to administer your system). I even still refer to it.
And, when the new Linux user becomes familiar with the "new" way of doing things, there are other choices she can make -- which office suite, mail agent, mailer, desktop, etc. to use. But, she will be making these decisions based on some real experience.
Ratboy.
Tom's was clear -- they promote that the new linux user install EVERYTHING.
The GTK comment just hammers this home.
Until the new user KNOWS what they want, they can try it all -- especially the target user here (knows Windows, and is curious).
The data migration tips alone make this an article worth reading -- how to move your IE bookmarks and PST mailbox over...
And "Oracle, Apache and SMB" come later. First, convert, bringing over your data, and start exploring the new environment.
I give Tom's a really big "Two Thumbs Up" for this article -- good to get more hobby users into Linux!
Ratboy
Ok.
If this is a civil matter, you *may* be right.
If this is a CRIMINAL matter, you are very VERY wrong. Nothing to do with "...beyond a reason doubt." either.
And, just for your information, the US (I assume you are in that jursdication), does allow acts if there is no statement preventing said act. And that's in your constitution.
Not so in Canada, but I sure hope that AC has an agreement in place with the ex-employee. Without a mention of web site usage, they are pretty much fucked. Of course, this could be a last-gasp attempt at increasing AC stock price (what is it now? 1.10 CDN or so?) at WestJets expense.
Now, the ex-employee in question may or may not be a "geek". I'll leave that question alone. But if *I* were given marketing data, updated for five years, for my use *without* a rider restricting that use, I would sure use it. And, if sued, take it to the limit. $5 Million and any profits? Why, the counter-suit would be for the whole fucking company.
And that's why I think this is a very stupid move by the AC CEO. (and I fully expect that he expects a bail-out, and to keep his job).
Ratboy.
Sorry, you just may be locked in to Windows.
I believe that there is a Ximian connector for Exchange, but I don't know for sure -- never seen it. I have a Windows 2000 box at work for that reason (and that one alone). Microsoft Exchange...
As to Visio -- you can view Visio inserts in Word docs, but as to editing? I don't know, and have never seen such a thing on Linux or Solaris. It may exist -- but I don't know.
As to VPN -- CISCO has a VPN client for Linux. Others? Don't know.
And, for me, it doesn't matter. Since I need W2K for Exchange, its there. If its "Windows only", I run it there. If not, it runs on my Linux box.
For my document preparation needs, I use LyX, OpenOffice.org, XFig, and some other tools. A lot of them don't go back to the MS world easily, either. Probably "reverse-lockin" -- the MS tools are considered "good enough", and anyone running MS who ACTUALLY wants to edit this stuff can run a Linux box, use VMware, or Cygwin.
Anyway, todays computers are plenty quick enough, and RAM is cheap enough that a VMware solution may be in order. You would still pay Microsoft for the OS (to run the Exchange client).
But, honestly, there isn't anything really wrong with W2K. It does work. I know more about Linux/Solaris though, and prefer those environments. So, my penalty is 2 general purpose boxes on my desk (or money to VMware and more RAM).
So, why are you stressing over not being able to use Linux anyway?
Ratboy.
What HyperThreading is...
Out of order execution takes the processor to a particular level of performance. Unfortunately, (and especially with the X86 IA), we run out of steam rather quickly, and the processor blocks waiting on registers or memory. The idea behind HT is that the processor's execution elements can then be reassigned to something else waiting in cache.
Of course, this means we need a big fat cache, and something else to execute. Could be another thread or process, but the important thing is that the second job be independent.
This can increase the utilization of the processor's compute elements.
So, yes, the "builds with multiple jobs running at the same time" test makes sense.
I would like to see a benchmark with CPU stalls and utilization summarized at the end. Can't do it myself, because I am far too cheap to replace my current system (and yes, it is an MP box - dual 200Mhz PPRO - and it still does quite nicely).
Anyway, it does look the the Intel took a hit in this benchmark; too bad for them. I looked over the methodology -- and it looked reasonable given the scope of the project.
Ratboy.
RazorX90:
I just want to point out that your meathead will always keep your (A)I contained and, by design, prevent learning beyond the meatheads initial design.
Where is the fundamental flaw?
Oh, and programs that are introspective can be written, and they can modify themselves, without the original programmer being involved.
I personally don't *want* my computer to simulate a human "Gee, Ratboy, I don't feel like looking up that information right now...". And the worst would be the "sullen, adolescent" version. "..." (that was a lot of silence in answer to my request), or "Why did you crash?" "I don't know. You are ALWAYS picking on me.".
ratboy (IAAPAKTTT - "I am a programmer and know the techical terms")
Linux is not a religion.
... zealots". Its passion man! Enjoy it. And it makes not one whit of difference.
...". Whoever it is will know why and when. Because, it doesn't matter to me, because I don't make dime 1 on this. This is my HOBBY. I do this for fun. I also write stuff that very few people read, rant, play very bad guitar, and ski -- for the same reason. There are people who write better than I do -- and I buy and read their books. There are people who rant better than I do, and I elect them. There are people who play better guitar, and I see them in concerts (and sometimes even buy CDs). There are people who ski better than I do, and I watch "World Cup Downhill" events on TV.
It's a hobby.
So fuck off, use your Windows, AIX, cha-cha-cha, and leave us ALONE!
I don't understand, and have never understood, what the big deal about "Linux adoption" is. Happens to be a fine kernel -- and I certainly don't mind if people use it, even make money from it. My contributions were/are never with commercial intent -- simply "this is good, I like it, maybe you will like it too".
So let Dan Farber preach to business. It isn't my job to tear down any "walls to adoption". Really, as a hobby programmer, I don't care.
Now, many companies DO see the point (to them) of using this "open source" stuff. And many don't. Personally, I work for a company that DOES use linux in an embedded role -- works better for me.
Sure, any hobby will attract "Rabid, frothing
And that is why OS vendors are essentially doomed. We do this stuff for fun, and we don't care. Can't really leave it alone. It becomes very hard to compete, given that the cost of distribution is almost 0. (Imagine, someone who can lovingly hand-build cars, and then can turn around and give the final product away for no cost! And, can produce AS MUCH product as any big auto maker! That is the "competition").
If the big software mills COULD produce much higher quality goods at reasonable cost, we wouldn't even be HAVING this conversation.
If (for example) Solaris or Windows really were hyper-secure and had a record of no breakins, AND didn't crash, then, sure, I (you, we) would be a fool to not use them in a business context. And there would be no discussion (NO, NONE, NADA) of "Linux". (or OpenBSD, NetBSD, etc.)
But they aren't. Tough luck for the OS vendors, but the "hobbyist" system is just as good (arguably better in many respects). So, OS vendors, suck it up.
OS vendors have a few choices:
1 - Fight
Sure, give us a better product. I am waiting for "Longhorn". Might be worth it (see above).
2 - Switch
Novel with Netware, IBM with Linux. Phase out your un-competitive products, and back Linux. Why not? Sell something other than an OS (services, hardware, support...)
3 - Leave
Go into a different business. OS vending won't cut it.
Adapt or die.
Note that the question is never "Why should I switch to Linux from
So, ask yourself, WHY isn't AIX, Solaris, Windows superior enough to Linux, OpenBSD, NetBSD so we wouldn't have this talk?
'cause you know what?
With the current state of affairs, *even* with my "fuck you" attitude, OS vendors are basically screwed.
Demand more!
Ratboy
wmeyer:
Thanks for your comments.
Yup, PS is commonly used on Video Servers. Even over unreliable transports. EG. PS over RTP/UDP. Over wireless. With transcoding (sometimes).
Whereas TS should be used in those cases. Makes error recovery easier. Oh, well, live and learn.
As to 720p being superior to 1080i -- honestly, I don't know. I did say "etc". I would imagine sports (action) on 720p is superior, but I just work with canned streams. And the products support all formats. At home, my set is basically left at 1080i -- why? -- because I honestly can't tell the difference, and no one else in my family can either, especially because the highest resolution input we have is a non-progressive scan Sony DVD player. In the sense that I do drivers for HDTV, I guess I am a "professional" of sorts; my opinion is -- I don't think it matters.
MPEG-2 IBBP GOP sequence? Not given. depends on the encoder. To save on broadcast space, there is a push for "...BBBBBBBBBB..." sequences. Of course, that leads to artifacts on transmission errors (which wouldn't be an issue for PS streams, so why not?). Of course, errors are a fact of life with sat, etc. But, the GOP is variable (depends on the encoder and the program).
Ratboy
Interesting - I just want to expand on your "fleshy" point:
Modern TVs use a colour gamut designed to improved flesh tones, yet they have a smaller gamut than the original colour specification. In other words, an antique original colour TV is able to represent MORE colours than your current set.. It just won't do hiqh quality pr0n as well.
As to "near-perfect" signal quality and picture integrity... I would argue that "digital" mpeg encoding reduces quality. The mpeg encode of course relies on "picture integrity" (actually, no, everything is bundled up into 188 byte packets, with the assumption that there will be lossage, and no retransmission).
As to resolution - 480i has been "good enough". Indeed, DVDs are 480i/p as well. Generally, few complaints.
1080i (etc.) HD formats. ARE a major step. Roughly, an order of magnitude improvement. But, for many, 480i/p is "good enough" (please note that HD has 6 times the datarate of a current DVD - and DVD *is not* an HDTV format. The only source of HD will be broadcast (possibly cable or sat..). And, you won't actually be able to *record* an HD signal using normal consumer gear).
And, I find that 480i/p is good enough for me. I do have a largish set, and still don't really have the urge for HDTV. If we had "super-DVD" out there, with 1080i format movies, THEN I would for it. But, I honestly don't care for broadcast formats.
That's probably just me, though...
Ratboy.
"Bearing in mind the fact that it is impossible to switch to Linux at home for a number of reasons and also that because of the business I need to be online 24/7/365 what the Slashdot community suggest"
So you need Windows. Which is ok -- put Linux on another box, and secure it. I just bought a Compaq with 128MB of memory, 20GB or so hard drive, 400Mhz processor for 100$ CDN (80$ US or so). Used.
Something like that would make a good firewall for you.
Alternately, home routers also have reasonably firewalling. My SMC Barricade (gasp, yet, I know that a REAL geek wouldn't use one) offers the ability to drop in-bound traffic, and only allow certain ports through. This can provide you 80% of what you need (it does for me). Staying on top of patches can bring you the rest of the way. Just don't enable the "DMZ" feature!
As you mentioned, you have external hosting -- which means that you don't have to allow incoming HTTP, or SMTP. If you don't need to administer externally (and since you use Windows, you *probably* don't), you don't need port 22. So, close off ALL inbound connections. Just leaves you with FTP as an issue -- some router boxes will accomodate, or you can learn to love the PASV command (and, AFAIK, MS browser FTP does that automagically).
If you AREN'T using a small home router, GET ONE. They are even cheaper (I have seen brand new units selling here for $20 CDN, approx. $15 US).
Don't forget a good backup plan, just in case you get rooted (or other disaster strikes).
Still, buying a cheap box or two is reasonable. One for a "real" firewall, and another for SAMBA, and other internal services (DNS).
Ratboy
LyX will do all that... really.
Not that it matters... If it works for you, keep on trucking with it!
Ratboy.
Bel:
Some observations...
"Users are not experts"
Actually, they are. May not be expert with computers, but you should allow for expertise in the problem area.
"burden on programmers to take responsibility of policing bad behavior"
No. If this is done, then functionality that the programmer has not explicitly thought of is lost. To give you an example -- I needed to introduce a LOT of Postscript to an Apple LaserWriter (the original, with only a megabyte of memory). The code could NOT fit into memory -- yet I was able to tell the interpreter to do a "stupid" thing -- execute the serial port. Saved the day, and it worked fine. I don't think that the programmer had this in mind at all...
As another example: If you allow "printing" to a file, DON'T check to make utterly sure its a file -- it may be a pipe to another program. I guess you could check, at the loss of an enormous amount of flexibility.
Also, users should be encouraged to have "multiple" views of data. An example would be equations. One view is as a typeset equation. Working with that representation is *extremly* tedious. A better model for editing would be a formula view. Or, say, pictures. "Circle, 1 cm diameter". vs. something hand-drawn (and at what resolution?).
Symbolic reasoning is a higher level of abstraction than concrete reasoning. Most of the output produced by the user will be concrete (it will be a tangible, in the sense that it can be experienced). Just don't make the mistake of forcing the user to work in the concrete domain all the time.
Ratboy
I run a DNS server for my little LAN. Internally, machines are added to a ".lan" domain, because they are NAT'd anyway. Can't be accessed from outside, so they don't get the external domain name. Traffic inbound goes to the correct machines.
In the past, I had a guarantee that ".lan" would not be a valid TLD. It can't be a country code, and it's not a standard TLD.
So... now the assurance is going to go away. The price of progress, I guess. I do hope that some TLDs remain reserved for internal use (same as the unroutable IP addresses.)
Of course we need only one such private TLD (I sure hope it's ".lan" because that's easiest for *me*).
Maybe I should go and read the article now...
Ratboy.
In a word - YES.
But note that the "somebody" who DOWNLOADED the song was NOT "gone after".
The party doing the publishing was brought down -- not the people who used the result of the publication.
It IS legal to download ROMs. May or may not be legal to copy them, and provide them on electronic sites (yes, legal, if copyright has been granted, else no).
I truly would like to see the fireworks if a DOWNLOADER was prosecuted. It would open a nasty can of worms.
And that leaves (1) emulator and (2) the legality of making a copy to run on the emulator.
An emulator is perfectly legal. There may be patent encumberance though... and this is what Nintendo seems to be claiming.
If the emulator is legal, it may be considered a NCESSARY step to copy a ROM to run it. That is allowed in my jurisdiction (Canada) and, I beleive US as well. As well, backup copy can be made.
And there is no difference between material provided on-line (downloaded) and an actual ROM.
YOU CAN'T GET INTO TROUBLE FOR DOWNLOADING. Sorry for the harshness, but it seems that many would have you believe otherwise. There are, of course, exceptions to this rule. In Canada, it is illegal to download child porn, and certain other categories of material. Please note that this has nothing to do with copyright.
YOU CAN GET INTO TROUBLE FOR COPYING OR "UPLOADING". Big time.
Enjoy your Emulation
Ratboy.
If someone places music on-line, there are two possibilities -- the first is that the person has permission to do so (copyright), the second is that the person does not.
If the person *has* permission, then you have the right to download and burn. The thing that you are copying *is* the original.
If the person does NOT have permission, they have broken copyright. Not you... As an example -- if there is a legal music download service, and they have a single illegal track (no copyright granted), the people who download that track are not liable. Only the provider. Big time.
Now -- if you download music from sites that are not located in Canada, it becomes murkier. Are the download sites even part of the Berne Copyright agreement? If so, the Canadian personal copying provision still holds. It is, of course, impossible to tell if the source is legal or not. One presumes that sources that DO NOT have copyright have been or will be shut down by actions of the copyright holder.
All CDs are "copies" to begin with. There is no such thing, effectively, as an "original" (there is the original performance, of course, and there is a part of the Canadian Copyright Act that deals with that).
If the user is ALSO responsible, we have much larger problems. Example -- a teacher copies some pages out of a textbook. The students are later fined for using the material. You would have to verify the copyright of every item that you buy or are given.
Reading this way -- leach away, my Canadian friends!
Ratboy