Greetings,

I recently replied to a parent post on Slashdot and engaged in a back-and-forth that I would like to here your opinions on the subject. It seems that everytime anything software related on Slashdot is mentioned somehow 80% of the posts end up touting (with no evidence to support thier claims) that open-source is the "superior" model and yada yada. The original link to the parent is below and all subsequent posts are thereafter.

Original Slashdot thread

Please read the replies (I'm Leabre)...

Let me know what you think, I'm I the one being dense and naive in there? What am I not seeing?


Thanks,
_Shawn
Posted on 2004-05-01 22:01:01 by _Shawn
Hi _Shawn,

First off, I think GPL is destroying the open-source movement, if it has not already done so. It restricts usage unecessarily and just brings alot of legal issues to wrap up what in some cases is fairly simplistic software. The open source movement will eventually die because of GPL and inherent flaws in the concept. First as you said some companies would not consider using open-source GPL software as any mods they do would be covered by the GPL as well, whether they wanted to retain it as their intellectual property or not. While an extendable application that allows plugins or their equivalent and handles them well does not have this limitation, what you write remains yours. Second with very few exceptions, all OSS packages tend to flounder and die when the original author looses interest, and that happens alot. SourceForge is packed full of projects that have not moved in years because the original author lost interest or moved on to other projects. With non-GPL/OSS apps there is always a chance that some company will see potential in the application and continue development, this cannot happen realistically with a GPL application. For myself at least I am anti-GPL and always have been, if I write software and intend to give it away I do so without restrictions of any kind, free is free. Lastly, I do not program to pay my bills, I do it for fun and if one of my apps is not up to snuff, well, I'm not going to go broke because of it. This has a phsycological effect on free software, at some point you start wondering if it is worth the trouble of dealing with users and consuming your free time if there is no profit in it, alot of users will get just as irate and insulting with you as they would for paid software (believe me, they all downloaded TBPaint). The tendancy is to just abandon the projects, and I have already pointed out where that leads. For example the recent spate of insults to my programming skills because I don't make TBPaint capable of editing full sized bitmaps, I keep saying it's a targetted application but that never seems to make a difference.

Case-in-point:

I saw a custom control in a post, source code was posted and there was no mention of GPL. I thought it would be nice to add to TBPaint so I emailed the author and asked if I could use it, with due acknowledgements. The source was about 20 lines long but was part of a GPL application. To make a long story short, about 10 or 15 emails changed hands and I would have had to make TBPaint GPL, something I refuse to do with any of my software. So I ended up having to write my own control but had to write one that did not use any of the programming concepts that were included in the GPL version I had seen. Having seen the source my choices were limited even though I had no intention of using it. I could have used peices anyway but it would have been dishonest so just by seeing the source I created much more work for myself. This is the main reason I will no longer download or examine GPL software that might have things of interest to me.
Posted on 2004-05-01 22:42:06 by donkey
As an example, I tried to contribute to GMP and was told the changes would not be incorporated because I didn't submit the code in AT&T syntax. Can you imagine a non-programmer trying to make a suggestion/bug-report!?

Another example, show me an OSS for MMX neural nets. There are none. ;) Same goes for many high-valued algorithms!

I have serious doubts about the quality of any OSS. If you pay for something you have the right to complain or get your money back!
Posted on 2004-05-01 23:09:02 by bitRAKE

If you pay for something you have the right to complain or get your money back!


Exactly,it gives the software company an incentive (the largest incentive in the corporate world) to meet or exceed your expectations and needs, this does not exist in the OSS world. All software there is offered on a take it or leave it basis with no penalty possible if it does not function as advertised. I have numerous examples of free software I have downloaded and discarded only to end up buying a commercial product that actually does the job I need.
Posted on 2004-05-01 23:43:53 by donkey
I have never seen any OSS that is actually superior to a commercial solution, which makes me believe that the OSS model is not capable of producing such. And the reason for this is lack of management, vision and a common set of goals for the projects, I think.
The way I see it, most OSS is a clone of commercial software at the level that it was 5-10 years ago.
The only reason why OSS gets away with this, is that 5-10 years ago, the commercial software was already at a decent level of sophistication.
This is fine, if you want to use it, but where the OSS community goes wrong is where they want to force OSS on everyone, make arrogant and false claims that OSS is superior software, and project all evil on MS (somehow other commercial software companies are never attacked, even though they are very similar to MS, perhaps just a bit smaller).

Another thing where they go wrong is claiming that the sourcecode is everything. It is the design, the manual, etc.
This is nonsense. A proper description of the design of software is often more valuable than the sourcecode itself, if you want to understand how the program works, and even alter it.

Their claims that MS should release their sourcecode is therefore also nonsense. No user would benefit from it, and programmers would not benefit much from it either, since Windows is very well-documented already, in MSDN, books like Inside Windows, etc.
The only thing it would do, is destroy MS as a commercial company. Then again, that may be exactly what they want. I cannot understand why, though.
Posted on 2004-05-02 03:27:40 by Scali
I have never seen any OSS that is actually superior to a commercial solution, which makes me believe that the OSS model is not capable of producing such. And the reason for this is lack of management, vision and a common set of goals for the projects, I think.
The way I see it, most OSS is a clone of commercial software at the level that it was 5-10 years ago.


There are exceptions to every rule. Mozilla Firefox is near the cutting edge in browser technology. It is one of the very few exceptions that I mentioned. IE has a long way to go to catch up with it.
Posted on 2004-05-02 04:54:44 by donkey
Thing is though, IE hasn't been updated in years. IE6 is from 2001, I believe. So while it's not 5-10 years old, 3 years is still an eternity in the world of IT. And browsers aren't exactly rocket science. We'll see if Firefox is still ahead when IE 7 comes along.
Besides, I prefer MyIE2 over Firefox, because it adds all the features of Firefox, without giving up the IE speed and compatibility. And it's free aswell, and no opensource nonsense.

But what I am talking about is REAL applications. Office, PhotoShop, AutoCAD, ProEngineer, 3DStudio MAX, etc, etc. Where are the superior opensource replacements for those? That's the stuff that people work with everyday. That's what they make money with. Not with their browser or email client.
Posted on 2004-05-02 04:59:47 by Scali
Well,

Outside of custom written business software, almost all of my work is done with a browser and email client so I geuss I'm biased. And ofcourse the office runs Linux now for the most part so the common interface is helpful. You would be surprised how many people's browser has become one of the central applications to doing business.
Posted on 2004-05-02 05:08:34 by donkey
Would you mind explaining me how you can make money with a browser or an email client? You don't actually produce anything.
A browser can only be used to access information, but you would still need to use that information somewhere in order for it to be valuable.
As for email... The only thing I can think of would be a helpdesk or such, where you communicate with your customers, who are charged for this (then again, an email client can actually produce something: emails).

So I think that you need an application that can produce something, in order to make money and run a business. And in general, such applications are commercial. At least, the best ones in their class are.
Posted on 2004-05-02 05:16:59 by Scali
Would you mind explaining me how you can make money with a browser or an email client? You don't actually produce anything.


I appraise and dispose of the assets from bankrupt mining companies. I do 90% of my research on the internet and about 50% of the brokerage via email. All information from the sites flows in from the internet and all information to the receivers flows out through the internet. Information is worth alot more than anything else in the business world and there is no tool better suited to it than the Internet. Email is a communication device that is inexpensive, and like all communication devices it is only lack of imagination that limits it's profitability.
Posted on 2004-05-02 06:15:19 by donkey
Ah I see. However, I will then argue that it is the information that makes the money, not the browser. Namely, the browser is just one way to access the information. With any other browser, you will access the exact same information. And perhaps there are even other methods of accessing the information (TV, ceefax, telephone...)?

So this is different from the kind of software that I was referring to. For example, with PhotoShop or AutoCAD you actually 'create information'. And an opensource alternative will not be able to do this in the exact same way, because of lack of functions or quality or such.
(Perhaps the information that you access through your browser is also created via commercial software?).
Posted on 2004-05-02 06:28:14 by Scali
Namely, the browser is just one way to access the information. With any other browser, you will access the exact same information. And perhaps there are even other methods of accessing the information (TV, ceefax, telephone...)?


Any browser would do but then I could also take a tricycle the 4 miles to work but a car gets me there faster. The same case could be made for your AutoCAD software or any other thing that you like, you could draw the design by hand. All software no matter what it does is an information tool, it can bring the information to you, send it to others or control machines but it is not software, OSS or commercial, that produces anything or does any work, it only processes information.
Posted on 2004-05-02 06:38:03 by donkey
Any browser would do but then I could also take a tricycle the 4 miles to work but a car gets me there faster. The same case could be made for your AutoCAD software or any other thing that you like, you could draw the design by hand.


That's where you go wrong. Time is money, and while AutoCAD may be much faster than drawing by hand, I don't think that using browser A instead of browser B will make any difference. The speed depends pretty much completely on that of the internet connection, and the hardware.

it is not software, OSS or commercial, that produces anything or does any work, it only processes information.


I disagree. If I save my work in AutoCAD, I have produced a new file containing information. There is no save-option in a browser, and webpages are read-only in general, if you get my drift.
Also, AutoCAD can actually do work aswell. You can use all kinds of options and macros to automate drawing and calculations, so you don't have to do them by hand.
Ultimately all that a computer does is processing information. However, you seem to forget that processing information actually IS work, most of the time, and that you can also produce information.

Conclusion: browsers are an exception, and should therefore not be compared with other software, because different rules apply.
An analogy: Browsers are like televisions: tune in and you get a stream of information. The television is but a means to access this information. The actual production of the information is done elsewhere, with other tools, and by other people. Any reasonable television-set will work equally well in accessing the information, but will any person and any studio be able to produce the exact same information, in the exact same time and cost?
Posted on 2004-05-02 06:47:16 by Scali
That's where you go wrong. Time is money, and while AutoCAD may be much faster than drawing by hand, I don't think that using browser A instead of browser B will make any difference. The speed depends pretty much completely on that of the internet connection, and the hardware.


No, it doesn't. Tabbed browsing allows me to open multiple websites easily and crosscheck information and perform multiple simultaneous searches. The bookmarks toolbar allows me to have 100's of bookmarks in easy access. Multiple search engines on the toolbar allow me to easily switch from one engine to another without having to search around for URLs. This is exactly the same as AutoCad, the better tool will always win, the variation between ability may be different but that does not matter, only that it is the best tool for the job. You do not understand my job or it's requirements yet you seem bent on explaining it to me in terms you understand. I don't look up something on the internet then enter it in a log, I have databases of millions of items and typically have to research 200 to 300 items a day, anything that speeds that up makes me money.

I disagree. If I save my work in AutoCAD, I have produced a new file containing information. There is no save-option in a browser, and webpages are read-only in general, if you get my drift.


That is completely wrong, I can also save and edit a web page if I want. But whats the point the information I use is static, in AutoCad it's dynamic, that is the application of information as related to the task and has nothing at all to do with the topic at hand. Also when I open a webpage, information is saved to my browser cache, does that mean I have produced something ? no it doesn't it just means I have saved some information. When I put the information to work then I have produced something.
Posted on 2004-05-02 07:05:44 by donkey
Afternoon, donkey.

Most people do not use an internet browser as their main application of use.

You do, however you are the exception ;) .

Office grunts use MS Word, MS Excel, MS Project, PSP, Photoshop, etc to do their work.
There are few (no?) OSS which come close to the typical application used by most of the workforce.

(Note: I'll switch this discusion to the Crusades. Everyone remember to abuse products and opinions but not members, m'kay?)

Cheers,
Scronty
Posted on 2004-05-02 08:44:17 by Scronty
No, it doesn't. Tabbed browsing allows me to open multiple websites easily and crosscheck information and perform multiple simultaneous searches. The bookmarks toolbar allows me to have 100's of bookmarks in easy access. Multiple search engines on the toolbar allow me to easily switch from one engine to another without having to search around for URLs.This is exactly the same as AutoCad


I disagree. A browser is still an information consumer, while AutoCAD is an information producer.
And any browser that supports the features you named, such as tabbed browsing, would allow you to access the information about equally fast and easy. And there actually are many such browsers, I suppose. MyIE2, Avant, Mozilla, Opera, etc.
Which one you pick, doesn't matter. Vanilla IE might be the one exception here, since it hasn't been updated in years.
AutoCAD is different.
An example... I can browse this forum with any browser that I like, and I will still be able to see the same threads, and the same people posting. It's the same information. Why? It's not created by the browser, but it's created elsewhere, independent from your browser.

There are no opensource tools which can replace AutoCAD. Same goes for PhotoShop, 3D Studio, and many others. Why not? They are not trivial applications.

You do not understand my job or it's requirements yet you seem bent on explaining it to me in terms you understand.


I merely said it was an exception. It doesn't apply in the general case.

That is completely wrong, I can also save and edit a web page if I want.


A browser can save a COPY of the webpage, this is where you got confused, and a browser can not edit a webpage, although there have been browsers with a built-in editor, such as Netscape.

But whats the point the information I use is static, in AutoCad it's dynamic


Now you're starting to make sense.
Posted on 2004-05-02 08:49:21 by Scali
The GNU world has The GIMP... but replacing Photoshop with GIMP for serious use? Naw. Even Torvald's have to use MS PowerPoint to do presentations, since there's no OSS alternative.

I'm not really against the concept of Open-Source as such, btw - but I'm radically anti-GPL, it's a viral disease and it has caused a lot of brainrot.
Posted on 2004-05-02 09:01:47 by f0dder
What about VirtualDub?
Posted on 2004-05-02 13:43:07 by death
For the 3DMax OSS alternative, many will be quick to point out "Blender". Perhaps its easy to forget that Blender started out as a commercial product and for whatever reason, closed shop. Then, the OSS community gathered up $250,000 and purchased it and opened it up. So... there you go, it wasn't the OSS community that produced the application but, they are actively maintaining it.

The same is the case with Mozilla/Firefox. The OSS community didn't produce it, Netscape did, and they, too, went under, and then "donated" it to the OSS community. However, the OSS people do maintain it. But they are hardly examples of OSS "superiority". In my book, they started out superior because they were priority and had financial incentive to be good. OSS'ing them was an afterthought.

Perhaps Apache and PHP and the Kernel are good examples of projects that the OSS birthed and nurtured and the turned out pretty good. I don't mention KDE or GNOME, not because they aren't good, but because they are 10 years behind he commerical counterparts in many respects, such as, plug-and-play and hardware comatibility (which can make or break its acceptance).

Most of the OSS projects that have any commercial support whatsoever is primarily because a large company is driving it and marketing it or embracing it themselves. But not because it is "superior". Businesses don't care about "superior". They care about having good support in terms of what it is supposed to do, and having someone they can call up and ask for help when they need it, and not having to deal with teenagers who want to criticise them for being "ignorant" because they didn't know how to recompile their kernel for the latest video card. So if a big company steps in and supports an OSS piece and charges for it, then the business world is likely to "accept" it.

Concerning making money with the browser. I, too, make money that way. I write Web Applications. We targer IE 6 only. I work for an insurance software startup and we are really shaking up the industry mammoths. The software I write (I'm the senior developer of the company) delivers its information through the browswer, we use C# and ASP.NET. But it can drive the insurance companies through a complete life-cycle. This is a huge undertaking (40+ developers 3+ years). But, that's how we make our money, we charge for "seats". We have thousands of them, monthly.

Anyway... in the spirit of the thread, I don't see OSS going anywhere soon as they tend to follow rather than lead. I also don't like the GPL because I feel it is, as Microsoft says, "viral". It contaminates everything it touches and takes your freedom away, it is not Free, as they say. LGPL is better, but still in the same vein. A business that is writing a custom solution isn't going to GPL even if it is only distributed internally. They may not want their source being open to anyone lest one should run off with it and spread it to the world and then what can the company do? It's GPL. They can't do anything. Businesses want to control their own IP (even it they are doing exactly what everyone else is doing and their software isn't special, they still like to "think" they are doing something different and "own" it). GPL isn't an option in this arena.


Thanks,
_Shawn
Posted on 2004-05-02 15:12:05 by _Shawn
You know, I can't help but think about the Genesis project. It is a 3D game engine, OSS. A good one, and it chugs along only slowly. But, it split off into Genesis3D, CrystalSpace, Destiny3D, and RealityFactory (those are just the ones I know about).

The problem with OSS in this case, is is the case with the Kernel, is that requiring your changes to be made available to the public doesn't help the "master" project for two reasons: 1) The owner may not accept your changes (Linus is strict about what he approves, for example) and that in turn, causes 2) It forks off and now you have two projects that are chugging along slowly. One of them might split off again and cause further fragmentation.

In the spirit of tyring compete against MS and all the other commercial products, they also end up fragmenting and competing against themselves. In that respect, they don't see to want to work together well. At least, at a company making profits, you can keep the team focused and in one vision. In OSS, it is too hard and sooner or later, a senior teammember will split off and create their own distro and then you have many projects that aren't getting anyway.

I've been following Genises3D, Destiny3D, and CrystalSpace for at least 4 years now and they have each hardly made any progress. Destiny3D looked promising until the lead developer started school and then 90% of the progress stopped. They the other team members started complaining. Then the lead chimed in and said "when I'm not around nothing gets done"... even though people have to "contribute" $30 to the project to have access to it, there is financial incentive, but not enough to press forward at even half the rate a commercial counterpart would.

So I think that's a good example of how the GPL is detrimental. Of course, there are many more examples. I am not sure how the Kernal and PHP (PHP has commercial interest behind it: Zen. That my be why; and the kernel is too big for it to fork off easily) managed to stay "pure" but, it's amazing. MySQL also has commercial interest behind it, which may account for why there aren't many forks of it, they charge for their licences, GPL or not. PostGRE SQL is probly too complex to fork off easily, also. Esclipse and JBoss and a few others are also probly too difficult for fork, so those are some rare exceptions, but there are countless more examples of how OSS projects fragment (look at the many GUI's out there)...


Thanks,
_Shawn
Posted on 2004-05-02 15:25:36 by _Shawn