Readers sound off.
groff is Great

I am writing to thank you for publishing Wayne Marshall's article on groff [December 2000]. Even though I have been using groff extensively for four years, I still found something new to learn in Wayne's piece.

I learned about groff some years before when I obtained a copy of UNIX Text Processing by Dougherty and O'Reilly. I found it intriguing and poked away with the ms macros for a while until I returned to school and decided that the courses for which I was paying deserved to have my handwritten (and largely illegible) notes transcribed for future reference.

Wishing to “give back” to the Linux community, I prepared a “Groff and Friends HOWTO” (http://www.ucalgary.ca/~dprovins/), which after reading Wayne's article might be more aptly titled a “groff and Friends ms HOWTO”, as it is focussed on that set of macros, and how to adjust them to meet one's needs.

Even with all that “experience”, I still found Wayne's article enlightening, and when it comes time to rewrite the how-to (which it could use), I hope to include some of his insights. Naturally I'll mention his article as a reference for new or not-so-new users.

May I add that there is an active mail list for groff at http://groff.ffii.org/ that I recommend for those with both questions about, and suggestions for, groff development.

—Dean Provins provinsd@ve6cta.cuug.ab.ca

What's up with Ogg?

In my January 2001 copy of Linux Journal, there is an article on Ogg Vorbis as an open-standard alternative to MP3 [“Ogg Vorbis-Open, Free Audio—Set Your Media Free”] The article states that Ogg Vorbis' main financial backer is iCast, an entertainment division of CMGion. However, it makes no mention of the fact that CMGion closed down iCast services as of the end of last year and is closing many other divisions of its company acquisitions.

I was just curious why such an article was run with information that was old news or at least did not include an update when it seems that the backbone of this standards support is suffering from osteoporosis.

I checked both the Vorbis site and the xiph site linked from the article, but neither of them have any news or apparent updates since mid November (when icast was shut down according to ZDNet news).

—Kathy Lynn

Kathy, As you say, iCast services were closed down as of the end of last year. The issue of the article is January 2001, meaning it was on the streets mid-December and at the printer in November. The article's author, Jack Moffitt, former VP of technology for iCast was laid off when it happened and had no idea at the time he wrote the article. There is, however, already much software that supports Ogg Vorbis available at www.vorbis.com/software.htm.


Changes to the Python Developer's Handbook

I just received my February 2001 issue of Linux Journal, and it was great to see a review of my book there. I liked every single word that you said.

However, it seems that the guys from SAMS didn't tell you about the last-minute changes in the book. On October 16, when version 2.0 was released, I called my editors to let them know about it, and they asked me to cover version 2.0 in the book too. Therefore, the book that went to the printer also has an Appendix D “Migrating to Python 2.0”. Besides that, I added a lot of information throughout the book concerning version 2.0.

—Andre Lessa


Frank LaMonica's “Streaming Media” (January 2001)

The calculation of the bandwidth 8000 clients (80% of 10,000) produce is not 30GBs but 300MBs. This amount can be handled by today's possibilities in contrast to the published number. The second error is in the RAID naming. RAID 0 is striping, which is increasing the size (and speed) of a volume by adding them together in a row. The RAID level Frank describes is RAID 1 (mirroring).

Mick Bauer's “Paranoid Penguin” (January 2001)

Bauer incorrectly refers to Tatu Yloenen as being associated with F-Secure; in fact, he's chairman of the board and chief technical officer of SSH Communications Security, Inc. Mick also wishes he had credited at least Markus Friedel, Niels Provos, Bob Beck and Aaron Campbell. He would like to correct his statement that SSH Communication's version of SSH v.2.3 must be purchased if used in a commercial setting. That is not true for users of open-source operating systems:

To qualify for a Non-Commercial Use License, You must: (1) use the Software solely on a system under the Linux, FreeBSD, NetBSD, or OpenBSD operating system (whether for commercial or noncommercial use);... [SSH Communications Security Corp SSH(R) Secure ShellTM License Agreement, paragraph 3]