Re:Stop being an a$$hole....
on
Free Be
·
· Score: 1
The GPL allows you to sell software you develop. It also allows you to sell software other people have developed. It also allows other people to sell software that you developed.
I think that people have a point here, though. Opensource (especially Linux) is kinda like Microsoft in some aspects (i know... flamebait....) in that other companies really have the leadership position, and open source products just lumber along, steamrolling everything that doesn't move fast enough...
And i share the sentiment that if Open Source trully does win, it'll be a sad day for us all... Without the companies we've had investing billions of dollars in R&D, I don't know where the industry will go. But really, who will even want to learn to code if they know that they can make no money off of it, unless they also want to engage in support, etc?
Not to beat a dead horse or anything, but for one, according to the FAQ page, slash is just about GPL. The theory goes that they don't need to release it because it's only being used internally. However, when it was brought up that the NSA was contracting with (or considering to) a company to create a secure version of Linux, the general consensus was that the NSA should be required to publish the modifications that they contracted to have done. It was generally assumed that their distro wouldn't ever be intended to be seen outside the walls of the contracting company and the NSA, yet everyone assumed that since someone was shipping the code to someone else, the code should be available for all to see.
It's not that hard to layer value on top of the software distribution. Make it a support group, users group, etc... So long as you're banded together with a common interest and can show that, no one can really penalize you for the perks that you provide to members of your "club".
And you would need to be providing a very VALUABLE service. Not just redistributing GPLed software, because everyoen will know that they can get it for free... so if you've got something that kick ass that no ones developed yet and you're trying to circumvent the GPL, just go and choose a new license for it...
I don't htink you could demand new license fees with every release, because that wouldn't really be a subscription. But you could charge say $50/3 months and $200/yr and guarentee 1 software update every 3 months one's a member of your organization, I suppose....
Re:Stallman's right IMHO. In this case that's good
on
Hole in GNU GPL?
·
· Score: 2
Employees of corporations generally have no rights. They do not own the computer they use, the programs installed on it, or the data they generate. They're merely tools. They don't agree to the licenses of any software they use, the corporation does. Therefore, it's not the employees right or responsibility to distributed altered GPLed software if that work was done in a "work for hire" fashion (the general way most work is done for company's... you retain no rights to it and the company retains all).
I'm just poking around at all sides of this argument... I'm sure i've contradicted myself at some point across the past weeks. But this seems plausible to me.
But in that light, I can see how a company could fire, *with cause*, an employee who distributed GPL'ed software outside of the company. Trade secrets. Yes, you have a loyalty to the license, but first and formost, most companies demand loyalty from you. If you can't agree to do as you're told (code this for internal use only) then you should say so, rather than ignoring their orders. You won't be held liable for a GPL violation if it's found to be one, the company will... You were acting under orders.
Re:Stallman's right IMHO. In this case that's good
on
Hole in GNU GPL?
·
· Score: 2
Can one company contract to another company to make modifications to a GPLed product under the terms that those modifications may only be distributed back to the original, hiring company?
I think that's the real hang up right now... If that is the case, then essentially, I can create a product based on GPLed software and then only sell the software to people that agree to my terms, which may override the GPL.
Otherwise, if a company has a GPLed product, everyone in the world would be theoretically allowed to phone, fax, mail, or email requests fot their modifications to the code, even if they weren't in distribution.
This whole confusion about who gets what binary and source, etc, is escalating. I'd always interpreted the GPL as being essentially, if someone wanted the source, they get it. If you didn't like that, you'd choose a different license. But now, it appears that one can develope under the GPL and withhold distribution from people who don't agree to your additional terms...
Oh... if only i had the money to burn, i'd cause a major infraction on the GPL tomorrow, just so everyone would know where they stand... But, I don't have the money, I don't have the product, and I don't have the time... Oh well....
Re:I don't think this is a problem
on
Hole in GNU GPL?
·
· Score: 2
Using the NSA's securing of Linux as an example, I can think of plenty of reasons that a company wouldn't want changes they've done to GPLed software to go public. First off, if they make an iron clad secure version, then of course it's going to be installed in every embassy, base, etc, of every government around the world. They'll all be able to look and verify that indeed they've plugged up tons of holes and not left any back doors laying around.
And especially in a setting such as the NSA, security by obscurity I think is perfectly acceptable. They have a huge amount of personel on staff that are capable of reviewing their implentations who will all sign NDA's (or not work on the project) and have all passed rigid background examinations.
Likewise, if I made a search engine based on, what? MySQL's not GPLed... Maybe Postgres?, but anyways if i made a search engine based on a GPLed database but I somehow found a way to accelerate queries 100 times normal, why would i want to give that back to the community? I'm not selling it, or distributing it, i'm just using it. And being that a search engine isn't the best of examples these days, but you really run the risk of doing R&D for your competitors as well yourself by using GPLed software if you need to actually turn over your changes back to the community.
I'm trying to figure out if he brought this up due to the discussion about slashdot's code, or the NSA contracting with a company to secure Linux. Either way, I think it is somehting that needs to be cleared up... And definetly, someone out there needs to follow the GPL to the letter yet violate it in some way, so as to set a legal precedent.
Right now the GPL seems completely theoretical. But so far, everyone's respected it enough to not have to actually see if it's worth the paper it's printed on.
I'm not jumping on board in defending W2K or anything, but in your saying that you haven't used it since RC2, you're also saying you haven't used the most current builds. RC2 was pulled back and replaced with RC3... And then they went gold with it. Lot's of things could have changed... Maybe not much in terms of API's at that late point in development, but bug fixes... Microsoft went through a huge public beta process and i have to believe that a significant percentage of testers did indeed report their bugs back to Microsoft.
Just as when people say bad about Mozilla, they're reminded that it's barely alpha. What everyone to date (not under an NDA) has seen has been beta. For all the slipups in terms of security, privacy, and reliablity, you have to think that Microsoft did indeed realize that if their new OS was even half as bad as NT4 they'ed take a beating in the press and from the users.
I won't be first in line to buy Win2000 when it arrives but I will buy it as soon as i've got an extra computer for it to live on. And I definetly think this is Microsofts last real stand in terms of can they lead the market or forever follow...
Newest first is aweful. You get none of the continuity in postings or anything like that. Slashdot's too big now.
Maybe if they used their newfound money to post more stories then people would get spread around the site a bit more, rather than have everyone conjest the newest story.
But that'd also require a revamped interface, because the current one pretty much sucks you into the first story.
Microsoft Windows users use their computers in multitudes of different ways than Linux users. While a Linux user would know to use "shutdown -n blah blah blah" a Windows user might just hit the power switch. That's not a direct comparison, but really, I'm sure that there are some flaws in Linux that just haven't been found because Linux users operate their computers (for the most part) in a different manor than Windows.
Also, it doesn't help that with their popularity, Microsoft draws the fire of every scriptkiddie, security wannabe, etc, who all want to be the first to find a new bug and either exploit it or publish the fix.
The article seems to be slashdotted right now, so i can only speculate here... But i'm not convinced that Open Source produces cleaner code. It just allows you to have multitudes of people available to fix flaws, AS THEY'RE FOUND. In other terms, though, I think that closed source still has some advantages in a completely different context from security. Just more direction, rather than having everyone run around coding whatever it is they want to code.
Here's hoping that Microsoft stood up to their promise of not shipping Win2000 til it was ready.
Maybe he genuinely feels that Twofish and/or Blowfish are sufficient alternatives to the already existing algoryhms. Maybe it's his ego that gets boosted. And I don't know for sure, but after reading all of his articles, i would be suprised if he stood to make a dime from any implentation of his algorithms. Why? He's always pushed for royalty free, patent free cryptosystems, saying that since they already exist and work fine, there's no incentive for anyone to spend money licensing a new technology.
So, yeah, his ego gets stroked a bit, and he's definetly sided with his algorithms in the AES submission process, but in the end his only motivation is for pride's sake, rather than blatantly chasing the dollar signs.
They had no problem going onto the internet. It took them a while because they had a huge amount of infrastructure centered around providing dial up access to their isolated systems. But once they did, they did become a true internet company. And no, AOL didn't make a push for e-commerce, but the clientelle they attracted are responsible for being the spending side of the ecommerce boom.
And like you said, you don't care for their content. That's fine. You don't need to use it. But for some people, it's nice to be able to type in recipes and see recipes, rather than recieving the smorgasboard of websites that you get when you type that into a search engine.
Yes they do charge per hour. But considering I've never needed to do it, if one month I had to dial up to them for 10 hours to fix stuff, yeah, it'll cost me, but still less than if i'd had their 21.95 a month account. Like I said, it's mainly there for convienence and emergencies. You can change your billing plan whenever you like, so if you know you're going somewhere in 3 weeks, you can change for that billing period and then switch back when you return.
I've had AOL for a long time too, and i intend to keep it for a while too... Yes, i use other ISP's, and connect to AOL through their BYOA plan for $9.95/month. The reason is is that they're a permanent entity. Unlike my ISP, i know they'll be in business 5 years from now. When name servers mess up, or if i'm transfering domains, etc, I use AOL, simply because i don't even need to have an internet connection in the case of an emergency. Just dial up straight to them.
Plus, when you travel, if you're using a regional ISP, you're going to get long distance charges from hell... AOL, you just get a new local number. Even in Europe!
So yeah, AOL's not the greatest ever since i outgrew it. But for $9.95 a month, i intend to keep it as my safety net...
Okay fool... Go make yourself a billion dollars and stop whining about how AOL's junk. FYI, AOL does have it's own content, and it's organized multitudes of times better than the internet at large.
Yeah, there are a relatively small number of techies in AOL's audience, but you know what? Who cares. You know that whole e-commerce thing? The reason that Amazon and Ebay and Priceline and every other major.com is where they are today is because of the consumers in AOL's ranks.
You can probable step further out and say that Redhat and VA Linux owes a debt of gratitude to AOL. If it weren't for AOL popularizing the internet for the newbie's, then the whole e-commerce thing wouldn't have happened, so we'd not have all the.com stocks we have with absurd P/E ratios, and investors would still be in their old mentality of valueing a company based on their earnings rather than their potential.
So blah blah blah. If you don't like AOL, don't use them. But that's fine. They cater to a crowd that doesn't want you... That's not meant in a bad way. It's just, they're all the normal folk, who don't know DHCP from TCP/IP and SMTP from SNMP, etc etc etc.
You over look the advantage that AOL brings to customers with their proprietary service. In order to get on the internet, you only need to click the AOL icon. AOL's content is well organized, and you can be 99.9% sure that, supposing you don't frequent chatrooms or engage in IM's, that you're not going to have to wade through loads of smut. You just go to the appropriate keyword, and viola, everything you wanted about that subject.
Users can use AOL's built in browser to connect to the internet, or they can use IE4 or 5, Opera, or Netscape. And if all your friends use AOL as well, you can put them on your buddy list and know when they're online.
Those are all major advantages that AOL offers that other ISP's can't (except for the browsers...). WIth AOL merging into Time-Warner, it goes to show that AOL isn't really an ISP. They're a content company, and just happen to offer internet access (at the same price as their competitors, so it's not really gauging) to their customers so they can retrieve the content.
I don't want to join one side of the fight or the other, but how is it that Kerberos is more suited to being implented on OpenBSD rather than FreeBSD? Kerberos is apart from the OS, afterall, and any information as to how in implement it on any OS should be treated equally. Yeah, OpenBSD is more secure out of the box, but that's about it in terms of how applicable this is to that...
IBM's been bullied by Microsoft for so long, I'm sure they'ed much rather support their own OS rather than deal with another potentially obnoxious vendor.
And the only business they're trying to get out of is the retail PC business... They make a killing with DB/2, support, Lotus Notes, etc... And supporting Linux would play right into their business plan. Rolling their own distro wouldn't be much of an issue: Grab a copy of Debian Update the kernel Update Apache Include a free 5 or 10 user version of DB/2, and all of a sudden they've got one of the most compelling web platforms in the world....
Mark my words, if anyone does it, it will be IBM that builds their own distribution to bundle with their systems. Redhat has a name amongst the geeks and to an extent wall street (who just stand there wondering how it could be worth that much, but might as well buy some because it'll probably go up some more), but for the layman (whom i judge as my parents), they don't know redhat from slackware from debian from caldera. All they know is that Linux exists and it's an alternative.
For those people, IBM Linux would be much more trust worthy than Redhat Linux.
I'm sure you know this, but bandwidth isn't free. The only places that will give it to you for free are the ones that are giving away all that bandwidth to every tom dick and jane that is interested.... You'd probably not even be able to send 30GB in a week with one of them, and even that seems to understate it.
Realize that you've got a monster of a site, and shop around a bit... Extra CPU's and hard drives are cheap, one time expenses for ISP's, and they're amortized across so many people that of course they make sense. But bandwidth... if someone's using all of it and there's no provision in the contract, most (of the super-cheap places) will simply cut that client loose than deal with the headaches.
NEver trust your data to anyone else. Only send out copies. That means html, perl, everything. If something happens to their drive, and they have 10,000 people all emailing and calling them, you'll be much happier to know you have a safe copy of your data on your hard drive....
Only in the Software Industry/.com World are companies valued at what their potential could possibly be. Most of the time, as companies announce earnings and growth, analysts up their expectations and set new price targets, etc... Everythings been turned upside down recently (the past few years) as people scurry around looking for the next Yahoo's and Amazons, etc...
As for your comment about Engineers being more important than sales... Engineers just make a product. Sales and marketing actually earn the money for the company. How many companies have gone under while having an awesome product but aweful sales. And how about everyone's least favorite company around here, which has an inferior product, yet killer marketing and therefore they're the most valuable company in the world...?
And as for their profitablility... if they stop their attempts at growing, their value will drop like a rock. There's not much profit you can have with $12 million in sales that makes a $20 billion dollar market cap look worth while.
Someone's getting confused here. It's not a secondary initial public offering. It's just a secondary offering. They've already gone public. They made $14/share, when they did that. They've since split, so those share were effectively sold for $7. Now they're going to sell some more shares at $150 or whatever the market will take. That's all. It's a secondary offering.
Why won't developers standardize behind one of the already existing GUI's for Linux? We already have Gnome, KDE, etc... What good does bringing Be's interface to the table to? It would seem to just fragment efforts even further, if you ask me.
PC Manufacturers support the Windows installations they sell, rather than Microsoft. If Redhat wants $25 a machine (1/2 of what MSFT recieves) and IBM believes they can sell 1 million of them, then that's $25 million down the drain. Especially with the bottom of the barrel systems, where that $25 could have turned the profit from a machine from $50 to $75 dollars. That's 50% extra profit, if they decide to bring support in house. They already have the skills and expertise to support the systems, they would just need to retrain a fraction of their employees.
They could be quite clear about their motives: The license permits them to do this, They feel they can provide the same product for a lesser price to their customers, etc. etc. etc. If they DO make changes from the norm with a proprietary angle on their systems, people won't like it and won't buy it. But if all their changes are GPLed, then their systems will simply be the first on the block with feature A.
All the manufacturers were whining in the Microsoft case that they could no longer differntiate themselves from their competition. Rolling their own Linux distribution affords them the opportunity to maintain complete control of the systems they sell.
Slashdot Mirror
The GPL allows you to sell software you develop. It also allows you to sell software other people have developed. It also allows other people to sell software that you developed.
I think that people have a point here, though. Opensource (especially Linux) is kinda like Microsoft in some aspects (i know... flamebait....) in that other companies really have the leadership position, and open source products just lumber along, steamrolling everything that doesn't move fast enough...
And i share the sentiment that if Open Source trully does win, it'll be a sad day for us all... Without the companies we've had investing billions of dollars in R&D, I don't know where the industry will go. But really, who will even want to learn to code if they know that they can make no money off of it, unless they also want to engage in support, etc?
Not to beat a dead horse or anything, but for one, according to the FAQ page, slash is just about GPL. The theory goes that they don't need to release it because it's only being used internally. However, when it was brought up that the NSA was contracting with (or considering to) a company to create a secure version of Linux, the general consensus was that the NSA should be required to publish the modifications that they contracted to have done. It was generally assumed that their distro wouldn't ever be intended to be seen outside the walls of the contracting company and the NSA, yet everyone assumed that since someone was shipping the code to someone else, the code should be available for all to see.
Thats all.
It's not that hard to layer value on top of the software distribution. Make it a support group, users group, etc... So long as you're banded together with a common interest and can show that, no one can really penalize you for the perks that you provide to members of your "club".
And you would need to be providing a very VALUABLE service. Not just redistributing GPLed software, because everyoen will know that they can get it for free... so if you've got something that kick ass that no ones developed yet and you're trying to circumvent the GPL, just go and choose a new license for it...
I don't htink you could demand new license fees with every release, because that wouldn't really be a subscription. But you could charge say $50/3 months and $200/yr and guarentee 1 software update every 3 months one's a member of your organization, I suppose....
Employees of corporations generally have no rights. They do not own the computer they use, the programs installed on it, or the data they generate. They're merely tools. They don't agree to the licenses of any software they use, the corporation does. Therefore, it's not the employees right or responsibility to distributed altered GPLed software if that work was done in a "work for hire" fashion (the general way most work is done for company's... you retain no rights to it and the company retains all).
I'm just poking around at all sides of this argument... I'm sure i've contradicted myself at some point across the past weeks. But this seems plausible to me.
But in that light, I can see how a company could fire, *with cause*, an employee who distributed GPL'ed software outside of the company. Trade secrets. Yes, you have a loyalty to the license, but first and formost, most companies demand loyalty from you. If you can't agree to do as you're told (code this for internal use only) then you should say so, rather than ignoring their orders. You won't be held liable for a GPL violation if it's found to be one, the company will... You were acting under orders.
Can one company contract to another company to make modifications to a GPLed product under the terms that those modifications may only be distributed back to the original, hiring company?
I think that's the real hang up right now... If that is the case, then essentially, I can create a product based on GPLed software and then only sell the software to people that agree to my terms, which may override the GPL.
Otherwise, if a company has a GPLed product, everyone in the world would be theoretically allowed to phone, fax, mail, or email requests fot their modifications to the code, even if they weren't in distribution.
This whole confusion about who gets what binary and source, etc, is escalating. I'd always interpreted the GPL as being essentially, if someone wanted the source, they get it. If you didn't like that, you'd choose a different license. But now, it appears that one can develope under the GPL and withhold distribution from people who don't agree to your additional terms...
Oh... if only i had the money to burn, i'd cause a major infraction on the GPL tomorrow, just so everyone would know where they stand... But, I don't have the money, I don't have the product, and I don't have the time... Oh well....
Using the NSA's securing of Linux as an example, I can think of plenty of reasons that a company wouldn't want changes they've done to GPLed software to go public. First off, if they make an iron clad secure version, then of course it's going to be installed in every embassy, base, etc, of every government around the world. They'll all be able to look and verify that indeed they've plugged up tons of holes and not left any back doors laying around.
And especially in a setting such as the NSA, security by obscurity I think is perfectly acceptable. They have a huge amount of personel on staff that are capable of reviewing their implentations who will all sign NDA's (or not work on the project) and have all passed rigid background examinations.
Likewise, if I made a search engine based on, what? MySQL's not GPLed... Maybe Postgres?, but anyways if i made a search engine based on a GPLed database but I somehow found a way to accelerate queries 100 times normal, why would i want to give that back to the community? I'm not selling it, or distributing it, i'm just using it. And being that a search engine isn't the best of examples these days, but you really run the risk of doing R&D for your competitors as well yourself by using GPLed software if you need to actually turn over your changes back to the community.
I'm trying to figure out if he brought this up due to the discussion about slashdot's code, or the NSA contracting with a company to secure Linux. Either way, I think it is somehting that needs to be cleared up... And definetly, someone out there needs to follow the GPL to the letter yet violate it in some way, so as to set a legal precedent.
Right now the GPL seems completely theoretical. But so far, everyone's respected it enough to not have to actually see if it's worth the paper it's printed on.
I'm not jumping on board in defending W2K or anything, but in your saying that you haven't used it since RC2, you're also saying you haven't used the most current builds. RC2 was pulled back and replaced with RC3... And then they went gold with it. Lot's of things could have changed... Maybe not much in terms of API's at that late point in development, but bug fixes... Microsoft went through a huge public beta process and i have to believe that a significant percentage of testers did indeed report their bugs back to Microsoft.
Just as when people say bad about Mozilla, they're reminded that it's barely alpha. What everyone to date (not under an NDA) has seen has been beta. For all the slipups in terms of security, privacy, and reliablity, you have to think that Microsoft did indeed realize that if their new OS was even half as bad as NT4 they'ed take a beating in the press and from the users.
I won't be first in line to buy Win2000 when it arrives but I will buy it as soon as i've got an extra computer for it to live on. And I definetly think this is Microsofts last real stand in terms of can they lead the market or forever follow...
Newest first is aweful. You get none of the continuity in postings or anything like that. Slashdot's too big now.
Maybe if they used their newfound money to post more stories then people would get spread around the site a bit more, rather than have everyone conjest the newest story.
But that'd also require a revamped interface, because the current one pretty much sucks you into the first story.
Microsoft Windows users use their computers in multitudes of different ways than Linux users. While a Linux user would know to use "shutdown -n blah blah blah" a Windows user might just hit the power switch. That's not a direct comparison, but really, I'm sure that there are some flaws in Linux that just haven't been found because Linux users operate their computers (for the most part) in a different manor than Windows.
Also, it doesn't help that with their popularity, Microsoft draws the fire of every scriptkiddie, security wannabe, etc, who all want to be the first to find a new bug and either exploit it or publish the fix.
The article seems to be slashdotted right now, so i can only speculate here... But i'm not convinced that Open Source produces cleaner code. It just allows you to have multitudes of people available to fix flaws, AS THEY'RE FOUND. In other terms, though, I think that closed source still has some advantages in a completely different context from security. Just more direction, rather than having everyone run around coding whatever it is they want to code.
Here's hoping that Microsoft stood up to their promise of not shipping Win2000 til it was ready.
Maybe he genuinely feels that Twofish and/or Blowfish are sufficient alternatives to the already existing algoryhms. Maybe it's his ego that gets boosted. And I don't know for sure, but after reading all of his articles, i would be suprised if he stood to make a dime from any implentation of his algorithms. Why? He's always pushed for royalty free, patent free cryptosystems, saying that since they already exist and work fine, there's no incentive for anyone to spend money licensing a new technology.
So, yeah, his ego gets stroked a bit, and he's definetly sided with his algorithms in the AES submission process, but in the end his only motivation is for pride's sake, rather than blatantly chasing the dollar signs.
They had no problem going onto the internet. It took them a while because they had a huge amount of infrastructure centered around providing dial up access to their isolated systems. But once they did, they did become a true internet company. And no, AOL didn't make a push for e-commerce, but the clientelle they attracted are responsible for being the spending side of the ecommerce boom.
And like you said, you don't care for their content. That's fine. You don't need to use it. But for some people, it's nice to be able to type in recipes and see recipes, rather than recieving the smorgasboard of websites that you get when you type that into a search engine.
Yes they do charge per hour. But considering I've never needed to do it, if one month I had to dial up to them for 10 hours to fix stuff, yeah, it'll cost me, but still less than if i'd had their 21.95 a month account. Like I said, it's mainly there for convienence and emergencies. You can change your billing plan whenever you like, so if you know you're going somewhere in 3 weeks, you can change for that billing period and then switch back when you return.
I've had AOL for a long time too, and i intend to keep it for a while too... Yes, i use other ISP's, and connect to AOL through their BYOA plan for $9.95/month. The reason is is that they're a permanent entity. Unlike my ISP, i know they'll be in business 5 years from now. When name servers mess up, or if i'm transfering domains, etc, I use AOL, simply because i don't even need to have an internet connection in the case of an emergency. Just dial up straight to them.
Plus, when you travel, if you're using a regional ISP, you're going to get long distance charges from hell... AOL, you just get a new local number. Even in Europe!
So yeah, AOL's not the greatest ever since i outgrew it. But for $9.95 a month, i intend to keep it as my safety net...
Okay fool... Go make yourself a billion dollars and stop whining about how AOL's junk. FYI, AOL does have it's own content, and it's organized multitudes of times better than the internet at large.
.com is where they are today is because of the consumers in AOL's ranks.
.com stocks we have with absurd P/E ratios, and investors would still be in their old mentality of valueing a company based on their earnings rather than their potential.
Yeah, there are a relatively small number of techies in AOL's audience, but you know what? Who cares. You know that whole e-commerce thing? The reason that Amazon and Ebay and Priceline and every other major
You can probable step further out and say that Redhat and VA Linux owes a debt of gratitude to AOL. If it weren't for AOL popularizing the internet for the newbie's, then the whole e-commerce thing wouldn't have happened, so we'd not have all the
So blah blah blah. If you don't like AOL, don't use them. But that's fine. They cater to a crowd that doesn't want you... That's not meant in a bad way. It's just, they're all the normal folk, who don't know DHCP from TCP/IP and SMTP from SNMP, etc etc etc.
You over look the advantage that AOL brings to customers with their proprietary service. In order to get on the internet, you only need to click the AOL icon. AOL's content is well organized, and you can be 99.9% sure that, supposing you don't frequent chatrooms or engage in IM's, that you're not going to have to wade through loads of smut. You just go to the appropriate keyword, and viola, everything you wanted about that subject.
Users can use AOL's built in browser to connect to the internet, or they can use IE4 or 5, Opera, or Netscape. And if all your friends use AOL as well, you can put them on your buddy list and know when they're online.
Those are all major advantages that AOL offers that other ISP's can't (except for the browsers...). WIth AOL merging into Time-Warner, it goes to show that AOL isn't really an ISP. They're a content company, and just happen to offer internet access (at the same price as their competitors, so it's not really gauging) to their customers so they can retrieve the content.
I don't want to join one side of the fight or the other, but how is it that Kerberos is more suited to being implented on OpenBSD rather than FreeBSD? Kerberos is apart from the OS, afterall, and any information as to how in implement it on any OS should be treated equally. Yeah, OpenBSD is more secure out of the box, but that's about it in terms of how applicable this is to that...
IBM's been bullied by Microsoft for so long, I'm sure they'ed much rather support their own OS rather than deal with another potentially obnoxious vendor.
And the only business they're trying to get out of is the retail PC business... They make a killing with DB/2, support, Lotus Notes, etc... And supporting Linux would play right into their business plan. Rolling their own distro wouldn't be much of an issue:
Grab a copy of Debian
Update the kernel
Update Apache
Include a free 5 or 10 user version of DB/2, and all of a sudden they've got one of the most compelling web platforms in the world....
Mark my words, if anyone does it, it will be IBM that builds their own distribution to bundle with their systems. Redhat has a name amongst the geeks and to an extent wall street (who just stand there wondering how it could be worth that much, but might as well buy some because it'll probably go up some more), but for the layman (whom i judge as my parents), they don't know redhat from slackware from debian from caldera. All they know is that Linux exists and it's an alternative.
For those people, IBM Linux would be much more trust worthy than Redhat Linux.
I'm sure you know this, but bandwidth isn't free. The only places that will give it to you for free are the ones that are giving away all that bandwidth to every tom dick and jane that is interested.... You'd probably not even be able to send 30GB in a week with one of them, and even that seems to understate it.
Realize that you've got a monster of a site, and shop around a bit... Extra CPU's and hard drives are cheap, one time expenses for ISP's, and they're amortized across so many people that of course they make sense. But bandwidth... if someone's using all of it and there's no provision in the contract, most (of the super-cheap places) will simply cut that client loose than deal with the headaches.
NEver trust your data to anyone else. Only send out copies. That means html, perl, everything. If something happens to their drive, and they have 10,000 people all emailing and calling them, you'll be much happier to know you have a safe copy of your data on your hard drive....
Only in the Software Industry/.com World are companies valued at what their potential could possibly be. Most of the time, as companies announce earnings and growth, analysts up their expectations and set new price targets, etc... Everythings been turned upside down recently (the past few years) as people scurry around looking for the next Yahoo's and Amazons, etc...
As for your comment about Engineers being more important than sales... Engineers just make a product. Sales and marketing actually earn the money for the company. How many companies have gone under while having an awesome product but aweful sales. And how about everyone's least favorite company around here, which has an inferior product, yet killer marketing and therefore they're the most valuable company in the world...?
And as for their profitablility... if they stop their attempts at growing, their value will drop like a rock. There's not much profit you can have with $12 million in sales that makes a $20 billion dollar market cap look worth while.
Someone's getting confused here. It's not a secondary initial public offering. It's just a secondary offering. They've already gone public. They made $14/share, when they did that. They've since split, so those share were effectively sold for $7. Now they're going to sell some more shares at $150 or whatever the market will take. That's all. It's a secondary offering.
Why won't developers standardize behind one of the already existing GUI's for Linux? We already have Gnome, KDE, etc... What good does bringing Be's interface to the table to? It would seem to just fragment efforts even further, if you ask me.
PC Manufacturers support the Windows installations they sell, rather than Microsoft. If Redhat wants $25 a machine (1/2 of what MSFT recieves) and IBM believes they can sell 1 million of them, then that's $25 million down the drain. Especially with the bottom of the barrel systems, where that $25 could have turned the profit from a machine from $50 to $75 dollars. That's 50% extra profit, if they decide to bring support in house. They already have the skills and expertise to support the systems, they would just need to retrain a fraction of their employees.
They could be quite clear about their motives: The license permits them to do this, They feel they can provide the same product for a lesser price to their customers, etc. etc. etc. If they DO make changes from the norm with a proprietary angle on their systems, people won't like it and won't buy it. But if all their changes are GPLed, then their systems will simply be the first on the block with feature A.
All the manufacturers were whining in the Microsoft case that they could no longer differntiate themselves from their competition. Rolling their own Linux distribution affords them the opportunity to maintain complete control of the systems they sell.