Actually it's not a troll, and I didn't leave Apple out for any particular reason other than I don't believe their contributions back have been that substantial, although perhaps our definitions of "substantial" are simply at odds.
Ha-ha, I suppose it depends on what you consider "traditional". Sure, there's little likelihood of Linux displacing Windows on desktops, but it's doing just fine at replacing "traditional" UNIXen. Moreover, it's doing rather well in embedded systems.
As for whether or not the feedback loop exists, all you have to do is look to see how much code is being contributed by commercial concerns. Quite a lot, at this point.
Ask yourself this question as well: if it were not for the proliferation of UNIX softwares (as a result of the BSD forks), would Mr. Torvalds have created Linux?
Well, that is a rather complicated question, although I recall Torvalds saying he wrote UNIX because he couldn't get an affordable version for the PC. It is exceptionally likely that had BSD managed to be disassociated from AT&T a few years earlier Linux would never have happened. (And, in that world, it seems very likely that Microsoft would have won outright.)
But as for your intimation that UNIX was successful as a result of BSD forks, I would argue otherwise. In fact, only one BSD variant ever had significant commercial success (SunOS) and its success was short-lived (less than 10 years on the market). SysV was, however, astoundingly successful.
The reason why UNIX succeeded was entirely economic: You could buy it very cheaply relative to building your own. In the competitive workstation market (remember, UNIX never had more than a tiny niche on minis) that was clearly visible in the price of the machine. And as I am fond of saying, the cheapest thing that gets the job done wins.
You list a number of corporations that distributed BSD-forked software without contribution. I would love to do the same for GPL software, but unfortunately, that is a difficult proposition.
Whether or not they exist (I'm sure some do) is really not the point. The fact that there is a plethora of examples of corporations that did donate back to Linux, via the GPL, is proof that the feedback system works. Moreover, the GPL has been used, repeatedly, to enforce the feedback (eg in the case of Linksys).
Some of those donations have been very substantial, like the NUMA and logical volume support IBM donated. When was the last time someone donated code of that nature to BSD? I am pretty sure the answer to that is "never."
You seem to make the assumption that these same evil corporations that forked BSD would also fork and contribute to the same software if GPL-licensed.
I make no such assumption, as I believe the GPL does in fact scare corporations away. However it is clear by observation that it is the case that some corporations who did not donate code back to BSD-licensed codebases did do so for GPL-licensed codebases.
You missed the really big one, OpenOffice, although we all know that they did that solely to put a kink in Microsoft's cash cow. By and large if Sun has "open sourced" a significant body of code (like NFS) it was not under a BSD-style license, and certainly they never gave much in the way of code back to the BSD UNIX project despite obtaining very significant value from BSD UNIX code.
I don't mean to pick on Sun, simply to use them as an example of what naturally happens with a very loose license like BSD uses. If you don't have to give anything back, you most likely won't.
the whole point of ESR argument is that most companies don't like the viral nature of GPL so they'd perfer not to contribute at all but given a BSDed project they can contribute directly to that project and not worry about having to give up everything else they use it with. Everybody wins.
I'm well aware of his point, it has been made ad-infinitum over the last two decades. I've been in the middle of commercial products that had to deal with both licenses and, for sure, it's easier to deal with BSD than the twisty legalisms of the GPL.
I don't think anyone argues that the BSD license isn't preferable to businesses over GPL. And why would it not be? They literally get something for nothing. Everyone likes to get something for nothing!
The thing is, at some point the value of the codebase is greater than the cost of giving up some of your rights in order to use it. When that happens then the GPL has a major, major advantage in that it creates a feedback loop. The more code gets donated the more incentive there is to use it and as a result donate more code and make the code base more valuable.
Back ten years ago you could have argued that that loop wouldn't happen because of the unwillingness of people to give up their rights, and it would not have been possible to point to counterexamples. But Linux proves that there is a point at which the value overcomes the resistance.
No BSD-licensed product has ever had that happen, nor does it seem likely that it ever will happen -- there is no incentive whatsoever to feed code back, and plenty of counter-incentive.
Actually BSD is the project I use to show exactly the opposite. While it's true there have been many individuals who have contributed to BSD, many major corporations have taken very significant code out of it and given back... nothing.
Consider Sun Microsystems, whose SunOS operating system was based on BSD. What did they give back? Other than a few bug fixes early on, nothing.
Ultrix, from Digital Equipment, was BSD-based. Little to nothing came back to BSD from DEC.
Remember OSF/1, which was based on Mach/BSD? How much of their work went back? Next to nothing.
Microsoft used the BSD TCP stack as the basis of their TCP stack. What did they give back? Nothing.
FTP software based their whole product suite on the BSD codebase. How much came back? Nothing.
I don't know of any major corporation which has made significant donations back to the BSD core. There may be the rare exception, but the bulk of corporate back-donations has been some bug fixes. That has left the development almost entirely to individual developers or very small groups, and thereby limited how much could be done.
Lots of people think of the GPL as a "communist" license, but in fact it is BSD that is the free-for-all. The BSD license attaches no value to what it is licensing, and as a result you a software "tragedy of the commons" where everyone is happy to use it but almost nobody ever gives anything back. I know that there are going to be people who vehemently disagree with what I'm going to say, but: It has been my observation that the BSD source base has been relatively stagnant over more than a decade. If you look at what a modern BSD provides and compare it to what BSD 4.3 provided you'll find little that is new. A similar comparison with any major commercial UNIX will yield a great many such features (like working SMP support, journalled filesystems, NUMA support, logical volume management, realtime support, etc).
The GPL, on the other hand, leverages the fact that the source base is valuable. It is not a "give away" as so many people claim but rather an intellectual property trade very much like the patent sharing agreements so common in the proprietary world. While businesses would rather get something for nothing, if what they're getting in trade is valuable enough it is an incentive to give up some of their own rights.
If you think of the GPL as an intellectual property collective agreement you have the right idea. The thing about that kind of agreement is that the more IP that is covered by it the more valuable the collective becomes -- and therefore the more likely others are to join it.
In Linux' case the source base is exceptionally valuable at this point, worth literally billions of dollars, and for the better part of a decade has been receiving significant code donations from corporations. Remember the list of features modern UNIXen have that BSD doesn't? Did you notice how many of them Linux does support? All of them. For something like a decade corporations have been making major code donations back to the Linux codebase and it has advanced tremendously as a result. While Linux certainly has its rough edges it has seriously outgrown its tinkerer beginnings.
So Raymond could not be more wrong about this point. Oh, I agree that the development structure that Torvalds set up was a principal contributor to its success. To be sure, one of the major limitations in the BSD codebase has been the reluctance of the BSD principals to accept code they didn't write. But BSD has branched enough times that it has also seen conditions similar to what Linux enjoyed and it still never turned the corner.
What made Linux win was simply that large corporations had to give to get, and the more times that happens the more likely it becomes.
This is an old idea, but it's good to see it getting some traction because I think it's a good one. But if you buy one of these things be aware that you absolutely do not want to use it for paging; flash memory has a finite cycle count, somewhere in the area of a few hundred thousand writes. A system that is paging heavily can burn up flash memory in a matter of hours.
Is it amusing because you believe Apple locked everyone else of the market, despite the large number of competitive devices and services that existed both before and after Apple hit the scene? Or because I believe, based on the large number of competitive devices I've owned over the years, that Apple's product is generally quite a bit superior?
I'm curious.
I can't help but be amused at this when my iPod is stuffed to the max, plus a whole lot, and I've never made a purchase on iTunes. I did have to rip every song on all of my hundreds of CDs, but that's only because the recording industry would not allow anyone to sell me the music in a form that I could use on any extant player.
Jobs was unique in finding a way to make the harsh restrictions placed on downloaded music by the recording industry palatable to a wide audience and profitable to boot.
Seeing as Apple took the risk and won, I think it's unreasonable to ask them to give up the fruits of their labors. As an Apple shareholder I'd hold Jobs culpable if he ever did such a thing. I say to Jobs: "Milk it for all it's worth." Especially since there are plenty of competitors out there to keep him honest. The iPod doesn't have a monopoly because Apple locked everyone else out of the market, ala Microsoft, it has one because it's better.
If you don't like the fact that you can't play your Windows Media songs on the iPod, buy a different player... or do what I do and buy the physical CD and convert it into whatever format you prefer. I get my CDs primarily from Amazon.com, but never from Apple.
And Hillary, if you don't like the myriad proprietary forms of DRM on downloaded music, consider the fact that it's your fault it's there in the first place.
I also know alot of excellent grad students, also out of top schools, who have to settle for intern like positions. They are so overqualified, companies seriously don't know how to fit them in. Companies want young guys coming in fixing bugs, not architecting major projects.
Not to put too fine a point on it, but almost nobody coming out of school is qualified to do much at all. Schools rarely teach the things you most need to know in business -- even simple things like "using a version control system" and "scheduling", to say nothing of making significant changes to multimillion-line codebases without busting things.
It used to be much worse since even a decade and a half ago even the basic software tools used by business were typically not the tools students would have been exposed to. Thankfully broad availability of Windows and Linux and related tools has at least helped in that regard.
But strong knowledge of the kinds of things you find in a university environment is still not proper preparation for an architect role in a major project where you'll likely be piecing together a variety of proprietary technologies, many of which a university couldn't realistically afford to expose you to. So those graduates are not "overqualified" by any stretch of the imagination.
So business does more of an apprenticeship kind of thing: You come in working on shit projects and if you do well you'll get more and more interesting stuff to do -- and in the process of doing those crappy tasks you'll learn application structure and process that aren't taught at university.
How fast the move from lousy stuff to interesting stuff happens depends a lot on the kind of company you work for, but it's been my experience that the talented people shine so bright that they're hard to miss. So long as they aren't stuck-up assholes who are hard to work with (unfortunately you get a lot of that right out of school, especially from postgrads, until they figure out they don't know everything) people want to pull them into projects. There's more work to do than people to do it, always, in software.
If I were to give advice to an upcoming graduate, my advice would be to look at the company rather than the job. Good-paying jobs right out of school are probably going to be with larger companies who need to attract talent with money because the work is no fun. Challenging, fun jobs tend to come with smaller companies who don't have a lot of money and attract talent by doing cool things -- and these are the same companies that will advance you rapidly if you're capable because they can't afford to have talent sitting around doing makework.
The drop in CS degrees is coming about because people no longer believe it's easy money. But, really, it never was easy money, no matter the impression a lot of people got during the tech boom.
I dunno about scanning irises, but I do know that a lot of people are considering using the RF transmitter that almost everyone in the US carries (aka their cellphone) to target advertising and sales at them.
It's been possible for a few years. The "hard" part is associating the cell phone's ID to the person, but the second you run a credit card through the cash register....
I'd return the game to the manufacturer and tell them that was not one of the requirements on the outside of the box and you do not have access to play the game under an admin account. There's no reason a game should have free reign of a system.
I ran into this issue when I was setting up a computer for my 18 month old daughter. Most of her games required administrator privilege to run. I went to the point of calling Disney, who manufactured a couple of the games, to find out how to get it to work as a non-administrator account. "You can't" was their response. There was no plan to fix that. "You can return the software" they told me, except you can't because the retailers won't take it back if it has been opened and Disney had no process for sending it back to them.
So I gave her Admin privileges. The only upside to this is that her machine is hers alone so if it gets toasted I can just regen it without causing undue pain to the rest of us.
I know DLPs flicker, but I can't see it -- whereas I have no problem seeing flicker in most CRTs. The only issue I've seen with DLPs is color breakup and even that is rare.
As for LCDs, you're mistaken -- if it's analog (and most still are) then a poor quality signal decoder can result in bleed between pixels, which tends to make them look blurry. Another possibility when using low-quality decoders, or refresh frequencies higher than the decoder can tolerate, is pixel shimmer.
I can't tell you which employer it is, but the brand of LCD is CTX. Or, as I like to call them, POS. They stopped buying them shortly after I got mine, and now buy nice Sonys instead.
I don't know about everyone else, but I switched from CRTs to other technologies (LCD, DLP) as fast as I could not because LCDs and whatnot are smaller (although that was nice) but because they don't flicker and are very sharp in comparison.
Unfortunately my employer has found a source of LCD panels that are blurry, but at least they don't flicker.
To my knowledge, you cannot share a DRM'ed Ebook with anyone else. It is keyed specifically to your activation account on both MS reader and Adobe reader.
That's true of those DRM formats, and is why I prefer not to buy content in them, but it isn't true of all DRM formats. I mentioned specifically the Palm Reader format, which doesn't work that way at all and actually allows lending.
For my purposes if I have to buy a DRMed book I buy it in Palm Reader format if possible, Mobipocket if not, or not at all if I have to use the Microsoft or Adobe formats.
The other problem is that both of those readers also support expiration on the books. If you read the fine print on the ebooks you purchase, you will see that they will expire after a period of time and you will no longer be able to access the file.
I know that's possible for them to do, but I've never actually seen anyone use that capability. Have you? I have a hundred-odd ebooks purchased since 1997, in three different DRM formats, and I can still read all of them. There was a point where Palm changed the encryption in their reader ('round about 2000 I think), and I worried that all the books would be unreadable, but the vendor was happy to re-encode them.
I am not against DRM and I want to see eBooks take off more. But the current implemetation reeks of greed as the publishing house try to garner even more control over their products than they ever had in the past.
I'm not sure it's entirely greed, I think there's more paranoia in it than greed, but certainly some of the ebook prices seem rather high. On the other hand it's still very much an early adopter technology; they don't sell a lot of ebooks yet.
For my purposes I don't much care if they lock it down really tightly -- but if they want me to buy it, they need to price it accordingly. Give me a novel for $1 and I'll consider it a read-and-throw-away experience. But if you're going to sell it to me for $20 (many best-seller ebooks are priced that high!) then you better give me flexibility. Many books available in the Microsoft and Adobe formats are priced high and yet highly restricted. So I don't buy them, and I don't find it peculiar that they have had poor uptake.
You can install the reader, but you will be disappointed when you find that you cannot open your Adobe ebooks in the Adobe reader on your Pocket PC.
That's what I get for believing their website I guess, it sure seems like ebooks are supported. I guess that makes my argument as to why those early efforts failed so badly even stronger:-).
As far as DRM making things more accessable? What planet do you live on anyway?
I did not intend to say that ebook content, and particularly DRMed ebook content, was generally more accessible than paper. I pointed out that there is one mode in which ebooks allow accessibility that is not possible with paper.
If I have a book I can lend it to at most one person at a time, and I can't read it while they have it. With ebooks, even with most of the DRM systems out there, I can "lend" it to multiple people simultaneously. In that respect even DRMed content is more "accessible" than paper.
How flexible the content is to copying like this depends a lot on the DRM format. The least restrictive I've seen is that of the Palm Reader where you can beam the book to whoever you want, open it in the reader, and type in your credit card number to unlock it. I can therefore "lend" it to as many people as I want subject to having to type the key. They can't lend it to anyone else, though, unless I'm willing to give them my credit card number (fat chance). This is a very reasonable DRM system, the only issue I have with it is that I have to use their reader -- which while available on a fairly wide range of systems has not been universally available on even the ones I personally use.
But even much more restrictive DRM systems can be made to work like this. It can be managed with Mobipocket by registering their device on your account and re-downloading, although that has a number of bad side effects (like further restricting how many devices you yourself can ever use with the content). Many DRM systems similarly allow limited copies or limited generations.
Generally speaking I agree with you that I'd prefer no DRM at all, most especially because DRM systems tend to limit the longevity of the content for my own use. But there do exist DRM systems where the DRM exists and yet still allows accessibility modes that paper does not.
I don't like being forced to use a non-MS platform to read books (as far as PDA's go, Adobe ebooks cannot be read on MS OS PDAs)
You are mistaken, Adobe ebooks can be read on both MS and Palm PDAs as well as Windows and Mac PCs. In terms of document portability it's rivalled only by the Palm Reader. (Mobipocket actually runs on a wider variety of platforms but does not support Macs.) It's too bad its reader sucks so bad on some devices.
Just make the new formats as accessable as the previous and you won't run into the same problems as the music industry.
The problem is that even with DRM it is usually more accessible than paper in at least one way: The ability for multiple copies to exist simultaneously. That is especially problematic to publishers because books tend to be faddish, with lots of people reading the same book at once. Since the only thing they can limit is how many copies get made that's what they do, limiting not only copy breadth but also depth.
I note that serial DRM is not impossible or impractical in some cases. Libraries, for instance, can create versions of the documents that time out at the end of the lending period. If I had access to such a technology I might be happier about DRMed purchases.
I've been using DRM'ed ebooks for something like seven years now. Long enough to have tried a variety of different formats. The thing I keep coming back to is that I don't so much mind DRM so long as it's DRM where I am allowed to copy between my own devices an unlimited number of times.
Perhaps the simplest, most effective DRM I've seen is the one used in the Palm Reader where your credit card number is the key. There is an inherent tendency not to give that out willy-nilly, yet nothing stops you from moving it around other than not wanting to give out the key.
That kind of DRM is effective for the publisher and yet loose enough that I can move the content around on my own devices without restriction.
I have also purchased books with the Mobipocket DRM. I don't like it because it restricts the number of devices I can transfer it to. This is problematic not because I want to run it on a lot of devices in parallel, but instead because the reader device itself is somewhat ephemeral. I have had six different PDAs since I bought my first ebook back in 1997 and my current PDA will likely need replacement within a year or two just from wear-and-tear. With device-locked DRM mechanisms the content becomes inaccessible once you hit the device maximum even if some of the devices no longer exist. This I don't like.
Once I purchased a book in Adobe's ebook DRM format because no other format was available. That pointed out the most serious weakness in DRM: It only worked with Adobe's reader, and Adobe's reader is almost unusably horrible. To describe all the things it does wrong would take a whole article in and of itself, starting from an unreliable download process to a very time consuming transfer process to reader display that was so nonsensical as to make you wonder what they were smoking when they let it out the door. On top of that Adobe has even more restrictive device locking than Mobipocket. I hated the experience so much that I will never, ever buy another book in that format no matter its price.
That experience also led me to wonder if part of the problem with the acceptance of ebooks was that the mainstream businesses (B&N, Amazon) who offered ebooks only did it in two formats: Adobe's and Microsoft's. Adobe's has the benefit of working across a fairly wide range of devices but is so horrible that I just don't want to use it. Microsoft's is locked to a very limited set of Windows devices. I don't know about you, but I haven't found anyone who likes reading novels on their PC or laptop, so that pretty much limited the market for Microsoft DRMed books to PocketPC devices. A few years ago those were few and far between and even today that means spending several hundred dollars (whereas a PalmOS PDA can be had for under $100) and has only a few hours of battery life. If that's what my DRM choices look like I'll opt out entirely thanks.
I've also considered a number of dedicated ebook readers, but those have historically made the PocketPC scenario look good. Until a few weeks ago I'd never seen a dedicated ebook reader for less than $300. Today you can buy one for about $100 from fictionwise.com, a big improvement, but not enough to work around the other huge problem with these readers: Document portability. I have probably about a hundred ebooks at this point, in something like four different document formats. Not even one of those formats is supported by any of the dedicated ebook readers. Not Adobe. Not Mobipocket. Not Palm Reader. Not Palm Doc. I'm supposed to pay money for that?
If given the choice I would prefer no DRM if only because that gives me greater flexibility in which software I can use to read the books (or music, in the case of something like iTunes). But if you're going to use DRM, and frankly I can see why many publishers demand it, there needs to be a tradeoff between what is good for the publisher and whether or not the DRMed document will be long-term usable by the consumer. I worry about stuff purchased v
My point is that normal users typically have write permissions to that directory, one way or another.
I just checked the box I'm on (a generic WinXP Pro install) and found that c:\windows is writable by "administrators" and "power users". The former is appropriate, the latter isn't, but the whole thing is rendered moot by the fact that the accounts are, by default, created with administrator privileges.
That's largely of necessity, I realize. On one of my home XP boxes I decided that my 2 year old daughter's account really shouldn't be privileged, so I didn't make it so. The result? Nearly all of her children's games failed to operate. When I called vendors about that, I was told that I'd just have to give the account the necessary privileges. (Can't return the software, of course, nobody allows software returns.)
So: We have a system that, if configured securely, doesn't work very well -- and if configured so it works, is so wide open that any little application error can lead to a compromised system.
It's a disaster and the only solution to it is going to be to have Microsoft turn the security way up by default so the software vendors are forced to write their code accordingly. Like, say, every other major OS out there.
The transition is going to suck, but until it's made Windows is going to remain a really easy target.
This is kind of ridiculous. Oh, sure, malware on OS X is possible and perhaps even really growing in numbers. But the problem is not and cannot be anywhere near as severe as Windows because Apple, like all the other UNIX vendors, ships their systems in a (reasonably) secure state by default.
The malware problem on Windows is not primarily the result of the system's popularity, no matter how many times Microsoft claims that is so. Early attacks on the Internet did not target the most popular system; rather, the most attacks have always targetted the easiest systems to crack. That started out with SunOS and, by the mid-90s, was Linux. (If you think Windows has much better penetration that Linux today, just think how much more lopsided the numbers were in 1995-2000 when Linux was the most popular target.) These days Windows systems are easiest by far because at this point they are the only systems which ship without basic filesystem protections (now that it finally has a halfway decent firewall, a mere five years after everyone else).
If Windows had basic filesystem protection enabled by default on all critical filesystem areas, mandated nonprivileged user accounts, and an installer that required a password, suddenly Windows wouldn't get infected every time you sneezed in its general direction.
Maybe the future will prove me wrong but I will be very surprised to find OS X malware become a serious problem no matter how popular the OS gets. I don't suspect that its users are any smarter, but the barriers are a lot higher.
...until the standard configuration does not give (or applications require) normal users to run as administrators, or leave the filesystem and registry wide open to modification.
So long as installers run without requiring passwords, and I have to give my daughter administrator privileges to run Disney games, Windows is in for a lot of hurt in the security domain because there's really no way to control what users, and by proxy the programs they run, muck with.
I mean, it's so bad right now that whole markets spawned to supply band-aids for the lack of basic protections (anti-virus, anti-spyware), and to rebuild broken systems as quickly as possible (ghost). That's pathetic, particularly since Microsoft had the ability to do a much better job of securing their systems since the release of Windows NT in 1993, and it's been mainstream since XP. It's not that they couldn't do it, it's that they didn't.
If driver support is there, Linux is much more "integrated" and easier than Windows because everything works right out of the box, while you have to feed driver CDs when installing Windows. Of course if there is no Windows-driver all "easy support" goes to hell, because there is no community to create it.
I was thinking of the applications on top; so far as I'm concerned both systems have very good device driver support.
And.NET is just Java with more language bindings and less platform support.
That's very true; after Sun sued them, Microsoft took their VM, extended it slightly, transmogrified the libraries slightly, added a lot of new features (particularly around network protocol and XML support), and wrote new front-ends for it.
It's kind of funny to me that Sun's lawsuit ultimately hurt them more than Microsoft. But then again Sun never seemed to have figured out that Microsoft was not their enemy, Intel was. If anything Microsoft did them quite the favor by proving either unwilling or unable to make Windows scale well enough to allow Dell et al to compete head-to-head.
Anyway the one place where Microsoft just did a better job than Sun and all the rest is in the IDE. It really is a nice IDE, at least for small to medium size applications -- point it at a web service and it generates the code; tell it to deploy to a web server and next thing you know your application is up and running. It's neat and clean and easy.
I spent eight of the last nine years writing Java code, and I'm here to tell you: Microsoft did a better job in a number of places. In particular, J2EE sucks goat balls. It's almost like they went out of their way to make it as painful as possible to use even for the simplest things. Microsoft didn't try to do all the things in J2EE, they just tried to do a really good job at the things people do all the time (to whit: Writing web pages and integrating them with databases). Not that we didn't see the same kind of stuff in Java years earlier, but that stuff unfortunately didn't become the standard.
But whatever, it is a fact that Microsoft has at least the perception of easier configuration and better integration. Fact too, sometimes. And that alone will make them a certain number of sales.
If your law applies (and I think it does) then commercial software will be finding it harder to battle against Open Source Software.
Indeed. In order for commercial software to compete in the long term it must provide significant value-add, and even then its pricing will be depressed. You mention IIS as being a negative, but I disagree. There is quite a tools ecosystem built around it, and it's much easier to set up and work with than Apache (I've run both, but I'm not crazy enough to expose IIS to the net at large).
But while everyone is looking at Microsoft vs. Linux now, that's not going to be the only big name to see open source significantly impact their business.
Pay attention to Oracle vs. MySQL. Nobody's really going to start talking about that until 2007 when you'll start seeing significant vendor support for it (this in response to developers using it as a cheap and lightweight alternative). Oracle will have to react in the 2009-2010 timeframe and their financials will start being negatively impacted as soon as 2013 and no later than 2015.
The Mozilla effort is going to bear significant fruit, too, although not so much on desktops -- unless we see Firefox get bundled by major PC vendors, which I would bet against. I figure 10% maximum desktop penetration on PCs, nearly 100% on Linux, and somewhere around 20-30% on Macs (less if Safari gets better).
The real win is going to be outside of desktops on more purposed devices, especially media devices. In fact, these will be where open source (especially Linux) makes its biggest impact -- royalty free is too big an advantage for those vendors to pass up.
The really good part of this from the consumer's point of view is that if it's running Linux, you can get at least some of the device's source code. That will lead to a lot of third-party product enhancements (ala Tivo hacks and Sveasoft).
Outside of infrastructure code like this, though, I think open source will have marginal impact. You need a certain critical mass of users and developers to sustain and nourish it, and that's hard to do as you get more vertical. Still, we're talking change-the-world market penetration in a number of horizontals. Not half bad results stemming from the ravings of a greasy hacker.
Slashdot Mirror
That, my friend, is not what I would call it. My guess is that you never actually tried to use it under stress.
As for whether or not the feedback loop exists, all you have to do is look to see how much code is being contributed by commercial concerns. Quite a lot, at this point.
Well, that is a rather complicated question, although I recall Torvalds saying he wrote UNIX because he couldn't get an affordable version for the PC. It is exceptionally likely that had BSD managed to be disassociated from AT&T a few years earlier Linux would never have happened. (And, in that world, it seems very likely that Microsoft would have won outright.)
But as for your intimation that UNIX was successful as a result of BSD forks, I would argue otherwise. In fact, only one BSD variant ever had significant commercial success (SunOS) and its success was short-lived (less than 10 years on the market). SysV was, however, astoundingly successful.
The reason why UNIX succeeded was entirely economic: You could buy it very cheaply relative to building your own. In the competitive workstation market (remember, UNIX never had more than a tiny niche on minis) that was clearly visible in the price of the machine. And as I am fond of saying, the cheapest thing that gets the job done wins.
You list a number of corporations that distributed BSD-forked software without contribution. I would love to do the same for GPL software, but unfortunately, that is a difficult proposition.
Whether or not they exist (I'm sure some do) is really not the point. The fact that there is a plethora of examples of corporations that did donate back to Linux, via the GPL, is proof that the feedback system works. Moreover, the GPL has been used, repeatedly, to enforce the feedback (eg in the case of Linksys).
Some of those donations have been very substantial, like the NUMA and logical volume support IBM donated. When was the last time someone donated code of that nature to BSD? I am pretty sure the answer to that is "never."
You seem to make the assumption that these same evil corporations that forked BSD would also fork and contribute to the same software if GPL-licensed.
I make no such assumption, as I believe the GPL does in fact scare corporations away. However it is clear by observation that it is the case that some corporations who did not donate code back to BSD-licensed codebases did do so for GPL-licensed codebases.
You missed the really big one, OpenOffice, although we all know that they did that solely to put a kink in Microsoft's cash cow. By and large if Sun has "open sourced" a significant body of code (like NFS) it was not under a BSD-style license, and certainly they never gave much in the way of code back to the BSD UNIX project despite obtaining very significant value from BSD UNIX code.
I don't mean to pick on Sun, simply to use them as an example of what naturally happens with a very loose license like BSD uses. If you don't have to give anything back, you most likely won't.
I'm well aware of his point, it has been made ad-infinitum over the last two decades. I've been in the middle of commercial products that had to deal with both licenses and, for sure, it's easier to deal with BSD than the twisty legalisms of the GPL.
I don't think anyone argues that the BSD license isn't preferable to businesses over GPL. And why would it not be? They literally get something for nothing. Everyone likes to get something for nothing!
The thing is, at some point the value of the codebase is greater than the cost of giving up some of your rights in order to use it. When that happens then the GPL has a major, major advantage in that it creates a feedback loop. The more code gets donated the more incentive there is to use it and as a result donate more code and make the code base more valuable.
Back ten years ago you could have argued that that loop wouldn't happen because of the unwillingness of people to give up their rights, and it would not have been possible to point to counterexamples. But Linux proves that there is a point at which the value overcomes the resistance.
No BSD-licensed product has ever had that happen, nor does it seem likely that it ever will happen -- there is no incentive whatsoever to feed code back, and plenty of counter-incentive.
Consider Sun Microsystems, whose SunOS operating system was based on BSD. What did they give back? Other than a few bug fixes early on, nothing.
Ultrix, from Digital Equipment, was BSD-based. Little to nothing came back to BSD from DEC.
Remember OSF/1, which was based on Mach/BSD? How much of their work went back? Next to nothing.
Microsoft used the BSD TCP stack as the basis of their TCP stack. What did they give back? Nothing.
FTP software based their whole product suite on the BSD codebase. How much came back? Nothing.
I don't know of any major corporation which has made significant donations back to the BSD core. There may be the rare exception, but the bulk of corporate back-donations has been some bug fixes. That has left the development almost entirely to individual developers or very small groups, and thereby limited how much could be done.
Lots of people think of the GPL as a "communist" license, but in fact it is BSD that is the free-for-all. The BSD license attaches no value to what it is licensing, and as a result you a software "tragedy of the commons" where everyone is happy to use it but almost nobody ever gives anything back. I know that there are going to be people who vehemently disagree with what I'm going to say, but: It has been my observation that the BSD source base has been relatively stagnant over more than a decade. If you look at what a modern BSD provides and compare it to what BSD 4.3 provided you'll find little that is new. A similar comparison with any major commercial UNIX will yield a great many such features (like working SMP support, journalled filesystems, NUMA support, logical volume management, realtime support, etc).
The GPL, on the other hand, leverages the fact that the source base is valuable. It is not a "give away" as so many people claim but rather an intellectual property trade very much like the patent sharing agreements so common in the proprietary world. While businesses would rather get something for nothing, if what they're getting in trade is valuable enough it is an incentive to give up some of their own rights.
If you think of the GPL as an intellectual property collective agreement you have the right idea. The thing about that kind of agreement is that the more IP that is covered by it the more valuable the collective becomes -- and therefore the more likely others are to join it.
In Linux' case the source base is exceptionally valuable at this point, worth literally billions of dollars, and for the better part of a decade has been receiving significant code donations from corporations. Remember the list of features modern UNIXen have that BSD doesn't? Did you notice how many of them Linux does support? All of them. For something like a decade corporations have been making major code donations back to the Linux codebase and it has advanced tremendously as a result. While Linux certainly has its rough edges it has seriously outgrown its tinkerer beginnings.
So Raymond could not be more wrong about this point. Oh, I agree that the development structure that Torvalds set up was a principal contributor to its success. To be sure, one of the major limitations in the BSD codebase has been the reluctance of the BSD principals to accept code they didn't write. But BSD has branched enough times that it has also seen conditions similar to what Linux enjoyed and it still never turned the corner.
What made Linux win was simply that large corporations had to give to get, and the more times that happens the more likely it becomes.
Is it amusing because you believe Apple locked everyone else of the market, despite the large number of competitive devices and services that existed both before and after Apple hit the scene? Or because I believe, based on the large number of competitive devices I've owned over the years, that Apple's product is generally quite a bit superior? I'm curious.
Jobs was unique in finding a way to make the harsh restrictions placed on downloaded music by the recording industry palatable to a wide audience and profitable to boot.
Seeing as Apple took the risk and won, I think it's unreasonable to ask them to give up the fruits of their labors. As an Apple shareholder I'd hold Jobs culpable if he ever did such a thing. I say to Jobs: "Milk it for all it's worth." Especially since there are plenty of competitors out there to keep him honest. The iPod doesn't have a monopoly because Apple locked everyone else out of the market, ala Microsoft, it has one because it's better.
If you don't like the fact that you can't play your Windows Media songs on the iPod, buy a different player ... or do what I do and buy the physical CD and convert it into whatever format you prefer. I get my CDs primarily from Amazon.com, but never from Apple.
And Hillary, if you don't like the myriad proprietary forms of DRM on downloaded music, consider the fact that it's your fault it's there in the first place.
Not to put too fine a point on it, but almost nobody coming out of school is qualified to do much at all. Schools rarely teach the things you most need to know in business -- even simple things like "using a version control system" and "scheduling", to say nothing of making significant changes to multimillion-line codebases without busting things.
It used to be much worse since even a decade and a half ago even the basic software tools used by business were typically not the tools students would have been exposed to. Thankfully broad availability of Windows and Linux and related tools has at least helped in that regard.
But strong knowledge of the kinds of things you find in a university environment is still not proper preparation for an architect role in a major project where you'll likely be piecing together a variety of proprietary technologies, many of which a university couldn't realistically afford to expose you to. So those graduates are not "overqualified" by any stretch of the imagination.
So business does more of an apprenticeship kind of thing: You come in working on shit projects and if you do well you'll get more and more interesting stuff to do -- and in the process of doing those crappy tasks you'll learn application structure and process that aren't taught at university.
How fast the move from lousy stuff to interesting stuff happens depends a lot on the kind of company you work for, but it's been my experience that the talented people shine so bright that they're hard to miss. So long as they aren't stuck-up assholes who are hard to work with (unfortunately you get a lot of that right out of school, especially from postgrads, until they figure out they don't know everything) people want to pull them into projects. There's more work to do than people to do it, always, in software.
If I were to give advice to an upcoming graduate, my advice would be to look at the company rather than the job. Good-paying jobs right out of school are probably going to be with larger companies who need to attract talent with money because the work is no fun. Challenging, fun jobs tend to come with smaller companies who don't have a lot of money and attract talent by doing cool things -- and these are the same companies that will advance you rapidly if you're capable because they can't afford to have talent sitting around doing makework.
The drop in CS degrees is coming about because people no longer believe it's easy money. But, really, it never was easy money, no matter the impression a lot of people got during the tech boom.
It's been possible for a few years. The "hard" part is associating the cell phone's ID to the person, but the second you run a credit card through the cash register....
I ran into this issue when I was setting up a computer for my 18 month old daughter. Most of her games required administrator privilege to run. I went to the point of calling Disney, who manufactured a couple of the games, to find out how to get it to work as a non-administrator account. "You can't" was their response. There was no plan to fix that. "You can return the software" they told me, except you can't because the retailers won't take it back if it has been opened and Disney had no process for sending it back to them.
So I gave her Admin privileges. The only upside to this is that her machine is hers alone so if it gets toasted I can just regen it without causing undue pain to the rest of us.
As for LCDs, you're mistaken -- if it's analog (and most still are) then a poor quality signal decoder can result in bleed between pixels, which tends to make them look blurry. Another possibility when using low-quality decoders, or refresh frequencies higher than the decoder can tolerate, is pixel shimmer.
I can't tell you which employer it is, but the brand of LCD is CTX. Or, as I like to call them, POS. They stopped buying them shortly after I got mine, and now buy nice Sonys instead.
Unfortunately my employer has found a source of LCD panels that are blurry, but at least they don't flicker.
That's true of those DRM formats, and is why I prefer not to buy content in them, but it isn't true of all DRM formats. I mentioned specifically the Palm Reader format, which doesn't work that way at all and actually allows lending.
For my purposes if I have to buy a DRMed book I buy it in Palm Reader format if possible, Mobipocket if not, or not at all if I have to use the Microsoft or Adobe formats.
The other problem is that both of those readers also support expiration on the books. If you read the fine print on the ebooks you purchase, you will see that they will expire after a period of time and you will no longer be able to access the file.
I know that's possible for them to do, but I've never actually seen anyone use that capability. Have you? I have a hundred-odd ebooks purchased since 1997, in three different DRM formats, and I can still read all of them. There was a point where Palm changed the encryption in their reader ('round about 2000 I think), and I worried that all the books would be unreadable, but the vendor was happy to re-encode them.
I am not against DRM and I want to see eBooks take off more. But the current implemetation reeks of greed as the publishing house try to garner even more control over their products than they ever had in the past.
I'm not sure it's entirely greed, I think there's more paranoia in it than greed, but certainly some of the ebook prices seem rather high. On the other hand it's still very much an early adopter technology; they don't sell a lot of ebooks yet.
For my purposes I don't much care if they lock it down really tightly -- but if they want me to buy it, they need to price it accordingly. Give me a novel for $1 and I'll consider it a read-and-throw-away experience. But if you're going to sell it to me for $20 (many best-seller ebooks are priced that high!) then you better give me flexibility. Many books available in the Microsoft and Adobe formats are priced high and yet highly restricted. So I don't buy them, and I don't find it peculiar that they have had poor uptake.
That's what I get for believing their website I guess, it sure seems like ebooks are supported. I guess that makes my argument as to why those early efforts failed so badly even stronger :-).
As far as DRM making things more accessable? What planet do you live on anyway?
I did not intend to say that ebook content, and particularly DRMed ebook content, was generally more accessible than paper. I pointed out that there is one mode in which ebooks allow accessibility that is not possible with paper.
If I have a book I can lend it to at most one person at a time, and I can't read it while they have it. With ebooks, even with most of the DRM systems out there, I can "lend" it to multiple people simultaneously. In that respect even DRMed content is more "accessible" than paper.
How flexible the content is to copying like this depends a lot on the DRM format. The least restrictive I've seen is that of the Palm Reader where you can beam the book to whoever you want, open it in the reader, and type in your credit card number to unlock it. I can therefore "lend" it to as many people as I want subject to having to type the key. They can't lend it to anyone else, though, unless I'm willing to give them my credit card number (fat chance). This is a very reasonable DRM system, the only issue I have with it is that I have to use their reader -- which while available on a fairly wide range of systems has not been universally available on even the ones I personally use.
But even much more restrictive DRM systems can be made to work like this. It can be managed with Mobipocket by registering their device on your account and re-downloading, although that has a number of bad side effects (like further restricting how many devices you yourself can ever use with the content). Many DRM systems similarly allow limited copies or limited generations.
Generally speaking I agree with you that I'd prefer no DRM at all, most especially because DRM systems tend to limit the longevity of the content for my own use. But there do exist DRM systems where the DRM exists and yet still allows accessibility modes that paper does not.
You are mistaken, Adobe ebooks can be read on both MS and Palm PDAs as well as Windows and Mac PCs. In terms of document portability it's rivalled only by the Palm Reader. (Mobipocket actually runs on a wider variety of platforms but does not support Macs.) It's too bad its reader sucks so bad on some devices.
Just make the new formats as accessable as the previous and you won't run into the same problems as the music industry.
The problem is that even with DRM it is usually more accessible than paper in at least one way: The ability for multiple copies to exist simultaneously. That is especially problematic to publishers because books tend to be faddish, with lots of people reading the same book at once. Since the only thing they can limit is how many copies get made that's what they do, limiting not only copy breadth but also depth.
I note that serial DRM is not impossible or impractical in some cases. Libraries, for instance, can create versions of the documents that time out at the end of the lending period. If I had access to such a technology I might be happier about DRMed purchases.
Perhaps the simplest, most effective DRM I've seen is the one used in the Palm Reader where your credit card number is the key. There is an inherent tendency not to give that out willy-nilly, yet nothing stops you from moving it around other than not wanting to give out the key.
That kind of DRM is effective for the publisher and yet loose enough that I can move the content around on my own devices without restriction.
I have also purchased books with the Mobipocket DRM. I don't like it because it restricts the number of devices I can transfer it to. This is problematic not because I want to run it on a lot of devices in parallel, but instead because the reader device itself is somewhat ephemeral. I have had six different PDAs since I bought my first ebook back in 1997 and my current PDA will likely need replacement within a year or two just from wear-and-tear. With device-locked DRM mechanisms the content becomes inaccessible once you hit the device maximum even if some of the devices no longer exist. This I don't like.
Once I purchased a book in Adobe's ebook DRM format because no other format was available. That pointed out the most serious weakness in DRM: It only worked with Adobe's reader, and Adobe's reader is almost unusably horrible. To describe all the things it does wrong would take a whole article in and of itself, starting from an unreliable download process to a very time consuming transfer process to reader display that was so nonsensical as to make you wonder what they were smoking when they let it out the door. On top of that Adobe has even more restrictive device locking than Mobipocket. I hated the experience so much that I will never, ever buy another book in that format no matter its price.
That experience also led me to wonder if part of the problem with the acceptance of ebooks was that the mainstream businesses (B&N, Amazon) who offered ebooks only did it in two formats: Adobe's and Microsoft's. Adobe's has the benefit of working across a fairly wide range of devices but is so horrible that I just don't want to use it. Microsoft's is locked to a very limited set of Windows devices. I don't know about you, but I haven't found anyone who likes reading novels on their PC or laptop, so that pretty much limited the market for Microsoft DRMed books to PocketPC devices. A few years ago those were few and far between and even today that means spending several hundred dollars (whereas a PalmOS PDA can be had for under $100) and has only a few hours of battery life. If that's what my DRM choices look like I'll opt out entirely thanks.
I've also considered a number of dedicated ebook readers, but those have historically made the PocketPC scenario look good. Until a few weeks ago I'd never seen a dedicated ebook reader for less than $300. Today you can buy one for about $100 from fictionwise.com, a big improvement, but not enough to work around the other huge problem with these readers: Document portability. I have probably about a hundred ebooks at this point, in something like four different document formats. Not even one of those formats is supported by any of the dedicated ebook readers. Not Adobe. Not Mobipocket. Not Palm Reader. Not Palm Doc. I'm supposed to pay money for that?
If given the choice I would prefer no DRM if only because that gives me greater flexibility in which software I can use to read the books (or music, in the case of something like iTunes). But if you're going to use DRM, and frankly I can see why many publishers demand it, there needs to be a tradeoff between what is good for the publisher and whether or not the DRMed document will be long-term usable by the consumer. I worry about stuff purchased v
I just checked the box I'm on (a generic WinXP Pro install) and found that c:\windows is writable by "administrators" and "power users". The former is appropriate, the latter isn't, but the whole thing is rendered moot by the fact that the accounts are, by default, created with administrator privileges.
That's largely of necessity, I realize. On one of my home XP boxes I decided that my 2 year old daughter's account really shouldn't be privileged, so I didn't make it so. The result? Nearly all of her children's games failed to operate. When I called vendors about that, I was told that I'd just have to give the account the necessary privileges. (Can't return the software, of course, nobody allows software returns.)
So: We have a system that, if configured securely, doesn't work very well -- and if configured so it works, is so wide open that any little application error can lead to a compromised system.
It's a disaster and the only solution to it is going to be to have Microsoft turn the security way up by default so the software vendors are forced to write their code accordingly. Like, say, every other major OS out there.
The transition is going to suck, but until it's made Windows is going to remain a really easy target.
So, can you modify files in c:\windows in that XP installation? Yes? Then the system is an open book to anything that can get even a toehold.
The malware problem on Windows is not primarily the result of the system's popularity, no matter how many times Microsoft claims that is so. Early attacks on the Internet did not target the most popular system; rather, the most attacks have always targetted the easiest systems to crack. That started out with SunOS and, by the mid-90s, was Linux. (If you think Windows has much better penetration that Linux today, just think how much more lopsided the numbers were in 1995-2000 when Linux was the most popular target.) These days Windows systems are easiest by far because at this point they are the only systems which ship without basic filesystem protections (now that it finally has a halfway decent firewall, a mere five years after everyone else).
If Windows had basic filesystem protection enabled by default on all critical filesystem areas, mandated nonprivileged user accounts, and an installer that required a password, suddenly Windows wouldn't get infected every time you sneezed in its general direction.
Maybe the future will prove me wrong but I will be very surprised to find OS X malware become a serious problem no matter how popular the OS gets. I don't suspect that its users are any smarter, but the barriers are a lot higher.
So long as installers run without requiring passwords, and I have to give my daughter administrator privileges to run Disney games, Windows is in for a lot of hurt in the security domain because there's really no way to control what users, and by proxy the programs they run, muck with.
I mean, it's so bad right now that whole markets spawned to supply band-aids for the lack of basic protections (anti-virus, anti-spyware), and to rebuild broken systems as quickly as possible (ghost). That's pathetic, particularly since Microsoft had the ability to do a much better job of securing their systems since the release of Windows NT in 1993, and it's been mainstream since XP. It's not that they couldn't do it, it's that they didn't.
I was thinking of the applications on top; so far as I'm concerned both systems have very good device driver support.
And .NET is just Java with more language bindings and less platform support.
That's very true; after Sun sued them, Microsoft took their VM, extended it slightly, transmogrified the libraries slightly, added a lot of new features (particularly around network protocol and XML support), and wrote new front-ends for it.
It's kind of funny to me that Sun's lawsuit ultimately hurt them more than Microsoft. But then again Sun never seemed to have figured out that Microsoft was not their enemy, Intel was. If anything Microsoft did them quite the favor by proving either unwilling or unable to make Windows scale well enough to allow Dell et al to compete head-to-head.
Anyway the one place where Microsoft just did a better job than Sun and all the rest is in the IDE. It really is a nice IDE, at least for small to medium size applications -- point it at a web service and it generates the code; tell it to deploy to a web server and next thing you know your application is up and running. It's neat and clean and easy.
I spent eight of the last nine years writing Java code, and I'm here to tell you: Microsoft did a better job in a number of places. In particular, J2EE sucks goat balls. It's almost like they went out of their way to make it as painful as possible to use even for the simplest things. Microsoft didn't try to do all the things in J2EE, they just tried to do a really good job at the things people do all the time (to whit: Writing web pages and integrating them with databases). Not that we didn't see the same kind of stuff in Java years earlier, but that stuff unfortunately didn't become the standard.
But whatever, it is a fact that Microsoft has at least the perception of easier configuration and better integration. Fact too, sometimes. And that alone will make them a certain number of sales.
Indeed. In order for commercial software to compete in the long term it must provide significant value-add, and even then its pricing will be depressed. You mention IIS as being a negative, but I disagree. There is quite a tools ecosystem built around it, and it's much easier to set up and work with than Apache (I've run both, but I'm not crazy enough to expose IIS to the net at large).
But while everyone is looking at Microsoft vs. Linux now, that's not going to be the only big name to see open source significantly impact their business.
Pay attention to Oracle vs. MySQL. Nobody's really going to start talking about that until 2007 when you'll start seeing significant vendor support for it (this in response to developers using it as a cheap and lightweight alternative). Oracle will have to react in the 2009-2010 timeframe and their financials will start being negatively impacted as soon as 2013 and no later than 2015.
The Mozilla effort is going to bear significant fruit, too, although not so much on desktops -- unless we see Firefox get bundled by major PC vendors, which I would bet against. I figure 10% maximum desktop penetration on PCs, nearly 100% on Linux, and somewhere around 20-30% on Macs (less if Safari gets better).
The real win is going to be outside of desktops on more purposed devices, especially media devices. In fact, these will be where open source (especially Linux) makes its biggest impact -- royalty free is too big an advantage for those vendors to pass up.
The really good part of this from the consumer's point of view is that if it's running Linux, you can get at least some of the device's source code. That will lead to a lot of third-party product enhancements (ala Tivo hacks and Sveasoft).
Outside of infrastructure code like this, though, I think open source will have marginal impact. You need a certain critical mass of users and developers to sustain and nourish it, and that's hard to do as you get more vertical. Still, we're talking change-the-world market penetration in a number of horizontals. Not half bad results stemming from the ravings of a greasy hacker.