Under The Radar 39
| Under The Radar | |
| author | Robert Young and Wendy Goldman Rohm |
| pages | 197 |
| publisher | The Coriolis Group, 1999 |
| rating | 7/10 |
| reviewer | Malcolm Tredinnick |
| ISBN | 1576105067 |
| summary | A summary, designed for an outsider |
The Scenario
This book is subtitled How Red Hat changed the software business - and took Microsoft by surprise. However, that is a slightly misleading statement about the contents. More accurately, this book is a series of stories about both Red Hat and the Open Source/free software movement in general (and Linux in particular) woven together very naturally to make an interesting tale.
The book starts out at the point when Red Hat were trying to secure some venture capital in the early months of 1998. This naturally leads to a recounting of meetings with Benchmark Capital (a Silicon Valley venture capital firm), Intel and Netscape. Chapters 2, 3 and 4 cover this story quite naturally, giving an introduction to the concept of free (as in speech not beer) software along the way. This is not the story of how everything went swimmingly and was done over a cup of coffee. There are accounts of how the VC firms tried to push Intel out of the deal and then how Netscape stepped in and was willing to compromise. Chapter 4, in particular, covers (in a few brief pages) how first IBM, then Dell and Compaq courted companies offering services such as Red Hat (OK .. the actual chronology may have been different, but that's the order they're presented in the book).
Chapters 5 and 6 are basically all about Netscape. Beginning with accounts of executive level meetings at the companies where the decision to release the source code was first discussed, we are led through the whole saga up to the present day. This includes a very coherent discussion about licensing issues and the proprietary software that was already in the Netscape code. What we have here is a layman's account of why it is taking so long for Mozilla to walk out the door.
There then follows two densely packed chapters about various figures in the Open Source world. It is only at this point that Linus Torvald's is mentioned in more than passing. We don't get an account of Linux from day one, rather an insight into how Linux and Linus operate today. For example, there is little on Linus' days as a student at the University of Finland writing Linux, but quite a lot about his current work porting Linux to Intel's Merced chip (and for those seeking information about Transmeta, there is a big secret contained herein: Linus' office measures 10 foot by 10 foot and has a window). Capsule introductions to figures like Eric Allman (Mr. Sendmail), Larry Wall (Perl) and Bruce Behlendorf (Apache) are given, along with slightly longer passages on Eric Raymond and Richard Stallman. Nearly a whole chapter is devoted to Larry McVoy (formerly of Sun and SGI) and his Open Source advocacy.
Chapter 9 is a short introduction to Linux in "other products", such as disk drives and video recorders. It gives a very good idea of why such ideas are both possible and practical, but is mostly of a hypothetical nature, since nothing like this has quite hit the mainstream yet. This is followed by another "once was hypothetical, now is real" theme: Gnome. We get a run through of current work in making Linux a little more end-user friendly and the usual capsule introductions to various key figures (can you give an introduction to Miguel de Icaza in three paragraphs? Bob Young does).
Under the Radar concludes with a chapter on the recent IPO (this is an up to date book) - quite effectively conveying Young's distaste (I gathered) for the whole process, if not the concept. Anybody who was upset at Red Hat's slightly confusing attempts to offer shares to those who had contributed software would do well to at least read this chapter. Here we have a clear explanation of what was involved from Red Hat's end and what Young thought went wrong. There is also an appendix explaining the whole IPO concept for the non-financial wizards amongst us, which I found quite useful.
Having now peformed my duty of showing you the contents from a height of 10,000 metres, let me now give my impressions of how well this content was actually conveyed (with apologies to Clint Eastwood)...
The Good ...
Bob Young is obviously a man who loves and deeply understands the concept of open sourced code. This comes across on almost every page of the book where he stops telling stories and starts explaining the philosophy behind various companies' actions. Probably half the book is an advanced sales pitch on the advantages of the Open Source movement. He is also, obviously, very well plugged in to other companies attempts at introducing such ideas into their own work practices. So we are told a lot of stories of "real world" applications of free software. And these stories and philosophising will provide interest for readers at all levels.
If you were a person who had not heard of Open Source software, except maybe in passing and was wondering what the whole Linux thing was about, this would be quite a good introduction (except for a couple of problems I mention below). Statements such as "...the world naturally assumes that software is written by commercial companies to end up on the shelves of CompUSA and other software retailers. But, in fact, most software is written by professionals in order to solve their organizations' internal computing needs." (p.16) go a long way towards explaining why anybody would want to tinker with source code (or contemplate giving it away).
On the other hand, if you have any interest in the various debates going on with regards to the different forms of licenses available within the community, there is food for thought here. There is one of the best descriptions of the "viral" nature of the GPL license (and its advantages and disadvantages in different situations) that I have read beginning on page 77. Earlier in the book, the authors point out the motivation behind the original BSD license and the X Consortium's licencse. These are not technical discussion of legalities and sub-clauses; rather they are a look at the motivations behind each license and where they are useful and deficient. I came away with a real feeling for the agonising that must have gone one at Netscape prior to the creation of the Netscape Public License (and the subsequent Mozilla Public License). In fact, the whole account of Netscape's freeing of it's source code was fascinating (and relatively unknown to me - I knew the result, but hadn't heard about the deliberations leading up to it).
For those with an interest in financial machinations, this book is peppered with accounts of Red Hat's search for funding over the last couple of years. The initial chapters covering their first round of funding is consistent with what I understand of many startup companies' experiences. The final chapter on the recent IPO is interesting for its "insider" viewpoint on something that many Slashdot readers would have followed closely as it happened. Of particular interest were passages like "We did consider what Microsoft could do to use, as an exercise. For example, a week before our IPO was to register, what if Microsoft were to fund a bunch of little developers to sue us claiming ownership of some piece of code we were shipping." (p.165) To think of something this scary with an impending IPO would certainly cause me to lose a lot of sleep!
There is also a nice summary of the various business models a company basing itself around open source software might use: Support sellers, widget frosting, accessorizing, service enablors, brand licensing, sell it - free it and software franchising (the names used by Netscape salesman Frank Hecker - see p.78).
... the Bad ...
Oh dear ... where to begin? It is difficult to pin down the intended audience of this book. On the whole, this is not a bad thing; as I mentioned above, it will appeal to many. However, in quite a few cases, this lack of focus will leave readers confused or annoyed. For example, if you were in the category of readers who was looking for an introduction to Open Source, you would possibly be hard pressed to actually say what Red Hat's business was by the end of the book. The information is in there, but you sort of have to read between the lines a bit. At least, that was my feeling and I knew what I was looking for (and simply trying to confirm it was there). I'm not sure what impression a complete newcomer to this field would get about Red Hat.
Similarly, at one point there is a discussion of the push to get the 2.2 version of the kernel out the door and the schedule that Linus has proposed for version 2.4. However, at no point in all the discussions about how so many people contribute effectively to the kernel is the situation with the "stable" and "developers" (2.1, 2.3, etc) versions of the kernel even mentioned. If I was plugging software that was continually being developed, I would try to emphasise how current and future developments are avoiding breaking earlier releases by operating on an entirely seperate code tree. This may seem like an unnecessary delve into technicalities, but it would appear to have fit with the story.
My main criticism of this text is that it seems a bit superficial and even incomplete in some parts. This is partly forgivable, since Young mentions in his foreword that it is not meant to be a comprehensive account. Nevertheless, we often get to a point that most people would consider to have succesfully introduced a story and then ... nothing! It just stops. Next topic, please. One prime example is an account of a meeting at Larry McVoy's place (p.119) which was meant to help settle a potential dispute between Linus and David Miller. Miller apparently felt that some of his (good) patches to the 2.1 kernels were being rejected by Linus and he was "dropping them on the floor as he got overloaded." Great! Now we get to see how this was resolved and the lessons that were learnt (you might think). Wrong. Now we get to have a whole chapter extolling the virtues of Larry McVoy. We are given a one paragraph summary of the conclusion to the problem (David Miller became the collector of Sparc related patches and forwards them to Linus). However, there are many such examples of temporary differences within the Linux community (one only has to browse the kernel-developers archives to see this) and a few more examples and their succesful resolutions would have brought home the point. As it went, McVoy ended up getting more space than Linus.
While on this subject (Larry McVoy), there is a little inconsistency here. McVoy's BitKeeper source code management software is mentioned more or less in passing as a good idea. However, in a later chapter, Young points out that the real problem with the KDE was the slightly less than open Qt license. Since BitKeeper suffers from a similar sort of problem (note, "similar", not "identical"), is Young guilty of glossing over something here (since KDE is more or less a competitor to Gnome, which Red Hat has put its weight behind, the drawback of KDE is mentioned)? I'm not sure.
... and the Ugly.
Under the Radar was obviously produced in a hurry. The last page of the main text talks about events that happened on August 11, 1999, yet I was able to purchase a copy of this book in my local (not particularly large) bookstore in Sydney, Australia on October 31. Even allowing for the fact that most things would have been already written by then, this is a quick turnaround time. And it shows, sadly.
There are many simple typographical mistakes in the text. For example, Young talks about the rush to get the 2.1 kernel out the door (p.119). I assume he means the 2.2 kernel, since 2.1 was a development series of kernels. Similarly, the Qt library is called the Q+ library (p.140). Not big errors, but ones that should have been picked up by an editor with greater knowledge of the subject matter (dare one say, O'Reilly?). Slightly more annoying is the nine and half pages of sans serif font (pp.74-83). At various points this font is used to indicate email extracts, however in this case it is just gratuitous (and mistaken). It is not my intention to conduct a page by page post-mortem on the production of this book, but there are similar errors throughout the book.
A couple of errors of another variety are possibly a result of how the book was produced. I hope the authors will forgive me if I am completely mistaken, but here is my guess of how this book was written: Bob Young sits down on a number of ocassions with Wendy Goldman Rohm and starts talking. From what I understand, he is an excellent talker and salesman (I have never heard him speak). Goldman Rohm would then tape these chats. Young would also provide his co-author with copies of relevant emails and introductions to others in the business that were present at various meetings, etc. Finally (and I'm still guessing here), Young would follow up his talks with some email to Goldman Rohm including points that he had forgotten to mention or ideas that he had subsequently had. From her end, Goldman Rohm would pull everything together into a nice narrative, conduct interviews with other key players (Linus Torvalds, etc) and basically fill out the text. Given that this scenario is even slightly accurate, something got lost in the process. It was not always the case that the background research was done (and included) that should have been. As I mentioned above, some topics came across as decidedly incomplete.
Lest I get accused of spouting unfounded accusations, some examples: the three infamous Mindcraft studies are mentioned. The first one is covered in some detail, including a couple of technical details and reports on how the fiasco was uncovered (Yay! Slashdot gets mentioned). The second study is mentioned in about one sentence and the third, public, one is also mentioned. Unfortunately, no conclusion about the third study is drawn and the results are not mentioned at all. Since these studies were mentioned not in the context of Microsoft-bashing (of which, thankfully, there is little without foundation), but as an example of how the Linux community rose above it and benefitted from the experience, there was ample opportunity (and need?) to draw some conclusions.
Or take the case of Andrea Arkangeli, an active contributor to the kernel. The point is made that he is one of the people Linus includes amongst his top tier of kernel developers (together with Alan Cox, David Miller, Ingo Molnar, Steven Tweedy and Ted T'so). Further, the great advantage of the internet as a devlopment environment is espoused (Linus and Andrea have never met). But to then quote Linus as saying "I think he's from somewhere in Europe"(p.117) and not actually track down the answer seens a bit careless to me (another example of an interesting tidbit that remains unresolved).
Conclusion
This is a nice book to own just for the ability to be able to lend it to friends. As I hope I indicated above, it does an excellent job of explaining the whole "community" aspect in simple, non-patronising terms. It is also a nice, relaxing read. However, the more technically interested readers will probably find themselves frustrated by the lack of details and incompleteness at some points. Waiting for the cheaper softcover version may be a good idea here.
For those looking for a business oriented discussion of Red Hat the company, this is probably not the right book. Bob Young did much better job of that in his submission to Open Sources - Voices from the Open Source Revolution , published earlier this year by O'Reilly and Associates.
To purchase this book, head over fatbrain
Condensed version (Score:1)
Under the Radar is a sloppy attempt to cash in on the open source revolution by releasing a disorganized book in the hopes of having middle management types pick it up and read it for "business strategies". It's long term usefulness in your house will be determined only by how high you need to prop your monitor up to bring it to eye level.
Ultimate Rating: *
1 star: Gratuitous lack of content
Isn't it Brian Behlendorf? (Score:1)
--
Didn't see it mentioned in the review... (Score:4)
But remember, "Red Hat is just releasing this to make a buck".
Nice review (Score:2)
The worst glaring error in the book (Score:3)
migration, and referred to it as one of Linux's
"graphics libraries". I think its on page 126,
left hand page, towards the bottom.....
Andrea Arcangeli lives in Imola, Italy (Score:2)
--
How they took Microsoft by suprise? (Score:1)
Re:Condensed version (Score:1)
Why do people insist on crying that Red Hat and Bob Young are just out for a buck when EVERYTHING they've done has served to further the intrests of Free Software? All the software they produce is GPLed. They've been very generous about funding development (that means PAYING programmers to write free software). They pioneered "the letter".
It's reasonable that not everyone is going to like their distribution. Fine. Don't use it. But to accuse them of base motives when their actions CLEARLY indicate otherwise is just slander.
Re:Condensed version (Score:1)
And you're way too paranoid. I'm attacking the book not the company.
Re:Condensed version (Score:1)
Open Sources (Score:1)
Open Sources - Voices from the Revolution is a better all-round read and the whole book is available online from here [oreilly.com].
Re:7/10 ????????? (Score:1)
The problem is anyone who could review this book would allready be bies tword it.
Over all I'd say this was a very good effort to review the book.
Personally I'm not fond of thies CEO bios. Oh wow the CEO of StupidThings inc has put out a book. They make lots of money doing stupid things let's get the book and see how they do it.
Re:Condensed version (Score:1)
Marketing (Score:1)
Re:Marketing (Score:1)
Re:Condensed version (Score:1)
Re:Condensed version (Score:1)
the problem is yours... (Score:1)
I have _never_ experienced any distribution that I could not make do _exactly_ as I wanted.
If you have trouble with RedHat, use caldera.. I hear it's 'easier' for newbies.
Not for OpenSource community (Score:1)
I think O'Reilly's OpenSources is a _much_ better book because it contains the _opinions_ of the open source people and doesn't focus on redhat's history.
It's not bad, but the audience is a completely different. - It was interesting to read and I learned much more about Red Hat, but nothing I really need or want to know
All that RedHat's done?? (Score:1)
[sarcasm]I guess that's all Debian and Suse do for that matter too[/sarcasm]
Re:Didn't see it mentioned in the review... (Score:1)
Save your money (Score:1)
The chapter on Netscape and releasing Mozilla was interesting, the rest is a self-aggrandizing account of Bob Young et al. playing "Let's make a deal".
Recommendation: save yourself $30 odd and give it a pass.
About time someone is getting the story out... (Score:2)
Linux's success is dependant upon REAL credibility. While this book is unlikely to be discussed in grocery stores, I would be shocked at any technical manager who doesn't read it. It will serve to get the actual story (more or less) about Open Source... something that has been lacking.
Lots of hype about Linux is great, but people want to understand where it comes from. People watched MS evolve, they know who big blue is, but Linus? He's a non-American (taboo in an American culture celebrating Americans) who got annoyed with Minix and accidentally started a revolution of sorts.
I wonder what impact Linus's background will come to play over the next few years... Part of the reason that VHS beat Beta was that more American companies were on-line with VHS so the government looked the other way at illegal (dumping) practices in an attempt to monopolize a market. We'll see what happens if the country decides that the US Gov't toppled an American Success Story and brought in foreign competition.
Re:Open Sources (Score:1)
what "with" means, various comments (Score:3)
When you see a book written by "A with B" it usually means that A didn't write a word: A was interviewed by B, and then B wrote it.
For example, the Freeing the Source: The Story of Mozilla [oreilly.com] chapter of O'Reilly's Open Sources [oreilly.com] book is credited as ``by Jim Hamerly and Tom Paquin with Susan Walton.'' In fact, that document was written entirely by Susan Walton (and I think she did a good job of it.) She based it on extensive interviews with Hammerly, Paquin, and myself. I pissed off a lot of people (at Netscape and O'Reilly) by refusing to allow them to list me as an author, because, quite simply, I hadn't written a word of it, and I didn't want to take credit for something I didn't write.
``This is how the publishing industry works,'' they told me. ``Everyone knows what `with' really means.'' Well, I hadn't known that, and I found it to be deceptive, so I wouldn't play along. I thought it should have been credited as ``by Susan Walton based on interviews with ...'' or ``...as told to...'' or something, but they hated those.
Anyway, about the Red Hat book...
I thought the history of Red Hat was interesting. But I'll comment on the parts I have direct knowlege of:
The inside-Netscape history is fairly accurate, including the deliberations about the licenses, but as far as I can tell most of it is lifted directly from Frank Hecker's writings [hecker.org] on the subject. If you're interested in this part, you should read Hecker's papers, because they are a much better explanation in their pre-condensed form.
I thought their summary of why we didn't use GPL left out a very important detail. Anyone who understands copyright law would respond to the reasons they gave (``we want to bundle with other proprietary software'') by pointing out that the copyright holder always has the right to do things like that. The important point that they failed to mention is that if the copyright holder does so, they eliminate their ability to take contributions from the outside, which is pretty much a deal-breaker. They mentioned that we had a very hard time trying to find a solution to our various license-related problems, but I don't think they did a very good job of explaining what those problems were, or why we reached the decisions we did. Maybe this wasn't an important detail to a book about Red Hat, but if not, they shouldn't have included it at all. What they did include is fairly muddled.
Throughout the book, there were quite a few things that jumped out at me, like saying in one paragraph about Cryptozilla on page 98 ``less than a month after the source code was released ... the group added full encryption,'' and then saying two paragraphs later, ``fifteen hours after the source code release, a fully crypto-enabled version of Mozilla for Linux was released.''
So yeah, 15 hours is less than a month. But the book contains a lot of strange errors like this. It's as if there were no review copies distributed at all. (Were there?)
The entirety of the first year of mozilla.org, between April 1998 and April 1999, seems to be skipped over somewhere in the middle of page 100. Then they say, ``it took over a year for Netscape to ship Netscape Navigator version 5.0.''
When did they ship that exactly? For those keeping score at home, it has now been 18 months, and will certainly have been two years, if it ships at all.
Also they consistently misspelled my name (for which Bob was extremely apologetic -- he mailed me about it a few days before the book was released.) Not that I particularly care about trivia like the spelling of my name, but getting that wrong in a book is really a rookie move, and reflects badly on the book that something so basic slipped through.
I think that all of these problems stem from sloppy editing and lack of review. This book would have been much better if they had taken another month or two to distribute review copies and get feedback and corrections.
``Release early, release often'' doesn't work so well with physical media.
Re:7/10 ????????? (Score:1)
Firstly, the 7/10 is very dodgy, but for those who think you can accurately sum up an entire book with one number between 0 and 10, I suggest you go and rent "Dead Poets Society" on video and look at the scene where Robin Williams first meets his new class. Initially, I wanted to give it a score of about 3/10 for technical accuracy, 5 or 6/10 for interest to technical types (some of the chapters *were* interesting, despite their lack of complete accuracy) and about 8/10 for those who are not heavily involved "into" Linux already. I decided on the final number by looking back at some past Slashdot reviews and comparing my opinions of those books (I have read most of them) with the numbers they got. The numbers are slightly high, in general, but if you going to buy/read a book based on the number, rather than the review, then you probably deserve to be punished.
I believe I made my true feelings about the book and its shortcomings known in the body of the review. However, I was also interested in trying to view it from the point of view of somebody who may not (yet) be a Linux user.
As to the Felinoid's comment that the book would get a good review based on "open source is good"
Re:what "with" means, various comments (Score:2)
Last Post! (Score:1)
Bob Young and his unimpressive script-writer friends at redhat are becoming centimillionaires off of your work. They do not write the software; they write installation scripts and burn CDs. You are, to put it succinctly, being shafted.
Your open source gift to the world is being exploited to make others undeserved money. If you're not outraged, you're not paying attention.
Don't be placated by the peanuts redhat tosses the open source community. In proper perspective, think of yourself as a beggar accepting pocket change from someone driving a porsche.
_.......................__
||.....__...._._||_..||-\\..._...._._||_
||......_\\.(/_'..||....||-//.//.\\.(/_'..||
||__((_||_,_/).||_..||....\\_//.,_/).\\_
HAHA! LAST POST! Anything following is redundant.
Moderate that post up! (Score:1)
I wanted a Red Hat distro because I wanted a company with
Red Hat brought Linux to the non-UNIX types, and don't ever forget it.
That sums up my experience with Linux exactly. When I started playing with Linux about 2 years ago, it was becuase I borrowed a friend's RedHat 4.x CD.
I had known about Linux before, but all I ever knew was you had to download the floppy images and it was a pain to install anyway. Being a newbie, all I was interested in was throwing it on a PC to play with for a couple weekends to tinker and see how it worked. The whole download-copy-install ordeal was not something I was prepared to do. My bad for being lazy, but I had other things to spend my time on.
Over the last two years, every time I got a new computer, I would install whatever the current Red Hat was on it just to see for myself how far Linux had come. It wasn't until Red hat 6 that I found something I considered worth building a PC around.
I would never have done that if I still had to download floppy images (although I now realize how neat it is that an OS exists where you can do that and it's actually encouraged!)
-James
Re:Last Post! (Score:1)
It's never been about money and financial gain for the developers! It's about writing Free Software. By definition that means it's OK for others to make money off the software. (provided they follow the license terms, source code must be available, derivative works must remain GPL, etc.)
If Linus was only interested in making money off his software, and stopping others from doing the same, he never would have used the GPL.
I'm not a developer myself so I don't presume to speak for them, but everythign I've read seems to indicate they are ecstatic that Linux has become so popular, and it seems they mostly appreciate the role RHAT has had in that.
Everything I've read indicates they've been an upstanding member of the Open Source community, they encourage (and more importantly, fund) a lot of good packages and developers, and they have always followed the GPL with their own code.
And what's with that dumb-ass last post remark in your sig, anyway? Lousy troll...
-James
Re:what "with" means, various comments (Score:2)
If, as Jamie suggests, the Netscape chapters are just a rewrite of Frank Hecker's writings, then a credit is almost required.
I don't have a copy of the book handy, but I believe that I was in fact credited for the substantial chunks of material than Wendy Goldman Rohm quoted from my paper. Unfortunately, if I remember correctly the way she quoted me gives a misleading impression to the reader, because the paper she's quoting is the public paper that I wrote after the Communicator source release, not the original paper I wrote for internal consumption at Netscape. Thus, for example, the book quotes me as discussing choices between various licenses, business models, etc., and (to my mind at least) gives the impression that this stuff went into Netscape's decision to release source. However that material was written after the Mozilla source release, based on both the Mozilla license discussions and some of the opensource.org [opensource.org] material.
The complete and authoritative history of the Communicator source code release has not yet been written. In my opinion the best sources so far are the last chapter of Josh Quittner's book Speeding the Net, which covers the period up to the January 1998 source code announcement, and the chapter in Open Sources, which covers the period between the January announcement and the actual release of the code at the end of March.