"opening a database connection for every visitor on every page is suicide if you only have a single server.. and your site is featured on slashdot.."
Um, no.
Real database connection pools (say, JDBC) running through a real application server (say, Resin) instead of a compiled-into-Apache-interpreter (PHP), accessing a robust database (Oracle) can and will run at an extremely high speed even when heavily loaded via Slashdot, etc.
All of these "create your own database-driven site!" beginner books are really nothing more than terrible instructions about how to build half-assed applications that don't scale worth a damn. It's because of these "learn PHP and MySQL in 30 days!" pseudo-manuals that we even have the slashdot effect. If web application programmers were to take the time to learn real enterprise-level application coding, you wouldn't see absolutely false statements like "opening a database connection for every visitor on every page is suicide".
I'm sorry, but that's like saying catching exceptions in your JSP code is an advanced feature, and you don't really need to use it.
Commit/rollback is an essential component to any decent data management system. Until you are absolutely sure that your data is correct -- that is to say, until you have done all of your transactions on the page successfully -- you should never actually write data to your database. Without using commit/rollback, you are stuck with the haphazard method of trying to manage potentially disastrous records of data showing up in your db.
Your other quote: "And MySQL will blow the doors off of Oracle and other databases in terms of raw speed" is similarly incorrect: MySQL may be faster when you are dealing with a small amount of connections, but as soon as your application starts getting any amount of concurrent users, MySQL is famous for falling down rather rapidly as it strains to write its data to disk.
MySQL cannot scale reliably, period. Having two database systems act as a pool, under MySQL, is a crapshoot at best. Unless you like designing single points of failure into your web applications, stay away from MySQL.
Simply put, if you expect your web application to get any amount of decent traffic (say 100,000 pageviews+ per day), then MySQL is simply not an option. Oracle is simply the standard upon which others can only attempt to compare themselves to. MySQL may be fine for the low-end "check out my k00l dynamic site!!11!!" crowd, but for professional web applications, MySQL has a long, long ways to go.
This is a bad, bad idea. Sorenson-encoded video is already extremely compressed; by re-encoding it as DivX, you have to go through another round of compression, making the video even worse quality.
What you really want is to be able to import uncompressed video via Firewire (or DV-compressed video, like what the story mentions) and edit it from there.
"I can't wait until the day I plug a Firewire dv cam into a mandrake box, a dv cam icon pops up on the desktop and allows joe to edit away to his heart's content."
If you're willing to shell out $999 for an iBook, you can have this today. Cheers! Enjoy Gnome 2.2 (snicker)....
Remember, Quicktime is a platform for serving up media of various codecs. Whenever you play "Quicktime video" in all likelihood you're playing Sorenson-compressed video through the Quicktime platform. Sorenson Video is the codec in this instance.
Quit demanding that Apple should make you a movie player for free. If you want a high-quality Quicktime experience, buy a Mac. If you want to remain five years behind the times with regards to audio and video, stick with Linux.
I have news for you: you aren't editing (or "manipulating" as you call it) video. You're manipulating DVD VOB files. There is a huge difference.
"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."
Similarly, my eight-year-old cousin would get "stopped dead in his tracks" the moment he had to apt-get or rpm get any kind of package or driver required to even play a Quicktime movie. Get it through your head: LINUX USABILITY IS AN UTTER SHAM.
"Now, it is true that there probably exist underground firmware flashes to circumvent the region restrictions,"
You're absolutely correct. There are all sorts of DVD-duplicating, region-switching, etc software for the Mac, all of which will happily do what you want to do, in about one tenth the time that it takes for you to do the same thing on Linux, because you don't have to read a 50k README and compile ten different packages in order to get them to work!
"I want a computer that does what I want, not what Hollywood wants, and to achieve this goal I have to use free software."
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. I imagine it would take about 5 minutes to download and install and use the correct utilities under OS X.
Taking this back to the jwz rant, like everything else in desktop Linux, video manipulation has a huge learning curve because of the lack of standardization on the part of interface designers, driver coders, and overall system-management-utilities.
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.
"Free software" exists on more than one platform, for your information. OS X isn't just a collection of ten applications from Apple, you know-- it's a bad argument, but "anything you can do I can do better."
Couple of points, before you start feeling too good about yourself. (er, whoops, too late)
- First of all, Jamie was talking about just trying to play video. If he has to do any kind of configuration or compilation at all, he's done too much. Video playing is the easiest thing on the planet! Why is it so damned time-consuming under Linux?
For the rest of your self-praising argument,
- How long did it take you to compile and configure vcr, avilib support, transcode, mpeg2enc, and all the various patches that are required to get your video to record?
- "Also I create videos with a (...) camcorder (...) and Kino (...)"
Congratulations. Kino provides the same functionality that even the shittiest NLE, Adobe Premiere, had in version 1.0 (in... 1994?). Way to be on the cutting edge there.
From the Kino site: "It does not support multiple layers or tracks of video and audio." Huh? How can it be an NLE if it doesn't do more than one track? Have you ever sit in front of an Avid bay and done any kind of real video editing? Because I have news for you: Kino is to a NLE as a Kia is to a Ferrari.
- "I then save to mpeg2 and encode 9kbit video (...)"
I'm going to assume you meant 9Mbit video, because 9kbit video is like looking at an old, worn-out three-quarter-inch reel-to-reel tape from the 1950s. Still, though, let's review: you're shooting with a single-chip camera, importing as a lossy format, editing with a one-track editor, and exporting as a lossy format. Again, way to be on the cutting edge.
...all to do the most basic of tasks, record video from line in and encode to mpeg2?
What you're doing can be done by my eight-year-old cousin on his iMac, using iMovie and iDVD, which (last I checked) doesn't take any time to install, because they come with OS X. And I'd bet the quality of his resulting video is completely superlative to yours, because the tools he's using are actually modern software (where the engineers have spent more than five minutes on the interface).
Jamie needs to bite the bullet and spend the cash for a good Powerbook or G4 tower. Linux on the desktop is dead. It will never get to the level that OS X is currently at. Face the facts.
Wouldn't one of the reasons that iCommune got pulled is because Apple is probably building in Rendezvous support for iTunes into iTunes 4? They don't want to be beaten to the punch, and a third party offering "Rendezvous-like" functionality goes against Apple's plan.
I suppose my requirements for a content management system are different than yours. I have yet to see an XSL-based solution able to expire content. Until I can see a system that reliably expires content without some sort of song-and-dance with 0-length files, cron jobs, and the like, I'll stick with "update article set available = 0 where id = blah" and run a dynamic site.
That "breaking news mode" that CNN busts into where their front page is static only happens at a threshhold of 100,000 simultaneous requests. Anything underneath that and they remain in their "regular news mode" of running a dynamic site.
Additionally, the only reason that they don't continue to run in "regular news mode" above that 100k threshhold is cost. Their server farm is perfectly capable of handling 10x the requests, but CNN has mandated that they aren't willing to spend the money in maintenance costs.
Your example of the charts on finance.yahoo.com is a bad example; "pre-generated files" is a misnomer-- Yahoo is simply telling their application servers to cache that information for a few minutes. This is, once again, a great example as to why a well-tuned application server is far superior to serving flat content.
XSL is not yet mature enough to replace good ol'-fashioned Java (that was a weak attempt at humor). It simply is too time-consuming to build templates to handle anything more than the simplest error-checking. XSL only has the most basic of variable support, and writing a "function" in XSL is somewhat akin to using an iPod as a PDA-- you can do it, but why not just use the real thing?
"The website is using dynamic content with PHP. Already we know the site isn't setup for high concurrency."
What exactly makes you think that? I'm not a huge PHP fan, but the site could be running any number of load balancers (software or hardware) in front of its backend. Additionally, PHP can scale to very high concurrency, it just needs to be tuned correctly. Your precious "flat-file-serving" Apache can only serve 150 simultaneous clients out of the box if you haven't tuned it.
"News flash: RDBMS do not scale."
Now I know you're talking out your ass, because if you had any experience at all with Oracle, you wouldn't have stated this. I suppose Amazon, Yahoo, CNN, et al aren't able to scale? The simple fact of the matter is that, like any other piece of software, how you tune your applications goes a long way to how many simultaneous threads you can serve up.
I think you really underestimate the true power of the Slashdot effect. What's likely happening in this situation is that Apache is buckling under the thousands of simultaneous requests, many of whom are coming from modem users -- using up precious processes to download the content. I seriously doubt that, if this person is running a 500MHz or greater machine, that processor time is an issue at all. What he/she needs to do is re-tune Apache to serve more simultaneous requests.
"Try generating your content to a static file and serving that."
That is one of the harder things to do, believe it or not -- you're venturing into the realm of a more fully-featured content management system at that point. What if your content changes on a regular basis? What if, like Slashdot, the comments are dynamic?
"Even Ye Olde Apache 1.3 can serve tens of thousands of static files per second."
Not out of the box it won't. Read your own httpd.conf and look up "MaxClients" and "KeepAliveTimeout". Out of the box, MaxClients is set to 150, and KeepAliveTimeout is set to 15 seconds. Once you hit the 150 MaxClient threshhold, the performance of your webserver will be terrible, because you'll be waiting 15 seconds between unique client requests.
I seriously question the article, not because of the benchmarks, but because of the scope by which the author defines a good editing station.
Firstly, you simply cannot edit using After Effects. Forget it. Your workflow is so amazingly hindered within the program. I will admit that it is probably the industry standard (for low- to medium-end stations) to do titling, chyron, graphics, etc, but to do day-to-day editing work, it is next to useless.
That said, the choices for editing software in the Windows environment are horrifically bad compared to the choices for the Mac. Other than the high-end Avid system, the Windows platform has absolutely nothing. Adobe Premiere is an atrocity that passes for software; instability, terrible interface, doesn't play well with others. Vegas Video is marginally better.
The Mac, on the other hand, has all sorts of quality hardware and software solutions. Take the Media100i system, for example. They just recently have ported the editing system to OS X. I have found that the Media 100 is the best mid-end editing station out there. Broadcast video, hardware codecs, plays well with a Beta SP deck or your firewire deck, etc.
Additionally, Final Cut Pro is rapidly becoming the standard for low-end stations. The USC film school is switching to an almost all-DV program, and the unofficial word is that students should go out and get FCP if they want to edit. It doesn't offer the speed that a Media100 station offers, but for an all-software solution, it blows the doors off anything Adobe or Sonic Foundry has ever made.
If these guys are so concerned about a $3500 Dell PC outperforming a $5500 Mac, perhaps they shouldn't be in the video editing business. I would rather spend the extra $2k, then spend an additional ~$5k for a good Medea RAID system, ~$5k for a Media 100 system, and be able to create broadcast video for $15k. (Nb: that is an almost unheard-of low cost of entry to the broadcast arena) Alternatively, if I were on a student's budget, I'd go for the $2500 Mac, a $999 (or cheaper for students, correct?) copy of Final Cut, and be safe in the knowledge that I was using a high-quality, reliable package, rather than spending $2000 on a PC and struggling with Premiere.
Does the book have a chapter on optimizing Tomcat's threads to provide better performance than the out-of-the-box installation? If not, then don't bother buying the book.
Instead, use the money to license a copy of Resin which is, for lack of a better description, Tomcat on Nitro. It follows the reference implementation of JSP and Servlets just as well as Tomcat does, and even the default configuration, which is tuned for development, outperforms Tomcat.
The configuration of Resin is almost exactly like the config of Tomcat, so I honestly don't see why you'd pick Tomcat over Resin (unless you were having trouble getting the 1.2 or 1.3 JDK installed on your FreeBSD box, something that is historically difficult to do).
IMHO, Corel has gotten itself into this rut because it has tried to create too much with too little.
Draw, Wordperfect, Office, etc etc. All the while they're creating ports of.Net to FreeBSD (that won't generate any revenue) and other various frivolous projects. This is a little bit like the plight of Sonic Foundry; getting into video and creating five different audio suites really dilutes the manpower to create great applications.
What Corel needs to do is concentrate on one product and make sure it's the best in the business. Go after Photoshop. Go after Office (well, on second thought, don't). But don't go after both at the same time.
In order to add great sound support to Debian, you have to:
- Make sure you have a 2.5.x kernel or above.
- Select your card from a dropdown.
- Retrieve, unpack, and compile source code.
- Install resulting software with a strange command-line utility.
- Retrieve, unpack, and install even more software.
- Edit a configuration file.
- Edit another configuration file.
- Run a script.
- Start a daemon.
Wow. That's so easy! I can see why OS X is the number one selling Unix:
- Go to System Preferences.
- Select "Sound".
- Select "Output".
- Select your high-end audio card.
- Select "Input".
- Select your high-end audio card.
Let's assume that your time is worth $50/hr. After a few hours of struggling to set up sound under Linux, that extra cash for Apple hardware doesn't sound so bad...
Slashdot Mirror
Um, no.
Real database connection pools (say, JDBC) running through a real application server (say, Resin) instead of a compiled-into-Apache-interpreter (PHP), accessing a robust database (Oracle) can and will run at an extremely high speed even when heavily loaded via Slashdot, etc.
All of these "create your own database-driven site!" beginner books are really nothing more than terrible instructions about how to build half-assed applications that don't scale worth a damn. It's because of these "learn PHP and MySQL in 30 days!" pseudo-manuals that we even have the slashdot effect. If web application programmers were to take the time to learn real enterprise-level application coding, you wouldn't see absolutely false statements like "opening a database connection for every visitor on every page is suicide".
Commit/rollback is an essential component to any decent data management system. Until you are absolutely sure that your data is correct -- that is to say, until you have done all of your transactions on the page successfully -- you should never actually write data to your database. Without using commit/rollback, you are stuck with the haphazard method of trying to manage potentially disastrous records of data showing up in your db.
Your other quote: "And MySQL will blow the doors off of Oracle and other databases in terms of raw speed" is similarly incorrect: MySQL may be faster when you are dealing with a small amount of connections, but as soon as your application starts getting any amount of concurrent users, MySQL is famous for falling down rather rapidly as it strains to write its data to disk.
MySQL cannot scale reliably, period. Having two database systems act as a pool, under MySQL, is a crapshoot at best. Unless you like designing single points of failure into your web applications, stay away from MySQL.
Simply put, if you expect your web application to get any amount of decent traffic (say 100,000 pageviews+ per day), then MySQL is simply not an option. Oracle is simply the standard upon which others can only attempt to compare themselves to. MySQL may be fine for the low-end "check out my k00l dynamic site!!11!!" crowd, but for professional web applications, MySQL has a long, long ways to go.
...because in about 10 years, you're going to be incredibly disappointed.
What you really want is to be able to import uncompressed video via Firewire (or DV-compressed video, like what the story mentions) and edit it from there.
"I can't wait until the day I plug a Firewire dv cam into a mandrake box, a dv cam icon pops up on the desktop and allows joe to edit away to his heart's content."
If you're willing to shell out $999 for an iBook, you can have this today. Cheers! Enjoy Gnome 2.2 (snicker)....
Quit demanding that Apple should make you a movie player for free. If you want a high-quality Quicktime experience, buy a Mac. If you want to remain five years behind the times with regards to audio and video, stick with Linux.
I have news for you: you aren't editing (or "manipulating" as you call it) video. You're manipulating DVD VOB files. There is a huge difference.
"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."
Similarly, my eight-year-old cousin would get "stopped dead in his tracks" the moment he had to apt-get or rpm get any kind of package or driver required to even play a Quicktime movie. Get it through your head: LINUX USABILITY IS AN UTTER SHAM.
"Now, it is true that there probably exist underground firmware flashes to circumvent the region restrictions,"
You're absolutely correct. There are all sorts of DVD-duplicating, region-switching, etc software for the Mac, all of which will happily do what you want to do, in about one tenth the time that it takes for you to do the same thing on Linux, because you don't have to read a 50k README and compile ten different packages in order to get them to work!
"I want a computer that does what I want, not what Hollywood wants, and to achieve this goal I have to use free software."
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. I imagine it would take about 5 minutes to download and install and use the correct utilities under OS X.
Taking this back to the jwz rant, like everything else in desktop Linux, video manipulation has a huge learning curve because of the lack of standardization on the part of interface designers, driver coders, and overall system-management-utilities.
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.
"Free software" exists on more than one platform, for your information. OS X isn't just a collection of ten applications from Apple, you know-- it's a bad argument, but "anything you can do I can do better."
- First of all, Jamie was talking about just trying to play video. If he has to do any kind of configuration or compilation at all, he's done too much. Video playing is the easiest thing on the planet! Why is it so damned time-consuming under Linux?
For the rest of your self-praising argument,
- How long did it take you to compile and configure vcr, avilib support, transcode, mpeg2enc, and all the various patches that are required to get your video to record?
- "Also I create videos with a (...) camcorder (...) and Kino (...)"
Congratulations. Kino provides the same functionality that even the shittiest NLE, Adobe Premiere, had in version 1.0 (in... 1994?). Way to be on the cutting edge there.
From the Kino site: "It does not support multiple layers or tracks of video and audio." Huh? How can it be an NLE if it doesn't do more than one track? Have you ever sit in front of an Avid bay and done any kind of real video editing? Because I have news for you: Kino is to a NLE as a Kia is to a Ferrari.
- "I then save to mpeg2 and encode 9kbit video (...)"
I'm going to assume you meant 9Mbit video, because 9kbit video is like looking at an old, worn-out three-quarter-inch reel-to-reel tape from the 1950s. Still, though, let's review: you're shooting with a single-chip camera, importing as a lossy format, editing with a one-track editor, and exporting as a lossy format. Again, way to be on the cutting edge.
- "This is TOO simple."
I question your definition of simple. Check:
- Compile
- Find driver
- Compile
- Compile
- Install
- Try to find package
- Compile
- Install
- Cross fingers
- Compile
- Install
What you're doing can be done by my eight-year-old cousin on his iMac, using iMovie and iDVD, which (last I checked) doesn't take any time to install, because they come with OS X. And I'd bet the quality of his resulting video is completely superlative to yours, because the tools he's using are actually modern software (where the engineers have spent more than five minutes on the interface).
Jamie needs to bite the bullet and spend the cash for a good Powerbook or G4 tower. Linux on the desktop is dead. It will never get to the level that OS X is currently at. Face the facts.
Wouldn't one of the reasons that iCommune got pulled is because Apple is probably building in Rendezvous support for iTunes into iTunes 4? They don't want to be beaten to the punch, and a third party offering "Rendezvous-like" functionality goes against Apple's plan.
I suppose my requirements for a content management system are different than yours. I have yet to see an XSL-based solution able to expire content. Until I can see a system that reliably expires content without some sort of song-and-dance with 0-length files, cron jobs, and the like, I'll stick with "update article set available = 0 where id = blah" and run a dynamic site.
Additionally, the only reason that they don't continue to run in "regular news mode" above that 100k threshhold is cost. Their server farm is perfectly capable of handling 10x the requests, but CNN has mandated that they aren't willing to spend the money in maintenance costs.
Your example of the charts on finance.yahoo.com is a bad example; "pre-generated files" is a misnomer-- Yahoo is simply telling their application servers to cache that information for a few minutes. This is, once again, a great example as to why a well-tuned application server is far superior to serving flat content.
XSL is not yet mature enough to replace good ol'-fashioned Java (that was a weak attempt at humor). It simply is too time-consuming to build templates to handle anything more than the simplest error-checking. XSL only has the most basic of variable support, and writing a "function" in XSL is somewhat akin to using an iPod as a PDA-- you can do it, but why not just use the real thing?
What exactly makes you think that? I'm not a huge PHP fan, but the site could be running any number of load balancers (software or hardware) in front of its backend. Additionally, PHP can scale to very high concurrency, it just needs to be tuned correctly. Your precious "flat-file-serving" Apache can only serve 150 simultaneous clients out of the box if you haven't tuned it.
Now I know you're talking out your ass, because if you had any experience at all with Oracle, you wouldn't have stated this. I suppose Amazon, Yahoo, CNN, et al aren't able to scale? The simple fact of the matter is that, like any other piece of software, how you tune your applications goes a long way to how many simultaneous threads you can serve up.
I think you really underestimate the true power of the Slashdot effect. What's likely happening in this situation is that Apache is buckling under the thousands of simultaneous requests, many of whom are coming from modem users -- using up precious processes to download the content. I seriously doubt that, if this person is running a 500MHz or greater machine, that processor time is an issue at all. What he/she needs to do is re-tune Apache to serve more simultaneous requests.
That is one of the harder things to do, believe it or not -- you're venturing into the realm of a more fully-featured content management system at that point. What if your content changes on a regular basis? What if, like Slashdot, the comments are dynamic?Not out of the box it won't. Read your own httpd.conf and look up "MaxClients" and "KeepAliveTimeout". Out of the box, MaxClients is set to 150, and KeepAliveTimeout is set to 15 seconds. Once you hit the 150 MaxClient threshhold, the performance of your webserver will be terrible, because you'll be waiting 15 seconds between unique client requests.
She gives five stars to a can opener on August 1st.
She gives five stars to _another_ can opener on the very next day, August 2nd.
She gives five stars to _another_ men's electric razor on November 26.
She gives five stars to an iron on October 1.
She gives five stars to _another_ iron on November 23.
She gives five stars to a cordless vaccuum on August 11.
She gives five stars to _another_ cordless vaccuum on September 7.
She gives five stars to a regular vaccuum on August 6.
She gives five stars to _another_ regular vaccuum on October 13.
Come to your own conclusions. My feeling is that she is either:
A: a professional product reviewer, in which case Amazon should include a disclaimer that she is being paid for her reviews,
B: a compulsive liar / attention-seeker,
C: a collection of reviewers all publishing under one pseudonym, in which case Amazon should include a disclaimer that she is not a real person.
D: the marketing department for Amazon / Target, in which case Amazon should include a disclaimer that she is being paid and is not a real person.
Firstly, you simply cannot edit using After Effects. Forget it. Your workflow is so amazingly hindered within the program. I will admit that it is probably the industry standard (for low- to medium-end stations) to do titling, chyron, graphics, etc, but to do day-to-day editing work, it is next to useless.
That said, the choices for editing software in the Windows environment are horrifically bad compared to the choices for the Mac. Other than the high-end Avid system, the Windows platform has absolutely nothing. Adobe Premiere is an atrocity that passes for software; instability, terrible interface, doesn't play well with others. Vegas Video is marginally better.
The Mac, on the other hand, has all sorts of quality hardware and software solutions. Take the Media100i system, for example. They just recently have ported the editing system to OS X. I have found that the Media 100 is the best mid-end editing station out there. Broadcast video, hardware codecs, plays well with a Beta SP deck or your firewire deck, etc.
Additionally, Final Cut Pro is rapidly becoming the standard for low-end stations. The USC film school is switching to an almost all-DV program, and the unofficial word is that students should go out and get FCP if they want to edit. It doesn't offer the speed that a Media100 station offers, but for an all-software solution, it blows the doors off anything Adobe or Sonic Foundry has ever made.
If these guys are so concerned about a $3500 Dell PC outperforming a $5500 Mac, perhaps they shouldn't be in the video editing business. I would rather spend the extra $2k, then spend an additional ~$5k for a good Medea RAID system, ~$5k for a Media 100 system, and be able to create broadcast video for $15k. (Nb: that is an almost unheard-of low cost of entry to the broadcast arena) Alternatively, if I were on a student's budget, I'd go for the $2500 Mac, a $999 (or cheaper for students, correct?) copy of Final Cut, and be safe in the knowledge that I was using a high-quality, reliable package, rather than spending $2000 on a PC and struggling with Premiere.
Instead, use the money to license a copy of Resin which is, for lack of a better description, Tomcat on Nitro. It follows the reference implementation of JSP and Servlets just as well as Tomcat does, and even the default configuration, which is tuned for development, outperforms Tomcat.
The configuration of Resin is almost exactly like the config of Tomcat, so I honestly don't see why you'd pick Tomcat over Resin (unless you were having trouble getting the 1.2 or 1.3 JDK installed on your FreeBSD box, something that is historically difficult to do).
Draw, Wordperfect, Office, etc etc. All the while they're creating ports of .Net to FreeBSD (that won't generate any revenue) and other various frivolous projects. This is a little bit like the plight of Sonic Foundry; getting into video and creating five different audio suites really dilutes the manpower to create great applications.
What Corel needs to do is concentrate on one product and make sure it's the best in the business. Go after Photoshop. Go after Office (well, on second thought, don't). But don't go after both at the same time.
- Make sure you have a 2.5.x kernel or above.
- Select your card from a dropdown.
- Retrieve, unpack, and compile source code.
- Install resulting software with a strange command-line utility.
- Retrieve, unpack, and install even more software.
- Edit a configuration file.
- Edit another configuration file.
- Run a script.
- Start a daemon.
Wow. That's so easy! I can see why OS X is the number one selling Unix:
- Go to System Preferences.
- Select "Sound".
- Select "Output".
- Select your high-end audio card.
- Select "Input".
- Select your high-end audio card.
Let's assume that your time is worth $50/hr. After a few hours of struggling to set up sound under Linux, that extra cash for Apple hardware doesn't sound so bad...