ZDnet needs to shut up about Linus
|
Author | Content |
---|---|
Grishnakh Jul 22, 2011 6:44 PM EDT |
What an utterly moronic article from the idiots at ZDnet, who consistently write drivel. The idiot author, who I'm sure is no programmer himself, compares a few bullet points about the Apple filesystem and Linux's current leading filesystem (which will be replaced soon by btrfs), and thinks that proves anything? There's a lot more to a FS than how large the files it stores are, and if it does or does not have journaling. Reiser3 has journaling too, but every comparison against other filesystems shows it to generally have very poor performance. How does HFS+ stack up? I don't know, but idiot Ken's article didn't address this. This Ken Hess moron reminds me of recruiters, who just look at a few keywords on someone's resume, and some keywords in a job description, match them up, and think that they're an expert on hiring and know whether a candidate is a good fit for a job or not. If he were a true journalist, he would have consulted other kernel and filesystem programmers, and asked them for their opinions. Linus is a kernel programmer himself; I'll give more weight to his opinion about a filesystem than someone who's only a journalist^Hwriter^Hhack, and not a technical worker or software engineer at all. |
khess Jul 22, 2011 7:18 PM EDT |
Dude, or dudette, get a frickin' grip. It's humor. And, I am a tech person. |
jackd2 Jul 22, 2011 9:41 PM EDT |
KHess, take a look in the mirror before you get bent out of shape. I honestly would not have known (part of?) the article was supposed to be humorous. Or maybe your retort is more humour, and I'm not getting this joke either? |
dinotrac Jul 22, 2011 11:24 PM EDT |
khess -- Well, you could answer 1 question: Was your intent to point out that most people don't share David Gerwitz's sense that 3 year old items are news? That part is pretty funny, I'll admit, but... This whole "techie" claim is weakened by your little feature listing that compares the HFS to ext4. It's not just that ext4 is but one of many file system choices available to Linux users, it's that bullet points are pretty useless in real technical evaluations. I could make a bullet-pointed feature list that makes a Chevy Cruze look better than an Audi R8, but execution matters. And -- while I agree that Linus has a, ummmm, propensity to say unfortunately harsh things now and then, when it comes to evaluating system level software, he has a certain credibility that you (or I ) don't. |
tracyanne Jul 23, 2011 12:26 AM EDT |
Quoting:Dude, or dudette, get a frickin' grip. It's humor. And, I am a tech person. Yeah but you are no comedian, at least not intentionally. |
djohnston Jul 23, 2011 1:25 AM EDT |
Poor Ken. He takes such heat over the flamebait articles he writes. |
gus3 Jul 23, 2011 6:59 AM EDT |
@dino, actually, Linus doesn't have that authority. What he knows about software security, he has learned in the school of hard knocks. Fortunately for the rest of us, the GPL will keep the code available to those who do know how to write secure code, so that they can find and fix the holes. |
dinotrac Jul 23, 2011 12:40 PM EDT |
@gus3 - Silly statement. Never mind that Linus did study computer science in college, and that is the original genesis of Linux, what have you got against the school of hard knocks? I would take Linux over Ken Hess anyday, and his real-world knowledge/vision over many others. I still remember when people were calling him backwards for building Linux on a monolithic kernel. Seems to work OK. The Hurd? Not so much. Good engineering triumphs, even over "suboptimal" theoretical choices. |
gus3 Jul 23, 2011 7:31 PM EDT |
I studied computer science in college, too. Big whoop. Secure programming isn't part of the curriculum, unless you count "use strncpy, memmove, and SSL". Really, that's what our students are learning about security in a typical CS course of study. Nothing about demonstrable correctness, nothing about how to audit code for security. And given some of the security lapses that have made it into Linux, including lots of information leaks, the record isn't stellar by a long shot. Oh, and the school of hard knocks is a mighty poor way to learn security. What good is the lesson, once the information is stolen? |
dinotrac Jul 23, 2011 11:16 PM EDT |
@gus3 -- The school of hard knocks is not the most pleasant way to learn anything, but it is your failures that teach you the most. You, I presume, are much too successful to have learned much. And, with all due respect to your awesome prowess, Linus ain't exactly some guy named Joe. He's got quite the resume -- not to mention circle of friends. Truth is, if you try to tell me Linus knows squat, I will accept that somebody almost certainly doesn't know what he thinks he does. Wouldn't be a Finn, though. |
Jose_X Jul 24, 2011 2:16 AM EDT |
>> And given some of the security lapses that have made it into Linux, including lots of information leaks, the record isn't stellar by a long shot. Microsoft employs many PHDs, yet I feel very confident saying, ..And given some of the security lapses that have made it into Windows and other Microsoft products, including lots of information leaks, the record isn't stellar by a long shot. So what was your point? |
gus3 Jul 24, 2011 7:20 AM EDT |
@Jose, fair point, but there's a philosophical difference. At M$, security has always been an afterthought. Privilege enforcement and management played second fiddle to "just make it run, so we can ship it." Once they did start using a kernel with real MAC, their customers found cold comfort in the PR blitz. "Trust us, we got it right this time!" Uh-huh. Sure. Unix and VMS at least had a MAC infrastructure in place, even if it wasn't perfect. But MAC is no substitute for secure code, as the Morris worm demonstrated. Then along comes some undergrad who says, "I can do that!" while publicly stating that he refuses to heed some very well-founded advice from those who have gone before. The next thing you know, a whole bunch of people and orgs have deployed his stuff, thinking it's trustworthy. Yet, they're no more qualified to reach that conclusion than he is. Thinking like a criminal, while working for the good guys, is an art that takes years to master. And criminals know it. Linus does not think like a criminal; if he ever did take a software security course, either the instructor was just as unqualified, or Linus didn't pay attention in class. |
jdixon Jul 24, 2011 9:44 AM EDT |
> Linus does not think like a criminal; If he did, he'd probably be working for Microsoft. 90% of people don't think like a criminal, and don't want to. That doesn't mean they can't do good work, even in security. |
nikkels Jul 24, 2011 10:45 AM EDT |
>>>>90% of people don't think like a criminal, and don't want to. Only 3 days ago I have seen the results of a limited poll. ( probably very limited ) Question was if it's ok that the government was corrupt. Answer : 32% said NO-- 68% said YES, if I get my part of the benefits So, you don't mind if I disagree with you ? Of course, this has nothing to do with computers , but still...... |
tuxchick Jul 24, 2011 11:49 AM EDT |
This 'article' doesn't even rise to the level of piffle. It's piffle lite. |
Jose_X Jul 24, 2011 12:53 PM EDT |
@gus3, Linus is not the only one creating Linux. There are many lines of code where the primary decision-maker for that code getting in was not Linus. We have people with different ideas about security contributing and making their case to other gate-keepers. The ultimate produce isn't just Linux either. It's a full distro with security concerns in many places. Let us also not forget that Red Hat (and others) does significant commercial activity with Linux with customers who do worry and check to see if they are being attacked. Red Hat contributes a lot. They have lots of weight. We also have to keep in mind we are not building with security as the highest of all concerns [think Alcatraz or where/how the President is protected in event of homeland attack.] The vanilla product does not trade off left and right for security. The PC hardware itself has security issues in the first place. Where security is more important and something like Linux is desired, there are patches out there to help add layers of security and there are security experts that you can contract to help you build up a full secure solution. The biggest threat is not within Linux kernel but in other layers above (including with human manipulation). Open source means that people can point out security problems at any time. Do you want to give a list of significant security problems with Linux that exist today and which are being ignored? See, that would be a problem. All of this said, I agree that the design is important, but am not in a position to specifically say that Linux has some significant flaw. I don't know Linux source very much nor am I a security expert. |
tuxchick Jul 24, 2011 6:21 PM EDT |
I got lost on the turnoff to the comments on security -- where did that come from? These meatless throwaway "articles" are annoying. Ten minutes' work and whee, some sucker actually coughs up a paycheck for it. There is no real story here, just another link in the...er... (LXer TOS looms)...circle of lazy "journalist" self-gratification. First the original interview: http://www.smh.com.au/news/technology/q-and-a-with-linus-tor... Quoting: Q. Do you have a favourite between Leopard and Vista? I wish the interviewer had followed up-- what is about Apple's filesystem that is so bad? Maybe he didn't have a chance to follow up. Maybe Linus saying "crap" was good enough. Then we get an 'article' dashed off from a Google alert that links to a 3-year old summary of the interview, http://www.zdnet.com/blog/government/linux-linus-calls-apple... Then we get today's content-free flamebait. This is what I loathe about tech "journalism." It's junk. It's a waste of space and an abuse of one of the most wonderful technologies in human history. After reading all these "articles" I'm still wondering what is wrong with the Apple filesystem? There are maybe four thousand words expended in these four articles, and not one single word of substance addressing this question, just a bunch of Beavis and Butthead-style "hur hur, Linus said crap." The original interviewer does some actual work, and then a succession of lazy ripoffs follow. Which is nothing new, journalism has always been like that, the Internet just makes it easier. |
gus3 Jul 24, 2011 10:34 PM EDT |
dinotrac wrote:The school of hard knocks is not the most pleasant way to learn anything, but it is your failures that teach you the most. Would you trust someone claiming to be a lawyer, who learned his practice that way? Or a structural engineer? Or a physician? When others' lives and fortunes depend on you knowing how to do your job, the school of hard knocks is not a good entry to put on a resume. Besides, we all know at least one person who stubbornly refuses to learn from his/her failures, always hoping for different results from the same choices. |
jdixon Jul 24, 2011 11:40 PM EDT |
> So, you don't mind if I disagree with you ? Not at all. But since almost all government are corrupt, I'd merely note that 68% of the respondents have a firm grasp of reality. > Would you trust someone claiming to be a lawyer, who learned his practice that way? It's only in modern times that a law degree was required to be a lawyer. For a long time all you had to do was pass the bar exam, and you could do that by reading law on your own. Some of the best lawyers in history were self educated. > Besides, we all know at least one person who stubbornly refuses to learn from his/her failures, always hoping for different results from the same choices. Sure. But that's not a description of Linus Torvalds. |
dinotrac Jul 26, 2011 6:17 PM EDT |
@gus3 - I would never trust anybody who was afraid to fess up to mistakes. My experience is that there are two kinds of people: Those who make mistakes, and... wait. Make that one kind of people. |
nikkels Jul 26, 2011 8:33 PM EDT |
Those who make mistakes, and... .....those who immediately start a party on your head |
patrokov Jul 27, 2011 12:26 AM EDT |
@gus That's how all surgeons and interventional cardiologists learn. It's estimated a cardiologist needs to perform more than a 100 catheterizations to BASIC competence. And it takes a lot more to get good. So I hope all of those "practice" patients didn't have any serious problems. Of course that's about as pertinent as your statements about Linus. |
Posting in this forum is limited to members of the group: [ForumMods, SITEADMINS, MEMBERS.]
Becoming a member of LXer is easy and free. Join Us!