Actually that figure might be correct. For example, I recall reading that high tech companies in Bangalore pay significantly more than 5900/yr, depending on the job, for example 2K/month vs. 5k in the states, still a savings. But not just programmers, for example why pay a wall street whiz 250K when you can pay 12k or so in India?
Therefore, there is a price differential, not as much as you think. It is somewhat annoying to see the offshore blended hourly rate less, but not much less ($10-20) than what you'd pay if you did the project in the states. All in all, more profit for the big companies, whether Covansys or Wal-mart.
Just like it's cheaper to live in, say, Arizona/Texas, hence companies/jobs move there from Silicon Valley, same goes for India. Provided you aren't concerned about coming back to the U.S., you could live very very well on $5900 USD in India.
Eventually there will be some sort of equilibrium I suppose, but that seems to take a while (again look at Silicon Valley), right now you can capitalize on the differential, sort of a smuggler's economy.
Backpacking across Canada is fine. I met a Dutch student who had done just the same - hitchiking and such. Don't assume everybody is on the Europe - see 6 countries in a day, and think "backpacking" as sort ofa day hike. There are those who consider it a challenge to hike the Appalachian Trail (Maine to Florida) or the Pacific Crest Trail (Washington to California).
So, Canada is a great place to backpack! Have fun!
Im my shop we are moving fast towards completely open enviroment. In most cases only Windows itself is the last piece that must go but the market still requires us to have it around.
And this is, I think part of IBM's strategy in contributing Eclipse to the open source world - so that more tool vendors, using the plugin feature, would be able to provide cross-platform tools, instead of always focusing on the Windows-only platform, and porting to other platforms later, if that.
One of the irritating things about using VisualAge (an excellent ide) was that IBM build it around a compiler. So my Linux VAJ 3.0 was running 1.1.x, but VAJ 3.5 on windows ran 1.2. And IBM never built a VAJ 3.5 for Linux - that was probably part of the reasoning behind going to WSAD (in addition to allowing you to use different JDKs).
Take a look at the eclipse.org site to see the consortium of vendors supporting the eclipse ide.
First of all it's sort of a holdover from MSofts standalone PC past to require 1000 machines to be patched. Much better would be to use a thin client system, especially when you're talking about a mega-installation like this, but that's a whole different argument.
Anyway, the poster you're referring to makes a very good point - once you homogenize everything, you make *everybody* vulnerable. So let's say Machine 1 is running SQL Server, and is hit by the slammer worm, then everyother machine that has SQL Server running can also be hit, and in turn hit other servers. This same scenario nearly took down the internet a while back, certainly could do the same at a major gov. office.
A simpler analogy is in science: let's say you introduce a new engineered corn, designed to withstand such and such, and fast growing etc. Now a certain plant disease, that evolves to not be harmed by whatever chemicals are used to grow the corn, simply lays waste to *all* the plants. Or think of the potato blight in Ireland in the 1850s. A final example is in cheese making, where if *all* bacteria is destroyed, it makes it easier for a dangerous bacteria to spread like wildfire, since there is no "good" bacteria to keep it in check.
Homogenization definitely has it's advantages, simplifies the job of the purchaser (write one check), and computer support (just call DELL), training (just one manual) but it also has it's costs - no way would I say this makes for a more secure environment. It does indeed make the system more susceptible to mass failure, which seems to go against the Army/Government philosophy.
If the CEO's can figure out how to give themselves all sorts of stock options and pay raises, while the companies lose money (Enrons Worldcoms), with the stockholders saying nary a whit, they'll figure out how to outsource everything except their chair. But then the overseas companies won't need them as they gain higher level experience - architecture management and the like which companies like infosys and tata already surely have.
I recall reading somewhere that companies were starting to rely on India for jobs such as stock analysts etc, why pay $250k when someone with just as good education and skills takes in 10-12k? So yeah, it's not just high-tech by any means.
One Indian programmer told me that he was sent to Thailand to write a banking system for Oracle or someone. They didn't have anyone who could speak Thai, so no one understood the business. You can imagine the resulting mess. Similar sort of mess the americans' find themselves in Iraq, not understanding the culture or lanquage.
This is great stuff to alleviate everyday boredom. Remember the day by day news items about the Netscape vs. Microsoft (IE) trial? Or Apple vs. Microsoft? We've gotten bored with the everybody vs. Microsoft. Except for those radical New Englanders - must be some west coast / east coast thing.
As an aside, read the book "The New New Thing" all about Jim Clarke, founder of SGI, Netscape, etc, and how he got the whole Netscape trial rolling.
This is anything but the sticking their heads in the sand bit, that seems like a bit of FUD to frighten Linux/AIX users who are hanging tight. What are they supposed to do? Give SCO license money? Switch to Sun or HP?
What with all the FUD out there surrounding this case, a little pro-Linux stuff on the web - what the heck? One should take with a grain of salt pretty much anything written about this case, whether pro-SCO or not. Right now it's media jousting, nothing is settled in the courts, but it probably is a good idea for the Linux side to downplay this in the media in various ways, being careful not to give any ideas away to SCO.
Forbes is rather M$ biased - you know -"Capitalist Tool" - don't give it too much credence. Used to be a good mag. They had a big article awhile back about how M$ was working so hard in China, all about security, sharing the source etc. Somehow they had avoided the whole reason *why* they had to work so hard, which was namely Linux. So either they knew it but didn't mention it, or totally missed the real reason. That's part of the danger for M$ right now, in that if they are so anti-Linux, they end up making managers take notice of Linux, such that they say "what's all this stuff about Linux Ballmer keeps yelling about? We need to get one of those!"
That's all well and good except that the GPL, being a bit of a non-standard contract, has never been tested in court.
What doesn't kill us makes us stronger.
I would say that IE is the defacto standard, having overtaken Netscape way back (why download when it's installed already, and pretty good?), however I would not make the same statement about IIS, SQLServer, 2kserver, since, especially in larger organizations, it's been a Solaris/Oracle combination, with WebSphere or BEA. On the contrary, it's M$ that's playing catchup with.NET. The 'it's cheaper' card is harder to play now that orgs can swap out their Solaris server and use RedHat.
I remember reading, I think in one of Robert X. Cringley's books, that one of the PC pioneers, maybe Scott McNealy, kept an ancient workstation from PARC days, on his desktop, to keep his perspective. It had a GUI, ethernet connection, mouse, printer - pretty much everything modern workstations have. Cringley also noted a study where Lotus contacted some of the first buyers of Lotus 1-2-3 (or maybe the Apple version), and found that most were still perfectly happy with it.
I think the trend is driven more by wintel, but it's getting harder and harder to justify.
Getting linux to run on something it's not supposed to? nah, it's about "hacking". One of the cool things about Linux is it's ability to run on a variety of platforms, from handhelds to mainframes to TiVo boxes to you name it. So "not supposed to run on" is not part of the Linux vocabulary.
Secondly, getting an x-box for $179 and using that as a cheap pc is definitely a draw
Thirdly, the x-box does have a network interface, for online gaming.
Uh, AOL - since when is AOL free? Does Mandrake time out after 45 days or 100 hours use - hey good idea!
Here's the reality: You still need *some* customers to pay. As long as there are some paying customers, there can be freeloaders, but that model fails when *nobody* pays.
See the quote below, one way to support Mandrake is through their retail partners - this quote is directly from Mandrake - this is their business model. Freeloaders can still help by sending words of support, giving advice, customer leads without paying a thing, but how many freeloaders do that?
Mandrake is not supported by freeloaders, it is supported by cash paying customers, through the channels listed below. Notice that their business model is now geared to business customers - make that "paying" business customers. Too bad, this will surely have an impact on the direction and focus of Mandrake (do./s need clustering? No, that's not your average home-user requirement). But I'd rather see this great product stay afloat, then go under because no one supports it.
"Numerous people have honored us with their trust by:
# Sending words of support (which we value)
# Giving us advice (which we listen to)
# Providing customer leads (which we follow-up on)
# Establishing business partnership (which we make effective)
# Purchasing our products in retail stores, on the web, or directly from us
I'll agree that it is important to have an easy to use system, and while a windows plus is that Microsoft puts great emphasis on ease of use as far as the apps go, that does not necessarily mean it is easier to *administer* than something like Linux.
Windows boxes require constant care and feeding, fixpacks, IE patches and the like. This is due to the origins of DOS as a standalone system. This is why so many organizations have full time help desks that are simply swamped. Now compare that with running apps from the server. Certainly you have this capability with Windows, i.e. using Citrix, however with Unix/Linux, this is how it was designed to work from the ground up (for example, X-windows was specifically designed to run over a network connection, this is why you can actually have decent performance over a modem).
For very specific systems such as point-of-sales (think HomeDepot, or your basic chain department store, etc), etc, you will definitely find they run Unix or even on a mainframe, as it's vastly easier to administer.
Therefore, if it were my money (and in a way it is), and I had very specific applications that are required, I would go with Linux. But usually in huge deals like this, there is much unseen bargaining going on that we are not privy to.
The original poster was not just talking about whether or not they were "high performance", but about their quality. So from the looks of it, you will be needing those identical machines, when they start crapping out on you, and you have to replace 'em.
With this case on the headlines in front of every CTO and CIO, their plan might actually backfire on them. No software company (Sun, Microsoft) is immune to this same sort of thing, for example Microsoft got into quite a bit of trouble with one version of DOS because it used compression technology from another company, I believe Stack Technologies, and this case seems to be on a lot shakier ground (perfect fodder for the lawyers of course).
Therefore, in the long run, this is probably great advertising for Linux - think of all the CTOs and CIOs who can actually tell the difference between Linus and Linux now?
I'd say the original poster (anonymous coward) is the arm-chair theorist, you've nailed it on the head.
Look at how much companies are shaped by their bosses, i.e. is it Oracle or Larry Ellison, Microsoft or Bill Gates? Also take a look at the arm wrestling over PeopleSoft, Siebel and JDEdwards, with some of the execs called "Little Larrys" because they previously worked at Oracle. Now they are suing Oracle and hugely ticked at Larry, but Larry claims he's on good terms and is proud of his proteges.
As you correctly note, once you get to the top, there's a lot of ego involved.
Therefore it is really amusing to realize that "companies" are really just subject to the whims at the "boys" or "girls" (Carly) running the show. This is especially apparent with the Oracle bid for PeopleSoft. Is this just Larry throwing his weight around? Is there real business sense here, or a corporate clash? Either way, it's a brawl. I don't know Palmisano too well (he's more low profile than the rest), but I'd say he's a pretty tough cookie too.
Check out some of the excellent open source libraries available for Java as they are very easy to use. Furthermore, in terms of quality, power, performance, and support they undoubtably far exceed Microsoft's offering. I'm not sure which, except for JAXB, of the XML and SOAP tools you used, but apparently you had some difficulty with them. JAXB is actually for data binding, it is one many data binding tools available for Java, confusing maybe, but welcome to the freedom of choice provided by Java!
I'll give you the fact that integration is probably easier with M$ since you don't have the plethora of choices you have with Java, but once installed (a trivial task), they are a snap to use, including "XML serialization", (although that doesn't really even seem much of a test of an XML parser, with the competition within the Java world, it usually revolves around which parser is the fastest, has the smallest memory footprint, best supports the XML standard etc. Again, a non-issue if you're stuck with one parser.
So I'd conclude, you're probably seriously limiting yourself with M$, however, there is the no hassle benefit of not having to go and install stuff etc, sort of like using IE because it's right there, users don't have to download something off the net to use.
Also,.NET/C# - it's a Java clone alright, but it's a subset being limited to a specific OS, unlike Java. So it's kinda like Java for Microsoft boxes. MONO? I find it very hard to believe M$ will let Mono get past the training wheel stage. Nice idea though, but hey, this is M$, not some charity.
Exactly. And I think you missed one of the best examples: Cobol.NET. To extend your point then, with dot net we are tying ourselves to the least common denominator, or COBOL. Rock on, all you.NETers! COBOL rules!
SCO used to be partially owned by Microsoft, I think they had up to 25% or so. Back before NT, SCO had a product called Xenix, which was basically Unix for the x86 platform. Ran quite well.
A colleague who was pretty familiar with both Microsoft and SCO at the time said he figured M$ learned quite a bit about building drivers while assisting with Xenix (although M$ was always better at working with other companies to make sure their drivers worked well with Wintel, way more so than IBM for example.)
Interesting thought anyway - no doubt there were *some* Xenix influences on NT. If you are interested, read Robert X. Cringeleys excellent book on the billionaires of silicon valley (who still can't get a date) for a taste of the wicked going's on in the early days of M$, in which Bill purchased a copy of QDOS (quick & dirty dos), which was apparently included CP/M code, since the CP/M author had included odd hacks which were dutifully copied over to QDOS/DOS. Interesting precedent.
" And taking the offensive, Perens added, "Copying works both ways. I want to see some proof they haven't copied Linux source code into SCO Unix."
I find this statement has more merit than anything from SCO. Corporate driven programming is much different from Open Source. With OSS, there's really no pressure to get stuff done and out the door, or at least far less than in the corporate world. This idea is based on Linus's many comments about when such and such release will be shipped : "when it's done".
Now contrast that with UnixWare version whatever, with a crackpot like McBride at the helm whipping up the team to get the code out the door. You can imagine the chaos: see the Mythical Man Month about OS/360, and I believe there's a book about Dave Cutler's team at M$. This is not Extreme Programming, it's the death march.
To conclude, which programmer is more likely to grab some peice of code that just works?
Here's another point: From my experience, OSS code is revised and rewritten constantly. Look at Ingo's work with the scheduler, or the recent work by several folks on the VM. Or Apache Xerces (XML4J). I've been using that for a few years, since it had jp (originally written by some IBMers at one of IBM's Japan labs I think), and that has been rewritten from scratch at least once. One last example - Mozilla was rewritten entirely, sidelining it for several years. That was a questionable move, but I say they did a good job - Mozilla is awesome.
Again contrast that with corporate software - how often do you rewrite working code from scratch? You do if business rules change, but I'd argue that with corporate software (think COBOL) it's more if it ain't broke don't fix it. Case in point: at the organization I work at, we have a database (well, more of a filesystem) written in assembler that is basically 25 year old code. Why? It works, it is very fast, it runs mission-critical systems, and it's very hard to modify, let alone maintain it. (Yes it is being replaced, so they say).
So, Unix code in Linux, copied line by line? Doubtful, I side with Perens.
To conclude, Perens has a good point. But as can be seen by the Sun/Netscape vs. Microsoft lawsuit, anything can happen in the courts. They'll have fun figuring out the "obfuscated" code.
Slashdot Mirror
Therefore, there is a price differential, not as much as you think. It is somewhat annoying to see the offshore blended hourly rate less, but not much less ($10-20) than what you'd pay if you did the project in the states. All in all, more profit for the big companies, whether Covansys or Wal-mart.
Eventually there will be some sort of equilibrium I suppose, but that seems to take a while (again look at Silicon Valley), right now you can capitalize on the differential, sort of a smuggler's economy.
Too bad, when we've had such great presidents: Roosevelt (Teddy), Kennedy, Washington, Jefferson, Lincoln.
So, Canada is a great place to backpack! Have fun!
And this is, I think part of IBM's strategy in contributing Eclipse to the open source world - so that more tool vendors, using the plugin feature, would be able to provide cross-platform tools, instead of always focusing on the Windows-only platform, and porting to other platforms later, if that.
One of the irritating things about using VisualAge (an excellent ide) was that IBM build it around a compiler. So my Linux VAJ 3.0 was running 1.1.x, but VAJ 3.5 on windows ran 1.2. And IBM never built a VAJ 3.5 for Linux - that was probably part of the reasoning behind going to WSAD (in addition to allowing you to use different JDKs).
Take a look at the eclipse.org site to see the consortium of vendors supporting the eclipse ide.
Check out this article on SWT and GCG: http://www-106.ibm.com/developerworks/java/library /j-nativegui2/
Anyway, the poster you're referring to makes a very good point - once you homogenize everything, you make *everybody* vulnerable. So let's say Machine 1 is running SQL Server, and is hit by the slammer worm, then everyother machine that has SQL Server running can also be hit, and in turn hit other servers. This same scenario nearly took down the internet a while back, certainly could do the same at a major gov. office.
A simpler analogy is in science: let's say you introduce a new engineered corn, designed to withstand such and such, and fast growing etc. Now a certain plant disease, that evolves to not be harmed by whatever chemicals are used to grow the corn, simply lays waste to *all* the plants. Or think of the potato blight in Ireland in the 1850s. A final example is in cheese making, where if *all* bacteria is destroyed, it makes it easier for a dangerous bacteria to spread like wildfire, since there is no "good" bacteria to keep it in check.
Homogenization definitely has it's advantages, simplifies the job of the purchaser (write one check), and computer support (just call DELL), training (just one manual) but it also has it's costs - no way would I say this makes for a more secure environment. It does indeed make the system more susceptible to mass failure, which seems to go against the Army/Government philosophy.
If the CEO's can figure out how to give themselves all sorts of stock options and pay raises, while the companies lose money (Enrons Worldcoms), with the stockholders saying nary a whit, they'll figure out how to outsource everything except their chair. But then the overseas companies won't need them as they gain higher level experience - architecture management and the like which companies like infosys and tata already surely have.
I recall reading somewhere that companies were starting to rely on India for jobs such as stock analysts etc, why pay $250k when someone with just as good education and skills takes in 10-12k? So yeah, it's not just high-tech by any means.
One Indian programmer told me that he was sent to Thailand to write a banking system for Oracle or someone. They didn't have anyone who could speak Thai, so no one understood the business. You can imagine the resulting mess. Similar sort of mess the americans' find themselves in Iraq, not understanding the culture or lanquage.
As an aside, read the book "The New New Thing" all about Jim Clarke, founder of SGI, Netscape, etc, and how he got the whole Netscape trial rolling.
What with all the FUD out there surrounding this case, a little pro-Linux stuff on the web - what the heck? One should take with a grain of salt pretty much anything written about this case, whether pro-SCO or not. Right now it's media jousting, nothing is settled in the courts, but it probably is a good idea for the Linux side to downplay this in the media in various ways, being careful not to give any ideas away to SCO.
Forbes is rather M$ biased - you know -"Capitalist Tool" - don't give it too much credence. Used to be a good mag. They had a big article awhile back about how M$ was working so hard in China, all about security, sharing the source etc. Somehow they had avoided the whole reason *why* they had to work so hard, which was namely Linux. So either they knew it but didn't mention it, or totally missed the real reason. That's part of the danger for M$ right now, in that if they are so anti-Linux, they end up making managers take notice of Linux, such that they say "what's all this stuff about Linux Ballmer keeps yelling about? We need to get one of those!"
That's all well and good except that the GPL, being a bit of a non-standard contract, has never been tested in court. What doesn't kill us makes us stronger.
I would say that IE is the defacto standard, having overtaken Netscape way back (why download when it's installed already, and pretty good?), however I would not make the same statement about IIS, SQLServer, 2kserver, since, especially in larger organizations, it's been a Solaris/Oracle combination, with WebSphere or BEA. On the contrary, it's M$ that's playing catchup with .NET. The 'it's cheaper' card is harder to play now that orgs can swap out their Solaris server and use RedHat.
I think the trend is driven more by wintel, but it's getting harder and harder to justify.
Secondly, getting an x-box for $179 and using that as a cheap pc is definitely a draw
Thirdly, the x-box does have a network interface, for online gaming.
Here's the reality: You still need *some* customers to pay. As long as there are some paying customers, there can be freeloaders, but that model fails when *nobody* pays.
See the quote below, one way to support Mandrake is through their retail partners - this quote is directly from Mandrake - this is their business model. Freeloaders can still help by sending words of support, giving advice, customer leads without paying a thing, but how many freeloaders do that?
Mandrake is not supported by freeloaders, it is supported by cash paying customers, through the channels listed below. Notice that their business model is now geared to business customers - make that "paying" business customers. Too bad, this will surely have an impact on the direction and focus of Mandrake (do ./s need clustering? No, that's not your average home-user requirement). But I'd rather see this great product stay afloat, then go under because no one supports it.
"Numerous people have honored us with their trust by:
# Sending words of support (which we value)
# Giving us advice (which we listen to)
# Providing customer leads (which we follow-up on)
# Establishing business partnership (which we make effective)
# Purchasing our products in retail stores, on the web, or directly from us
# By subscribing to MandrakeClub "
Windows boxes require constant care and feeding, fixpacks, IE patches and the like. This is due to the origins of DOS as a standalone system. This is why so many organizations have full time help desks that are simply swamped. Now compare that with running apps from the server. Certainly you have this capability with Windows, i.e. using Citrix, however with Unix/Linux, this is how it was designed to work from the ground up (for example, X-windows was specifically designed to run over a network connection, this is why you can actually have decent performance over a modem).
For very specific systems such as point-of-sales (think HomeDepot, or your basic chain department store, etc), etc, you will definitely find they run Unix or even on a mainframe, as it's vastly easier to administer.
Therefore, if it were my money (and in a way it is), and I had very specific applications that are required, I would go with Linux. But usually in huge deals like this, there is much unseen bargaining going on that we are not privy to.
The original poster was not just talking about whether or not they were "high performance", but about their quality. So from the looks of it, you will be needing those identical machines, when they start crapping out on you, and you have to replace 'em.
Therefore, in the long run, this is probably great advertising for Linux - think of all the CTOs and CIOs who can actually tell the difference between Linus and Linux now?
Look at how much companies are shaped by their bosses, i.e. is it Oracle or Larry Ellison, Microsoft or Bill Gates? Also take a look at the arm wrestling over PeopleSoft, Siebel and JDEdwards, with some of the execs called "Little Larrys" because they previously worked at Oracle. Now they are suing Oracle and hugely ticked at Larry, but Larry claims he's on good terms and is proud of his proteges.
As you correctly note, once you get to the top, there's a lot of ego involved.
Therefore it is really amusing to realize that "companies" are really just subject to the whims at the "boys" or "girls" (Carly) running the show. This is especially apparent with the Oracle bid for PeopleSoft. Is this just Larry throwing his weight around? Is there real business sense here, or a corporate clash? Either way, it's a brawl. I don't know Palmisano too well (he's more low profile than the rest), but I'd say he's a pretty tough cookie too.
I'll give you the fact that integration is probably easier with M$ since you don't have the plethora of choices you have with Java, but once installed (a trivial task), they are a snap to use, including "XML serialization", (although that doesn't really even seem much of a test of an XML parser, with the competition within the Java world, it usually revolves around which parser is the fastest, has the smallest memory footprint, best supports the XML standard etc. Again, a non-issue if you're stuck with one parser.
So I'd conclude, you're probably seriously limiting yourself with M$, however, there is the no hassle benefit of not having to go and install stuff etc, sort of like using IE because it's right there, users don't have to download something off the net to use.
Also, .NET/C# - it's a Java clone alright, but it's a subset being limited to a specific OS, unlike Java. So it's kinda like Java for Microsoft boxes. MONO? I find it very hard to believe M$ will let Mono get past the training wheel stage. Nice idea though, but hey, this is M$, not some charity.
Exactly. And I think you missed one of the best examples: Cobol.NET. To extend your point then, with dot net we are tying ourselves to the least common denominator, or COBOL. Rock on, all you .NETers! COBOL rules!
A colleague who was pretty familiar with both Microsoft and SCO at the time said he figured M$ learned quite a bit about building drivers while assisting with Xenix (although M$ was always better at working with other companies to make sure their drivers worked well with Wintel, way more so than IBM for example.)
Interesting thought anyway - no doubt there were *some* Xenix influences on NT. If you are interested, read Robert X. Cringeleys excellent book on the billionaires of silicon valley (who still can't get a date) for a taste of the wicked going's on in the early days of M$, in which Bill purchased a copy of QDOS (quick & dirty dos), which was apparently included CP/M code, since the CP/M author had included odd hacks which were dutifully copied over to QDOS/DOS. Interesting precedent.
I find this statement has more merit than anything from SCO. Corporate driven programming is much different from Open Source. With OSS, there's really no pressure to get stuff done and out the door, or at least far less than in the corporate world. This idea is based on Linus's many comments about when such and such release will be shipped : "when it's done".
Now contrast that with UnixWare version whatever, with a crackpot like McBride at the helm whipping up the team to get the code out the door. You can imagine the chaos: see the Mythical Man Month about OS/360, and I believe there's a book about Dave Cutler's team at M$. This is not Extreme Programming, it's the death march.
To conclude, which programmer is more likely to grab some peice of code that just works?
Here's another point: From my experience, OSS code is revised and rewritten constantly. Look at Ingo's work with the scheduler, or the recent work by several folks on the VM. Or Apache Xerces (XML4J). I've been using that for a few years, since it had jp (originally written by some IBMers at one of IBM's Japan labs I think), and that has been rewritten from scratch at least once. One last example - Mozilla was rewritten entirely, sidelining it for several years. That was a questionable move, but I say they did a good job - Mozilla is awesome.
Again contrast that with corporate software - how often do you rewrite working code from scratch? You do if business rules change, but I'd argue that with corporate software (think COBOL) it's more if it ain't broke don't fix it. Case in point: at the organization I work at, we have a database (well, more of a filesystem) written in assembler that is basically 25 year old code. Why? It works, it is very fast, it runs mission-critical systems, and it's very hard to modify, let alone maintain it. (Yes it is being replaced, so they say).
So, Unix code in Linux, copied line by line? Doubtful, I side with Perens.
To conclude, Perens has a good point. But as can be seen by the Sun/Netscape vs. Microsoft lawsuit, anything can happen in the courts. They'll have fun figuring out the "obfuscated" code.
Indeed, about as funny as a reference to NP.