If you link GPL-licensed code in with your project, poof! Your project is now GPL-licensed as well, for better or for worse. Some people will argue it's better, some worse, but all agree that it's viral.
You're missing the point.
First of all, what you say is false. Linking GPL code in with your project does not compel your project to be GPL, unless you distribute the resulting project. It is amazing how often incorrect myths like your statement are perpetuated.
Second of all, you conveniently neglect to mention the fact that if you link MS code in with your project and distribute it, you go to jail.
I would much much much rather work with GPL code than MS code.
to this day almost all advances in pure math are made by single people working alone.
No way. Do you have a research degree in mathematics (e.g. Ph.D.)? If not, then you're not in a position to know how research mathematics is done.
I would certainly agree that pure math is more amenable to solo progress than any other science, but to say "almost all advances" are done solo is going way too far.
These days, even if you work alone, you are still utterly dependent on conferences, seminars, and publications by others in the community. No mathematician can get far today without other people helping. If nothing else, you need to know what others are doing so that you do not duplicate their work.
Now that they are making enterprise level IDE drives your statement is no longer correct.
You mean, when enterprise-level IDE drives become available for purchase, then my statement will no longer be correct. Right now you can't get one of these nice IDE drives unless you're a hardware review site.
In any case, the enterprise-level IDE drives cost just about as much to make as enterprise-level SCSI drives, so I will be very interested in seeing how aggressively the manufacturers compete on price with these drives. Which, of course, won't be clear until they're actually available.
Why would SCSI be less prone to heat and wear than IDE?
I don't know why SCSI drives have less heat than IDE drives, but they just do.
I have a web server with multiple 10k rpm SCSI drives inside it. Zero case fans. Zero drive fans. Approximately 1mm of vertical clearance between the drives.
If you try doing this with high performance IDE drives, even mere 7200 rpm drives, you could expect a quick heat death, but my SCSI drives in this server never get any hotter than warm to the touch.
SCSI has a lot more advantages over IDE than mere statistics and benchmarks can indicate.
Nonlinear history traversal -- if you mean jumping back more than one page at a time, that's been a feature since Netscape 4.0 at least. Or do you mean something else?
Sorry, I didn't really say what I meant here. What I meant to say was nonlinear site navigation, which includes of course nonlinear history navigation, but much much more.
Nonlinear means you are not limited to going backwards and forwards. I do not mean just jumping back more than one page at a time. I mean that you should be able to (among other things) view your entire browsing history as a tree and jump to any branch of it, even branches off the main line. Some of this (but not all) can be achieved in Mozilla using the History window (Ctrl-H).
I also would like to see IE adopt innovations such as site navigation tabs (as in this example). Navigation tabs are much more useful than mere navigation links, which are often off-screen or on a different page from the page you're viewing.
Well now, I'm not going to chew out this article entirely, because most of it is an interesting read, but there are some places where the author is just plain ignorant. For example:
Not yet under monopoly control, Internet technologies are a seething hotbed of innovation - with one exception. Once Microsoft's Internet Explorer reached 70% market share, the once blazing hot evolution of Web browsers came to a sudden screeching halt.
The only insight this quote reveals is that the author uses Internet Explorer. It would be more accurate to say that "the once blazing hot evolution of Internet Explorer came to a sudden screeching halt."
In almost every web browser other than Internet Explorer, innovation is alive and well -- we have popup blocking, ad blocking, tabbed browsing, fine grained scripting control, and nonlinear history traversal, all of which are genuinely valuable innovations that Internet Explorer would do well to adopt.
When it's called PGP it's good, when they call it Microsoft Something Something it's bad?
You're really missing the entire point if you think PGP is anything like DRM.
PGP is designed to keep something secret when both the sender and the recipient want to keep it secret.
DRM is designed to keep something secret when only the sender but not the recipient wants to keep it secret.
The first is a relatively easy problem with a good solution. The second is a completely impossible problem whose attempted solution will nevertheless cause a lot of grief to society as circumvention tools like digital cameras and copy machines get banned.
I fully agree with your general point, but your interpretation of MTBF is wrong.
MTBF is actually the mean time between non-age-related drive failures. For a drive to have a MTBF of 150000 hours means that if you took 150 drives and ran them for 1000 hours each, you would expect one drive failure.
The MTBF is not intended to indicate how long an individual drive is expected to last before failure -- for that, you need to look up the service life or (for the more cynical among us) the warranty length.
I hate one year drive warranties as much as you do, but MTBF has nothing to do with it.
While 3+GHz might be overkill right now, is it overkill few years down the road?
In most cases, it is far more cost effective to buy what you need for the moment, and upgrade later after
prices have dropped,
you actually need the upgrade.
Here's one example: in 1998 I bought a Pentium II 350 at about the same time my friend bought a Xeon 450. Four years later, I upgraded to a Pentium III 1.2GHz, while his Xeon slot doesn't support any cpu upgrade options. Today my system smokes his system for way less money.
Of course, if you actually need the Xeon right now then you should get it, but buying in anticipation of future needs makes no sense to me when prices are falling so fast.
You are absolutely correct that Linux is likely to remain a niche player for the forseeable future. What you don't seem to understand, however, is that Linux can thrive as a niche player.
A lot of people do not realize how revolutionary that fact is. It is the critical difference between free software and everything that came before it.
Most people automatically assume that a platform has to be popular in order to thrive. This assumption is correct for commercial platforms (which require corporate revenues, and thus end users, to survive). However, it is false for open source platforms such as Linux. Even if Linux attracts no end user support whatsoever, it will continue to do well as long as it can keep the users it already has right now, most of whom are extraordinarily talented developers with both the means and the desire to continue contributing to its progress.
The only way Linux will ever die is if laws like the SSSCA are passed to make it illegal.
Are there really any tolerable programs for Linux that'd do wav-based subtitling?
Since you asked...
My subtitling guide describes how I (used to) subtitle in linux and my timing guide describes how I do wav timing.
Both of these documents are unfortunately old, and I have not had the time to update them to reflect recent improvements. For the subtitling engine I have switched to using transcode instead of mplayer, because transcode has all the features I described in the previous post, including inverse pulldown which is very important for anime! (Probably not important for finnish TV, since pulldown is an NTSC-only abomination.)
For wav timing, I still use the same basic input setup as described on my timing page, but I have written a graphical interface which allows somewhat sane editing of the script without *having* to know SQL. A preliminary version can be seen in this screenshot. I plan to release this software as well, probably at the same time I update the guide.
At the present time my Ph.D thesis takes priority, so expect to wait a few months before I get back into fansubbing mode.
But freedom to do anything is nothing without tools to actually do something... video tools Just Aren't There Yet!
Part of what I mean by freedom is having the ability to easily create your tools where none existed before.
For most people, I agree, video tools in Linux are inadequate. In my case, for what I want to do (anime fansubbing), the existing tools are barely inadequate, and I have patched them up to be adequate.
I can do wav timing in Linux using an SQL database for script storage, giving me much more flexibility in script manipulation than any Windows or MacOS program I know. I can render arbitrarily many subtitle streams in transcode, with smooth scrolling, fades, and even color deltas. I can get almost perfect inverse pulldown (perfect is impossible, because of the way different sprites have different pulldown patterns) in transcode using custom distance metrics for frame decimation.
Everything I describe above required having access to the source code. The last two could be done in VirtualDub if you have a C compiler, but I don't know any Windows way to achieve the first short of writing a new program from scratch.
Did all of this take a lot of time? Sure it did. But anime fansubbing is a time consuming endeavor anyway. The time spent patching the tools is a drop in the bucket compared to the time spent subtitling. I'm glad I took the time to get it right.
I would have preferred to have kept it at the level of a civilized discussion instead of descending into base insults, as you have done.
Linux usability is terrible (I agree with this point, I will agree a thousand times over), but MacOS isn't such a usability paradise either. Just to name one thing: redhat gives me four virtual desktops out of the box; why oh why doesn't MacOS do that? I find one desktop unusable.
You're deluding yourself, and your blindness is only making your work/hobby/whatever miserable for you. Everything you listed can even be done on Windows, with the correct software installed.
Windows is even worse than MacOS. Please don't go there.
If "miserable" means having the full source code to everything I use, then I'll take misery any day. I could write a thesis on how many times having the source code has made my work easier, except that I already have another thesis that I have to write first.
Mac OS X is so amazingly superior to Linux in all of these fashions, I am constantly astounded when I encounter people such as yourselves who so blindly follow the belief "Linux equals freedom, and everything else is a tool of oppression." Get off your high horses and join the 21st century.
You think I'm blind? Wait until Stallman's right to read comes true (we're halfway there already), and then we'll see who's blind.
I am astounded that you would be so venomously antagonistic towards someone else who chooses to use Linux. It's not like I'm forcing you to use it. It's not like I'm saying it has no flaws. I simply have a different set of priorities that cause me to prefer Linux over MacOS.
"Free software" exists on more than one platform, for your information.
A platform such as MacOS where the OS is not free software does not do what I want.
To tell the truth, the issue goes beyond DVD editing. I'm not a stellar programmer by any means, but I do want the source code, so that I can do things like
patch my kerne to use IPsec encryption with the AES cipher
write an input plugin for transcode to handle subsonic format anime scripts
use glame for timing dialogue, exporting the resulting script directly into a mysql database
All of which are things that I have done, and that have required the source code.
Most people are better off with Mac OS than Linux, but there are some people for which it is the reverse, and I am one of them. I want the freedom to use Linux, and I certainly hope you're not suggesting I shouldn't have that freedom, because I'll defend it to the death.
Look, I'm not gonna sit here and say video work on linux is easy (it's not) or cutting edge (it's not). Everyone here knows that jwz despite the harsh ranting is basically right.
But I still use linux for manipulating videos. Why? Because freedom is priceless.
You rave about your eight year old cousin doing wonderful things on his iMac. But I bet your eight year old cousin would be stopped dead in his tracks the moment he has to play an out-of-region DVD.
Playing out of region DVDs is not just a theoretical problem for me: I have almost as many region 2 DVDs (forty) as region 1 DVDs (fifty). As great as the iMac is, I can't use it because it restricts me to one region.
Now, it is true that there probably exist underground firmware flashes to circumvent the region restrictions, but the principle at stake goes way beyond this simple technical problem. I want a computer that does what I want, not what Hollywood wants, and to achieve this goal I have to use free software.
I want to be able to:
copy commercial DVDs for fair use purposes
in particular, add my own subtitles to DVDs for my own private viewing pleasure, and store copies of the results
do so on any DVD without regard to what region it was sold in
override fast-forward lockouts and menu button lockouts
Your facts are all correct, but what you are forgetting in this instance is that free software, unlike commercial proprietary software, does not need corporate financial support in order to progress and improve.
A commercially backed OS like MacOS or OS/2 or NextStep will die without corporate success, but Linux already has more developers working for it for free than MacOS or OS/2 or NextStep ever did for pay.
The only way Linux and its free software friends will ever die is if laws like the SSSCA are passed to make it illegal.
not many websites cater to the deal-seekers that use SCSI.
Ignoring for the moment the fact that the phrase "deal-seekers that use SCSI" is practically an oxymoron...
Surplus Computers has as good a deal as any you'll ever find on 10k rpm SCSI drives. I've had two of the 36.7 GB drives for six months and they've been nothing but a pleasure to use.
I have no relationship to Surplus Computers other than that of satisfied customer.
As for quality, I have to admit, it's not on par with tmpgenc yet. If you are working with untelecined 24 fps film source then Linux can produce marginally acceptable SVCDs. However, 30 fps TV source NTSC will look really bad at SVCD bitrates unless it is one of the rare such streams that detelecines perfectly. For these you will have to bump up to DVD bitrates to get decent quality. See LG83 for how to author DVDs in Linux.
Avisynth has also been useful for various NLE and filtering tasks...is something similar available for Linux?
Okay, here's the beauty of Linux. You don't need it. If you simply want to frameserve an AVI, a named pipe (man mkfifo) will do just fine. If you want to do fancy stuff like overlay two AVIs, check out the subtitler plugin in the transcode software I mentioned above, which can do overlays, fades, and scrolling of many types of objects including text, pictures, and video.
scsi vs. ide: from someone who knows
on
IDE RAID Examined
·
· Score: 3, Informative
I can easily tell that 90% of the people spouting off here have never used both modern SCSI and modern IDE. Well, I have actually used both. So take it from someone who knows.
There are valid performance and reliability reasons for using SCSI drives instead of IDE drives; the question is whether these gains are worth the cost, not whether they are there at all.
Reasons why SCSI might be worth it:
Spin rate. Until IDE drives gain 10k and 15k spin rates, SCSI drives will always be king in multitasking and random-access situations. 3ms seek time is so much better than 10ms that you have to use it to believe the difference.
Reliability. IDE drives have one year or at best three year warranties. SCSI drives have five year warranties. You can run modern 15k scsi drives stacked next to each other with zero additional case fans and expect to outlast your warranty. Try that with IDE.
Hot swap. Does anyone here know of a hot-swap IDE raid solution? I think not.
Tagged command queuing. A SCSI drive can collect multiple drive requests and reorder them to optimize the actual physical retrieval of the bits in question. IDE drives, even if the box lists this feature, have never done TCQ particularly well. This kind of thing is impossible to benchmark because its benefits only show up under heavy multitasking, not single-tasking benchmarks.
For most people, I would agree that you would be better off buying 2GB ram or two CPUs before spending money on SCSI. However, if you already have 2GB ram and two CPUs, and you still need more, then that's when you should look into high end SCSI.
You can have two products with the same names but in different markets without having a trademark problem.
This may be true on paper (just like, on paper, laws have to be constitutional), but it is certainly not the way trademark law is actually being used today.
For example VISA the credit card company is currently (successfully) suppressing the web site evisa.com even though the latter is in a totally different market than the market for credit cards. Also note another similarity to this case: visa is a common english word that existed long before the credit card company obtained its trademark.
But now that it's clear what the [DVD] license holders are demanding
The problem is that it is NOT clear.
When I buy a DVD, I am not presented with a contract from the license holders that I have to sign.
It has never been illegal to use DeCSS or libdvdcss to play DVDs in most of Europe or Australia or indeed anywhere in the world except the United States. Until that changes, I think people in the United States have a legitimate gripe.
If you don't want me to use free software to play DVDs, then say so in a contract and give me the choice of whether or not to sign it. Buying a special-interest law like the DMCA is, frankly, more sickening than anything that anyone on the free software side has ever done.
Of course, Red Hat and Mandrake's solutions aren't much better.
This blanket statement is unjustified.
Fetching the public key from the same server as the packages is useless. But Redhat and Mandrake don't do it that way. With Redhat and Mandrake the public key is available on the physical installation media (i.e. retail CDs). Trojaning a retail CD takes a lot more work than trojaning a public key on a remote server.
I personally think the danger from trojans is much less widespread than you and others think, but that's a subject for another day. It certainly is no excuse for not updating openssl, since the old openssl already has a known confirmed remote root hole.
I haven't yet found a way of persuading Linux of this fact, I would prefer the kernel to use the lower eight megs preferentially.
This is actually possible to do in linux 2.4. Believe me, I am an expert on this kind of thing, as my situation is similar -- the Intel pentium chipsets only cache the lower 64MB of RAM, so on systems with more RAM you would like to tell linux to use the lower RAM preferentially.
The secret is the slram.o module in the MTD section of the linux kernel. See here for the full scoop. This module doesn't solve everything; in particular, linux doesn't disk cache as effectively when slow ram is in use. You'll just have to see for yourself whether it's worth it or not.
Slashdot Mirror
You're missing the point.
First of all, what you say is false. Linking GPL code in with your project does not compel your project to be GPL, unless you distribute the resulting project. It is amazing how often incorrect myths like your statement are perpetuated.
Second of all, you conveniently neglect to mention the fact that if you link MS code in with your project and distribute it, you go to jail.
I would much much much rather work with GPL code than MS code.
320 megabytes is about 2.5 gigabits ... which is a lot closer to 3 gigabits than the erroneous 320 megabits figure.
No way. Do you have a research degree in mathematics (e.g. Ph.D.)? If not, then you're not in a position to know how research mathematics is done.
I would certainly agree that pure math is more amenable to solo progress than any other science, but to say "almost all advances" are done solo is going way too far.
These days, even if you work alone, you are still utterly dependent on conferences, seminars, and publications by others in the community. No mathematician can get far today without other people helping. If nothing else, you need to know what others are doing so that you do not duplicate their work.
You mean, when enterprise-level IDE drives become available for purchase, then my statement will no longer be correct. Right now you can't get one of these nice IDE drives unless you're a hardware review site.
In any case, the enterprise-level IDE drives cost just about as much to make as enterprise-level SCSI drives, so I will be very interested in seeing how aggressively the manufacturers compete on price with these drives. Which, of course, won't be clear until they're actually available.
Wait and see!
Agreed. I am saying that SCSI drives are superior in build to IDE drives, not that the interface difference is responsible for the superior build.
I don't know why SCSI drives have less heat than IDE drives, but they just do.
I have a web server with multiple 10k rpm SCSI drives inside it. Zero case fans. Zero drive fans. Approximately 1mm of vertical clearance between the drives.
If you try doing this with high performance IDE drives, even mere 7200 rpm drives, you could expect a quick heat death, but my SCSI drives in this server never get any hotter than warm to the touch.
SCSI has a lot more advantages over IDE than mere statistics and benchmarks can indicate.
If it's so damn easy, why won't Microsoft include it?
Sorry, I didn't really say what I meant here. What I meant to say was nonlinear site navigation, which includes of course nonlinear history navigation, but much much more.
Nonlinear means you are not limited to going backwards and forwards. I do not mean just jumping back more than one page at a time. I mean that you should be able to (among other things) view your entire browsing history as a tree and jump to any branch of it, even branches off the main line. Some of this (but not all) can be achieved in Mozilla using the History window (Ctrl-H).
I also would like to see IE adopt innovations such as site navigation tabs (as in this example). Navigation tabs are much more useful than mere navigation links, which are often off-screen or on a different page from the page you're viewing.
In almost every web browser other than Internet Explorer, innovation is alive and well -- we have popup blocking, ad blocking, tabbed browsing, fine grained scripting control, and nonlinear history traversal, all of which are genuinely valuable innovations that Internet Explorer would do well to adopt.
You're really missing the entire point if you think PGP is anything like DRM.
PGP is designed to keep something secret when both the sender and the recipient want to keep it secret.
DRM is designed to keep something secret when only the sender but not the recipient wants to keep it secret.
The first is a relatively easy problem with a good solution. The second is a completely impossible problem whose attempted solution will nevertheless cause a lot of grief to society as circumvention tools like digital cameras and copy machines get banned.
MTBF is actually the mean time between non-age-related drive failures. For a drive to have a MTBF of 150000 hours means that if you took 150 drives and ran them for 1000 hours each, you would expect one drive failure.
The MTBF is not intended to indicate how long an individual drive is expected to last before failure -- for that, you need to look up the service life or (for the more cynical among us) the warranty length.
I hate one year drive warranties as much as you do, but MTBF has nothing to do with it.
In most cases, it is far more cost effective to buy what you need for the moment, and upgrade later after
- prices have dropped,
- you actually need the upgrade.
Here's one example: in 1998 I bought a Pentium II 350 at about the same time my friend bought a Xeon 450. Four years later, I upgraded to a Pentium III 1.2GHz, while his Xeon slot doesn't support any cpu upgrade options. Today my system smokes his system for way less money.Of course, if you actually need the Xeon right now then you should get it, but buying in anticipation of future needs makes no sense to me when prices are falling so fast.
You are absolutely correct that Linux is likely to remain a niche player for the forseeable future. What you don't seem to understand, however, is that Linux can thrive as a niche player.
A lot of people do not realize how revolutionary that fact is. It is the critical difference between free software and everything that came before it.
Most people automatically assume that a platform has to be popular in order to thrive. This assumption is correct for commercial platforms (which require corporate revenues, and thus end users, to survive). However, it is false for open source platforms such as Linux. Even if Linux attracts no end user support whatsoever, it will continue to do well as long as it can keep the users it already has right now, most of whom are extraordinarily talented developers with both the means and the desire to continue contributing to its progress.
The only way Linux will ever die is if laws like the SSSCA are passed to make it illegal.
Since you asked...
My subtitling guide describes how I (used to) subtitle in linux and my timing guide describes how I do wav timing.
Both of these documents are unfortunately old, and I have not had the time to update them to reflect recent improvements. For the subtitling engine I have switched to using transcode instead of mplayer, because transcode has all the features I described in the previous post, including inverse pulldown which is very important for anime! (Probably not important for finnish TV, since pulldown is an NTSC-only abomination.)
For wav timing, I still use the same basic input setup as described on my timing page, but I have written a graphical interface which allows somewhat sane editing of the script without *having* to know SQL. A preliminary version can be seen in this screenshot. I plan to release this software as well, probably at the same time I update the guide.
At the present time my Ph.D thesis takes priority, so expect to wait a few months before I get back into fansubbing mode.
Part of what I mean by freedom is having the ability to easily create your tools where none existed before.
For most people, I agree, video tools in Linux are inadequate. In my case, for what I want to do (anime fansubbing), the existing tools are barely inadequate, and I have patched them up to be adequate.
I can do wav timing in Linux using an SQL database for script storage, giving me much more flexibility in script manipulation than any Windows or MacOS program I know. I can render arbitrarily many subtitle streams in transcode, with smooth scrolling, fades, and even color deltas. I can get almost perfect inverse pulldown (perfect is impossible, because of the way different sprites have different pulldown patterns) in transcode using custom distance metrics for frame decimation.
Everything I describe above required having access to the source code. The last two could be done in VirtualDub if you have a C compiler, but I don't know any Windows way to achieve the first short of writing a new program from scratch.
Did all of this take a lot of time? Sure it did. But anime fansubbing is a time consuming endeavor anyway. The time spent patching the tools is a drop in the bucket compared to the time spent subtitling. I'm glad I took the time to get it right.
Linux usability is terrible (I agree with this point, I will agree a thousand times over), but MacOS isn't such a usability paradise either. Just to name one thing: redhat gives me four virtual desktops out of the box; why oh why doesn't MacOS do that? I find one desktop unusable.
You're deluding yourself, and your blindness is only making your work/hobby/whatever miserable for you. Everything you listed can even be done on Windows, with the correct software installed.
Windows is even worse than MacOS. Please don't go there.
If "miserable" means having the full source code to everything I use, then I'll take misery any day. I could write a thesis on how many times having the source code has made my work easier, except that I already have another thesis that I have to write first.
Mac OS X is so amazingly superior to Linux in all of these fashions, I am constantly astounded when I encounter people such as yourselves who so blindly follow the belief "Linux equals freedom, and everything else is a tool of oppression." Get off your high horses and join the 21st century.
You think I'm blind? Wait until Stallman's right to read comes true (we're halfway there already), and then we'll see who's blind.
I am astounded that you would be so venomously antagonistic towards someone else who chooses to use Linux. It's not like I'm forcing you to use it. It's not like I'm saying it has no flaws. I simply have a different set of priorities that cause me to prefer Linux over MacOS.
"Free software" exists on more than one platform, for your information.
A platform such as MacOS where the OS is not free software does not do what I want.
To tell the truth, the issue goes beyond DVD editing. I'm not a stellar programmer by any means, but I do want the source code, so that I can do things like
- patch my kerne to use IPsec encryption with the AES cipher
- write an input plugin for transcode to handle subsonic format anime scripts
- use glame for timing dialogue, exporting the resulting script directly into a mysql database
All of which are things that I have done, and that have required the source code.Most people are better off with Mac OS than Linux, but there are some people for which it is the reverse, and I am one of them. I want the freedom to use Linux, and I certainly hope you're not suggesting I shouldn't have that freedom, because I'll defend it to the death.
But I still use linux for manipulating videos. Why? Because freedom is priceless.
You rave about your eight year old cousin doing wonderful things on his iMac. But I bet your eight year old cousin would be stopped dead in his tracks the moment he has to play an out-of-region DVD.
Playing out of region DVDs is not just a theoretical problem for me: I have almost as many region 2 DVDs (forty) as region 1 DVDs (fifty). As great as the iMac is, I can't use it because it restricts me to one region.
Now, it is true that there probably exist underground firmware flashes to circumvent the region restrictions, but the principle at stake goes way beyond this simple technical problem. I want a computer that does what I want, not what Hollywood wants, and to achieve this goal I have to use free software.
I want to be able to:
- copy commercial DVDs for fair use purposes
- in particular, add my own subtitles to DVDs for my own private viewing pleasure, and store copies of the results
- do so on any DVD without regard to what region it was sold in
- override fast-forward lockouts and menu button lockouts
and to do all that, I need free software.A commercially backed OS like MacOS or OS/2 or NextStep will die without corporate success, but Linux already has more developers working for it for free than MacOS or OS/2 or NextStep ever did for pay.
The only way Linux and its free software friends will ever die is if laws like the SSSCA are passed to make it illegal.
Ignoring for the moment the fact that the phrase "deal-seekers that use SCSI" is practically an oxymoron...
Surplus Computers has as good a deal as any you'll ever find on 10k rpm SCSI drives. I've had two of the 36.7 GB drives for six months and they've been nothing but a pleasure to use.
I have no relationship to Surplus Computers other than that of satisfied customer.
I wrote the Linux Digital Fansubbing Guide. I have a section in there on SVCDs. So I know a thing or two about making SVCDs in Linux.
Here's a couple of Linux programs that can encode mpeg-2:
- mjpegtools
- transcode (see here for its SVCD documentation section)
As for quality, I have to admit, it's not on par with tmpgenc yet. If you are working with untelecined 24 fps film source then Linux can produce marginally acceptable SVCDs. However, 30 fps TV source NTSC will look really bad at SVCD bitrates unless it is one of the rare such streams that detelecines perfectly. For these you will have to bump up to DVD bitrates to get decent quality. See LG83 for how to author DVDs in Linux.Avisynth has also been useful for various NLE and filtering tasks...is something similar available for Linux?
Okay, here's the beauty of Linux. You don't need it. If you simply want to frameserve an AVI, a named pipe (man mkfifo) will do just fine. If you want to do fancy stuff like overlay two AVIs, check out the subtitler plugin in the transcode software I mentioned above, which can do overlays, fades, and scrolling of many types of objects including text, pictures, and video.
There are valid performance and reliability reasons for using SCSI drives instead of IDE drives; the question is whether these gains are worth the cost, not whether they are there at all.
Reasons why SCSI might be worth it:
- Spin rate. Until IDE drives gain 10k and 15k spin rates, SCSI drives will always be king in multitasking and random-access situations. 3ms seek time is so much better than 10ms that you have to use it to believe the difference.
- Reliability. IDE drives have one year or at best three year warranties. SCSI drives have five year warranties. You can run modern 15k scsi drives stacked next to each other with zero additional case fans and expect to outlast your warranty. Try that with IDE.
- Hot swap. Does anyone here know of a hot-swap IDE raid solution? I think not.
- Tagged command queuing. A SCSI drive can collect multiple drive requests and reorder them to optimize the actual physical retrieval of the bits in question. IDE drives, even if the box lists this feature, have never done TCQ particularly well. This kind of thing is impossible to benchmark because its benefits only show up under heavy multitasking, not single-tasking benchmarks.
For most people, I would agree that you would be better off buying 2GB ram or two CPUs before spending money on SCSI. However, if you already have 2GB ram and two CPUs, and you still need more, then that's when you should look into high end SCSI.This may be true on paper (just like, on paper, laws have to be constitutional), but it is certainly not the way trademark law is actually being used today.
For example VISA the credit card company is currently (successfully) suppressing the web site evisa.com even though the latter is in a totally different market than the market for credit cards. Also note another similarity to this case: visa is a common english word that existed long before the credit card company obtained its trademark.
The problem is that it is NOT clear.
When I buy a DVD, I am not presented with a contract from the license holders that I have to sign.
It has never been illegal to use DeCSS or libdvdcss to play DVDs in most of Europe or Australia or indeed anywhere in the world except the United States. Until that changes, I think people in the United States have a legitimate gripe.
If you don't want me to use free software to play DVDs, then say so in a contract and give me the choice of whether or not to sign it. Buying a special-interest law like the DMCA is, frankly, more sickening than anything that anyone on the free software side has ever done.
This blanket statement is unjustified.
Fetching the public key from the same server as the packages is useless. But Redhat and Mandrake don't do it that way. With Redhat and Mandrake the public key is available on the physical installation media (i.e. retail CDs). Trojaning a retail CD takes a lot more work than trojaning a public key on a remote server.
I personally think the danger from trojans is much less widespread than you and others think, but that's a subject for another day. It certainly is no excuse for not updating openssl, since the old openssl already has a known confirmed remote root hole.
This is actually possible to do in linux 2.4. Believe me, I am an expert on this kind of thing, as my situation is similar -- the Intel pentium chipsets only cache the lower 64MB of RAM, so on systems with more RAM you would like to tell linux to use the lower RAM preferentially.
The secret is the slram.o module in the MTD section of the linux kernel. See here for the full scoop. This module doesn't solve everything; in particular, linux doesn't disk cache as effectively when slow ram is in use. You'll just have to see for yourself whether it's worth it or not.