Oh, please. The people who choose to use the RBLs are accountable to their users, who are usually their employers or customers. The people you send mail to may not be accountable to you, but it's not clear to me that's a problem.
Google is making so much money because of its keen business sense [...]
That's a bold assertion, but you give no data to back it up. And there's plenty of evidence on the other side. Take their UI, especially the way they handled ads. At the time, they were considered insane: all the search engines were moving the other way. It could be that they're two CS PhDs just happened to have business sense better than everybody else in the valley put together. But it's more likely that they were being honest: they put quality first, and ignored revenue maximization.
The clearest example is their IPO. There's no business justification for the risks they took in doing it such an unorthodox way. They must have taken fantastic heat from their VC firms, and VCs really know heat. But they did it anyhow. Why? I think it's because they see themselves as geek revolutionaries.
[...]and superior technology, not their "Do no evil" clause.
You really missed it here. I know a number of people who have taken jobs at Google. I also know people who work at their direct competitors. There's no question that Google gets first pick of the hardcore geeks. Why? Because Google as a company believes in the things that those geeks believe in: doing the right thing, doing that thing right, and making beautiful technology. In comparison, Yahoo's just a regular job.
Google's strong corporate philosophy isn't just window dressing, and it's why they have such great technology. If they abandon that, they'll lose their best troops. And I think they know that.
But is it "cynical" for them to move around enough money to actually pay for all that great stuff? Come on, folks, we can't have it both ways.
You should tell the Google founders that. It has been their frequent claim that they can have massive success because of their philosophy, that they can do well by doing good. So far, they've lived up to that, and become billionaires in the bargain.
There's nothing "evil" about growing the company. And all of you Google stock holders had better come clean now if your preference is that the stock stays low!
Yes, yes. That's the conventional MBA dogma. A growing company and high stock values validate any behavior that gets you there. The end justifies the means. Et cetera, ad nauseam.
But Google claimed to be different. And so far, they've acted like it. Hopefully this is just a temporary abberation. But if you're right, and they're changing course to be just like every other company, whose to say it will help them? I'd think a move to be more like other companies would harm their brand and lose them their competitive advantage.
Frankly the story about the cell phone I find iffy at best.
Yeah, I agree. According to a recent article in The Economist, your average transatlantic flight has a about a dozen cellphones left on. If they really a safety issue, you think they'd be a little more rigorous.
However, I firmly agree with the requirement that they be shut off. I can't imagine the pain of being trapped in a middle seat next to Inconsiderate Cellphone Man for five hours.
Design documents aren't just for programmers, though; they can also be useful for sysadmins.
I completely agree that sysadmins deserve high-quality documentation with every release, documentation that specifically addresses their needs and concerns. I don't think the design documents are the right ones for that, though. I also think programmers should work closely with operations staff *before* the final release, so that the product is built to be easy to maintain.
When managers say, "We need documentation for X", it makes me crazy if they can't say who the documents should be for, or why those people will read them. Often that this means that the programmers are forced to write a pile of documents that serve no need and are read by nobody. Your request is, thankfully, just the opposite: a clear need that programmers should specifically address.
Now I'm still using FF 0.9, same reason. [...] I pass. IMO a 1.x codebase should be mature and stable enough to be installed over an existing earlier version.
I think you're missing the point of an 0.x series. It's for early adopters who don't mind things not being quite perfect. If you want a solid product, you should probably wait until 1.0, which is their way of telling you that they think they have a solid product.
I know you were trying to be funny, but I think that's exactly the right question to ask.
The poster starts out with the assumption that the best way to get up to speed on a project is reading documentation. For most projects, I think that's wrong.
I have seen a lot of projects over the years, and I've never seen one with good, up-to-date design docs. I used to think that lack of docs was wrong, and that people needed to reform. But lately I've come to think that maybe the people were right, and my idea was wrong. If the goal is to bring somebody up to speed quickly, a week or two of pair programming with other team members will do much more than a lot of time wasted reading and, worse, writing docs.
If that isn't enough to get somebody started, I have to wonder if there's something wrong with the code itself. I'd much rather see the effort people normally spend on docs go on writing beautiful code and tests. On too many projects, the docs try to compensate for opaque, poorly structured code. It's like somebody has a dog that poops all over their house, and their solution is to wrap the turds in Christmas paper.
The one time I think programmer-written design docs are still valuable is at the end of a project. If I won't be around to introduce the next people to pick up the code, I like to write something explaining the big picture: what technologies we chose and why, what the architecture is like and how we got there. If we've done our job right, the details should be in the code and the test cases. But a good overview can help the new team get oriented.
I could create a document which included every design decision and every little facet of the project, but the document would wind up so huge, it'd be impossible to work with.
Exactly! Documents should be written for particular audiences who have particular purposes. The document that included everything would have enough information that you could just write a compiler to build the project, completely eliminating the coding phase. And it would be a terribly boring document that almost nobody would read.
The best resource I know of is the project guru.
Even better, you should arrange your project so that everybody has an opportunity to become a guru. Getting everybody (including the product manager) to work in a single project-focused team room is great for that. Also valuable are test-driven development and pair programming.
Eventually, the project will go dormant and the team will scatter. That's the time to write documentation.
The problem is when the prototype becomes the final product because TPTB don't want to waste money re-doing it.
That's indeed a huge problem. It's worth making the prototype look and act like a prototype. Fine ways to do that include Paper Prototyping, click-through mockups in plain HTML, and the Napkin Look-and-Feel. Also consider making the prototype obviously flawed in other ways. E.g., leaving stuff out, making graphics very low-res, adding long sleeps to make it slower, and so on.
Troll pts for that? I see we have a consultant mod in the house.
Or they could just be recognizing that some of the people here are consultants. For example, I do both contract development and consulting.
Of course, I think the common stereotype of consultants is often true, just like I think the stereotypes about programmers are often true. But the sterotypes are also often false. Stereotypes are like that.
Which is why acting as if a sterotype is a universal (e.g., "All women are... ") would be seen by some as trolling. Me, I think you were just being an insensitive clod; Slashdotters are all like that.
I think he's looking for the best way to get the point across.
I think the very best way is to tie it back to things the boss cares about: money and productivity.
Go through the report and come up with solutions that cover all the points, at least the ones that aren't bogus. Explain what each solution will cost (both in cash and in business impact), and what, in business terms, the benefits are.
If your instincts are right, your boss will say something like "Better security is well and good, but I'm not doubling the IT budget and inconveniencing our staff for so little improvement." And if it turns out there are some things that they're willing to pay extra for, then that's great: you get more budget and new toys.
Note that if they suggest you do more stuff without changing the budget, then you should be ready to say, "Oh, ok! Which things were you thinking of cutting? I recommend X, Y, and Z." Never let them get the idea that they can just heap unfunded mandates on you. That's not an option, just like haggling with the clerk at WalMart isn't an option. It's not that you refuse; it's just that it isn't an option.
Correct me if I am wrong--but Larry's oh-so public hissy fit is all anyone arguing the opposite side of the argument has.
Exactly true. We have partial stories from three different people, all of them with an interest in the outcome. I'd love to know what really happened. Often when I dig into disputes like this, all concerned have reasonable tales to tell, and it's not so much that one person is wrong and the other is right, as that they have different priorities and goals, and different opinions about what's really important.
(Or, at the very least, not *my* hyperbole.)
Heh. Well put. I look forward to Tridgell and OSDL speaking up on this!
I can't stand it when people act as though my belief in my own opinions is a negative thing (or even a potentially negative thing).
One would hope that you yourself believe your own opinions... in fact one might argue that the very definition implies it.
I try to take my opinions, like I take everybody else's, with a grain of salt. I generally avoid dealing with people who take their own opinions too seriously, as I find it makes getting at the truth unnecessarily arduous.
Take this case, for example. You said:
Tridge did not use BitKeeper **AT ALL**.
But your strong statement of a conclusion turns out to be backed up only by Tridgell's claim. Hopefully Tridgell is telling the truth, but he could also be lying or be mistaken. An all-caps-plus-double-bolding opinion expressed as a fact makes me suspect that, at least when you wrote the post I initially responded to, you were more intrested in winning an argument than you were in finding out the truth.
And hey, that's ok; on Slashdot, rants are accepted, and great ones get modded to the stars. But it's good to let people know whether you're ranting, and when you want to have reasonable discussions. Mixing the two doesn't work so well.
seriously people. mp3 is going to be around for a very long long time. why are you even bothering with any other format?
Aside from the good reasons others have mentioned, it's because there isn't just one MP3 format. Sure, anything can play most any MP3s, but there are many options for bit rate, and lots of different encoders with different encoding characteristics.
A friend of mine does exactly this (giant array with all CDs losslessly compressed), and then encoding what he needs when he needs it. That way when he's DJing or putting together a mix, he gets high quality. When he wants to squeeze a bunch of stuff on his iPod, he optimizes for space. And when the latest encoders improve things, he just writes a batch job and lets the box chug away for a couple of weeks.
"I'm ripping my entire audio collection to lossless audio files and I need a cheap large-volume storage solution...."
I'll be on vacation next week, so let me add one valuable reminder now.
As a habit now, any box I build has two drives in a RAID-1 mirror set; compared with the pain of recovering from a failure, the cost of the extra drive was good insurance. And I thought RAID-1 was fine; how likely was it that two drives would fail at the same time?
It turns out that if you buy the same model of drive at the same time, drive failures are much more correlated than random chance. Out of four drives I had bought at the same time, three of them failed within a two week period. (They were the now-infamous IBM Deskstars.) It made me very sad.
Now I'm inclined to mix drives from different manufacturers, and I like to keep a spare drive at hand, so that I can keep the high-risk period as small as possible. Or for RAID-5, I'm a big fan of the hot spare.
You could look at it that way. But given that the problem only cropped up when someone, apparently out of fundamentalist software-must-be-free beliefs, started building a tool that Linus didn't want and asked him not to build, it's at least as reasonable to say that the real problem is with people starting holy wars.
the whole _concept_ of having a software company is to charge people for the software you develop
And this is not just the way that the companies making the software think. Consumer behavior around pricing is just as weird. I can't name the company, but there's a company that sells a regular version and an enterprise version of their product. The two versions are identical in functionality, but one costs ten times as much. A significant percentage of purchases are the enterprise version.
Similarly, some friends of mine make an on-line game. It's well-reviewed, really fun, and appeals to a broad range of people. But a lot of potential users won't take it seriously because you can just download the client. So they've given up arguing and are making a retail version in a pretty box. It's the same client you could download in ten minutes over your broadband connection, but all their research suggests that a big percentage of people will be much happier paying $25 for it than getting if for free.
And people do this in other realms, too. I know fuck-all about wine, and so when I'm faced with the store's wall of near-identical bottles, a good portion of my judgement is based on price. In asking around, it turns out most people do the same.
Linus clearly believes that if someone gives you a gift, you don't bend them over and ram them up the rear for their generosity, even if it's perfectly legal for you to do so.
What the hell are you talking about?? Tridge was not given a gift.
You seem a bit in a lather about this. I'm hoping you've cooled down a little by the time you read this, as I think you're not trying to look at this from McVoy's perspective. Note here that I don't really agree with McVoy, and I'll say more about that at the end.
McVoy has built what many people think is the best SCM system out there; certainly Linus thinks that. But BitMover, his company, is tiny compared with the other major SCM vendors, who include giants like Microsoft and IBM. Building a company is hard work, and I'm sure he feels very protective of it. Further, I imagine he thinks that his main edge over his competitors is his superior technology, which is embodied in his IP. (Certainly it's not his winning personality.)
Now I'm sure he thought he was doing something really good for the Linux kernel development community by giving them free BitKeeper licenses. Then, from out of the blue, somebody in that community starts trying to reverse-engineer his product. And even worse, it turns out to be a coworker of Linus's, somebody working for a company who has signed the BitKeeper license that specifically forbids that. Not only does McVoy have to worry about the IP threat, but he has to worry about somebody accidentally screwing up existing repositories as they hack their way to a working solution. To him, this would feel like a slap in the face after the time and money he's spent. He even tries to negotiate, and only gets more frustration out of it. Eventually he decides that he doesn't need the hassle, takes his ball and bat, and goes home.
Now personally, I think McVoy's a bit of a crackpot, and that his desire to control things so tightly is harming his company. But I never built a first-class SCM system from scratch, so it's not like I can offer Linus anything better.
OSDL "agreed" to not use BitKeeper for reverse engineering. Tridge did not use BitKeeper **AT ALL**.
Since you seem to be pretty sure of your opinion, perhaps you can explain how he did it? I'm not being snarky, by the way, I'd actually like to know. My presumption was that he had access to protocol captures because of his OSDL access.
Thanks for a reasonable and thoughtful contribution to a discussion that hasn't seen much of that. But I think you've missed something.
The made-up quote has the same gist, even if it's critically wrong in (a) the file format, and (b) the fact that Linus is talking about somebody else's beliefs, not his own. This gist, however, is clear that Linus believes roughly the same thing.
That's not the only possible interpretation. When Linus says he can't argue with that, I think all he means is that McVoy has a right to take that position, and that it's understandable and logically consistent. Linus didn't say that he shares McVoy's view or that McVoy's is the only reasonable position one could have on the topic of reverse engineering.
My guess is that Linus is perfectly fine with reverse-engineering file formats in general. I think his opposition to this particular case has nothing to do with high-minded princples of freedom and the legal right to reverse-engineer, but rather with pragmatic issues specific to his situation. Whatever Tridgell's intentions, he cause Linus (and presumably OSDL) a lot of trouble, and I think Linus wanted to avoid that trouble.
Slashdot Mirror
Given a choice of reading comments or reading well structured tests, I will read tests any day. I at least know if they run, that they are true.
Yes! Good tests are documentation that a computer can automatically validate.
So in the end no one is accountable.
Oh, please. The people who choose to use the RBLs are accountable to their users, who are usually their employers or customers. The people you send mail to may not be accountable to you, but it's not clear to me that's a problem.
Google is making so much money because of its keen business sense [...]
That's a bold assertion, but you give no data to back it up. And there's plenty of evidence on the other side. Take their UI, especially the way they handled ads. At the time, they were considered insane: all the search engines were moving the other way. It could be that they're two CS PhDs just happened to have business sense better than everybody else in the valley put together. But it's more likely that they were being honest: they put quality first, and ignored revenue maximization.
The clearest example is their IPO. There's no business justification for the risks they took in doing it such an unorthodox way. They must have taken fantastic heat from their VC firms, and VCs really know heat. But they did it anyhow. Why? I think it's because they see themselves as geek revolutionaries.
[...]and superior technology, not their "Do no evil" clause.
You really missed it here. I know a number of people who have taken jobs at Google. I also know people who work at their direct competitors. There's no question that Google gets first pick of the hardcore geeks. Why? Because Google as a company believes in the things that those geeks believe in: doing the right thing, doing that thing right, and making beautiful technology. In comparison, Yahoo's just a regular job.
Google's strong corporate philosophy isn't just window dressing, and it's why they have such great technology. If they abandon that, they'll lose their best troops. And I think they know that.
But is it "cynical" for them to move around enough money to actually pay for all that great stuff? Come on, folks, we can't have it both ways.
You should tell the Google founders that. It has been their frequent claim that they can have massive success because of their philosophy, that they can do well by doing good. So far, they've lived up to that, and become billionaires in the bargain.
There's nothing "evil" about growing the company. And all of you Google stock holders had better come clean now if your preference is that the stock stays low!
Yes, yes. That's the conventional MBA dogma. A growing company and high stock values validate any behavior that gets you there. The end justifies the means. Et cetera, ad nauseam.
But Google claimed to be different. And so far, they've acted like it. Hopefully this is just a temporary abberation. But if you're right, and they're changing course to be just like every other company, whose to say it will help them? I'd think a move to be more like other companies would harm their brand and lose them their competitive advantage.
Frankly the story about the cell phone I find iffy at best.
Yeah, I agree. According to a recent article in The Economist, your average transatlantic flight has a about a dozen cellphones left on. If they really a safety issue, you think they'd be a little more rigorous.
However, I firmly agree with the requirement that they be shut off. I can't imagine the pain of being trapped in a middle seat next to Inconsiderate Cellphone Man for five hours.
Design documents aren't just for programmers, though; they can also be useful for sysadmins.
I completely agree that sysadmins deserve high-quality documentation with every release, documentation that specifically addresses their needs and concerns. I don't think the design documents are the right ones for that, though. I also think programmers should work closely with operations staff *before* the final release, so that the product is built to be easy to maintain.
When managers say, "We need documentation for X", it makes me crazy if they can't say who the documents should be for, or why those people will read them. Often that this means that the programmers are forced to write a pile of documents that serve no need and are read by nobody. Your request is, thankfully, just the opposite: a clear need that programmers should specifically address.
Now I'm still using FF 0.9, same reason. [...] I pass. IMO a 1.x codebase should be mature and stable enough to be installed over an existing earlier version.
I think you're missing the point of an 0.x series. It's for early adopters who don't mind things not being quite perfect. If you want a solid product, you should probably wait until 1.0, which is their way of telling you that they think they have a solid product.
Dee..zin...dok...u...ment...?
I know you were trying to be funny, but I think that's exactly the right question to ask.
The poster starts out with the assumption that the best way to get up to speed on a project is reading documentation. For most projects, I think that's wrong.
I have seen a lot of projects over the years, and I've never seen one with good, up-to-date design docs. I used to think that lack of docs was wrong, and that people needed to reform. But lately I've come to think that maybe the people were right, and my idea was wrong. If the goal is to bring somebody up to speed quickly, a week or two of pair programming with other team members will do much more than a lot of time wasted reading and, worse, writing docs.
If that isn't enough to get somebody started, I have to wonder if there's something wrong with the code itself. I'd much rather see the effort people normally spend on docs go on writing beautiful code and tests. On too many projects, the docs try to compensate for opaque, poorly structured code. It's like somebody has a dog that poops all over their house, and their solution is to wrap the turds in Christmas paper.
The one time I think programmer-written design docs are still valuable is at the end of a project. If I won't be around to introduce the next people to pick up the code, I like to write something explaining the big picture: what technologies we chose and why, what the architecture is like and how we got there. If we've done our job right, the details should be in the code and the test cases. But a good overview can help the new team get oriented.
I could create a document which included every design decision and every little facet of the project, but the document would wind up so huge, it'd be impossible to work with.
Exactly! Documents should be written for particular audiences who have particular purposes. The document that included everything would have enough information that you could just write a compiler to build the project, completely eliminating the coding phase. And it would be a terribly boring document that almost nobody would read.
The best resource I know of is the project guru.
Even better, you should arrange your project so that everybody has an opportunity to become a guru. Getting everybody (including the product manager) to work in a single project-focused team room is great for that. Also valuable are test-driven development and pair programming.
Eventually, the project will go dormant and the team will scatter. That's the time to write documentation.
The problem is when the prototype becomes the final product because TPTB don't want to waste money re-doing it.
That's indeed a huge problem. It's worth making the prototype look and act like a prototype. Fine ways to do that include Paper Prototyping, click-through mockups in plain HTML, and the Napkin Look-and-Feel. Also consider making the prototype obviously flawed in other ways. E.g., leaving stuff out, making graphics very low-res, adding long sleeps to make it slower, and so on.
There's a (American?) fable about a tortise and a hare racing.
Well, actually, they're a tiny bit older than that. They predate Columbus's discovery of America by about two thousand years.
Troll pts for that? I see we have a consultant mod in the house.
Or they could just be recognizing that some of the people here are consultants. For example, I do both contract development and consulting.
Of course, I think the common stereotype of consultants is often true, just like I think the stereotypes about programmers are often true. But the sterotypes are also often false. Stereotypes are like that.
Which is why acting as if a sterotype is a universal (e.g., "All women are... ") would be seen by some as trolling. Me, I think you were just being an insensitive clod; Slashdotters are all like that.
I think he's looking for the best way to get the point across.
I think the very best way is to tie it back to things the boss cares about: money and productivity.
Go through the report and come up with solutions that cover all the points, at least the ones that aren't bogus. Explain what each solution will cost (both in cash and in business impact), and what, in business terms, the benefits are.
If your instincts are right, your boss will say something like "Better security is well and good, but I'm not doubling the IT budget and inconveniencing our staff for so little improvement." And if it turns out there are some things that they're willing to pay extra for, then that's great: you get more budget and new toys.
Note that if they suggest you do more stuff without changing the budget, then you should be ready to say, "Oh, ok! Which things were you thinking of cutting? I recommend X, Y, and Z." Never let them get the idea that they can just heap unfunded mandates on you. That's not an option, just like haggling with the clerk at WalMart isn't an option. It's not that you refuse; it's just that it isn't an option.
But I don't get the sense he's [Linus] thought through the implications fully.
That's certainly plausible. And I'm sure McVoy's panic attack didn't make it any easier to think things through clearly.
Correct me if I am wrong--but Larry's oh-so public hissy fit is all anyone arguing the opposite side of the argument has.
Exactly true. We have partial stories from three different people, all of them with an interest in the outcome. I'd love to know what really happened. Often when I dig into disputes like this, all concerned have reasonable tales to tell, and it's not so much that one person is wrong and the other is right, as that they have different priorities and goals, and different opinions about what's really important.
(Or, at the very least, not *my* hyperbole.)
Heh. Well put. I look forward to Tridgell and OSDL speaking up on this!
Yes, but nobody would write a joke in Base 13. It just isn't funny.
But you shouldn't take that to mean that only jokes in base 10 are funny.
You know why programmers confuse Halloween and Christmas? Because OCT 31 = DEC 25.
And no, you're wrong, that really is funny.
I can't stand it when people act as though my belief in my own opinions is a negative thing (or even a potentially negative thing).
One would hope that you yourself believe your own opinions... in fact one might argue that the very definition implies it.
I try to take my opinions, like I take everybody else's, with a grain of salt. I generally avoid dealing with people who take their own opinions too seriously, as I find it makes getting at the truth unnecessarily arduous.
Take this case, for example. You said:
Tridge did not use BitKeeper **AT ALL**.
But your strong statement of a conclusion turns out to be backed up only by Tridgell's claim. Hopefully Tridgell is telling the truth, but he could also be lying or be mistaken. An all-caps-plus-double-bolding opinion expressed as a fact makes me suspect that, at least when you wrote the post I initially responded to, you were more intrested in winning an argument than you were in finding out the truth.
And hey, that's ok; on Slashdot, rants are accepted, and great ones get modded to the stars. But it's good to let people know whether you're ranting, and when you want to have reasonable discussions. Mixing the two doesn't work so well.
XCode and VC both have the good things and bad things.
Say, does XCode have any of the automated refactorings that something like IntelliJ's IDEA has? I'm completely hooked on those.
seriously people. mp3 is going to be around for a very long long time. why are you even bothering with any other format?
Aside from the good reasons others have mentioned, it's because there isn't just one MP3 format. Sure, anything can play most any MP3s, but there are many options for bit rate, and lots of different encoders with different encoding characteristics.
A friend of mine does exactly this (giant array with all CDs losslessly compressed), and then encoding what he needs when he needs it. That way when he's DJing or putting together a mix, he gets high quality. When he wants to squeeze a bunch of stuff on his iPod, he optimizes for space. And when the latest encoders improve things, he just writes a batch job and lets the box chug away for a couple of weeks.
"I'm ripping my entire audio collection to lossless audio files and I need a cheap large-volume storage solution...."
I'll be on vacation next week, so let me add one valuable reminder now.
As a habit now, any box I build has two drives in a RAID-1 mirror set; compared with the pain of recovering from a failure, the cost of the extra drive was good insurance. And I thought RAID-1 was fine; how likely was it that two drives would fail at the same time?
It turns out that if you buy the same model of drive at the same time, drive failures are much more correlated than random chance. Out of four drives I had bought at the same time, three of them failed within a two week period. (They were the now-infamous IBM Deskstars.) It made me very sad.
Now I'm inclined to mix drives from different manufacturers, and I like to keep a spare drive at hand, so that I can keep the high-risk period as small as possible. Or for RAID-5, I'm a big fan of the hot spare.
Everyone say after me: RMS was right.
You could look at it that way. But given that the problem only cropped up when someone, apparently out of fundamentalist software-must-be-free beliefs, started building a tool that Linus didn't want and asked him not to build, it's at least as reasonable to say that the real problem is with people starting holy wars.
the whole _concept_ of having a software company is to charge people for the software you develop
And this is not just the way that the companies making the software think. Consumer behavior around pricing is just as weird. I can't name the company, but there's a company that sells a regular version and an enterprise version of their product. The two versions are identical in functionality, but one costs ten times as much. A significant percentage of purchases are the enterprise version.
Similarly, some friends of mine make an on-line game. It's well-reviewed, really fun, and appeals to a broad range of people. But a lot of potential users won't take it seriously because you can just download the client. So they've given up arguing and are making a retail version in a pretty box. It's the same client you could download in ten minutes over your broadband connection, but all their research suggests that a big percentage of people will be much happier paying $25 for it than getting if for free.
And people do this in other realms, too. I know fuck-all about wine, and so when I'm faced with the store's wall of near-identical bottles, a good portion of my judgement is based on price. In asking around, it turns out most people do the same.
You seem a bit in a lather about this. I'm hoping you've cooled down a little by the time you read this, as I think you're not trying to look at this from McVoy's perspective. Note here that I don't really agree with McVoy, and I'll say more about that at the end.
McVoy has built what many people think is the best SCM system out there; certainly Linus thinks that. But BitMover, his company, is tiny compared with the other major SCM vendors, who include giants like Microsoft and IBM. Building a company is hard work, and I'm sure he feels very protective of it. Further, I imagine he thinks that his main edge over his competitors is his superior technology, which is embodied in his IP. (Certainly it's not his winning personality.)
Now I'm sure he thought he was doing something really good for the Linux kernel development community by giving them free BitKeeper licenses. Then, from out of the blue, somebody in that community starts trying to reverse-engineer his product. And even worse, it turns out to be a coworker of Linus's, somebody working for a company who has signed the BitKeeper license that specifically forbids that. Not only does McVoy have to worry about the IP threat, but he has to worry about somebody accidentally screwing up existing repositories as they hack their way to a working solution. To him, this would feel like a slap in the face after the time and money he's spent. He even tries to negotiate, and only gets more frustration out of it. Eventually he decides that he doesn't need the hassle, takes his ball and bat, and goes home.
Now personally, I think McVoy's a bit of a crackpot, and that his desire to control things so tightly is harming his company. But I never built a first-class SCM system from scratch, so it's not like I can offer Linus anything better.
OSDL "agreed" to not use BitKeeper for reverse engineering. Tridge did not use BitKeeper **AT ALL**.
Since you seem to be pretty sure of your opinion, perhaps you can explain how he did it? I'm not being snarky, by the way, I'd actually like to know. My presumption was that he had access to protocol captures because of his OSDL access.
Thanks for a reasonable and thoughtful contribution to a discussion that hasn't seen much of that. But I think you've missed something.
The made-up quote has the same gist, even if it's critically wrong in (a) the file format, and (b) the fact that Linus is talking about somebody else's beliefs, not his own. This gist, however, is clear that Linus believes roughly the same thing.
That's not the only possible interpretation. When Linus says he can't argue with that, I think all he means is that McVoy has a right to take that position, and that it's understandable and logically consistent. Linus didn't say that he shares McVoy's view or that McVoy's is the only reasonable position one could have on the topic of reverse engineering.
My guess is that Linus is perfectly fine with reverse-engineering file formats in general. I think his opposition to this particular case has nothing to do with high-minded princples of freedom and the legal right to reverse-engineer, but rather with pragmatic issues specific to his situation. Whatever Tridgell's intentions, he cause Linus (and presumably OSDL) a lot of trouble, and I think Linus wanted to avoid that trouble.