Month: October 2001

Good article on building a large-scale e-commerce site with Apache and mod_perl, for what that’s worth nowadays, at perl.com. This bit was especially pleasing to my SQL-database-phobic mindset:

Since Perl code executes so quickly under mod_perl, the performance bottleneck is usually at the database. We applied all the documented tricks for improving DBD::Oracle performance. We used bind variables, prepare_cached(), Apache::DBI, and adjustments to the RowCache buffer size.

The big win of course is avoiding going to the database in the first place. The caching work we did had a huge impact on performance. Fetching product data from the Berkeley DB cache was about 10 times faster than fetching it from the database. Serving a product page from the proxy cache was about 10 times faster than generating it on the application server from cached data. Clearly, the site would never have survived under heavy load without the caching.

Ha! Take that, database-backed-website fans! ;)

Great story from Jon Callas about the history of CD-ROM drives, and why they can play audio CDs in the first place.

Date: Thu, 27 Sep 2001 15:28:26 -0700
From: Jon Callas (spam-protected)
Subject: Re: ah, been waiting for this

At 5:38 PM -0400 9/27/01, t byfield wrote:

> i’ve never bought or even seen a CD of any kind that warrants
> that it will be playable in any physical device capable of in-
> terpreting a CD. and it was only a few years ago that the dis-
> tinction between audio CDs and data CDs was clear: one worked
> in certain kinds of devices, the other worked in other kinds,
> and at the time the twain did not meet. then for a while they
> did, and now manufacturers are reintroducing the distinction.
> i’m not endorsing their actions; but CDs have been around for
> ~20 years, and the ability to use a computer to read data CDs
> has been around for more like ~6 years. this doesn’t have the
> makings of a solid legal argument; it wouldn’t take a rocket
> scientists litigator to ask why no one seems to object to the
> fact that data CDs don’t function in audio equipment. and even
> if it did have the makings of a solid legal argument, pursuing
> a ‘truth in labeling’ strategy may very well end up with audio
> CDs that are truthfully labeled. brilliant.

I disagree on some of the facts.

CD-ROM players have always been able to play audio CDs. The very first CD manufacturing plant to be owned by someone other than the Sony/Philps consortium was built by Digital Equipment Corporation to made CD-ROMs for software distribution, in about ’87-88, as I remember it. While ISO9660 came after the audio formats, it was intended from the start to interoperate with audio CDs in the sense that a player can easily tell the difference.

Now then, early audio players couldn’t tell a data CD and would play them as audio noise. Also, some early DEC CD-ROMs did not have audio jacks. But this is a tale of corporate stupidity only. You see, someone brainiac decided that a CD-ROM was a Serious Business Device and not for entertainment. So they went to Sony and requested that Sony make a model that did not include the audio jack. Sony said, “Sure, no problem, it’s just a manufacturing line change, which we charge $3 million for, so pay us and it’s yours.” So DEC did.

Then customers started complaining about the lack of an audio jack. Their argument was, “Look, I paid $1000 for this CD player, and while I’m in the machine room, it would be really great if I can pop in a CD and listen to it.” The DEC response was, “This isn’t an entertainment device, it’s a Serious Business Device.” The customer response was, “Ummm, look at these schematics. This Serious Business Device that I paid a kilobuck for has all the audio circuitry. If I put an audio CD in it, it mounts up and spins. All it’s missing is an audio jack, which costs about a quarter at Radio Shack. So here’s a quarter. I’ll pay $1000.25 for one, okay?” At one DECUS, there was even a session on how to unscrew the cabinet and which Radio Shack part number would solder directly to the circuit board, and what I/O calls to the device would operate the play/stop/pause/etc. buttons.

Finally, DEC went back to Sony and said, “We’ve changed our minds. We want all our future CD-ROM drives to have audio jacks. Sony said, “Sure, no problem, it’s just a manufacturing line change, which we charge $3 million for, so pay us and it’s yours.” So DEC did.

Now until recently, CD-ROM drives would operate in either “data” mode or “audio” mode. In audio mode, you could punch the buttons and all, but you couldn’t get the digital audio bits off of it. So you could have a software-controlled CD front panel, but not digital music. The first one of those was the Apple double-speed CD-ROM, which for reasons that I don’t know allowed direct access to the digital bits. It is my belief that this occurred because by this time, Sony owned both hardware companies and record companies, and thought it would be cool to allow computers to process music. I do know, however, that for a couple of years, if you happened to have one of those Apple drives, people who were into digital recording would pay you a lot of money for one. And *that* happened around ’96.

*Ripping* is only about six years old. But there has never, ever been a CD-ROM that would not play an audio CD, with the exception of the early DEC fiasco, and even they would, really. The same is true with DVD players, that they’ve been transparently compatible with CD players, both data and audio. Consequently, consumers have the reasonable expectation that if they buy an audio CD that they can go to any CD-ROM in the world and it will play audio. I think they should put a label on the protected CDs that’s at least as big as the naughty word advisory, “WARNING: DOES NOT PLAY IN CD-ROM OR DVD PLAYERS.”

Jon