• We’re currently investigating an issue related to the forum theme and styling that is impacting page layout and visual formatting. The problem has been identified, and we are actively working on a resolution. There is no impact to user data or functionality, this is strictly a front-end display issue. We’ll post an update once the fix has been deployed. Thanks for your patience while we get this sorted.

Binary drivers are great!

Page 2 - Seeking answers? Join the AnandTech community: where nearly half-a-million members share solutions and discuss the latest tech.
Originally posted by: Markbnj
...what about the other 99+% who are running Windows XP on their laptops? Will open source drivers solve all their problems too?

Yep.

Provided they respect the terms of the license, the manufacturer could port those same high quality, open source drivers to XP as well.
 
Provided they respect the terms of the license, the manufacturer could port those same high quality, open source drivers to XP as well.

That's a different thing alltogether. So if the OEM wanted to rely on the open source community to develop drivers for their hardware, and potentially pick one distribution and say "this is the one we provide support for," and port it, then that might work fine. Some sort of creative relationship like that might actually be a good solution.

But then what happens when there is a bug, the open source community patches the drivers, but the OEM doesn't get around to porting the patch and updating their support right away?

I reiterate: the average consumer-level user needs a single point of support for replacement parts for their systems and applications. The freewheeling technical community of open source advocates will not provide a solution for the average user. My Dad is 70, and very computer literate, but I can't get him to search forums for answers to problems. He calls me or the OEM 🙂.
 
Originally posted by: Markbnj
Provided they respect the terms of the license, the manufacturer could port those same high quality, open source drivers to XP as well.

That's a different thing alltogether. So if the OEM wanted to rely on the open source community to develop drivers for their hardware, and potentially pick one distribution and say "this is the one we provide support for," and port it, then that might work fine. Some sort of creative relationship like that might actually be a good solution.

But then what happens when there is a bug, the open source community patches the drivers, but the OEM doesn't get around to porting the patch and updating their support right away?

I reiterate: the average consumer-level user needs a single point of support for replacement parts for their systems and applications. The freewheeling technical community of open source advocates will not provide a solution for the average user. My Dad is 70, and very computer literate, but I can't get him to search forums for answers to problems. He calls me or the OEM 🙂.

Lazy OEMs (in theory) get left in the dust in future purchases. Some things never change.
 
Originally posted by: Markbnj
We're talking past each other here, and I have better things to do... ok, no I don't, but still 😉.

Ok, I will stipulate that open source drivers solve all of the problems that users of open source operating systems encounter with proprietary binary drivers.

Whew. Well, now that we've taken care of the .03% of people affected by the problem reported in the OP who are using Linux, what about the other 99+% who are running Windows XP on their laptops? Will open source drivers solve all their problems too?

I thought they didn't have problems. Except crappy drivers, which only the OEMs and Microsoft can solve. Maybe if Microsoft got the documentation for the hardware they would help their users out. 🙂
 
Originally posted by: Markbnj
But then what happens when there is a bug, the open source community patches the drivers, but the OEM doesn't get around to porting the patch and updating their support right away?

Why wouldn't an OEM take advantage of free bug and security updates? I have no idea. My guess is they would be very happy to.


 
Why wouldn't an OEM take advantage of free bug and security updates? I have no idea. My guess is they would be very happy to.

Because there is no chain of accountability that they have control over. There are no legal obligations, binding contractual promises, employment conditions, etc. How do they know some disgruntled coder isn't going to embed something nasty into a driver? They don't. What if that happens? They'll get sued. Who do they turn to to recover then? Predatory lawyers aren't going to go after a bunch of open source programmers. They go after the money. Have any of you guys ever worked in a business before? 😉
 
Originally posted by: Markbnj
We're talking past each other here, and I have better things to do... ok, no I don't, but still 😉.

Ok, I will stipulate that open source drivers solve all of the problems that users of open source operating systems encounter with proprietary binary drivers.

Whew. Well, now that we've taken care of the .03% of people affected by the problem reported in the OP who are using Linux, what about the other 99+% who are running Windows XP on their laptops? Will open source drivers solve all their problems too?

Ya sure.

Just have them uninstall Windows and install OpenBSD or Debian Linux. Pretty simple. 😛


I reiterate: the average consumer-level user needs a single point of support for replacement parts for their systems and applications. The freewheeling technical community of open source advocates will not provide a solution for the average user. My Dad is 70, and very computer literate, but I can't get him to search forums for answers to problems. He calls me or the OEM

Why assume that Open Source == IRC/Forums only help and Propriatory == Quick Phone call and answer?

There are plenty of people that support Open source systems professionally. You can pay for help with Debian or Redhat or FreeBSD and probably even find professional support options for OpenBSD if you felt like it.

Dell and HP do offer support for running Redhat or Suse Linux on a veriaty of hardware. Even pre-installed. Of course it's all for server and workstation level stuff, but that doesn't preclude the fact that they can't do this later on for Desktops.

If you want professional support options for Desktop Linux you can go for Linspire which caters specificly to people like your father.

Or if you just want, say a notebook, with Linux that 'just works' there are a veriaty of places you can go:
Linux Certified has Linux preinstalled on a veriaty of notebooks.
http://www.linuxcertified.com/linux_laptops.html

As does lots of other places:
http://system76.com/index.php/cPath/1?gclid=COuM07LL4oUCFQsESAodygg7QQ
http://www.aslab.com/products/laptops/laptops.html

For instance this place seems to do a pretty good job. They have a veriaty of Notebooks with Fedora Core 5 preinstalled.

They setup a nice swap partition.. Setup a small Fat32 partition for doing BIOS updates and such. Will setup Windows XP dual boot if you want. They get all the power management stuff working, get all the special laptop buttons working. Get suspend-to-disk working. Make it easy to control power saving modes via the desktop.

Then they setup extra software for XOSD (display overlay, probably for showing response to special buttons or display/power management changes). Setup DVD Author, Kino. Setup Java for you. Adobe acrobat reader. Setup Macromedia flash. They make sure that wireless is working, that you can do one-click-installs of most commonly used media applications. Make sure audio is working. Scribus desktop publishing, Q-cad.

So they do what most Linux users would do when you first install a distro anyways.. so I think that's pretty neat.

And there are about a dozen other places that offer similar levels of support for pre-installed Linux. So if you don't like that place there are other choices.

Actually one of the big selling points for Redhat is it's level of support that it's able to provide. Since everything is open source, not just the core operating system, but the drivers, and the application stack, Redhat can offer much higher level of support then they would otherwise be able to provide if they were a closed source system distributer.

Of course since Linux is relatively unpopular for desktop use it's generally a lot easier to find Windows support (just run down to your local Best Buy, for instance), but it's aviable if you want to pay for it.
 
Originally posted by: Markbnj
Why wouldn't an OEM take advantage of free bug and security updates? I have no idea. My guess is they would be very happy to.

Because there is no chain of accountability that they have control over. There are no legal obligations, binding contractual promises, employment conditions, etc. How do they know some disgruntled coder isn't going to embed something nasty into a driver? They don't. What if that happens? They'll get sued. Who do they turn to to recover then? Predatory lawyers aren't going to go after a bunch of open source programmers. They go after the money. Have any of you guys ever worked in a business before? 😉

But with the fact it's open source they can audit it to their hearts' content. 😉
 
But with the fact it's open source they can audit it to their hearts' content.

Which version should they audit? Who is responsible for it? How do they manage their customer's expectations when there are different people working on different distributions of source code for their hardware? Look at how many versions of Linux there are in the short time its been around, and look at all the infighting and political bullshit that goes on in those groups. What does the company that is about to release a new piece of hardware do when the group they were relying on to develop the drivers for it decides to break up over a design argument? Nobody is legally bound to do anything, or deliver anything. You think that's a model you would risk a few tens of millions of development costs on?

Open source software is amateur software, regardless of whether the people developing it are themselves professionals, and that's the bottom line imho. It's a hobby for them. There's nothing wrong with it, or new about it, and some of it exceeds the engineering standards of a lot of proprietary stuff. But I wouldn't risk a business on some open source group delivering, nor recommend that my customers use open source drivers to run my hardware. There's going to be a supported driver that I develop in-house, or that I hire someone to develop under a contract that binds them to accomplish certain goals, and adhere to certain standards. That's going to be the one I support and distribute. If my customers want to use someone else's driver that's fine, but if they call me about it I am going to tell them to go to the person that wrote it.

 
There are no legal obligations, binding contractual promises, employment conditions, etc.

Have you ever read the license agreements for any piece of software? They all say that the company who produced them isn't responsible if anything goes wrong, so what's the difference?

How do they manage their customer's expectations when there are different people working on different distributions of source code for their hardware?

Have you ever seen a machine fresh from an OEM restore disc? From that I'd say they're not terribly concerned with customer expectations.

What does the company that is about to release a new piece of hardware do when the group they were relying on to develop the drivers for it decides to break up over a design argument?

What do they do when their lead/only developer quits?

You think that's a model you would risk a few tens of millions of development costs on?

IBM, HP, Dell, SGI, Intel, etc all seem to be working with it pretty well.

Open source software is amateur software, regardless of whether the people developing it are themselves professionals, and that's the bottom line imho.

Wow, I haven't seen someone say something like that since like 1995. And I'm pretty sure all of the people at IBM, HP, Dell, SGI, Intel, etc who are paid to work on FOSS software think otherwise.
 
Have you ever read the license agreements for any piece of software? They all say that the company who produced them isn't responsible if anything goes wrong, so what's the difference?

That statment wasnt about liability for events on the user side, although that's an issue too. Ask Sony. I was referring to the contractual and practical bonds between employees/vendors/partners and their employers or counterparts, bonds which are totaly absent from the open source world.

Have you ever seen a machine fresh from an OEM restore disc? From that I'd say they're not terribly concerned with customer expectations.

Whether all current OEMs are doing a good job now, or not, is beside the point. You say the model is broken, and we need to turn to a wikiware model. I say the model isn't broken, just because some low-margin OEMs don't do a good job.

What do they do when their lead/only developer quits?

Promote the next guy, but I'd guess the lead is a lot less likely to wake up one day and think "screw it" if it's how he makes a living. That's called having some skin in the game, which most open source developers do not. I'll get to your later point about people being paid to develop open source projects below.

Let's just collapse these last two points:

IBM, HP, Dell, SGI, Intel, etc all seem to be working with it pretty well.
Wow, I haven't seen someone say something like that since like 1995. And I'm pretty sure all of the people at IBM, HP, Dell, SGI, Intel, etc who are paid to work on FOSS software think otherwise.

So a few very large companies paying some people to work on FOSS is evidence of what, exactly? They're paying people to write software to a certain design, and against certain standards and milestones. If they open the source up, great. They are risking nothing in the scheme of things. Nothing being done by volunteers is critical to a product launch timeline. Nothing is being done by volunteers that represents proprietary interests of the company. They're big enough to do this for the hell of it, for the PR value, or to try to chip away at Microsoft (the motivation for 90% of investment in FOSS by the companies you mention, with the exception of Dell).

Besides, since when was "open source" about a few huge companies releasing some of their low-value, low-risk work into an open source environment? I must have missed the last ten years of Stallman-esque ranting against the evils of proprietary software business models. Whatever the failings of the current model, which depends on the profit-motive to create value, the open source world has them worse. The primary value proposition of 95% of the open source software out there is that it is free and college students don't have to dip into the meal fund to acquire it.

The whole thing is just silly to me. It was silly in 1995, and it's still silly. Not the idea of people getting together to write cool software. I didn't think that was silly in 1988 when I was doing it with friends over Compuserve, and I don't think it's silly now. But the idea that the mere existence of proprietary (Microsoft) software is evil, and we need a socio-technical movement to replace it (Microsoft) with happy little collectives producing organic code without the use of smelly dollars, that's the part that's silly. Microsoft has been managing and evolving its word processor for twenty years. I'll be interested to see where Open Office is twenty years from now. I have every reason to think MS will still have a word processor then, and it will be able to open and read the documents I am creating now.
 
Originally posted by: Markbnj
That statment wasnt about liability for events on the user side, although that's an issue too. Ask Sony. I was referring to the contractual and practical bonds between employees/vendors/partners and their employers or counterparts, bonds which are totaly absent from the open source world.

I haven't read what you guys are arguing about so I'm skipping this part.

Originally posted by: Markbnj
Have you ever seen a machine fresh from an OEM restore disc? From that I'd say they're not terribly concerned with customer expectations.

Whether all current OEMs are doing a good job now, or not, is beside the point. You say the model is broken, and we need to turn to a wikiware model. I say the model isn't broken, just because some low-margin OEMs don't do a good job.

Are we talking about software or hardware here? I prefer to deal with companies that provide open-source drivers when referring to hardware. If the ODM dies (low chance I know), I know the ability of someone or some group being able to take the source code and continue to work on it is there. If companies like Novell feel the hardware is important to provide they can pay developers to keep coding drivers and the cycle continues or smaller groups of coders can take it up however motivated. If the ODM doesn't open-source its drivers, I feel less assured that my hardware can or will be supported in the future and in my platform.

Originally posted by: Markbnj
IBM, HP, Dell, SGI, Intel, etc all seem to be working with it pretty well.
Wow, I haven't seen someone say something like that since like 1995. And I'm pretty sure all of the people at IBM, HP, Dell, SGI, Intel, etc who are paid to work on FOSS software think otherwise.

So a few very large companies paying some people to work on FOSS is evidence of what, exactly?

That it benefits them somehow and is somehow worth it. Most likely involving linux.

Originally posted by: Markbnj
They're paying people to write software to a certain design, and against certain standards and milestones. If they open the source up, great. They are risking nothing in the scheme of things.

They're risking their code being stolen/viewed by their competitors. Unless you mean paying third party developers to code open-source drivers. Maybe I misunderstand what you're trying to say.

Originally posted by: Markbnj
Nothing being done by volunteers is critical to a product launch timeline. Nothing is being done by volunteers that represents proprietary interests of the company. They're big enough to do this for the hell of it, for the PR value, or to try to chip away at Microsoft (the motivation for 90% of investment in FOSS by the companies you mention, with the exception of Dell).

Almost agree with you here. Unless the volunteers work is being included in the linux kernel and have that being used by Novell, Red Hat, and a number of other corporations who base their operating systems off the linux kernel. Then they have more stake in it.

Lol, your last comment made no sense. At least when referring to hardware manufacturers. First, no hardware manufacturers would open-source their drivers 'for the hell of it'. Most have to make sure that no licenses are being violated. Sometimes companies borrow and license code from others. Plus when a company open-sources their code they have to be worried what their competition will gain from their code. When you think about it, its usually a harder decision than just doing it for the hell of it. There is no reason to do it 'for the hell of it'.

PR value? Why would they even have PR value from open-sourcing their drivers? It has to be at least somewhat important and it is.

When it comes to open-sourcing drivers, hardware manufacturers can offload their work by allowing their bsd customers to depend on bsd developers and linux customers to linux developers. Now what has this done? It makes their customers who run alternative operating systems happy. Why do hardware manufacturers care at all about linux? The ONLY reason should be is because they have customers who use linux-based software and operating systems. They don't care if Microsoft fails or stays afloat, the same with any other corporation providing an operating system. They only care about selling their product and making sure their customer can use their product.

Originally posted by: Markbnj
Besides, since when was "open source" about a few huge companies releasing some of their low-value, low-risk work into an open source environment? I must have missed the last ten years of Stallman-esque ranting against the evils of proprietary software business models. Whatever the failings of the current model, which depends on the profit-motive to create value, the open source world has them worse. The primary value proposition of 95% of the open source software out there is that it is free and college students don't have to dip into the meal fund to acquire it.

Sorry I'm not sure what you're trying to say or argue.

Originally posted by: Markbnj
The whole thing is just silly to me. It was silly in 1995, and it's still silly. Not the idea of people getting together to write cool software. I didn't think that was silly in 1988 when I was doing it with friends over Compuserve, and I don't think it's silly now. But the idea that the mere existence of proprietary (Microsoft) software is evil, and we need a socio-technical movement to replace it (Microsoft) with happy little collectives producing organic code without the use of smelly dollars, that's the part that's silly.

Well its happening right now whether its silly or not. Some software developers feel they shouldn't force people to pay for their software. Open-source isn't the best model when expecting to make profit, but people use it anyway. Guess some people don't always care about money. And if you write quality software people will support you. Take a look at K3B's fundraiser.

Originally posted by: Markbnj
Microsoft has been managing and evolving its word processor for twenty years. I'll be interested to see where Open Office is twenty years from now. I have every reason to think MS will still have a word processor then, and it will be able to open and read the documents I am creating now.

This has to be the funniest thing you've said this whole time. Let's list the scenarios:

1) If Microsoft continues to succeed you're able to pay for their software to write documents and access their proprietary format which doesn't work that well in other software or hope someone can find a way to access their proprietary format. If OpenOffice continues to succeed you're able to have free software and freely access their format from anyone who includes the code which wouldn't be hard to find since the format is open-source.

2) If Microsoft continues to succeed you're able to pay for their software to write documents and access their proprietary format which doesn't work that well in other software or hope someone can find a way to access their proprietary format. If OpenOffice fails, someone can fork their code and continue to work on the software or you can use any other software that includes their format which wouldn't be hard to find since the format is open-source.

3) If Microsoft fails, you will have to somehow find their older software to access your documents, hope Microsoft either opens the format or sells it to another company to sell you a word processor do access your files, hope someone can find a way to access their proprietary format, or be SOL. If OpenOffice continues to succeed you're able to have free software and freely access their format from anyone who includes the code which wouldn't be hard to find since the format is open-source.

4) If Microsoft fails, you will have to somehow find their older software to access your documents, hope Microsoft either opens the format or sells it to another company to sell you a word processor do access your files, hope someone can find a way to access their proprietary format, or be SOL. If OpenOffice fails, someone can fork their code and continue to work on the software or you can use any other software that includes their format which wouldn't be hard to find since the format is open-source.

For your sake, I hope Microsoft succeeds so you can still spend $400 to read your documents.
 
Microsoft is going to stay around for a long long ass time. they aren't going anywere..

But it's not uncommon for businesses like banks or governments who were very early adopters of computer technology to have YEARS of data locked away in a propriatory format.

REALY. I am not joking. In the midwest it's common for older banks to have information locked away in computer systems original built in the late 80's. File formats, hardware, tapes, and even networking equipment whose original makers have long since gone out of business. Now they can spend easily tends of thousands of dollars just to get information that they already have archived. Sometimes it can turn into quite a nightmare.

Governments have faced this also and it's not uncommon situation to find yourself in if your dealing with managing digital data more then ten years old. That stuff needs to be accessable and it gets expensive.

This is why we have things like the 'OpenGroup'. And 'Open' this and 'Open' that. It's why 'propriatory' is a dirty word in computer-land were in other industries its used as a 'good thing' to set their product apart from their competitors.

 
Originally posted by: Markbnj
Which version should they audit? Who is responsible for it? How do they manage their customer's expectations when there are different people working on different distributions of source code for their hardware? Look at how many versions of Linux there are in the short time its been around, and look at all the infighting and political bullshit that goes on in those groups. What does the company that is about to release a new piece of hardware do when the group they were relying on to develop the drivers for it decides to break up over a design argument? Nobody is legally bound to do anything, or deliver anything. You think that's a model you would risk a few tens of millions of development costs on?

I don't think open source works for everyone. If they want to officially support Win32 people only, I'm fine with that. I get the support I need from open channels. But releasing documentation is plenty of support for many of the open source developers out there.

Open source software is amateur software, regardless of whether the people developing it are themselves professionals, and that's the bottom line imho.

I wonder how Sun, IBM, Microsoft, Novell, and RedHat, Checkpoint, Nokia, the US federal government, and just about any company that has ever used a TCP/IP stack feel about statements like that. 😕

There's going to be a supported driver that I develop in-house, or that I hire someone to develop under a contract that binds them to accomplish certain goals, and adhere to certain standards. That's going to be the one I support and distribute. If my customers want to use someone else's driver that's fine, but if they call me about it I am going to tell them to go to the person that wrote it.

Good. I have no problems with that. That's a good model, IMO. But that model should not stop the hardware company from providing a minimum of support to all of their customers, which can be done with no work (excluding bureaucracratic BS) on their part.
 
Originally posted by: Markbnj
That statment wasnt about liability for events on the user side, although that's an issue too. Ask Sony. I was referring to the contractual and practical bonds between employees/vendors/partners and their employers or counterparts, bonds which are totaly absent from the open source world.

Things that have absolutely no relation to software what so ever.

Whether all current OEMs are doing a good job now, or not, is beside the point. You say the model is broken, and we need to turn to a wikiware model. I say the model isn't broken, just because some low-margin OEMs don't do a good job.

The model isn't broken for everyone, just people that care about quality.

Promote the next guy, but I'd guess the lead is a lot less likely to wake up one day and think "screw it" if it's how he makes a living.

I've seen people get up on the desk and dance in the middle of a work day. Saying "screw it" is pretty normal at most jobs.

That's called having some skin in the game, which most open source developers do not.

Please expand.

So a few very large companies paying some people to work on FOSS is evidence of what, exactly? They're paying people to write software to a certain design, and against certain standards and milestones.

What standards are being broken by open source development?

If they open the source up, great. They are risking nothing in the scheme of things. Nothing being done by volunteers is critical to a product launch timeline.

We're talking about paid workers, not the normal "volunteer." We're talking about employees now. 😉

Nothing is being done by volunteers that represents proprietary interests of the company. They're big enough to do this for the hell of it, for the PR value, or to try to chip away at Microsoft (the motivation for 90% of investment in FOSS by the companies you mention, with the exception of Dell).

What do volunteers have to do with employees at IBM, SGI, US Federal Government, Sourcefire, Nokia, Sun, and a dozen other companies creating FOSS?

Besides, since when was "open source" about a few huge companies releasing some of their low-value, low-risk work into an open source environment?

Solaris has become low-value and low-risk now? 😕

I must have missed the last ten years of Stallman-esque ranting against the evils of proprietary software business models. Whatever the failings of the current model, which depends on the profit-motive to create value, the open source world has them worse. The primary value proposition of 95% of the open source software out there is that it is free and college students don't have to dip into the meal fund to acquire it.

The whole thing is just silly to me. It was silly in 1995, and it's still silly. Not the idea of people getting together to write cool software. I didn't think that was silly in 1988 when I was doing it with friends over Compuserve, and I don't think it's silly now. But the idea that the mere existence of proprietary (Microsoft) software is evil, and we need a socio-technical movement to replace it (Microsoft) with happy little collectives producing organic code without the use of smelly dollars, that's the part that's silly. Microsoft has been managing and evolving its word processor for twenty years. I'll be interested to see where Open Office is twenty years from now. I have every reason to think MS will still have a word processor then, and it will be able to open and read the documents I am creating now.

Closed source software wouldn't be bad, if it wasn't bad. There are too many flaws out there being exploited, and the users don't have a chance. At least with open source software people can audit it, they can review it, they can fix it on their own without having to rely on a corporation that only cares about the bottom line.

Who is responsible when there is a flaw in a Microsoft program?

Hardware makers don't give us the time of day. Go ask them how to write a driver for their hardware, many won't tell you. I can go buy a manual on my car engine, but not on a card I put inside my computer? That's ridiculous.

I'd like to know how many of the features the majority of people use in Word. 😉
 
Microsoft and Apple have accepted certain models of software development, largely ignoring security along the way. They give away plastic discs and sell expensive pieces of paper allowing people to use certain binaries. The whole thing from the top down is closed to the consumer. You cannot know what is going on, do not look behind the curtain, accept these conditions and the fact that you have no control over the future of this software. If there is a bug no one is responsible for fixing it, and no one will fix it unless it makes the original company look bad.

These are the kinds of companies who make people think that everything is intellectual property, no matter how ridiculous. Hardware companies produce hardware, and the software that interacts between said hardware and the underlying software. These hardware companies produce a real product (hardware) and a virtual product (driver). Many have decided it is all intellectual property that should be defended with a rabid pack of lawyers. The consumer cannot know how the hardware works, and cannot know how the driver works. It's one thing when you buy a virtual thing to be totally oblivious and kept in the dark about how it works, it's another when you buy something physical.

The hardware companies have decided that the interface between the software and the hardware, something that should give away no secrets, is worth keeping close to their chests. In doing so they tell their customers that they can only use their hardware how they want you to use it with the software and other hardware that they approve. If someone wants to accept this by using Microsoft Windows, Apple Mac OS X, IBM AIX, etc. that is their choice and it's a choice they should have.

But, just because the majority of people choose this direction does not mean that these hardware companies can ignore the rest of their customers. Remember, we're talking about people using the hardware they have purchased with the hardware and software they want to use it with. Companies have a number of choices at this point:
1. Release closed source/partially closed drivers for a choice group of OSes.
2. Release open source drivers for a choice group of OSes.
3. Ignore the OSes.
4. Release documentation to the world, letting people know how to interface with the hardware they have purchased and let the communities create the driver.

All companies (including nVidia and ATI) are going to do #1, even if those drivers are Apple or Microsoft specific. A number of companies (mostly small companies, like Intel) have done #2. Both of these take a lot of work, and probably a decent amount of money to write the drivers and keep them up to date. These options draw a little ire from the community because source code can be difficult to debug when dealing with mystery hardware.

A lot of companies ignore everything except Windows. They just refuse to admit it exists. They gain no customers from the groups they ignore, and probably lose sales due to word of mouth.

A few companies have gone with #4. They release the documentation on how to interface with their hardware to the world, and they gain sales and support as the various communities do the work for them. The drivers are written and supported. No one gets confused and tries to use a Linux driver on their Windows machine and the linux people will go to the linux community for support, so their tech support personell won't be busy trying to fix anything relating to the linux driver, or the BSD driver, or the Solaris driver. They will only be supporting the software released by the company. A lot of benefit (increased sales, good word of mouth) for almost no work (releasing documentation). Hell, some companies (mostly non-US) see so much benefit from this they ship hardware along with the documentation.

There is no downside to #4, there is almost no work involved (excluding the bureaucracy that can only hold the company back). There is no need for chain of custody, business partners, ceritification, verification, exorcism, or anything else. The community will take care of itself, if given the freedom of knowing how the products they have purchased works.
 
So a few very large companies paying some people to work on FOSS is evidence of what, exactly?

It's evidence that it's a viable business model.

They're paying people to write software to a certain design, and against certain standards and milestones.

Milestones maybe, since a lot of FOSS projects don't care to release on some random company's schedule. But what standards are they breaking?

They are risking nothing in the scheme of things.

They risk losing all of the money that they paid those developers if the project doesn't pan out. And they risk losing face if they make a big deal of the project and then it flops. But neither of those things are tied to FOSS development.

Nothing being done by volunteers is critical to a product launch timeline.

So I guess TiVo, CyberGuard (now Secure Computing), Barracuda, etc don't have any timelines for their products?

Besides, since when was "open source" about a few huge companies releasing some of their low-value, low-risk work into an open source environment?

It's not, this whole thread is about companies releasing specs, like they did in the old days, so that drivers can be written even if the company doesn't want to support FOSS. The point is that releasing specs won't hurt them at all, so why not do it?
 
Ok, guys. New work week, I have a bunch of developers downstairs who want to rip the heads off the operations guys, so this is it for me. You'll just have to accept that you're wrong. 🙂. Some quick responses to some of the key points.

For your sake, I hope Microsoft succeeds so you can still spend $400 to read your documents.

I'm a business user. We get Office on a volume license for a _lot_ less than that, and it is an insignificant cost of onboarding a new employee. I appreciate your good wishes, but I believe there is a lot less risk of MS going away and leaving me stranded than there is of any open source alternative withering on the vine. Looking around, I guess about 99.98% of the business community agrees with me. Watch how the Massachusetts thing goes.

This is why we have things like the 'OpenGroup'. And 'Open' this and 'Open' that. It's why 'propriatory' is a dirty word in computer-land were in other industries its used as a 'good thing' to set their product apart from their competitors.

You have a very good point with regard to people getting stranded by proprietary formats. Some people bought betamax too. But the software business gives lipservice to open source for the reasons I've already mentioned. The vast majority of all software technology is still proprietary, and sold for profit. Also, there is nothing like the churn in the business that there used to be, when you could build your network on Netware and five years later it had almost disappeared. Microsoft now has a 20+ year track record of continuous support of their O/S and applications. Aside from lipservice paid for other reasons, nobody in business feels seriously fearful of being stranded on Word.

wonder how Sun, IBM, Microsoft, Novell, and RedHat, Checkpoint, Nokia, the US federal government, and just about any company that has ever used a TCP/IP stack feel about statements like that.

Haha, please, give me a break. You been in this business long? The TCP/IP stack was an open source _idea_ maybe (because it came from a time before there was much commercial software). There are as many different closed, proprietary implementations of that idea as there are companies pursuing one of... oh... a zillion different closed, proprietary versions of the Unix and other operating systems. I love when open source-heads point to Unix and its substrata and talk about "open systems." Bullshit.

Things that have absolutely no relation to software what so ever.

Those things have very strong relations to business planning, business risks, delivering value, and generating returns for shareholders. But then you guys obviously aren't very interested in businesss.

The model isn't broken for everyone, just people that care about quality.

You know, this statement sort of pisses me off. Here's my advice to the open source "community": stop mocking established, successful companies for being less than perfect. It's cheap as hell, because you have never had to deliver against the expectations of millions of users, almost nobody is using your crap, and you have nothing invested in whether it succeeds or not other than your own egos. You sound like a bunch of thirteen year-olds who are certain their parents are idiots, until someday they grow up and learn what the real world is like. When your software runs critical apps for a couple hundred million people you can come back and lecture us all on how it was done.

I've seen people get up on the desk and dance in the middle of a work day. Saying "screw it" is pretty normal at most jobs.

Now you're just being silly again.

Solaris has become low-value and low-risk now?

Haha, you said it, I didn't, but yes. In fact it is getting lower in value all the time. And WTF does Solaris have to do with anything? The vast majority of that platform is closed, proprietary property of Sun (much like Java). My point in that section was simply this: a few big companies paying a very small percentage of their workforce to spend time on FOSS because it is politically attractive to them doesn't mean ******. It doesn't change anything.

Closed source software wouldn't be bad, if it wasn't bad. There are too many flaws out there being exploited, and the users don't have a chance. At least with open source software people can audit it, they can review it, they can fix it on their own without having to rely on a corporation that only cares about the bottom line.

Dude, you can crow to me about how well the open source world does at preventing exploits when you have: a) hundreds of millions of consumers using it; and b) every scumbag on the planet writing malware to target it. You guys aren't even in the race and you're laughing at the lead runner for stumbling from time to time. Obnoxious. Yeah, you might have more people looking at open source stuff, but guess what? It's a crapload easier to find exploitable holes when I can download the source code, too. If you could snap your fingers and convert every desktop in the world to Linux tomorrow, within three months it would be a flaming wreck.

Microsoft and Apple have accepted certain models of software development, largely ignoring security along the way. They give away plastic discs and sell expensive pieces of paper allowing people to use certain binaries. The whole thing from the top down is closed to the consumer. You cannot know what is going on, do not look behind the curtain, accept these conditions and the fact that you have no control over the future of this software. If there is a bug no one is responsible for fixing it, and no one will fix it unless it makes the original company look bad.

Bullshit. First, they didn't "accept" the model. Might as well blame Dan Bricklin, or anyone else who sold their work because, you know, that's what we do here. Second, they haven't "ignored security along the way." They've been evolving their products successfully for twenty+ years. The security landscape is much different now than it was then. The consumer doesn't care what "is going on" in their computers or their source code. They just want it to work. These companies evolve their products as quickly as they can given the demands on them.

These are the kinds of companies who make people think that everything is intellectual property, no matter how ridiculous.

Give me an example of something ridiculous that is considered intellectual property, and I'll describe the investment that had to be made to create it.

There is no downside to #4, there is almost no work involved (excluding the bureaucracy that can only hold the company back). There is no need for chain of custody, business partners, ceritification, verification, exorcism, or anything else. The community will take care of itself, if given the freedom of knowing how the products they have purchased works.

There is a tremendous amount of work involved, and a lot of responsibility. I've described it several times but you just don't want to hear it.

It's evidence that it's a viable business model.

It's not a business model at all. They're all losing money on their investment in open source, but can afford to do so because it is such a small piece of what they do, and it has the political benefits that I won't describe for the 14th time in this thread.

They risk losing all of the money that they paid those developers if the project doesn't pan out. And they risk losing face if they make a big deal of the project and then it flops. But neither of those things are tied to FOSS development.

Oh yeah, you're right. There is that. It's probably nearly as much as they spent on napkins for the cafeteria last year.

So I guess TiVo, CyberGuard (now Secure Computing), Barracuda, etc don't have any timelines for their products?

I didn't say there were "no volunteers working on products with timelines", did I? Is that the same thing as saying "nothing being done by volunteers is critical to a product timeline?" Try to read the words one at a time. If you have a concrete example of a company outsourcing a key piece of a project to a volunteer group, by all means offer it up.

There you go. I'm done. Have fun with the thread.
 
Originally posted by: Markbnj
Ok, guys. New work week, I have a bunch of developers downstairs who want to rip the heads off the operations guys, so this is it for me. You'll just have to accept that you're wrong. 🙂. Some quick responses to some of the key points.

No one is wrong, most of this stuff is opinion. 🙂

Haha, please, give me a break. You been in this business long? The TCP/IP stack was an open source _idea_ maybe (because it came from a time before there was much commercial software). There are as many different closed, proprietary implementations of that idea as there are companies pursuing one of... oh... a zillion different closed, proprietary versions of the Unix and other operating systems. I love when open source-heads point to Unix and its substrata and talk about "open systems." Bullshit.

TCP/IP is an open standard. It was developed by a group of FOSS developers. The original implimentation was a FOSS implimentation, and many of the implimentations out there sprang from this original source. It was more than just an open source idea.

Those things have very strong relations to business planning, business risks, delivering value, and generating returns for shareholders. But then you guys obviously aren't very interested in businesss.

That's business, maybe even software business, but it isn't software. 😉

You know, this statement sort of pisses me off. Here's my advice to the open source "community": stop mocking established, successful companies for being less than perfect. It's cheap as hell, because you have never had to deliver against the expectations of millions of users, almost nobody is using your crap, and you have nothing invested in whether it succeeds or not other than your own egos. You sound like a bunch of thirteen year-olds who are certain their parents are idiots, until someday they grow up and learn what the real world is like. When your software runs critical apps for a couple hundred million people you can come back and lecture us all on how it was done.

It'a already happening. DE-IX uses FOSS for a lot of their stuff. I'd bet that most of the root name servers out there are running FOSS be it BIND, Linux, BSD, or what have you.

Haha, you said it, I didn't, but yes. In fact it is getting lower in value all the time. And WTF does Solaris have to do with anything? The vast majority of that platform is closed, proprietary property of Sun (much like Java). My point in that section was simply this: a few big companies paying a very small percentage of their workforce to spend time on FOSS because it is politically attractive to them doesn't mean ******. It doesn't change anything.

Supposedly Sun is working on opening up Java too. 😉

The Open Solaris Project has opened up quite a bit of it at this point, with more to come.

Dude, you can crow to me about how well the open source world does at preventing exploits when you have: a) hundreds of millions of consumers using it; and b) every scumbag on the planet writing malware to target it. You guys aren't even in the race and you're laughing at the lead runner for stumbling from time to time. Obnoxious. Yeah, you might have more people looking at open source stuff, but guess what? It's a crapload easier to find exploitable holes when I can download the source code, too. If you could snap your fingers and convert every desktop in the world to Linux tomorrow, within three months it would be a flaming wreck.

There are hundreds of millions of users using FOSS every day, and plenty of people trying to break it. And yes, it gets broken, but there is usually a fix out there pretty quickly and chances are it could have been prevented in the first place with a little extra work.

Bullshit. First, they didn't "accept" the model. Might as well blame Dan Bricklin, or anyone else who sold their work because, you know, that's what we do here. Second, they haven't "ignored security along the way." They've been evolving their products successfully for twenty+ years. The security landscape is much different now than it was then. The consumer doesn't care what "is going on" in their computers or their source code. They just want it to work. These companies evolve their products as quickly as they can given the demands on them.

When did Microsoft start implimenting real memory protection? Non-executable heaps and stacks? When did they really start to push the "you don't have to be an administrator to use our OS easily" idea? When did Microsoft add a firewall to its OS? Which Windows version starts up secure by default?

Give me an example of something ridiculous that is considered intellectual property, and I'll describe the investment that had to be made to create it.

Interfaces to the hardware so that people can write drivers. People who will recommend the products to the people that will service those hundreds of thousands of users.

There is a tremendous amount of work involved, and a lot of responsibility. I've described it several times but you just don't want to hear it.

There is no work involved beyond a small amount of HTML editting, unless the company does not have anyone that can make a decision to support their customers.

What responsibility? They release documentation, and that is where the responsibility ends. They aren't responsible for the eventual drivers that are written, just for making the hardware they are in business to make.

I didn't say there were "no volunteers working on products with timelines", did I? Is that the same thing as saying "nothing being done by volunteers is critical to a product timeline?" Try to read the words one at a time. If you have a concrete example of a company outsourcing a key piece of a project to a volunteer group, by all means offer it up.

OpenSSH is critical to the well being of just about every network out there.

There you go. I'm done. Have fun with the thread.

I am. Good luck with the issues at work, it's been fun. 🙂
 
The vast majority of that platform is closed, proprietary property of Sun (much like Java).

It was my understanding that the majority of the system was opened under the CDL and the bits that they couldn't, like drivers, were being worked on.

I didn't say there were "no volunteers working on products with timelines", did I? Is that the same thing as saying "nothing being done by volunteers is critical to a product timeline?" Try to read the words one at a time.

Read it again yourself. "Nothing being done by volunteers" rules out 99% of FOSS software and is definitely critical to the products those companies sell.

If you have a concrete example of a company outsourcing a key piece of a project to a volunteer group, by all means offer it up.

RedHat? Novell? Yes, they both employ some of the developers on the larger products that they use but neither of them employ all of the developers for a project AFAIK and a lot of the pieces of their products are completely maintained by developers outside of their respective companies.
 
Originally posted by: n0cmonkey
Hardware makers don't give us the time of day. Go ask them how to write a driver for their hardware, many won't tell you. I can go buy a manual on my car engine, but not on a card I put inside my computer? That's ridiculous.

Most true thing said in this thread. 🙂
 
Originally posted by: Markbnj
For your sake, I hope Microsoft succeeds so you can still spend $400 to read your documents.

I'm a business user. We get Office on a volume license for a _lot_ less than that, and it is an insignificant cost of onboarding a new employee. I appreciate your good wishes, but I believe there is a lot less risk of MS going away and leaving me stranded than there is of any open source alternative withering on the vine. Looking around, I guess about 99.98% of the business community agrees with me. Watch how the Massachusetts thing goes.

I'm curious, if it wasn't for being able to read the proprietary formats in your business what advantages does Microsoft Office have over OpenOffice justifing its cost?

Originally posted by: Markbnj
The model isn't broken for everyone, just people that care about quality.

You know, this statement sort of pisses me off. Here's my advice to the open source "community": stop mocking established, successful companies for being less than perfect. It's cheap as hell, because you have never had to deliver against the expectations of millions of users, almost nobody is using your crap, and you have nothing invested in whether it succeeds or not other than your own egos. You sound like a bunch of thirteen year-olds who are certain their parents are idiots, until someday they grow up and learn what the real world is like. When your software runs critical apps for a couple hundred million people you can come back and lecture us all on how it was done.

Yeah I know. Nobody takes linux seriously... No one.. Maybe if we keep saying this it'll come true?.. Ok I'll stop it. Tired of governments interested in linux? What about AOL, Amazon.com, Meriyll Lynch, Dreamworks, Goodyear, Blue Cross Blue Shield, Michelin, Nisan, Southwest Airlines, The Washington Post, France Telecom, or thousands more who choose linux and open-source. Wow look at all those people who run Apache as their web server. What a mistake!
 
I better tell our DBA's to get ready for a swift platform change since we're running our transaction systems and databases on an amateur OS that no one takes seriously :Q
 
Yeah I know. Nobody takes linux seriously... No one.. Maybe if we keep saying this it'll come true?.. Ok I'll stop it. Tired of governments interested in linux? What about AOL, Amazon.com, Meriyll Lynch, Dreamworks, Goodyear, Blue Cross Blue Shield, Michelin, Nisan, Southwest Airlines, The Washington Post, France Telecom, or thousands more who choose linux and open-source. Wow look at all those people who run Apache as their web server. What a mistake!

That's an excellent attempt to post a bunch of crap that has nothing to do with the point I made, but then that is the normal experience of debating anything with Linux-heads.

I don't care how many people are "interested" in Linux, or how many web servers run Apache, since we're talking about Microsoft's performance, and the performance of MS OEM partners, in the broader computing space, which contains more individual clients by several orders of magnitude than there are Linux desktops or web servers in the world.

To repeat, when Linux is in use by a few hundred million people and is managed and evolved to satisfy their expectations over 20+ years, then you guys will have earned the right to lecture everyone else about how to produce quality software. Until then you're just yapping.
 
Originally posted by: Markbnj
That's an excellent attempt to post a bunch of crap that has nothing to do with the point I made, but then that is the normal experience of debating anything with Linux-heads.

I don't care how many people are "interested" in Linux, or how many web servers run Apache, since we're talking about Microsoft's performance, and the performance of MS OEM partners, in the broader computing space, which contains more individual clients by several orders of magnitude than there are Linux desktops or web servers in the world.

To repeat, when Linux is in use by a few hundred million people and is managed and evolved to satisfy their expectations over 20+ years, then you guys will have earned the right to lecture everyone else about how to produce quality software. Until then you're just yapping.

So it revolves around Linux not being old enough (since it has been used by hundreds of millions of people for years now)? 😕
 
Back
Top